* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc: (23 commits)
[POWERPC] Remove leftover printk in isa-bridge.c
[POWERPC] Remove duplicate #include
[POWERPC] Initialize lockdep earlier
[POWERPC] Document when printk is useable
[POWERPC] Fix bogus paca->_current initialization
[POWERPC] Fix of_i2c include for module compilation
[POWERPC] Make default cputable entries reflect selected CPU family
[POWERPC] spufs: lockdep annotations for spufs_dir_close
[POWERPC] spufs: don't requeue victim contex in find_victim if it's not in spu_run
[POWERPC] 4xx: Fix PCI mem in sequoia DTS
[POWERPC] 4xx: Add endpoint support to 4xx PCIe driver
[POWERPC] 4xx: Fix problem with new TLB storage attibute fields on 440x6 core
[POWERPC] spufs: spu_create should send inotify IM_CREATE event
[POWERPC] spufs: handle faults while the context switch pending flag is set
[POWERPC] spufs: fix concurrent delivery of class 0 & 1 exceptions
[POWERPC] spufs: try to route SPU interrupts to local node
[POWERPC] spufs: set SPU_CONTEXT_SWITCH_PENDING before synchronising SPU irqs
[POWERPC] spufs: don't acquire state_mutex interruptible while performing callback
[POWERPC] spufs: update master runcntl with context lock held
[POWERPC] spufs: fix post-stopped update of MFC_CNTL register
...
kgdb is a source level debugger for linux kernel. It is used along
with gdb to debug a linux kernel. The expectation is that gdb can
be used to "break in" to the kernel to inspect memory, variables
- and look through a cal stack information similar to what an
+ and look through call stack information similar to what an
application developer would use gdb for. It is possible to place
breakpoints in kernel code and perform some limited execution
stepping.
<chapter id="CompilingAKernel">
<title>Compiling a kernel</title>
<para>
- To enable <symbol>CONFIG_KGDB</symbol>, look under the "Kernel debugging"
- and then select "KGDB: kernel debugging with remote gdb".
+ To enable <symbol>CONFIG_KGDB</symbol> you should first turn on
+ "Prompt for development and/or incomplete code/drivers"
+ (CONFIG_EXPERIMENTAL) in "General setup", then under the
+ "Kernel debugging" select "KGDB: kernel debugging with remote gdb".
</para>
<para>
Next you should choose one of more I/O drivers to interconnect debugging
void (*destroy_inode)(struct inode *);
void (*dirty_inode) (struct inode *);
int (*write_inode) (struct inode *, int);
- void (*put_inode) (struct inode *);
void (*drop_inode) (struct inode *);
void (*delete_inode) (struct inode *);
void (*put_super) (struct super_block *);
destroy_inode: no
dirty_inode: no (must not sleep)
write_inode: no
-put_inode: no
drop_inode: no !!!inode_lock!!!
delete_inode: no
put_super: yes yes no
void (*dirty_inode) (struct inode *);
int (*write_inode) (struct inode *, int);
- void (*put_inode) (struct inode *);
void (*drop_inode) (struct inode *);
void (*delete_inode) (struct inode *);
void (*put_super) (struct super_block *);
inode to disc. The second parameter indicates whether the write
should be synchronous or not, not all filesystems check this flag.
- put_inode: called when the VFS inode is removed from the inode
- cache.
-
drop_inode: called when the last access to the inode is dropped,
with the inode_lock spinlock held.
mac5380= [HW,SCSI] Format:
<can_queue>,<cmd_per_lun>,<sg_tablesize>,<hostid>,<use_tags>
- mac53c9x= [HW,SCSI] Format:
- <num_esps>,<disconnect>,<nosync>,<can_queue>,<cmd_per_lun>,<sg_tablesize>,<hostid>,<use_tags>
-
machvec= [IA64] Force the use of a particular machine-vector
(machvec) in a generic kernel.
Example: machvec=hpzx1_swiotlb
This is normally done in pci_enable_device(),
so this option is a temporary workaround
for broken drivers that don't call it.
+ skip_isa_align [X86] do not align io start addr, so can
+ handle more pci cards
firmware [ARM] Do not re-enumerate the bus but instead
just use the configuration from the
bootloader. This is currently used on
Enable logging of debug information in case of ccw device timeouts.
-
-* cio_msg = yes | no
-
- Determines whether information on found devices and sensed device
- characteristics should be shown during startup or when new devices are
- found, i. e. messages of the types "Detected device 0.0.4711 on subchannel
- 0.0.0042" and "SenseID: Device 0.0.4711 reports: ...".
-
- Default is off.
-
-
* cio_ignore = {all} |
{<device> | <range of devices>} |
{!<device> | !<range of devices>}
+++ /dev/null
- Goals, Design and Implementation of the
- new ultra-scalable O(1) scheduler
-
-
- This is an edited version of an email Ingo Molnar sent to
- lkml on 4 Jan 2002. It describes the goals, design, and
- implementation of Ingo's new ultra-scalable O(1) scheduler.
- Last Updated: 18 April 2002.
-
-
-Goal
-====
-
-The main goal of the new scheduler is to keep all the good things we know
-and love about the current Linux scheduler:
-
- - good interactive performance even during high load: if the user
- types or clicks then the system must react instantly and must execute
- the user tasks smoothly, even during considerable background load.
-
- - good scheduling/wakeup performance with 1-2 runnable processes.
-
- - fairness: no process should stay without any timeslice for any
- unreasonable amount of time. No process should get an unjustly high
- amount of CPU time.
-
- - priorities: less important tasks can be started with lower priority,
- more important tasks with higher priority.
-
- - SMP efficiency: no CPU should stay idle if there is work to do.
-
- - SMP affinity: processes which run on one CPU should stay affine to
- that CPU. Processes should not bounce between CPUs too frequently.
-
- - plus additional scheduler features: RT scheduling, CPU binding.
-
-and the goal is also to add a few new things:
-
- - fully O(1) scheduling. Are you tired of the recalculation loop
- blowing the L1 cache away every now and then? Do you think the goodness
- loop is taking a bit too long to finish if there are lots of runnable
- processes? This new scheduler takes no prisoners: wakeup(), schedule(),
- the timer interrupt are all O(1) algorithms. There is no recalculation
- loop. There is no goodness loop either.
-
- - 'perfect' SMP scalability. With the new scheduler there is no 'big'
- runqueue_lock anymore - it's all per-CPU runqueues and locks - two
- tasks on two separate CPUs can wake up, schedule and context-switch
- completely in parallel, without any interlocking. All
- scheduling-relevant data is structured for maximum scalability.
-
- - better SMP affinity. The old scheduler has a particular weakness that
- causes the random bouncing of tasks between CPUs if/when higher
- priority/interactive tasks, this was observed and reported by many
- people. The reason is that the timeslice recalculation loop first needs
- every currently running task to consume its timeslice. But when this
- happens on eg. an 8-way system, then this property starves an
- increasing number of CPUs from executing any process. Once the last
- task that has a timeslice left has finished using up that timeslice,
- the recalculation loop is triggered and other CPUs can start executing
- tasks again - after having idled around for a number of timer ticks.
- The more CPUs, the worse this effect.
-
- Furthermore, this same effect causes the bouncing effect as well:
- whenever there is such a 'timeslice squeeze' of the global runqueue,
- idle processors start executing tasks which are not affine to that CPU.
- (because the affine tasks have finished off their timeslices already.)
-
- The new scheduler solves this problem by distributing timeslices on a
- per-CPU basis, without having any global synchronization or
- recalculation.
-
- - batch scheduling. A significant proportion of computing-intensive tasks
- benefit from batch-scheduling, where timeslices are long and processes
- are roundrobin scheduled. The new scheduler does such batch-scheduling
- of the lowest priority tasks - so nice +19 jobs will get
- 'batch-scheduled' automatically. With this scheduler, nice +19 jobs are
- in essence SCHED_IDLE, from an interactiveness point of view.
-
- - handle extreme loads more smoothly, without breakdown and scheduling
- storms.
-
- - O(1) RT scheduling. For those RT folks who are paranoid about the
- O(nr_running) property of the goodness loop and the recalculation loop.
-
- - run fork()ed children before the parent. Andrea has pointed out the
- advantages of this a few months ago, but patches for this feature
- do not work with the old scheduler as well as they should,
- because idle processes often steal the new child before the fork()ing
- CPU gets to execute it.
-
-
-Design
-======
-
-The core of the new scheduler contains the following mechanisms:
-
- - *two* priority-ordered 'priority arrays' per CPU. There is an 'active'
- array and an 'expired' array. The active array contains all tasks that
- are affine to this CPU and have timeslices left. The expired array
- contains all tasks which have used up their timeslices - but this array
- is kept sorted as well. The active and expired array is not accessed
- directly, it's accessed through two pointers in the per-CPU runqueue
- structure. If all active tasks are used up then we 'switch' the two
- pointers and from now on the ready-to-go (former-) expired array is the
- active array - and the empty active array serves as the new collector
- for expired tasks.
-
- - there is a 64-bit bitmap cache for array indices. Finding the highest
- priority task is thus a matter of two x86 BSFL bit-search instructions.
-
-the split-array solution enables us to have an arbitrary number of active
-and expired tasks, and the recalculation of timeslices can be done
-immediately when the timeslice expires. Because the arrays are always
-access through the pointers in the runqueue, switching the two arrays can
-be done very quickly.
-
-this is a hybride priority-list approach coupled with roundrobin
-scheduling and the array-switch method of distributing timeslices.
-
- - there is a per-task 'load estimator'.
-
-one of the toughest things to get right is good interactive feel during
-heavy system load. While playing with various scheduler variants i found
-that the best interactive feel is achieved not by 'boosting' interactive
-tasks, but by 'punishing' tasks that want to use more CPU time than there
-is available. This method is also much easier to do in an O(1) fashion.
-
-to establish the actual 'load' the task contributes to the system, a
-complex-looking but pretty accurate method is used: there is a 4-entry
-'history' ringbuffer of the task's activities during the last 4 seconds.
-This ringbuffer is operated without much overhead. The entries tell the
-scheduler a pretty accurate load-history of the task: has it used up more
-CPU time or less during the past N seconds. [the size '4' and the interval
-of 4x 1 seconds was found by lots of experimentation - this part is
-flexible and can be changed in both directions.]
-
-the penalty a task gets for generating more load than the CPU can handle
-is a priority decrease - there is a maximum amount to this penalty
-relative to their static priority, so even fully CPU-bound tasks will
-observe each other's priorities, and will share the CPU accordingly.
-
-the SMP load-balancer can be extended/switched with additional parallel
-computing and cache hierarchy concepts: NUMA scheduling, multi-core CPUs
-can be supported easily by changing the load-balancer. Right now it's
-tuned for my SMP systems.
-
-i skipped the prev->mm == next->mm advantage - no workload i know of shows
-any sensitivity to this. It can be added back by sacrificing O(1)
-schedule() [the current and one-lower priority list can be searched for a
-that->mm == current->mm condition], but costs a fair number of cycles
-during a number of important workloads, so i wanted to avoid this as much
-as possible.
-
-- the SMP idle-task startup code was still racy and the new scheduler
-triggered this. So i streamlined the idle-setup code a bit. We do not call
-into schedule() before all processors have started up fully and all idle
-threads are in place.
-
-- the patch also cleans up a number of aspects of sched.c - moves code
-into other areas of the kernel where it's appropriate, and simplifies
-certain code paths and data constructs. As a result, the new scheduler's
-code is smaller than the old one.
-
- Ingo
S: Maintained
INTEL ETHERNET DRIVERS (e100/e1000/e1000e/igb/ixgb/ixgbe)
-P: Auke Kok
-M: auke-jan.h.kok@intel.com
-P: Jesse Brandeburg
-M: jesse.brandeburg@intel.com
P: Jeff Kirsher
M: jeffrey.t.kirsher@intel.com
+P: Jesse Brandeburg
+M: jesse.brandeburg@intel.com
P: Bruce Allan
M: bruce.w.allan@intel.com
P: John Ronciak
void __init orion5x_eth_init(struct mv643xx_eth_platform_data *eth_data)
{
+ eth_data->shared = &orion5x_eth_shared;
orion5x_eth.dev.platform_data = eth_data;
+
platform_device_register(&orion5x_eth_shared);
platform_device_register(&orion5x_eth);
}
#include <asm/uaccess.h>
#include <asm/segment.h>
-/*
- * sys_pipe() is the normal C calling standard for creating
- * a pipe. It's not the way Unix traditionally does this, though.
- */
-asmlinkage int sys_pipe(unsigned long __user * fildes)
-{
- int fd[2];
- int error;
-
- lock_kernel();
- error = do_pipe(fd);
- unlock_kernel();
- if (!error) {
- if (copy_to_user(fildes, fd, 2*sizeof(int)))
- error = -EFAULT;
- }
- return error;
-}
-
/* common code for old and new mmaps */
static inline long
do_mmap2(unsigned long addr, unsigned long len, unsigned long prot,
return oldval;
}
-/*
- * sys_pipe() is the normal C calling standard for creating
- * a pipe. It's not the way Unix traditionally does this, though.
- */
-asmlinkage int
-sys_pipe(unsigned long r0, unsigned long r1, unsigned long r2,
- unsigned long r3, unsigned long r4, unsigned long r5,
- unsigned long r6, struct pt_regs regs)
-{
- int fd[2];
- int error;
-
- error = do_pipe(fd);
- if (!error) {
- if (copy_to_user((void __user *)r0, fd, 2*sizeof(int)))
- error = -EFAULT;
- }
- return error;
-}
-
asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
unsigned long prot, unsigned long flags,
unsigned long fd, unsigned long pgoff)
* (if do_page_fault didn't fix the mapping,
* the writeback won't do good)
*/
+disable_wb:
#ifdef DEBUG
printk(".. disabling wb2\n");
#endif
if (fp->un.fmt7.wb2a == fp->un.fmt7.faddr)
fp->un.fmt7.wb2s &= ~WBV_040;
+ if (fp->un.fmt7.wb3a == fp->un.fmt7.faddr)
+ fp->un.fmt7.wb3s &= ~WBV_040;
}
- } else if (send_fault_sig(&fp->ptregs) > 0) {
- printk("68040 access error, ssw=%x\n", ssw);
- trap_c(fp);
+ } else {
+ /* In case of a bus error we either kill the process or expect
+ * the kernel to catch the fault, which then is also responsible
+ * for cleaning up the mess.
+ */
+ current->thread.signo = SIGBUS;
+ current->thread.faddr = fp->un.fmt7.faddr;
+ if (send_fault_sig(&fp->ptregs) >= 0)
+ printk("68040 bus error (ssw=%x, faddr=%lx)\n", ssw,
+ fp->un.fmt7.faddr);
+ goto disable_wb;
}
do_040writebacks(fp);
struct mac_booter_data mac_bi_data;
int mac_bisize = sizeof mac_bi_data;
-struct mac_hw_present mac_hw_present;
-EXPORT_SYMBOL(mac_hw_present);
-
/* New m68k bootinfo stuff and videobase */
extern int m68k_num_memory;
m68k_ramdisk.addr, m68k_ramdisk.size);
#endif
- /*
- * TODO: set the various fields in macintosh_config->hw_present here!
- */
- switch (macintosh_config->scsi_type) {
- case MAC_SCSI_OLD:
- MACHW_SET(MAC_SCSI_80);
- break;
- case MAC_SCSI_QUADRA:
- case MAC_SCSI_QUADRA2:
- case MAC_SCSI_QUADRA3:
- MACHW_SET(MAC_SCSI_96);
- if ((macintosh_config->ident == MAC_MODEL_Q900) ||
- (macintosh_config->ident == MAC_MODEL_Q950))
- MACHW_SET(MAC_SCSI_96_2);
- break;
- default:
- printk(KERN_WARNING "config.c: wtf: unknown scsi, using 53c80\n");
- MACHW_SET(MAC_SCSI_80);
- break;
- }
-
iop_init();
via_init();
oss_init();
static struct mv643xx_eth_platform_data eth0_pd = {
+ .shared = &mv643xx_eth_shared_device,
.port_number = 0,
+
.tx_sram_addr = PEGASOS2_SRAM_BASE_ETH0,
.tx_sram_size = PEGASOS2_SRAM_TXRING_SIZE,
.tx_queue_size = PEGASOS2_SRAM_TXRING_SIZE/16,
};
static struct mv643xx_eth_platform_data eth1_pd = {
+ .shared = &mv643xx_eth_shared_device,
.port_number = 1,
+
.tx_sram_addr = PEGASOS2_SRAM_BASE_ETH1,
.tx_sram_size = PEGASOS2_SRAM_TXRING_SIZE,
.tx_queue_size = PEGASOS2_SRAM_TXRING_SIZE/16,
memset(&pdata, 0, sizeof(pdata));
+ pdata.shared = shared_pdev;
+
prop = of_get_property(np, "reg", NULL);
if (!prop)
return -ENODEV;
};
static struct mv643xx_eth_platform_data eth0_pd = {
+ .shared = &mv64x60_eth_shared_device;
.port_number = 0,
};
};
static struct mv643xx_eth_platform_data eth1_pd = {
+ .shared = &mv64x60_eth_shared_device;
.port_number = 1,
};
};
static struct mv643xx_eth_platform_data eth2_pd = {
+ .shared = &mv64x60_eth_shared_device;
.port_number = 2,
};
Select this option to enable the special message interface to
the cooperative memory management.
+config PAGE_STATES
+ bool "Unused page notification"
+ help
+ This enables the notification of unused pages to the
+ hypervisor. The ESSA instruction is used to do the states
+ changes between a page that has content and the unused state.
+
config VIRT_TIMER
bool "Virtual CPU timer support"
help
lgfr %r3,%r3 # long
llgtr %r4,%r4 # long
llgfr %r5,%r5 # long
- jg sys_ptrace # branch to system call
+ jg compat_sys_ptrace # branch to system call
.globl sys32_alarm_wrapper
sys32_alarm_wrapper:
st %r2,SP_R2(%r15) # store return value (change R2 on stack)
sysc_return:
- tm SP_PSW+1(%r15),0x01 # returning to user ?
- bno BASED(sysc_restore)
tm __TI_flags+3(%r9),_TIF_WORK_SVC
bnz BASED(sysc_work) # there is work to do (signals etc.)
sysc_restore:
# One of the work bits is on. Find out which one.
#
sysc_work:
+ tm SP_PSW+1(%r15),0x01 # returning to user ?
+ bno BASED(sysc_restore)
tm __TI_flags+3(%r9),_TIF_MCCK_PENDING
bo BASED(sysc_mcck_pending)
tm __TI_flags+3(%r9),_TIF_NEED_RESCHED
la %r2,SP_PTREGS(%r15) # address of register-save area
basr %r14,%r1 # branch to standard irq handler
io_return:
- tm SP_PSW+1(%r15),0x01 # returning to user ?
-#ifdef CONFIG_PREEMPT
- bno BASED(io_preempt) # no -> check for preemptive scheduling
-#else
- bno BASED(io_restore) # no-> skip resched & signal
-#endif
tm __TI_flags+3(%r9),_TIF_WORK_INT
bnz BASED(io_work) # there is work to do (signals etc.)
io_restore:
.long 0, io_restore_trace + 0x80000000
#endif
-#ifdef CONFIG_PREEMPT
-io_preempt:
+#
+# switch to kernel stack, then check the TIF bits
+#
+io_work:
+ tm SP_PSW+1(%r15),0x01 # returning to user ?
+#ifndef CONFIG_PREEMPT
+ bno BASED(io_restore) # no-> skip resched & signal
+#else
+ bnz BASED(io_work_user) # no -> check for preemptive scheduling
+ # check for preemptive scheduling
icm %r0,15,__TI_precount(%r9)
- bnz BASED(io_restore)
+ bnz BASED(io_restore) # preemption disabled
l %r1,SP_R15(%r15)
s %r1,BASED(.Lc_spsize)
mvc SP_PTREGS(__PT_SIZE,%r1),SP_PTREGS(%r15)
br %r1 # call schedule
#endif
-#
-# switch to kernel stack, then check the TIF bits
-#
-io_work:
+io_work_user:
l %r1,__LC_KERNEL_STACK
s %r1,BASED(.Lc_spsize)
mvc SP_PTREGS(__PT_SIZE,%r1),SP_PTREGS(%r15)
stg %r2,SP_R2(%r15) # store return value (change R2 on stack)
sysc_return:
- tm SP_PSW+1(%r15),0x01 # returning to user ?
- jno sysc_restore
tm __TI_flags+7(%r9),_TIF_WORK_SVC
jnz sysc_work # there is work to do (signals etc.)
sysc_restore:
# One of the work bits is on. Find out which one.
#
sysc_work:
+ tm SP_PSW+1(%r15),0x01 # returning to user ?
+ jno sysc_restore
tm __TI_flags+7(%r9),_TIF_MCCK_PENDING
jo sysc_mcck_pending
tm __TI_flags+7(%r9),_TIF_NEED_RESCHED
la %r2,SP_PTREGS(%r15) # address of register-save area
brasl %r14,do_IRQ # call standard irq handler
io_return:
- tm SP_PSW+1(%r15),0x01 # returning to user ?
-#ifdef CONFIG_PREEMPT
- jno io_preempt # no -> check for preemptive scheduling
-#else
- jno io_restore # no-> skip resched & signal
-#endif
tm __TI_flags+7(%r9),_TIF_WORK_INT
jnz io_work # there is work to do (signals etc.)
io_restore:
.quad 0, io_restore_trace
#endif
-#ifdef CONFIG_PREEMPT
-io_preempt:
+#
+# There is work todo, we need to check if we return to userspace, then
+# check, if we are in SIE, if yes leave it
+#
+io_work:
+ tm SP_PSW+1(%r15),0x01 # returning to user ?
+#ifndef CONFIG_PREEMPT
+#if defined(CONFIG_KVM) || defined(CONFIG_KVM_MODULE)
+ jnz io_work_user # yes -> no need to check for SIE
+ la %r1, BASED(sie_opcode) # we return to kernel here
+ lg %r2, SP_PSW+8(%r15)
+ clc 0(2,%r1), 0(%r2) # is current instruction = SIE?
+ jne io_restore # no-> return to kernel
+ lg %r1, SP_PSW+8(%r15) # yes-> add 4 bytes to leave SIE
+ aghi %r1, 4
+ stg %r1, SP_PSW+8(%r15)
+ j io_restore # return to kernel
+#else
+ jno io_restore # no-> skip resched & signal
+#endif
+#else
+ jnz io_work_user # yes -> do resched & signal
+#if defined(CONFIG_KVM) || defined(CONFIG_KVM_MODULE)
+ la %r1, BASED(sie_opcode)
+ lg %r2, SP_PSW+8(%r15)
+ clc 0(2,%r1), 0(%r2) # is current instruction = SIE?
+ jne 0f # no -> leave PSW alone
+ lg %r1, SP_PSW+8(%r15) # yes-> add 4 bytes to leave SIE
+ aghi %r1, 4
+ stg %r1, SP_PSW+8(%r15)
+0:
+#endif
+ # check for preemptive scheduling
icm %r0,15,__TI_precount(%r9)
- jnz io_restore
+ jnz io_restore # preemption is disabled
# switch to kernel stack
lg %r1,SP_R15(%r15)
aghi %r1,-SP_SIZE
jg preempt_schedule_irq
#endif
-#
-# switch to kernel stack, then check TIF bits
-#
-io_work:
+io_work_user:
lg %r1,__LC_KERNEL_STACK
aghi %r1,-SP_SIZE
mvc SP_PTREGS(__PT_SIZE,%r1),SP_PTREGS(%r15)
j io_restore
io_work_done:
+#if defined(CONFIG_KVM) || defined(CONFIG_KVM_MODULE)
+sie_opcode:
+ .long 0xb2140000
+#endif
+
#
# _TIF_MCCK_PENDING is set, call handler
#
return 0;
}
-static int
-do_ptrace_normal(struct task_struct *child, long request, long addr, long data)
+long arch_ptrace(struct task_struct *child, long request, long addr, long data)
{
ptrace_area parea;
int copied, ret;
return 0;
}
-static int
-do_ptrace_emu31(struct task_struct *child, long request, long addr, long data)
+long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
+ compat_ulong_t caddr, compat_ulong_t cdata)
{
- unsigned int tmp; /* 4 bytes !! */
+ unsigned long addr = caddr;
+ unsigned long data = cdata;
ptrace_area_emu31 parea;
int copied, ret;
switch (request) {
- case PTRACE_PEEKTEXT:
- case PTRACE_PEEKDATA:
- /* read word at location addr. */
- copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0);
- if (copied != sizeof(tmp))
- return -EIO;
- return put_user(tmp, (unsigned int __force __user *) data);
-
case PTRACE_PEEKUSR:
/* read the word at location addr in the USER area. */
return peek_user_emu31(child, addr, data);
- case PTRACE_POKETEXT:
- case PTRACE_POKEDATA:
- /* write the word at location addr. */
- tmp = data;
- copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 1);
- if (copied != sizeof(tmp))
- return -EIO;
- return 0;
-
case PTRACE_POKEUSR:
/* write the word at location addr in the USER area */
return poke_user_emu31(child, addr, data);
copied += sizeof(unsigned int);
}
return 0;
- case PTRACE_GETEVENTMSG:
- return put_user((__u32) child->ptrace_message,
- (unsigned int __force __user *) data);
- case PTRACE_GETSIGINFO:
- if (child->last_siginfo == NULL)
- return -EINVAL;
- return copy_siginfo_to_user32((compat_siginfo_t
- __force __user *) data,
- child->last_siginfo);
- case PTRACE_SETSIGINFO:
- if (child->last_siginfo == NULL)
- return -EINVAL;
- return copy_siginfo_from_user32(child->last_siginfo,
- (compat_siginfo_t
- __force __user *) data);
}
- return ptrace_request(child, request, addr, data);
+ return compat_ptrace_request(child, request, addr, data);
}
#endif
-long arch_ptrace(struct task_struct *child, long request, long addr, long data)
-{
- switch (request) {
- case PTRACE_SYSCALL:
- /* continue and stop at next (return from) syscall */
- case PTRACE_CONT:
- /* restart after signal. */
- if (!valid_signal(data))
- return -EIO;
- if (request == PTRACE_SYSCALL)
- set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
- else
- clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
- child->exit_code = data;
- /* make sure the single step bit is not set. */
- user_disable_single_step(child);
- wake_up_process(child);
- return 0;
-
- case PTRACE_KILL:
- /*
- * make the child exit. Best I can do is send it a sigkill.
- * perhaps it should be put in the status that it wants to
- * exit.
- */
- if (child->exit_state == EXIT_ZOMBIE) /* already dead */
- return 0;
- child->exit_code = SIGKILL;
- /* make sure the single step bit is not set. */
- user_disable_single_step(child);
- wake_up_process(child);
- return 0;
-
- case PTRACE_SINGLESTEP:
- /* set the trap flag. */
- if (!valid_signal(data))
- return -EIO;
- clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
- child->exit_code = data;
- user_enable_single_step(child);
- /* give it a chance to run. */
- wake_up_process(child);
- return 0;
-
- /* Do requests that differ for 31/64 bit */
- default:
-#ifdef CONFIG_COMPAT
- if (test_thread_flag(TIF_31BIT))
- return do_ptrace_emu31(child, request, addr, data);
-#endif
- return do_ptrace_normal(child, request, addr, data);
- }
- /* Not reached. */
- return -EIO;
-}
-
asmlinkage void
syscall_trace(struct pt_regs *regs, int entryexit)
{
select PREEMPT_NOTIFIERS
select ANON_INODES
select S390_SWITCH_AMODE
- select PREEMPT
---help---
Support hosting paravirtualized guest machines using the SIE
virtualization capability on the mainframe. This should work
static int handle_noop(struct kvm_vcpu *vcpu)
{
switch (vcpu->arch.sie_block->icptcode) {
+ case 0x0:
+ vcpu->stat.exit_null++;
+ break;
case 0x10:
vcpu->stat.exit_external_request++;
break;
struct kvm_stats_debugfs_item debugfs_entries[] = {
{ "userspace_handled", VCPU_STAT(exit_userspace) },
+ { "exit_null", VCPU_STAT(exit_null) },
{ "exit_validity", VCPU_STAT(exit_validity) },
{ "exit_stop_request", VCPU_STAT(exit_stop_request) },
{ "exit_external_request", VCPU_STAT(exit_external_request) },
vcpu->arch.guest_fpregs.fpc &= FPC_VALID_MASK;
restore_fp_regs(&vcpu->arch.guest_fpregs);
restore_access_regs(vcpu->arch.guest_acrs);
-
- if (signal_pending(current))
- atomic_set_mask(CPUSTAT_STOP_INT,
- &vcpu->arch.sie_block->cpuflags);
}
void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
obj-y := init.o fault.o extmem.o mmap.o vmem.o pgtable.o
obj-$(CONFIG_CMM) += cmm.o
obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o
+obj-$(CONFIG_PAGE_STATES) += page-states.o
/* clear the zero-page */
memset(empty_zero_page, 0, PAGE_SIZE);
+ /* Setup guest page hinting */
+ cmma_init();
+
/* this will put all low memory onto the freelists */
totalram_pages += free_all_bootmem();
--- /dev/null
+/*
+ * arch/s390/mm/page-states.c
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * Guest page hinting for unused pages.
+ *
+ * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+
+#define ESSA_SET_STABLE 1
+#define ESSA_SET_UNUSED 2
+
+static int cmma_flag;
+
+static int __init cmma(char *str)
+{
+ char *parm;
+ parm = strstrip(str);
+ if (strcmp(parm, "yes") == 0 || strcmp(parm, "on") == 0) {
+ cmma_flag = 1;
+ return 1;
+ }
+ cmma_flag = 0;
+ if (strcmp(parm, "no") == 0 || strcmp(parm, "off") == 0)
+ return 1;
+ return 0;
+}
+
+__setup("cmma=", cmma);
+
+void __init cmma_init(void)
+{
+ register unsigned long tmp asm("0") = 0;
+ register int rc asm("1") = -EOPNOTSUPP;
+
+ if (!cmma_flag)
+ return;
+ asm volatile(
+ " .insn rrf,0xb9ab0000,%1,%1,0,0\n"
+ "0: la %0,0\n"
+ "1:\n"
+ EX_TABLE(0b,1b)
+ : "+&d" (rc), "+&d" (tmp));
+ if (rc)
+ cmma_flag = 0;
+}
+
+void arch_free_page(struct page *page, int order)
+{
+ int i, rc;
+
+ if (!cmma_flag)
+ return;
+ for (i = 0; i < (1 << order); i++)
+ asm volatile(".insn rrf,0xb9ab0000,%0,%1,%2,0"
+ : "=&d" (rc)
+ : "a" ((page_to_pfn(page) + i) << PAGE_SHIFT),
+ "i" (ESSA_SET_UNUSED));
+}
+
+void arch_alloc_page(struct page *page, int order)
+{
+ int i, rc;
+
+ if (!cmma_flag)
+ return;
+ for (i = 0; i < (1 << order); i++)
+ asm volatile(".insn rrf,0xb9ab0000,%0,%1,%2,0"
+ : "=&d" (rc)
+ : "a" ((page_to_pfn(page) + i) << PAGE_SHIFT),
+ "i" (ESSA_SET_STABLE));
+}
unsigned long stack_size)
{
unsigned long parent_tid_ptr, child_tid_ptr;
+ unsigned long orig_i1 = regs->u_regs[UREG_I1];
+ long ret;
parent_tid_ptr = regs->u_regs[UREG_I2];
child_tid_ptr = regs->u_regs[UREG_I4];
- return do_fork(clone_flags, stack_start,
- regs, stack_size,
- (int __user *) parent_tid_ptr,
- (int __user *) child_tid_ptr);
+ ret = do_fork(clone_flags, stack_start,
+ regs, stack_size,
+ (int __user *) parent_tid_ptr,
+ (int __user *) child_tid_ptr);
+
+ /* If we get an error and potentially restart the system
+ * call, we're screwed because copy_thread() clobbered
+ * the parent's %o1. So detect that case and restore it
+ * here.
+ */
+ if ((unsigned long)ret >= -ERESTART_RESTARTBLOCK)
+ regs->u_regs[UREG_I1] = orig_i1;
+
+ return ret;
}
/* Copy a Sparc thread. The fork() return value conventions
static inline void __user *get_sigframe(struct sigaction *sa, struct pt_regs *regs, unsigned long framesize)
{
- unsigned long sp;
+ unsigned long sp = regs->u_regs[UREG_FP];
- sp = regs->u_regs[UREG_FP];
+ /*
+ * If we are on the alternate signal stack and would overflow it, don't.
+ * Return an always-bogus address instead so we will die with SIGSEGV.
+ */
+ if (on_sig_stack(sp) && !likely(on_sig_stack(sp - framesize)))
+ return (void __user *) -1L;
/* This is the X/Open sanctioned signal stack switching. */
if (sa->sa_flags & SA_ONSTACK) {
- if (!on_sig_stack(sp) && !((current->sas_ss_sp + current->sas_ss_size) & 7))
+ if (sas_ss_flags(sp) == 0)
sp = current->sas_ss_sp + current->sas_ss_size;
}
+
+ /* Always align the stack frame. This handles two cases. First,
+ * sigaltstack need not be mindful of platform specific stack
+ * alignment. Second, if we took this signal because the stack
+ * is not aligned properly, we'd like to take the signal cleanly
+ * and report that.
+ */
+ sp &= ~7UL;
+
return (void __user *)(sp - framesize);
}
{
if (ARCH_SUN4C_SUN4 &&
(len > 0x20000000 ||
- ((flags & MAP_FIXED) &&
- addr < 0xe0000000 && addr + len > 0x20000000)))
+ (addr < 0xe0000000 && addr + len > 0x20000000)))
return -EINVAL;
/* See asm-sparc/uaccess.h */
unsigned long stack_size)
{
int __user *parent_tid_ptr, *child_tid_ptr;
+ unsigned long orig_i1 = regs->u_regs[UREG_I1];
+ long ret;
#ifdef CONFIG_COMPAT
if (test_thread_flag(TIF_32BIT)) {
child_tid_ptr = (int __user *) regs->u_regs[UREG_I4];
}
- return do_fork(clone_flags, stack_start,
- regs, stack_size,
- parent_tid_ptr, child_tid_ptr);
+ ret = do_fork(clone_flags, stack_start,
+ regs, stack_size,
+ parent_tid_ptr, child_tid_ptr);
+
+ /* If we get an error and potentially restart the system
+ * call, we're screwed because copy_thread() clobbered
+ * the parent's %o1. So detect that case and restore it
+ * here.
+ */
+ if ((unsigned long)ret >= -ERESTART_RESTARTBLOCK)
+ regs->u_regs[UREG_I1] = orig_i1;
+
+ return ret;
}
/* Copy a Sparc thread. The fork() return value conventions
static inline void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, unsigned long framesize)
{
- unsigned long sp;
+ unsigned long sp = regs->u_regs[UREG_FP] + STACK_BIAS;
- sp = regs->u_regs[UREG_FP] + STACK_BIAS;
+ /*
+ * If we are on the alternate signal stack and would overflow it, don't.
+ * Return an always-bogus address instead so we will die with SIGSEGV.
+ */
+ if (on_sig_stack(sp) && !likely(on_sig_stack(sp - framesize)))
+ return (void __user *) -1L;
/* This is the X/Open sanctioned signal stack switching. */
if (ka->sa.sa_flags & SA_ONSTACK) {
- if (!on_sig_stack(sp) &&
- !((current->sas_ss_sp + current->sas_ss_size) & 7))
+ if (sas_ss_flags(sp) == 0)
sp = current->sas_ss_sp + current->sas_ss_size;
}
+
+ /* Always align the stack frame. This handles two cases. First,
+ * sigaltstack need not be mindful of platform specific stack
+ * alignment. Second, if we took this signal because the stack
+ * is not aligned properly, we'd like to take the signal cleanly
+ * and report that.
+ */
+ sp &= ~7UL;
+
return (void __user *)(sp - framesize);
}
regs->u_regs[UREG_FP] &= 0x00000000ffffffffUL;
sp = regs->u_regs[UREG_FP];
+ /*
+ * If we are on the alternate signal stack and would overflow it, don't.
+ * Return an always-bogus address instead so we will die with SIGSEGV.
+ */
+ if (on_sig_stack(sp) && !likely(on_sig_stack(sp - framesize)))
+ return (void __user *) -1L;
+
/* This is the X/Open sanctioned signal stack switching. */
if (sa->sa_flags & SA_ONSTACK) {
- if (!on_sig_stack(sp) && !((current->sas_ss_sp + current->sas_ss_size) & 7))
+ if (sas_ss_flags(sp) == 0)
sp = current->sas_ss_sp + current->sas_ss_size;
}
+
+ /* Always align the stack frame. This handles two cases. First,
+ * sigaltstack need not be mindful of platform specific stack
+ * alignment. Second, if we took this signal because the stack
+ * is not aligned properly, we'd like to take the signal cleanly
+ * and report that.
+ */
+ sp &= ~7UL;
+
return (void __user *)(sp - framesize);
}
void *info = call_data->info;
clear_softint(1 << irq);
-
- irq_enter();
-
- if (!call_data->wait) {
- /* let initiator proceed after getting data */
- atomic_inc(&call_data->finished);
- }
-
- func(info);
-
- irq_exit();
-
if (call_data->wait) {
/* let initiator proceed only after completion */
+ func(info);
atomic_inc(&call_data->finished);
+ } else {
+ /* let initiator proceed after getting data */
+ atomic_inc(&call_data->finished);
+ func(info);
}
}
void smp_receive_signal_client(int irq, struct pt_regs *regs)
{
- irq_enter();
clear_softint(1 << irq);
- irq_exit();
}
void smp_new_mmu_context_version_client(int irq, struct pt_regs *regs)
struct mm_struct *mm;
unsigned long flags;
- irq_enter();
-
clear_softint(1 << irq);
/* See if we need to allocate a new TLB context because
load_secondary_context(mm);
__flush_tlb_mm(CTX_HWBITS(mm->context),
SECONDARY_CONTEXT);
-
- irq_exit();
}
void smp_new_mmu_context_version(void)
{
clear_softint(1 << irq);
- irq_enter();
-
preempt_disable();
__asm__ __volatile__("flushw");
prom_world(0);
preempt_enable();
-
- irq_exit();
}
/* /proc/profile writes can call this, don't __init it please. */
if (len >= STACK_TOP32)
return -EINVAL;
- if ((flags & MAP_FIXED) && addr > STACK_TOP32 - len)
+ if (addr > STACK_TOP32 - len)
return -EINVAL;
} else {
if (len >= VA_EXCLUDE_START)
return -EINVAL;
- if ((flags & MAP_FIXED) && invalid_64bit_range(addr, len))
+ if (invalid_64bit_range(addr, len))
return -EINVAL;
}
/* 32-bit timeval and related flotsam. */
-static long get_tv32(struct timeval *o, struct compat_timeval __user *i)
-{
- return (!access_ok(VERIFY_READ, i, sizeof(*i)) ||
- (__get_user(o->tv_sec, &i->tv_sec) |
- __get_user(o->tv_usec, &i->tv_usec)));
-}
-
static inline long put_tv32(struct compat_timeval __user *o, struct timeval *i)
{
return (!access_ok(VERIFY_WRITE, o, sizeof(*o)) ||
return do_sys_settimeofday(tv ? &kts : NULL, tz ? &ktz : NULL);
}
-asmlinkage long sys32_utimes(char __user *filename,
- struct compat_timeval __user *tvs)
-{
- struct timespec tv[2];
-
- if (tvs) {
- struct timeval ktvs[2];
- if (get_tv32(&ktvs[0], tvs) ||
- get_tv32(&ktvs[1], 1+tvs))
- return -EFAULT;
-
- if (ktvs[0].tv_usec < 0 || ktvs[0].tv_usec >= 1000000 ||
- ktvs[1].tv_usec < 0 || ktvs[1].tv_usec >= 1000000)
- return -EINVAL;
-
- tv[0].tv_sec = ktvs[0].tv_sec;
- tv[0].tv_nsec = 1000 * ktvs[0].tv_usec;
- tv[1].tv_sec = ktvs[1].tv_sec;
- tv[1].tv_nsec = 1000 * ktvs[1].tv_usec;
- }
-
- return do_utimes(AT_FDCWD, filename, tvs ? tv : NULL, 0);
-}
-
/* These are here just in case some old sparc32 binary calls it. */
asmlinkage long sys32_pause(void)
{
/*120*/ .word compat_sys_readv, compat_sys_writev, sys32_settimeofday, sys32_fchown16, sys_fchmod
.word sys_nis_syscall, sys32_setreuid16, sys32_setregid16, sys_rename, sys_truncate
/*130*/ .word sys_ftruncate, sys_flock, compat_sys_lstat64, sys_nis_syscall, sys_nis_syscall
- .word sys_nis_syscall, sys32_mkdir, sys_rmdir, sys32_utimes, compat_sys_stat64
+ .word sys_nis_syscall, sys32_mkdir, sys_rmdir, compat_sys_utimes, compat_sys_stat64
/*140*/ .word sys32_sendfile64, sys_nis_syscall, sys32_futex, sys_gettid, compat_sys_getrlimit
.word compat_sys_setrlimit, sys_pivot_root, sys32_prctl, sys_pciconfig_read, sys_pciconfig_write
/*150*/ .word sys_nis_syscall, sys_inotify_init, sys_inotify_add_watch, sys_poll, sys_getdents64
initrd_end = ramdisk_image + sparc_ramdisk_size;
lmb_reserve(initrd_start, initrd_end);
+
+ initrd_start += PAGE_OFFSET;
+ initrd_end += PAGE_OFFSET;
}
#endif
}
__asm__ __volatile__("wrpr %0, 0, %%pstate"
: : "r" (pstate));
}
-
-#ifdef CONFIG_MEMORY_HOTPLUG
-
-void online_page(struct page *page)
-{
- ClearPageReserved(page);
- init_page_count(page);
- __free_page(page);
- totalram_pages++;
- num_physpages++;
-}
-
-#endif /* CONFIG_MEMORY_HOTPLUG */
KERNEL_DEFINES = $(strip -Derrno=kernel_errno -Dsigprocmask=kernel_sigprocmask \
-Dmktime=kernel_mktime $(ARCH_KERNEL_DEFINES))
KBUILD_CFLAGS += $(KERNEL_DEFINES)
-# Disable unit-at-a-time mode on pre-gcc-4.0 compilers, it makes gcc use
-# a lot more stack due to the lack of sharing of stacklots:
-KBUILD_CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then \
- echo $(call cc-option,-fno-unit-at-a-time); fi ;)
+KBUILD_CFLAGS += $(call cc-option,-fno-unit-at-a-time,)
PHONY += linux
line_flush_buffer(tty);
}
-void line_put_char(struct tty_struct *tty, unsigned char ch)
+int line_put_char(struct tty_struct *tty, unsigned char ch)
{
- line_write(tty, &ch, sizeof(ch));
+ return line_write(tty, &ch, sizeof(ch));
}
int line_write(struct tty_struct *tty, const unsigned char *buf, int len)
char *init, char **error_out);
extern int line_write(struct tty_struct *tty, const unsigned char *buf,
int len);
-extern void line_put_char(struct tty_struct *tty, unsigned char ch);
+extern int line_put_char(struct tty_struct *tty, unsigned char ch);
extern void line_set_termios(struct tty_struct *tty, struct ktermios * old);
extern int line_chars_in_buffer(struct tty_struct *tty);
extern void line_flush_buffer(struct tty_struct *tty);
### Arch settings
config X86
def_bool y
+ select HAVE_UNSTABLE_SCHED_CLOCK
select HAVE_IDE
select HAVE_OPROFILE
select HAVE_KPROBES
obj-$(CONFIG_KVM_CLOCK) += kvmclock.o
obj-$(CONFIG_PARAVIRT) += paravirt.o paravirt_patch_$(BITS).o
-ifdef CONFIG_INPUT_PCSPKR
-obj-y += pcspeaker.o
-endif
+obj-$(CONFIG_PCSPKR_PLATFORM) += pcspeaker.o
obj-$(CONFIG_SCx200) += scx200.o
scx200-y += scx200_32.o
__FIXADDR_TOP = -reserve - PAGE_SIZE;
__VMALLOC_RESERVE += reserve;
}
-
-int pmd_bad(pmd_t pmd)
-{
- WARN_ON_ONCE(pmd_bad_v1(pmd) != pmd_bad_v2(pmd));
-
- return pmd_bad_v1(pmd);
-}
#include <asm/numa.h>
#include "pci.h"
-static int __devinit can_skip_ioresource_align(const struct dmi_system_id *d)
-{
- pci_probe |= PCI_CAN_SKIP_ISA_ALIGN;
- printk(KERN_INFO "PCI: %s detected, can skip ISA alignment\n", d->ident);
- return 0;
-}
-
-static struct dmi_system_id acpi_pciprobe_dmi_table[] __devinitdata = {
-/*
- * Systems where PCI IO resource ISA alignment can be skipped
- * when the ISA enable bit in the bridge control is not set
- */
- {
- .callback = can_skip_ioresource_align,
- .ident = "IBM System x3800",
- .matches = {
- DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
- DMI_MATCH(DMI_PRODUCT_NAME, "x3800"),
- },
- },
- {
- .callback = can_skip_ioresource_align,
- .ident = "IBM System x3850",
- .matches = {
- DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
- DMI_MATCH(DMI_PRODUCT_NAME, "x3850"),
- },
- },
- {
- .callback = can_skip_ioresource_align,
- .ident = "IBM System x3950",
- .matches = {
- DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
- DMI_MATCH(DMI_PRODUCT_NAME, "x3950"),
- },
- },
- {}
-};
-
struct pci_root_info {
char *name;
unsigned int res_num;
int pxm;
#endif
- dmi_check_system(acpi_pciprobe_dmi_table);
-
if (domain && !pci_domains_supported) {
printk(KERN_WARNING "PCI: Multiple domains not supported "
"(dom %d, bus %d)\n", domain, busnum);
*/
DEFINE_SPINLOCK(pci_config_lock);
-static void __devinit pcibios_fixup_device_resources(struct pci_dev *dev)
+static int __devinit can_skip_ioresource_align(const struct dmi_system_id *d)
{
- struct resource *rom_r = &dev->resource[PCI_ROM_RESOURCE];
-
- if (rom_r->parent)
- return;
- if (rom_r->start)
- /* we deal with BIOS assigned ROM later */
- return;
- if (!(pci_probe & PCI_ASSIGN_ROMS))
- rom_r->start = rom_r->end = rom_r->flags = 0;
+ pci_probe |= PCI_CAN_SKIP_ISA_ALIGN;
+ printk(KERN_INFO "PCI: %s detected, can skip ISA alignment\n", d->ident);
+ return 0;
+}
+
+static struct dmi_system_id can_skip_pciprobe_dmi_table[] __devinitdata = {
+/*
+ * Systems where PCI IO resource ISA alignment can be skipped
+ * when the ISA enable bit in the bridge control is not set
+ */
+ {
+ .callback = can_skip_ioresource_align,
+ .ident = "IBM System x3800",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "x3800"),
+ },
+ },
+ {
+ .callback = can_skip_ioresource_align,
+ .ident = "IBM System x3850",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "x3850"),
+ },
+ },
+ {
+ .callback = can_skip_ioresource_align,
+ .ident = "IBM System x3950",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "x3950"),
+ },
+ },
+ {}
+};
+
+void __init dmi_check_skip_isa_align(void)
+{
+ dmi_check_system(can_skip_pciprobe_dmi_table);
}
/*
void __devinit pcibios_fixup_bus(struct pci_bus *b)
{
- struct pci_dev *dev;
-
pci_read_bridge_bases(b);
- list_for_each_entry(dev, &b->devices, bus_list)
- pcibios_fixup_device_resources(dev);
}
/*
{}
};
+void __init dmi_check_pciprobe(void)
+{
+ dmi_check_system(pciprobe_dmi_table);
+}
+
struct pci_bus * __devinit pcibios_scan_root(int busnum)
{
struct pci_bus *bus = NULL;
struct pci_sysdata *sd;
- dmi_check_system(pciprobe_dmi_table);
-
while ((bus = pci_find_next_bus(bus)) != NULL) {
if (bus->number == busnum) {
/* Already scanned */
} else if (!strcmp(str, "routeirq")) {
pci_routeirq = 1;
return NULL;
+ } else if (!strcmp(str, "skip_isa_align")) {
+ pci_probe |= PCI_CAN_SKIP_ISA_ALIGN;
+ return NULL;
}
return str;
}
pcibios_disable_irq(dev);
}
-struct pci_bus *pci_scan_bus_on_node(int busno, struct pci_ops *ops, int node)
+struct pci_bus * __devinit pci_scan_bus_on_node(int busno, struct pci_ops *ops, int node)
{
struct pci_bus *bus = NULL;
struct pci_sysdata *sd;
return bus;
}
-struct pci_bus *pci_scan_bus_with_sysdata(int busno)
+struct pci_bus * __devinit pci_scan_bus_with_sysdata(int busno)
{
return pci_scan_bus_on_node(busno, &pci_root_ops, -1);
}
*/
static void fam10h_pci_cfg_space_size(struct pci_dev *dev)
{
- dev->cfg_size = pci_cfg_space_size_ext(dev, 0);
+ dev->cfg_size = pci_cfg_space_size_ext(dev);
}
DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1200, fam10h_pci_cfg_space_size);
printk(KERN_ERR
"PCI: Fatal: No config space access function found\n");
+ dmi_check_pciprobe();
+
+ dmi_check_skip_isa_align();
+
return 0;
}
arch_initcall(pci_access_init);
pci_dmi_bf,
};
+extern void __init dmi_check_pciprobe(void);
+extern void __init dmi_check_skip_isa_align(void);
+
/* pci-i386.c */
extern unsigned int pcibios_max_latency;
static void drive_stat_acct(struct request *rq, int new_io)
{
+ struct hd_struct *part;
int rw = rq_data_dir(rq);
if (!blk_fs_request(rq) || !rq->rq_disk)
return;
- if (!new_io) {
- __all_stat_inc(rq->rq_disk, merges[rw], rq->sector);
- } else {
- struct hd_struct *part = get_part(rq->rq_disk, rq->sector);
+ part = get_part(rq->rq_disk, rq->sector);
+ if (!new_io)
+ __all_stat_inc(rq->rq_disk, part, merges[rw], rq->sector);
+ else {
disk_round_stats(rq->rq_disk);
rq->rq_disk->in_flight++;
if (part) {
**/
void generic_unplug_device(struct request_queue *q)
{
- spin_lock_irq(q->queue_lock);
- __generic_unplug_device(q);
- spin_unlock_irq(q->queue_lock);
+ if (blk_queue_plugged(q)) {
+ spin_lock_irq(q->queue_lock);
+ __generic_unplug_device(q);
+ spin_unlock_irq(q->queue_lock);
+ }
}
EXPORT_SYMBOL(generic_unplug_device);
}
if (blk_fs_request(req) && req->rq_disk) {
+ struct hd_struct *part = get_part(req->rq_disk, req->sector);
const int rw = rq_data_dir(req);
- all_stat_add(req->rq_disk, sectors[rw],
- nr_bytes >> 9, req->sector);
+ all_stat_add(req->rq_disk, part, sectors[rw],
+ nr_bytes >> 9, req->sector);
}
total_bytes = bio_nbytes = 0;
const int rw = rq_data_dir(req);
struct hd_struct *part = get_part(disk, req->sector);
- __all_stat_inc(disk, ios[rw], req->sector);
- __all_stat_add(disk, ticks[rw], duration, req->sector);
+ __all_stat_inc(disk, part, ios[rw], req->sector);
+ __all_stat_add(disk, part, ticks[rw], duration, req->sector);
disk_round_stats(disk);
disk->in_flight--;
if (part) {
rcu_read_lock();
if (ioc->aic && ioc->aic->dtor)
ioc->aic->dtor(ioc->aic);
- rcu_read_unlock();
cfq_dtor(ioc);
+ rcu_read_unlock();
kmem_cache_free(iocontext_cachep, ioc);
return 1;
static int blk_hw_contig_segment(struct request_queue *q, struct bio *bio,
struct bio *nxt)
{
- if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
+ if (!bio_flagged(bio, BIO_SEG_VALID))
blk_recount_segments(q, bio);
- if (unlikely(!bio_flagged(nxt, BIO_SEG_VALID)))
+ if (!bio_flagged(nxt, BIO_SEG_VALID))
blk_recount_segments(q, nxt);
if (!BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt)) ||
BIOVEC_VIRT_OVERSIZE(bio->bi_hw_back_size + nxt->bi_hw_front_size))
q->last_merge = NULL;
return 0;
}
- if (unlikely(!bio_flagged(req->biotail, BIO_SEG_VALID)))
+ if (!bio_flagged(req->biotail, BIO_SEG_VALID))
blk_recount_segments(q, req->biotail);
- if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
+ if (!bio_flagged(bio, BIO_SEG_VALID))
blk_recount_segments(q, bio);
len = req->biotail->bi_hw_back_size + bio->bi_hw_front_size;
if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(req->biotail), __BVEC_START(bio))
return 0;
}
len = bio->bi_hw_back_size + req->bio->bi_hw_front_size;
- if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
+ if (!bio_flagged(bio, BIO_SEG_VALID))
blk_recount_segments(q, bio);
- if (unlikely(!bio_flagged(req->bio, BIO_SEG_VALID)))
+ if (!bio_flagged(req->bio, BIO_SEG_VALID))
blk_recount_segments(q, req->bio);
if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(req->bio)) &&
!BIOVEC_VIRT_OVERSIZE(len)) {
unsigned long nm;
ssize_t ret = queue_var_store(&nm, page, count);
+ spin_lock_irq(q->queue_lock);
if (nm)
- set_bit(QUEUE_FLAG_NOMERGES, &q->queue_flags);
+ queue_flag_set(QUEUE_FLAG_NOMERGES, q);
else
- clear_bit(QUEUE_FLAG_NOMERGES, &q->queue_flags);
+ queue_flag_clear(QUEUE_FLAG_NOMERGES, q);
+ spin_unlock_irq(q->queue_lock);
return ret;
}
__blk_free_tags(bqt);
q->queue_tags = NULL;
- queue_flag_clear(QUEUE_FLAG_QUEUED, q);
+ queue_flag_clear_unlocked(QUEUE_FLAG_QUEUED, q);
}
/**
**/
void blk_queue_free_tags(struct request_queue *q)
{
- queue_flag_clear(QUEUE_FLAG_QUEUED, q);
+ queue_flag_clear_unlocked(QUEUE_FLAG_QUEUED, q);
}
EXPORT_SYMBOL(blk_queue_free_tags);
* @q: the request queue for the device
* @depth: the maximum queue depth supported
* @tags: the tag to use
+ *
+ * Queue lock must be held here if the function is called to resize an
+ * existing map.
**/
int blk_queue_init_tags(struct request_queue *q, int depth,
struct blk_queue_tag *tags)
* assign it, all done
*/
q->queue_tags = tags;
- queue_flag_set(QUEUE_FLAG_QUEUED, q);
+ queue_flag_set_unlocked(QUEUE_FLAG_QUEUED, q);
INIT_LIST_HEAD(&q->tag_busy_list);
return 0;
fail:
kmem_cache_free(cfq_pool, cfqq);
}
+static void
+__call_for_each_cic(struct io_context *ioc,
+ void (*func)(struct io_context *, struct cfq_io_context *))
+{
+ struct cfq_io_context *cic;
+ struct hlist_node *n;
+
+ hlist_for_each_entry_rcu(cic, n, &ioc->cic_list, cic_list)
+ func(ioc, cic);
+}
+
/*
* Call func for each cic attached to this ioc.
*/
call_for_each_cic(struct io_context *ioc,
void (*func)(struct io_context *, struct cfq_io_context *))
{
- struct cfq_io_context *cic;
- struct hlist_node *n;
-
rcu_read_lock();
- hlist_for_each_entry_rcu(cic, n, &ioc->cic_list, cic_list)
- func(ioc, cic);
+ __call_for_each_cic(ioc, func);
rcu_read_unlock();
}
* should be ok to iterate over the known list, we will see all cic's
* since no new ones are added.
*/
- call_for_each_cic(ioc, cic_free_func);
+ __call_for_each_cic(ioc, cic_free_func);
}
static void cfq_exit_cfqq(struct cfq_data *cfqd, struct cfq_queue *cfqq)
printk(KERN_ERR "cfq: bad prio %x\n", ioprio_class);
case IOPRIO_CLASS_NONE:
/*
- * no prio set, place us in the middle of the BE classes
+ * no prio set, inherit CPU scheduling settings
*/
cfqq->ioprio = task_nice_ioprio(tsk);
- cfqq->ioprio_class = IOPRIO_CLASS_BE;
+ cfqq->ioprio_class = task_nice_ioclass(tsk);
break;
case IOPRIO_CLASS_RT:
cfqq->ioprio = task_ioprio(ioc);
menuconfig ACCESSIBILITY
bool "Accessibility support"
---help---
- Enable a submenu where accessibility items may be enabled.
+ Accessibility handles all special kinds of hardware devices or
+ software adapters which help people with disabilities (e.g.
+ blindness) to use computers.
+
+ That includes braille devices, speech synthesis, keyboard
+ remapping, etc.
+
+ Say Y here to get to see options for accessibility.
+ This option alone does not add any kernel code.
+
+ If you say N, all options in this submenu will be skipped and disabled.
If unsure, say N.
If unsure, say N.
config SATA_INIC162X
- tristate "Initio 162x SATA support (HIGHLY EXPERIMENTAL)"
- depends on PCI && EXPERIMENTAL
+ tristate "Initio 162x SATA support"
+ depends on PCI
help
This option enables support for Initio 162x Serial ATA.
If unsure, say N.
+config PATA_SCH
+ tristate "Intel SCH PATA support"
+ depends on PCI
+ help
+ This option enables support for Intel SCH PATA on the Intel
+ SCH (US15W, US15L, UL11L) series host controllers.
+
+ If unsure, say N.
+
config PATA_BF54X
tristate "Blackfin 54x ATAPI support"
depends on BF542 || BF548 || BF549
obj-$(CONFIG_PATA_TRIFLEX) += pata_triflex.o
obj-$(CONFIG_PATA_IXP4XX_CF) += pata_ixp4xx_cf.o
obj-$(CONFIG_PATA_SCC) += pata_scc.o
+obj-$(CONFIG_PATA_SCH) += pata_sch.o
obj-$(CONFIG_PATA_BF54X) += pata_bf54x.o
obj-$(CONFIG_PATA_PLATFORM) += pata_platform.o
obj-$(CONFIG_PATA_OF_PLATFORM) += pata_of_platform.o
void __iomem *port_mmio = ahci_port_base(link->ap);
u8 status = readl(port_mmio + PORT_TFDATA) & 0xFF;
- if (!(status & ATA_BUSY))
- return 1;
- return 0;
+ return ata_check_ready(status);
}
static int ahci_softreset(struct ata_link *link, unsigned int *class,
if (dev->vendor == PCI_VENDOR_ID_AL)
ata_pci_bmdma_clear_simplex(dev);
+ if (dev->vendor == PCI_VENDOR_ID_ATI) {
+ int rc = pcim_enable_device(dev);
+ if (rc < 0)
+ return rc;
+ pcim_pin_device(dev);
+ }
return ata_pci_sff_init_one(dev, ppi, &generic_sht, NULL);
}
{
struct pci_dev *pdev = to_pci_dev(host->dev);
struct piix_host_priv *hpriv = host->private_data;
+ struct ata_device *dev0 = &host->ports[0]->link.device[0];
+ u32 scontrol;
int i;
/* check for availability */
return;
hpriv->sidpr = pcim_iomap_table(pdev)[PIIX_SIDPR_BAR];
+
+ /* SCR access via SIDPR doesn't work on some configurations.
+ * Give it a test drive by inhibiting power save modes which
+ * we'll do anyway.
+ */
+ scontrol = piix_sidpr_read(dev0, SCR_CONTROL);
+
+ /* if IPM is already 3, SCR access is probably working. Don't
+ * un-inhibit power save modes as BIOS might have inhibited
+ * them for a reason.
+ */
+ if ((scontrol & 0xf00) != 0x300) {
+ scontrol |= 0x300;
+ piix_sidpr_write(dev0, SCR_CONTROL, scontrol);
+ scontrol = piix_sidpr_read(dev0, SCR_CONTROL);
+
+ if ((scontrol & 0xf00) != 0x300) {
+ dev_printk(KERN_INFO, host->dev, "SCR access via "
+ "SIDPR is available but doesn't work\n");
+ return;
+ }
+ }
+
host->ports[0]->ops = &piix_sidpr_sata_ops;
host->ports[1]->ops = &piix_sidpr_sata_ops;
}
EXPORT_SYMBOL_GPL(ata_eh_thaw_port);
EXPORT_SYMBOL_GPL(ata_eh_qc_complete);
EXPORT_SYMBOL_GPL(ata_eh_qc_retry);
+EXPORT_SYMBOL_GPL(ata_eh_analyze_ncq_error);
EXPORT_SYMBOL_GPL(ata_do_eh);
EXPORT_SYMBOL_GPL(ata_std_error_handler);
* LOCKING:
* Kernel thread context (may sleep).
*/
-static void ata_eh_analyze_ncq_error(struct ata_link *link)
+void ata_eh_analyze_ncq_error(struct ata_link *link)
{
struct ata_port *ap = link->ap;
struct ata_eh_context *ehc = &link->eh_context;
{
u8 status = link->ap->ops->sff_check_status(link->ap);
- if (!(status & ATA_BUSY))
- return 1;
- if (status == 0xff)
- return -ENODEV;
- return 0;
+ return ata_check_ready(status);
}
/**
.port_ops = &pacpi_ops,
};
const struct ata_port_info *ppi[] = { &info, NULL };
+ if (pdev->vendor == PCI_VENDOR_ID_ATI) {
+ int rc = pcim_enable_device(pdev);
+ if (rc < 0)
+ return rc;
+ pcim_pin_device(pdev);
+ }
return ata_pci_sff_init_one(pdev, ppi, &pacpi_sht, NULL);
}
--- /dev/null
+/*
+ * pata_sch.c - Intel SCH PATA controllers
+ *
+ * Copyright (c) 2008 Alek Du <alek.du@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License 2 as published
+ * by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; see the file COPYING. If not, write to
+ * the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ */
+
+/*
+ * Supports:
+ * Intel SCH (AF82US15W, AF82US15L, AF82UL11L) chipsets -- see spec at:
+ * http://download.intel.com/design/chipsets/embedded/datashts/319537.pdf
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/dmi.h>
+
+#define DRV_NAME "pata_sch"
+#define DRV_VERSION "0.2"
+
+/* see SCH datasheet page 351 */
+enum {
+ D0TIM = 0x80, /* Device 0 Timing Register */
+ D1TIM = 0x84, /* Device 1 Timing Register */
+ PM = 0x07, /* PIO Mode Bit Mask */
+ MDM = (0x03 << 8), /* Multi-word DMA Mode Bit Mask */
+ UDM = (0x07 << 16), /* Ultra DMA Mode Bit Mask */
+ PPE = (1 << 30), /* Prefetch/Post Enable */
+ USD = (1 << 31), /* Use Synchronous DMA */
+};
+
+static int sch_init_one(struct pci_dev *pdev,
+ const struct pci_device_id *ent);
+static void sch_set_piomode(struct ata_port *ap, struct ata_device *adev);
+static void sch_set_dmamode(struct ata_port *ap, struct ata_device *adev);
+
+static const struct pci_device_id sch_pci_tbl[] = {
+ /* Intel SCH PATA Controller */
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_SCH_IDE), 0 },
+ { } /* terminate list */
+};
+
+static struct pci_driver sch_pci_driver = {
+ .name = DRV_NAME,
+ .id_table = sch_pci_tbl,
+ .probe = sch_init_one,
+ .remove = ata_pci_remove_one,
+#ifdef CONFIG_PM
+ .suspend = ata_pci_device_suspend,
+ .resume = ata_pci_device_resume,
+#endif
+};
+
+static struct scsi_host_template sch_sht = {
+ ATA_BMDMA_SHT(DRV_NAME),
+};
+
+static struct ata_port_operations sch_pata_ops = {
+ .inherits = &ata_bmdma_port_ops,
+ .cable_detect = ata_cable_unknown,
+ .set_piomode = sch_set_piomode,
+ .set_dmamode = sch_set_dmamode,
+};
+
+static struct ata_port_info sch_port_info = {
+ .flags = 0,
+ .pio_mask = ATA_PIO4, /* pio0-4 */
+ .mwdma_mask = ATA_MWDMA2, /* mwdma0-2 */
+ .udma_mask = ATA_UDMA5, /* udma0-5 */
+ .port_ops = &sch_pata_ops,
+};
+
+MODULE_AUTHOR("Alek Du <alek.du@intel.com>");
+MODULE_DESCRIPTION("SCSI low-level driver for Intel SCH PATA controllers");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, sch_pci_tbl);
+MODULE_VERSION(DRV_VERSION);
+
+/**
+ * sch_set_piomode - Initialize host controller PATA PIO timings
+ * @ap: Port whose timings we are configuring
+ * @adev: ATA device
+ *
+ * Set PIO mode for device, in host controller PCI config space.
+ *
+ * LOCKING:
+ * None (inherited from caller).
+ */
+
+static void sch_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+ unsigned int pio = adev->pio_mode - XFER_PIO_0;
+ struct pci_dev *dev = to_pci_dev(ap->host->dev);
+ unsigned int port = adev->devno ? D1TIM : D0TIM;
+ unsigned int data;
+
+ pci_read_config_dword(dev, port, &data);
+ /* see SCH datasheet page 351 */
+ /* set PIO mode */
+ data &= ~(PM | PPE);
+ data |= pio;
+ /* enable PPE for block device */
+ if (adev->class == ATA_DEV_ATA)
+ data |= PPE;
+ pci_write_config_dword(dev, port, data);
+}
+
+/**
+ * sch_set_dmamode - Initialize host controller PATA DMA timings
+ * @ap: Port whose timings we are configuring
+ * @adev: ATA device
+ *
+ * Set MW/UDMA mode for device, in host controller PCI config space.
+ *
+ * LOCKING:
+ * None (inherited from caller).
+ */
+
+static void sch_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+ unsigned int dma_mode = adev->dma_mode;
+ struct pci_dev *dev = to_pci_dev(ap->host->dev);
+ unsigned int port = adev->devno ? D1TIM : D0TIM;
+ unsigned int data;
+
+ pci_read_config_dword(dev, port, &data);
+ /* see SCH datasheet page 351 */
+ if (dma_mode >= XFER_UDMA_0) {
+ /* enable Synchronous DMA mode */
+ data |= USD;
+ data &= ~UDM;
+ data |= (dma_mode - XFER_UDMA_0) << 16;
+ } else { /* must be MWDMA mode, since we masked SWDMA already */
+ data &= ~(USD | MDM);
+ data |= (dma_mode - XFER_MW_DMA_0) << 8;
+ }
+ pci_write_config_dword(dev, port, data);
+}
+
+/**
+ * sch_init_one - Register SCH ATA PCI device with kernel services
+ * @pdev: PCI device to register
+ * @ent: Entry in sch_pci_tbl matching with @pdev
+ *
+ * LOCKING:
+ * Inherited from PCI layer (may sleep).
+ *
+ * RETURNS:
+ * Zero on success, or -ERRNO value.
+ */
+
+static int __devinit sch_init_one(struct pci_dev *pdev,
+ const struct pci_device_id *ent)
+{
+ static int printed_version;
+ const struct ata_port_info *ppi[] = { &sch_port_info, NULL };
+ struct ata_host *host;
+ int rc;
+
+ if (!printed_version++)
+ dev_printk(KERN_DEBUG, &pdev->dev,
+ "version " DRV_VERSION "\n");
+
+ /* enable device and prepare host */
+ rc = pcim_enable_device(pdev);
+ if (rc)
+ return rc;
+ rc = ata_pci_sff_prepare_host(pdev, ppi, &host);
+ if (rc)
+ return rc;
+ pci_set_master(pdev);
+ return ata_pci_sff_activate_host(host, ata_sff_interrupt, &sch_sht);
+}
+
+static int __init sch_init(void)
+{
+ return pci_register_driver(&sch_pci_driver);
+}
+
+static void __exit sch_exit(void)
+{
+ pci_unregister_driver(&sch_pci_driver);
+}
+
+module_init(sch_init);
+module_exit(sch_exit);
* right. Documentation is available at initio's website but it only
* documents registers (not programming model).
*
- * - ATA disks work.
- * - Hotplug works.
- * - ATAPI read works but burning doesn't. This thing is really
- * peculiar about ATAPI and I couldn't figure out how ATAPI PIO and
- * ATAPI DMA WRITE should be programmed. If you've got a clue, be
- * my guest.
- * - Both STR and STD work.
+ * This driver has interesting history. The first version was written
+ * from the documentation and a 2.4 IDE driver posted on a Taiwan
+ * company, which didn't use any IDMA features and couldn't handle
+ * LBA48. The resulting driver couldn't handle LBA48 devices either
+ * making it pretty useless.
+ *
+ * After a while, initio picked the driver up, renamed it to
+ * sata_initio162x, updated it to use IDMA for ATA DMA commands and
+ * posted it on their website. It only used ATA_PROT_DMA for IDMA and
+ * attaching both devices and issuing IDMA and !IDMA commands
+ * simultaneously broke it due to PIRQ masking interaction but it did
+ * show how to use the IDMA (ADMA + some initio specific twists)
+ * engine.
+ *
+ * Then, I picked up their changes again and here's the usable driver
+ * which uses IDMA for everything. Everything works now including
+ * LBA48, CD/DVD burning, suspend/resume and hotplug. There are some
+ * issues tho. Result Tf is not resported properly, NCQ isn't
+ * supported yet and CD/DVD writing works with DMA assisted PIO
+ * protocol (which, for native SATA devices, shouldn't cause any
+ * noticeable difference).
+ *
+ * Anyways, so, here's finally a working driver for inic162x. Enjoy!
+ *
+ * initio: If you guys wanna improve the driver regarding result TF
+ * access and other stuff, please feel free to contact me. I'll be
+ * happy to assist.
*/
#include <linux/kernel.h>
#include <scsi/scsi_device.h>
#define DRV_NAME "sata_inic162x"
-#define DRV_VERSION "0.3"
+#define DRV_VERSION "0.4"
enum {
- MMIO_BAR = 5,
+ MMIO_BAR_PCI = 5,
+ MMIO_BAR_CARDBUS = 1,
NR_PORTS = 2,
+ IDMA_CPB_TBL_SIZE = 4 * 32,
+
+ INIC_DMA_BOUNDARY = 0xffffff,
+
+ HOST_ACTRL = 0x08,
HOST_CTL = 0x7c,
HOST_STAT = 0x7e,
HOST_IRQ_STAT = 0xbc,
PORT_SIZE = 0x40,
/* registers for ATA TF operation */
- PORT_TF = 0x00,
- PORT_ALT_STAT = 0x08,
+ PORT_TF_DATA = 0x00,
+ PORT_TF_FEATURE = 0x01,
+ PORT_TF_NSECT = 0x02,
+ PORT_TF_LBAL = 0x03,
+ PORT_TF_LBAM = 0x04,
+ PORT_TF_LBAH = 0x05,
+ PORT_TF_DEVICE = 0x06,
+ PORT_TF_COMMAND = 0x07,
+ PORT_TF_ALT_STAT = 0x08,
PORT_IRQ_STAT = 0x09,
PORT_IRQ_MASK = 0x0a,
PORT_PRD_CTL = 0x0b,
PORT_PRD_ADDR = 0x0c,
PORT_PRD_XFERLEN = 0x10,
+ PORT_CPB_CPBLAR = 0x18,
+ PORT_CPB_PTQFIFO = 0x1c,
/* IDMA register */
PORT_IDMA_CTL = 0x14,
+ PORT_IDMA_STAT = 0x16,
+
+ PORT_RPQ_FIFO = 0x1e,
+ PORT_RPQ_CNT = 0x1f,
PORT_SCR = 0x20,
/* HOST_CTL bits */
HCTL_IRQOFF = (1 << 8), /* global IRQ off */
- HCTL_PWRDWN = (1 << 13), /* power down PHYs */
+ HCTL_FTHD0 = (1 << 10), /* fifo threshold 0 */
+ HCTL_FTHD1 = (1 << 11), /* fifo threshold 1*/
+ HCTL_PWRDWN = (1 << 12), /* power down PHYs */
HCTL_SOFTRST = (1 << 13), /* global reset (no phy reset) */
HCTL_RPGSEL = (1 << 15), /* register page select */
PIRQ_PENDING = (1 << 7), /* port IRQ pending (STAT only) */
PIRQ_ERR = PIRQ_OFFLINE | PIRQ_ONLINE | PIRQ_FATAL,
-
- PIRQ_MASK_DMA_READ = PIRQ_REPLY | PIRQ_ATA,
- PIRQ_MASK_OTHER = PIRQ_REPLY | PIRQ_COMPLETE,
+ PIRQ_MASK_DEFAULT = PIRQ_REPLY | PIRQ_ATA,
PIRQ_MASK_FREEZE = 0xff,
/* PORT_PRD_CTL bits */
IDMA_CTL_RST_IDMA = (1 << 5), /* reset IDMA machinary */
IDMA_CTL_GO = (1 << 7), /* IDMA mode go */
IDMA_CTL_ATA_NIEN = (1 << 8), /* ATA IRQ disable */
+
+ /* PORT_IDMA_STAT bits */
+ IDMA_STAT_PERR = (1 << 0), /* PCI ERROR MODE */
+ IDMA_STAT_CPBERR = (1 << 1), /* ADMA CPB error */
+ IDMA_STAT_LGCY = (1 << 3), /* ADMA legacy */
+ IDMA_STAT_UIRQ = (1 << 4), /* ADMA unsolicited irq */
+ IDMA_STAT_STPD = (1 << 5), /* ADMA stopped */
+ IDMA_STAT_PSD = (1 << 6), /* ADMA pause */
+ IDMA_STAT_DONE = (1 << 7), /* ADMA done */
+
+ IDMA_STAT_ERR = IDMA_STAT_PERR | IDMA_STAT_CPBERR,
+
+ /* CPB Control Flags*/
+ CPB_CTL_VALID = (1 << 0), /* CPB valid */
+ CPB_CTL_QUEUED = (1 << 1), /* queued command */
+ CPB_CTL_DATA = (1 << 2), /* data, rsvd in datasheet */
+ CPB_CTL_IEN = (1 << 3), /* PCI interrupt enable */
+ CPB_CTL_DEVDIR = (1 << 4), /* device direction control */
+
+ /* CPB Response Flags */
+ CPB_RESP_DONE = (1 << 0), /* ATA command complete */
+ CPB_RESP_REL = (1 << 1), /* ATA release */
+ CPB_RESP_IGNORED = (1 << 2), /* CPB ignored */
+ CPB_RESP_ATA_ERR = (1 << 3), /* ATA command error */
+ CPB_RESP_SPURIOUS = (1 << 4), /* ATA spurious interrupt error */
+ CPB_RESP_UNDERFLOW = (1 << 5), /* APRD deficiency length error */
+ CPB_RESP_OVERFLOW = (1 << 6), /* APRD exccess length error */
+ CPB_RESP_CPB_ERR = (1 << 7), /* CPB error flag */
+
+ /* PRD Control Flags */
+ PRD_DRAIN = (1 << 1), /* ignore data excess */
+ PRD_CDB = (1 << 2), /* atapi packet command pointer */
+ PRD_DIRECT_INTR = (1 << 3), /* direct interrupt */
+ PRD_DMA = (1 << 4), /* data transfer method */
+ PRD_WRITE = (1 << 5), /* data dir, rsvd in datasheet */
+ PRD_IOM = (1 << 6), /* io/memory transfer */
+ PRD_END = (1 << 7), /* APRD chain end */
};
+/* Comman Parameter Block */
+struct inic_cpb {
+ u8 resp_flags; /* Response Flags */
+ u8 error; /* ATA Error */
+ u8 status; /* ATA Status */
+ u8 ctl_flags; /* Control Flags */
+ __le32 len; /* Total Transfer Length */
+ __le32 prd; /* First PRD pointer */
+ u8 rsvd[4];
+ /* 16 bytes */
+ u8 feature; /* ATA Feature */
+ u8 hob_feature; /* ATA Ex. Feature */
+ u8 device; /* ATA Device/Head */
+ u8 mirctl; /* Mirror Control */
+ u8 nsect; /* ATA Sector Count */
+ u8 hob_nsect; /* ATA Ex. Sector Count */
+ u8 lbal; /* ATA Sector Number */
+ u8 hob_lbal; /* ATA Ex. Sector Number */
+ u8 lbam; /* ATA Cylinder Low */
+ u8 hob_lbam; /* ATA Ex. Cylinder Low */
+ u8 lbah; /* ATA Cylinder High */
+ u8 hob_lbah; /* ATA Ex. Cylinder High */
+ u8 command; /* ATA Command */
+ u8 ctl; /* ATA Control */
+ u8 slave_error; /* Slave ATA Error */
+ u8 slave_status; /* Slave ATA Status */
+ /* 32 bytes */
+} __packed;
+
+/* Physical Region Descriptor */
+struct inic_prd {
+ __le32 mad; /* Physical Memory Address */
+ __le16 len; /* Transfer Length */
+ u8 rsvd;
+ u8 flags; /* Control Flags */
+} __packed;
+
+struct inic_pkt {
+ struct inic_cpb cpb;
+ struct inic_prd prd[LIBATA_MAX_PRD + 1]; /* + 1 for cdb */
+ u8 cdb[ATAPI_CDB_LEN];
+} __packed;
+
struct inic_host_priv {
- u16 cached_hctl;
+ void __iomem *mmio_base;
+ u16 cached_hctl;
};
struct inic_port_priv {
- u8 dfl_prdctl;
- u8 cached_prdctl;
- u8 cached_pirq_mask;
+ struct inic_pkt *pkt;
+ dma_addr_t pkt_dma;
+ u32 *cpb_tbl;
+ dma_addr_t cpb_tbl_dma;
};
static struct scsi_host_template inic_sht = {
- ATA_BMDMA_SHT(DRV_NAME),
+ ATA_BASE_SHT(DRV_NAME),
+ .sg_tablesize = LIBATA_MAX_PRD, /* maybe it can be larger? */
+ .dma_boundary = INIC_DMA_BOUNDARY,
};
static const int scr_map[] = {
static void __iomem *inic_port_base(struct ata_port *ap)
{
- return ap->host->iomap[MMIO_BAR] + ap->port_no * PORT_SIZE;
-}
-
-static void __inic_set_pirq_mask(struct ata_port *ap, u8 mask)
-{
- void __iomem *port_base = inic_port_base(ap);
- struct inic_port_priv *pp = ap->private_data;
+ struct inic_host_priv *hpriv = ap->host->private_data;
- writeb(mask, port_base + PORT_IRQ_MASK);
- pp->cached_pirq_mask = mask;
-}
-
-static void inic_set_pirq_mask(struct ata_port *ap, u8 mask)
-{
- struct inic_port_priv *pp = ap->private_data;
-
- if (pp->cached_pirq_mask != mask)
- __inic_set_pirq_mask(ap, mask);
+ return hpriv->mmio_base + ap->port_no * PORT_SIZE;
}
static void inic_reset_port(void __iomem *port_base)
{
void __iomem *idma_ctl = port_base + PORT_IDMA_CTL;
- u16 ctl;
- ctl = readw(idma_ctl);
- ctl &= ~(IDMA_CTL_RST_IDMA | IDMA_CTL_ATA_NIEN | IDMA_CTL_GO);
+ /* stop IDMA engine */
+ readw(idma_ctl); /* flush */
+ msleep(1);
/* mask IRQ and assert reset */
- writew(ctl | IDMA_CTL_RST_IDMA | IDMA_CTL_ATA_NIEN, idma_ctl);
+ writew(IDMA_CTL_RST_IDMA, idma_ctl);
readw(idma_ctl); /* flush */
-
- /* give it some time */
msleep(1);
/* release reset */
- writew(ctl | IDMA_CTL_ATA_NIEN, idma_ctl);
+ writew(0, idma_ctl);
/* clear irq */
writeb(0xff, port_base + PORT_IRQ_STAT);
-
- /* reenable ATA IRQ, turn off IDMA mode */
- writew(ctl, idma_ctl);
}
static int inic_scr_read(struct ata_port *ap, unsigned sc_reg, u32 *val)
{
- void __iomem *scr_addr = ap->ioaddr.scr_addr;
+ void __iomem *scr_addr = inic_port_base(ap) + PORT_SCR;
void __iomem *addr;
if (unlikely(sc_reg >= ARRAY_SIZE(scr_map)))
static int inic_scr_write(struct ata_port *ap, unsigned sc_reg, u32 val)
{
- void __iomem *scr_addr = ap->ioaddr.scr_addr;
- void __iomem *addr;
+ void __iomem *scr_addr = inic_port_base(ap) + PORT_SCR;
if (unlikely(sc_reg >= ARRAY_SIZE(scr_map)))
return -EINVAL;
- addr = scr_addr + scr_map[sc_reg] * 4;
writel(val, scr_addr + scr_map[sc_reg] * 4);
return 0;
}
-/*
- * In TF mode, inic162x is very similar to SFF device. TF registers
- * function the same. DMA engine behaves similary using the same PRD
- * format as BMDMA but different command register, interrupt and event
- * notification methods are used. The following inic_bmdma_*()
- * functions do the impedance matching.
- */
-static void inic_bmdma_setup(struct ata_queued_cmd *qc)
+static void inic_stop_idma(struct ata_port *ap)
{
- struct ata_port *ap = qc->ap;
- struct inic_port_priv *pp = ap->private_data;
void __iomem *port_base = inic_port_base(ap);
- int rw = qc->tf.flags & ATA_TFLAG_WRITE;
-
- /* make sure device sees PRD table writes */
- wmb();
-
- /* load transfer length */
- writel(qc->nbytes, port_base + PORT_PRD_XFERLEN);
-
- /* turn on DMA and specify data direction */
- pp->cached_prdctl = pp->dfl_prdctl | PRD_CTL_DMAEN;
- if (!rw)
- pp->cached_prdctl |= PRD_CTL_WR;
- writeb(pp->cached_prdctl, port_base + PORT_PRD_CTL);
- /* issue r/w command */
- ap->ops->sff_exec_command(ap, &qc->tf);
+ readb(port_base + PORT_RPQ_FIFO);
+ readb(port_base + PORT_RPQ_CNT);
+ writew(0, port_base + PORT_IDMA_CTL);
}
-static void inic_bmdma_start(struct ata_queued_cmd *qc)
+static void inic_host_err_intr(struct ata_port *ap, u8 irq_stat, u16 idma_stat)
{
- struct ata_port *ap = qc->ap;
+ struct ata_eh_info *ehi = &ap->link.eh_info;
struct inic_port_priv *pp = ap->private_data;
- void __iomem *port_base = inic_port_base(ap);
+ struct inic_cpb *cpb = &pp->pkt->cpb;
+ bool freeze = false;
- /* start host DMA transaction */
- pp->cached_prdctl |= PRD_CTL_START;
- writeb(pp->cached_prdctl, port_base + PORT_PRD_CTL);
-}
+ ata_ehi_clear_desc(ehi);
+ ata_ehi_push_desc(ehi, "irq_stat=0x%x idma_stat=0x%x",
+ irq_stat, idma_stat);
-static void inic_bmdma_stop(struct ata_queued_cmd *qc)
-{
- struct ata_port *ap = qc->ap;
- struct inic_port_priv *pp = ap->private_data;
- void __iomem *port_base = inic_port_base(ap);
+ inic_stop_idma(ap);
- /* stop DMA engine */
- writeb(pp->dfl_prdctl, port_base + PORT_PRD_CTL);
-}
+ if (irq_stat & (PIRQ_OFFLINE | PIRQ_ONLINE)) {
+ ata_ehi_push_desc(ehi, "hotplug");
+ ata_ehi_hotplugged(ehi);
+ freeze = true;
+ }
-static u8 inic_bmdma_status(struct ata_port *ap)
-{
- /* event is already verified by the interrupt handler */
- return ATA_DMA_INTR;
+ if (idma_stat & IDMA_STAT_PERR) {
+ ata_ehi_push_desc(ehi, "PCI error");
+ freeze = true;
+ }
+
+ if (idma_stat & IDMA_STAT_CPBERR) {
+ ata_ehi_push_desc(ehi, "CPB error");
+
+ if (cpb->resp_flags & CPB_RESP_IGNORED) {
+ __ata_ehi_push_desc(ehi, " ignored");
+ ehi->err_mask |= AC_ERR_INVALID;
+ freeze = true;
+ }
+
+ if (cpb->resp_flags & CPB_RESP_ATA_ERR)
+ ehi->err_mask |= AC_ERR_DEV;
+
+ if (cpb->resp_flags & CPB_RESP_SPURIOUS) {
+ __ata_ehi_push_desc(ehi, " spurious-intr");
+ ehi->err_mask |= AC_ERR_HSM;
+ freeze = true;
+ }
+
+ if (cpb->resp_flags &
+ (CPB_RESP_UNDERFLOW | CPB_RESP_OVERFLOW)) {
+ __ata_ehi_push_desc(ehi, " data-over/underflow");
+ ehi->err_mask |= AC_ERR_HSM;
+ freeze = true;
+ }
+ }
+
+ if (freeze)
+ ata_port_freeze(ap);
+ else
+ ata_port_abort(ap);
}
static void inic_host_intr(struct ata_port *ap)
{
void __iomem *port_base = inic_port_base(ap);
- struct ata_eh_info *ehi = &ap->link.eh_info;
+ struct ata_queued_cmd *qc = ata_qc_from_tag(ap, ap->link.active_tag);
u8 irq_stat;
+ u16 idma_stat;
- /* fetch and clear irq */
+ /* read and clear IRQ status */
irq_stat = readb(port_base + PORT_IRQ_STAT);
writeb(irq_stat, port_base + PORT_IRQ_STAT);
+ idma_stat = readw(port_base + PORT_IDMA_STAT);
- if (likely(!(irq_stat & PIRQ_ERR))) {
- struct ata_queued_cmd *qc =
- ata_qc_from_tag(ap, ap->link.active_tag);
+ if (unlikely((irq_stat & PIRQ_ERR) || (idma_stat & IDMA_STAT_ERR)))
+ inic_host_err_intr(ap, irq_stat, idma_stat);
- if (unlikely(!qc || (qc->tf.flags & ATA_TFLAG_POLLING))) {
- ap->ops->sff_check_status(ap); /* clear ATA interrupt */
- return;
- }
+ if (unlikely(!qc))
+ goto spurious;
- if (likely(ata_sff_host_intr(ap, qc)))
- return;
+ if (likely(idma_stat & IDMA_STAT_DONE)) {
+ inic_stop_idma(ap);
- ap->ops->sff_check_status(ap); /* clear ATA interrupt */
- ata_port_printk(ap, KERN_WARNING, "unhandled "
- "interrupt, irq_stat=%x\n", irq_stat);
+ /* Depending on circumstances, device error
+ * isn't reported by IDMA, check it explicitly.
+ */
+ if (unlikely(readb(port_base + PORT_TF_COMMAND) &
+ (ATA_DF | ATA_ERR)))
+ qc->err_mask |= AC_ERR_DEV;
+
+ ata_qc_complete(qc);
return;
}
- /* error */
- ata_ehi_push_desc(ehi, "irq_stat=0x%x", irq_stat);
-
- if (irq_stat & (PIRQ_OFFLINE | PIRQ_ONLINE)) {
- ata_ehi_hotplugged(ehi);
- ata_port_freeze(ap);
- } else
- ata_port_abort(ap);
+ spurious:
+ ata_port_printk(ap, KERN_WARNING, "unhandled interrupt: "
+ "cmd=0x%x irq_stat=0x%x idma_stat=0x%x\n",
+ qc ? qc->tf.command : 0xff, irq_stat, idma_stat);
}
static irqreturn_t inic_interrupt(int irq, void *dev_instance)
{
struct ata_host *host = dev_instance;
- void __iomem *mmio_base = host->iomap[MMIO_BAR];
+ struct inic_host_priv *hpriv = host->private_data;
u16 host_irq_stat;
int i, handled = 0;;
- host_irq_stat = readw(mmio_base + HOST_IRQ_STAT);
+ host_irq_stat = readw(hpriv->mmio_base + HOST_IRQ_STAT);
if (unlikely(!(host_irq_stat & HIRQ_GLOBAL)))
goto out;
return IRQ_RETVAL(handled);
}
+static int inic_check_atapi_dma(struct ata_queued_cmd *qc)
+{
+ /* For some reason ATAPI_PROT_DMA doesn't work for some
+ * commands including writes and other misc ops. Use PIO
+ * protocol instead, which BTW is driven by the DMA engine
+ * anyway, so it shouldn't make much difference for native
+ * SATA devices.
+ */
+ if (atapi_cmd_type(qc->cdb[0]) == READ)
+ return 0;
+ return 1;
+}
+
+static void inic_fill_sg(struct inic_prd *prd, struct ata_queued_cmd *qc)
+{
+ struct scatterlist *sg;
+ unsigned int si;
+ u8 flags = 0;
+
+ if (qc->tf.flags & ATA_TFLAG_WRITE)
+ flags |= PRD_WRITE;
+
+ if (ata_is_dma(qc->tf.protocol))
+ flags |= PRD_DMA;
+
+ for_each_sg(qc->sg, sg, qc->n_elem, si) {
+ prd->mad = cpu_to_le32(sg_dma_address(sg));
+ prd->len = cpu_to_le16(sg_dma_len(sg));
+ prd->flags = flags;
+ prd++;
+ }
+
+ WARN_ON(!si);
+ prd[-1].flags |= PRD_END;
+}
+
+static void inic_qc_prep(struct ata_queued_cmd *qc)
+{
+ struct inic_port_priv *pp = qc->ap->private_data;
+ struct inic_pkt *pkt = pp->pkt;
+ struct inic_cpb *cpb = &pkt->cpb;
+ struct inic_prd *prd = pkt->prd;
+ bool is_atapi = ata_is_atapi(qc->tf.protocol);
+ bool is_data = ata_is_data(qc->tf.protocol);
+ unsigned int cdb_len = 0;
+
+ VPRINTK("ENTER\n");
+
+ if (is_atapi)
+ cdb_len = qc->dev->cdb_len;
+
+ /* prepare packet, based on initio driver */
+ memset(pkt, 0, sizeof(struct inic_pkt));
+
+ cpb->ctl_flags = CPB_CTL_VALID | CPB_CTL_IEN;
+ if (is_atapi || is_data)
+ cpb->ctl_flags |= CPB_CTL_DATA;
+
+ cpb->len = cpu_to_le32(qc->nbytes + cdb_len);
+ cpb->prd = cpu_to_le32(pp->pkt_dma + offsetof(struct inic_pkt, prd));
+
+ cpb->device = qc->tf.device;
+ cpb->feature = qc->tf.feature;
+ cpb->nsect = qc->tf.nsect;
+ cpb->lbal = qc->tf.lbal;
+ cpb->lbam = qc->tf.lbam;
+ cpb->lbah = qc->tf.lbah;
+
+ if (qc->tf.flags & ATA_TFLAG_LBA48) {
+ cpb->hob_feature = qc->tf.hob_feature;
+ cpb->hob_nsect = qc->tf.hob_nsect;
+ cpb->hob_lbal = qc->tf.hob_lbal;
+ cpb->hob_lbam = qc->tf.hob_lbam;
+ cpb->hob_lbah = qc->tf.hob_lbah;
+ }
+
+ cpb->command = qc->tf.command;
+ /* don't load ctl - dunno why. it's like that in the initio driver */
+
+ /* setup PRD for CDB */
+ if (is_atapi) {
+ memcpy(pkt->cdb, qc->cdb, ATAPI_CDB_LEN);
+ prd->mad = cpu_to_le32(pp->pkt_dma +
+ offsetof(struct inic_pkt, cdb));
+ prd->len = cpu_to_le16(cdb_len);
+ prd->flags = PRD_CDB | PRD_WRITE;
+ if (!is_data)
+ prd->flags |= PRD_END;
+ prd++;
+ }
+
+ /* setup sg table */
+ if (is_data)
+ inic_fill_sg(prd, qc);
+
+ pp->cpb_tbl[0] = pp->pkt_dma;
+}
+
static unsigned int inic_qc_issue(struct ata_queued_cmd *qc)
{
struct ata_port *ap = qc->ap;
+ void __iomem *port_base = inic_port_base(ap);
- /* ATA IRQ doesn't wait for DMA transfer completion and vice
- * versa. Mask IRQ selectively to detect command completion.
- * Without it, ATA DMA read command can cause data corruption.
- *
- * Something similar might be needed for ATAPI writes. I
- * tried a lot of combinations but couldn't find the solution.
- */
- if (qc->tf.protocol == ATA_PROT_DMA &&
- !(qc->tf.flags & ATA_TFLAG_WRITE))
- inic_set_pirq_mask(ap, PIRQ_MASK_DMA_READ);
- else
- inic_set_pirq_mask(ap, PIRQ_MASK_OTHER);
+ /* fire up the ADMA engine */
+ writew(HCTL_FTHD0, port_base + HOST_CTL);
+ writew(IDMA_CTL_GO, port_base + PORT_IDMA_CTL);
+ writeb(0, port_base + PORT_CPB_PTQFIFO);
+
+ return 0;
+}
+
+static void inic_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
+{
+ void __iomem *port_base = inic_port_base(ap);
+
+ tf->feature = readb(port_base + PORT_TF_FEATURE);
+ tf->nsect = readb(port_base + PORT_TF_NSECT);
+ tf->lbal = readb(port_base + PORT_TF_LBAL);
+ tf->lbam = readb(port_base + PORT_TF_LBAM);
+ tf->lbah = readb(port_base + PORT_TF_LBAH);
+ tf->device = readb(port_base + PORT_TF_DEVICE);
+ tf->command = readb(port_base + PORT_TF_COMMAND);
+}
- /* Issuing a command to yet uninitialized port locks up the
- * controller. Most of the time, this happens for the first
- * command after reset which are ATA and ATAPI IDENTIFYs.
- * Fast fail if stat is 0x7f or 0xff for those commands.
+static bool inic_qc_fill_rtf(struct ata_queued_cmd *qc)
+{
+ struct ata_taskfile *rtf = &qc->result_tf;
+ struct ata_taskfile tf;
+
+ /* FIXME: Except for status and error, result TF access
+ * doesn't work. I tried reading from BAR0/2, CPB and BAR5.
+ * None works regardless of which command interface is used.
+ * For now return true iff status indicates device error.
+ * This means that we're reporting bogus sector for RW
+ * failures. Eeekk....
*/
- if (unlikely(qc->tf.command == ATA_CMD_ID_ATA ||
- qc->tf.command == ATA_CMD_ID_ATAPI)) {
- u8 stat = ap->ops->sff_check_status(ap);
- if (stat == 0x7f || stat == 0xff)
- return AC_ERR_HSM;
- }
+ inic_tf_read(qc->ap, &tf);
- return ata_sff_qc_issue(qc);
+ if (!(tf.command & ATA_ERR))
+ return false;
+
+ rtf->command = tf.command;
+ rtf->feature = tf.feature;
+ return true;
}
static void inic_freeze(struct ata_port *ap)
{
void __iomem *port_base = inic_port_base(ap);
- __inic_set_pirq_mask(ap, PIRQ_MASK_FREEZE);
-
- ap->ops->sff_check_status(ap);
+ writeb(PIRQ_MASK_FREEZE, port_base + PORT_IRQ_MASK);
writeb(0xff, port_base + PORT_IRQ_STAT);
-
- readb(port_base + PORT_IRQ_STAT); /* flush */
}
static void inic_thaw(struct ata_port *ap)
{
void __iomem *port_base = inic_port_base(ap);
- ap->ops->sff_check_status(ap);
writeb(0xff, port_base + PORT_IRQ_STAT);
+ writeb(PIRQ_MASK_DEFAULT, port_base + PORT_IRQ_MASK);
+}
- __inic_set_pirq_mask(ap, PIRQ_MASK_OTHER);
+static int inic_check_ready(struct ata_link *link)
+{
+ void __iomem *port_base = inic_port_base(link->ap);
- readb(port_base + PORT_IRQ_STAT); /* flush */
+ return ata_check_ready(readb(port_base + PORT_TF_COMMAND));
}
/*
void __iomem *port_base = inic_port_base(ap);
void __iomem *idma_ctl = port_base + PORT_IDMA_CTL;
const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context);
- u16 val;
int rc;
/* hammer it into sane state */
inic_reset_port(port_base);
- val = readw(idma_ctl);
- writew(val | IDMA_CTL_RST_ATA, idma_ctl);
+ writew(IDMA_CTL_RST_ATA, idma_ctl);
readw(idma_ctl); /* flush */
msleep(1);
- writew(val & ~IDMA_CTL_RST_ATA, idma_ctl);
+ writew(0, idma_ctl);
rc = sata_link_resume(link, timing, deadline);
if (rc) {
struct ata_taskfile tf;
/* wait for link to become ready */
- rc = ata_sff_wait_after_reset(link, 1, deadline);
+ rc = ata_wait_after_reset(link, deadline, inic_check_ready);
/* link occupied, -ENODEV too is an error */
if (rc) {
ata_link_printk(link, KERN_WARNING, "device not ready "
return rc;
}
- ata_sff_tf_read(ap, &tf);
+ inic_tf_read(ap, &tf);
*class = ata_dev_classify(&tf);
}
static void inic_error_handler(struct ata_port *ap)
{
void __iomem *port_base = inic_port_base(ap);
- struct inic_port_priv *pp = ap->private_data;
- unsigned long flags;
- /* reset PIO HSM and stop DMA engine */
inic_reset_port(port_base);
-
- spin_lock_irqsave(ap->lock, flags);
- ap->hsm_task_state = HSM_ST_IDLE;
- writeb(pp->dfl_prdctl, port_base + PORT_PRD_CTL);
- spin_unlock_irqrestore(ap->lock, flags);
-
- /* PIO and DMA engines have been stopped, perform recovery */
ata_std_error_handler(ap);
}
inic_reset_port(inic_port_base(qc->ap));
}
-static void inic_dev_config(struct ata_device *dev)
-{
- /* inic can only handle upto LBA28 max sectors */
- if (dev->max_sectors > ATA_MAX_SECTORS)
- dev->max_sectors = ATA_MAX_SECTORS;
-
- if (dev->n_sectors >= 1 << 28) {
- ata_dev_printk(dev, KERN_ERR,
- "ERROR: This driver doesn't support LBA48 yet and may cause\n"
- " data corruption on such devices. Disabling.\n");
- ata_dev_disable(dev);
- }
-}
-
static void init_port(struct ata_port *ap)
{
void __iomem *port_base = inic_port_base(ap);
+ struct inic_port_priv *pp = ap->private_data;
- /* Setup PRD address */
+ /* clear packet and CPB table */
+ memset(pp->pkt, 0, sizeof(struct inic_pkt));
+ memset(pp->cpb_tbl, 0, IDMA_CPB_TBL_SIZE);
+
+ /* setup PRD and CPB lookup table addresses */
writel(ap->prd_dma, port_base + PORT_PRD_ADDR);
+ writel(pp->cpb_tbl_dma, port_base + PORT_CPB_CPBLAR);
}
static int inic_port_resume(struct ata_port *ap)
static int inic_port_start(struct ata_port *ap)
{
- void __iomem *port_base = inic_port_base(ap);
+ struct device *dev = ap->host->dev;
struct inic_port_priv *pp;
- u8 tmp;
int rc;
/* alloc and initialize private data */
- pp = devm_kzalloc(ap->host->dev, sizeof(*pp), GFP_KERNEL);
+ pp = devm_kzalloc(dev, sizeof(*pp), GFP_KERNEL);
if (!pp)
return -ENOMEM;
ap->private_data = pp;
- /* default PRD_CTL value, DMAEN, WR and START off */
- tmp = readb(port_base + PORT_PRD_CTL);
- tmp &= ~(PRD_CTL_DMAEN | PRD_CTL_WR | PRD_CTL_START);
- pp->dfl_prdctl = tmp;
-
/* Alloc resources */
rc = ata_port_start(ap);
- if (rc) {
- kfree(pp);
+ if (rc)
return rc;
- }
+
+ pp->pkt = dmam_alloc_coherent(dev, sizeof(struct inic_pkt),
+ &pp->pkt_dma, GFP_KERNEL);
+ if (!pp->pkt)
+ return -ENOMEM;
+
+ pp->cpb_tbl = dmam_alloc_coherent(dev, IDMA_CPB_TBL_SIZE,
+ &pp->cpb_tbl_dma, GFP_KERNEL);
+ if (!pp->cpb_tbl)
+ return -ENOMEM;
init_port(ap);
}
static struct ata_port_operations inic_port_ops = {
- .inherits = &ata_sff_port_ops,
+ .inherits = &sata_port_ops,
- .bmdma_setup = inic_bmdma_setup,
- .bmdma_start = inic_bmdma_start,
- .bmdma_stop = inic_bmdma_stop,
- .bmdma_status = inic_bmdma_status,
+ .check_atapi_dma = inic_check_atapi_dma,
+ .qc_prep = inic_qc_prep,
.qc_issue = inic_qc_issue,
+ .qc_fill_rtf = inic_qc_fill_rtf,
.freeze = inic_freeze,
.thaw = inic_thaw,
- .softreset = ATA_OP_NULL, /* softreset is broken */
.hardreset = inic_hardreset,
.error_handler = inic_error_handler,
.post_internal_cmd = inic_post_internal_cmd,
- .dev_config = inic_dev_config,
.scr_read = inic_scr_read,
.scr_write = inic_scr_write,
};
static struct ata_port_info inic_port_info = {
- /* For some reason, ATAPI_PROT_PIO is broken on this
- * controller, and no, PIO_POLLING does't fix it. It somehow
- * manages to report the wrong ireason and ignoring ireason
- * results in machine lock up. Tell libata to always prefer
- * DMA.
- */
.flags = ATA_FLAG_SATA | ATA_FLAG_PIO_DMA,
.pio_mask = 0x1f, /* pio0-4 */
.mwdma_mask = 0x07, /* mwdma0-2 */
{
struct ata_host *host = dev_get_drvdata(&pdev->dev);
struct inic_host_priv *hpriv = host->private_data;
- void __iomem *mmio_base = host->iomap[MMIO_BAR];
int rc;
rc = ata_pci_device_do_resume(pdev);
return rc;
if (pdev->dev.power.power_state.event == PM_EVENT_SUSPEND) {
- rc = init_controller(mmio_base, hpriv->cached_hctl);
+ rc = init_controller(hpriv->mmio_base, hpriv->cached_hctl);
if (rc)
return rc;
}
struct ata_host *host;
struct inic_host_priv *hpriv;
void __iomem * const *iomap;
+ int mmio_bar;
int i, rc;
if (!printed_version++)
host->private_data = hpriv;
- /* acquire resources and fill host */
+ /* Acquire resources and fill host. Note that PCI and cardbus
+ * use different BARs.
+ */
rc = pcim_enable_device(pdev);
if (rc)
return rc;
- rc = pcim_iomap_regions(pdev, 0x3f, DRV_NAME);
+ if (pci_resource_flags(pdev, MMIO_BAR_PCI) & IORESOURCE_MEM)
+ mmio_bar = MMIO_BAR_PCI;
+ else
+ mmio_bar = MMIO_BAR_CARDBUS;
+
+ rc = pcim_iomap_regions(pdev, 1 << mmio_bar, DRV_NAME);
if (rc)
return rc;
host->iomap = iomap = pcim_iomap_table(pdev);
+ hpriv->mmio_base = iomap[mmio_bar];
+ hpriv->cached_hctl = readw(hpriv->mmio_base + HOST_CTL);
for (i = 0; i < NR_PORTS; i++) {
struct ata_port *ap = host->ports[i];
- struct ata_ioports *port = &ap->ioaddr;
- unsigned int offset = i * PORT_SIZE;
-
- port->cmd_addr = iomap[2 * i];
- port->altstatus_addr =
- port->ctl_addr = (void __iomem *)
- ((unsigned long)iomap[2 * i + 1] | ATA_PCI_CTL_OFS);
- port->scr_addr = iomap[MMIO_BAR] + offset + PORT_SCR;
-
- ata_sff_std_ports(port);
-
- ata_port_pbar_desc(ap, MMIO_BAR, -1, "mmio");
- ata_port_pbar_desc(ap, MMIO_BAR, offset, "port");
- ata_port_desc(ap, "cmd 0x%llx ctl 0x%llx",
- (unsigned long long)pci_resource_start(pdev, 2 * i),
- (unsigned long long)pci_resource_start(pdev, (2 * i + 1)) |
- ATA_PCI_CTL_OFS);
- }
- hpriv->cached_hctl = readw(iomap[MMIO_BAR] + HOST_CTL);
+ ata_port_pbar_desc(ap, mmio_bar, -1, "mmio");
+ ata_port_pbar_desc(ap, mmio_bar, i * PORT_SIZE, "port");
+ }
/* Set dma_mask. This devices doesn't support 64bit addressing. */
rc = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
return rc;
}
- rc = init_controller(iomap[MMIO_BAR], hpriv->cached_hctl);
+ rc = init_controller(hpriv->mmio_base, hpriv->cached_hctl);
if (rc) {
dev_printk(KERN_ERR, &pdev->dev,
"failed to initialize controller\n");
#include <linux/platform_device.h>
#include <linux/ata_platform.h>
#include <linux/mbus.h>
+#include <linux/bitops.h>
#include <scsi/scsi_host.h>
#include <scsi/scsi_cmnd.h>
#include <scsi/scsi_device.h>
MV_IRQ_COAL_TIME_THRESHOLD = (MV_IRQ_COAL_REG_BASE + 0xd0),
MV_SATAHC0_REG_BASE = 0x20000,
- MV_FLASH_CTL = 0x1046c,
- MV_GPIO_PORT_CTL = 0x104f0,
- MV_RESET_CFG = 0x180d8,
+ MV_FLASH_CTL_OFS = 0x1046c,
+ MV_GPIO_PORT_CTL_OFS = 0x104f0,
+ MV_RESET_CFG_OFS = 0x180d8,
MV_PCI_REG_SZ = MV_MAJOR_REG_AREA_SZ,
MV_SATAHC_REG_SZ = MV_MAJOR_REG_AREA_SZ,
/* PCI interface registers */
PCI_COMMAND_OFS = 0xc00,
+ PCI_COMMAND_MRDTRIG = (1 << 7), /* PCI Master Read Trigger */
PCI_MAIN_CMD_STS_OFS = 0xd30,
STOP_PCI_MASTER = (1 << 2),
PCI_MASTER_EMPTY = (1 << 3),
GLOB_SFT_RST = (1 << 4),
- MV_PCI_MODE = 0xd00,
+ MV_PCI_MODE_OFS = 0xd00,
+ MV_PCI_MODE_MASK = 0x30,
+
MV_PCI_EXP_ROM_BAR_CTL = 0xd2c,
MV_PCI_DISC_TIMER = 0xd04,
MV_PCI_MSI_TRIGGER = 0xc38,
MV_PCI_SERR_MASK = 0xc28,
- MV_PCI_XBAR_TMOUT = 0x1d04,
+ MV_PCI_XBAR_TMOUT_OFS = 0x1d04,
MV_PCI_ERR_LOW_ADDRESS = 0x1d40,
MV_PCI_ERR_HIGH_ADDRESS = 0x1d44,
MV_PCI_ERR_ATTRIBUTE = 0x1d48,
PHY_MODE4 = 0x314,
PHY_MODE2 = 0x330,
SATA_IFCTL_OFS = 0x344,
+ SATA_TESTCTL_OFS = 0x348,
SATA_IFSTAT_OFS = 0x34c,
VENDOR_UNIQUE_FIS_OFS = 0x35c,
- FIS_CFG_OFS = 0x360,
- FIS_CFG_SINGLE_SYNC = (1 << 16), /* SYNC on DMA activation */
+ FISCFG_OFS = 0x360,
+ FISCFG_WAIT_DEV_ERR = (1 << 8), /* wait for host on DevErr */
+ FISCFG_SINGLE_SYNC = (1 << 16), /* SYNC on DMA activation */
MV5_PHY_MODE = 0x74,
- MV5_LT_MODE = 0x30,
- MV5_PHY_CTL = 0x0C,
- SATA_INTERFACE_CFG = 0x050,
+ MV5_LTMODE_OFS = 0x30,
+ MV5_PHY_CTL_OFS = 0x0C,
+ SATA_INTERFACE_CFG_OFS = 0x050,
MV_M2_PREAMP_MASK = 0x7e0,
EDMA_CMD_OFS = 0x28, /* EDMA command register */
EDMA_EN = (1 << 0), /* enable EDMA */
EDMA_DS = (1 << 1), /* disable EDMA; self-negated */
- ATA_RST = (1 << 2), /* reset trans/link/phy */
+ EDMA_RESET = (1 << 2), /* reset eng/trans/link/phy */
+
+ EDMA_STATUS_OFS = 0x30, /* EDMA engine status */
+ EDMA_STATUS_CACHE_EMPTY = (1 << 6), /* GenIIe command cache empty */
+ EDMA_STATUS_IDLE = (1 << 7), /* GenIIe EDMA enabled/idle */
- EDMA_IORDY_TMOUT = 0x34,
- EDMA_ARB_CFG = 0x38,
+ EDMA_IORDY_TMOUT_OFS = 0x34,
+ EDMA_ARB_CFG_OFS = 0x38,
+
+ EDMA_HALTCOND_OFS = 0x60, /* GenIIe halt conditions */
GEN_II_NCQ_MAX_SECTORS = 256, /* max sects/io on Gen2 w/NCQ */
MV_HP_GEN_II = (1 << 7), /* Generation II: 60xx */
MV_HP_GEN_IIE = (1 << 8), /* Generation IIE: 6042/7042 */
MV_HP_PCIE = (1 << 9), /* PCIe bus/regs: 7042 */
+ MV_HP_CUT_THROUGH = (1 << 10), /* can use EDMA cut-through */
/* Port private flags (pp_flags) */
MV_PP_FLAG_EDMA_EN = (1 << 0), /* is EDMA engine enabled? */
MV_PP_FLAG_NCQ_EN = (1 << 1), /* is EDMA set up for NCQ? */
+ MV_PP_FLAG_FBS_EN = (1 << 2), /* is EDMA set up for FBS? */
+ MV_PP_FLAG_DELAYED_EH = (1 << 3), /* delayed dev err handling */
};
#define IS_GEN_I(hpriv) ((hpriv)->hp_flags & MV_HP_GEN_I)
#define IS_GEN_II(hpriv) ((hpriv)->hp_flags & MV_HP_GEN_II)
#define IS_GEN_IIE(hpriv) ((hpriv)->hp_flags & MV_HP_GEN_IIE)
+#define IS_PCIE(hpriv) ((hpriv)->hp_flags & MV_HP_PCIE)
#define HAS_PCI(host) (!((host)->ports[0]->flags & MV_FLAG_SOC))
#define WINDOW_CTRL(i) (0x20030 + ((i) << 4))
unsigned int resp_idx;
u32 pp_flags;
+ unsigned int delayed_eh_pmp_map;
};
struct mv_port_signal {
static int mv5_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val);
static int mv_port_start(struct ata_port *ap);
static void mv_port_stop(struct ata_port *ap);
+static int mv_qc_defer(struct ata_queued_cmd *qc);
static void mv_qc_prep(struct ata_queued_cmd *qc);
static void mv_qc_prep_iie(struct ata_queued_cmd *qc);
static unsigned int mv_qc_issue(struct ata_queued_cmd *qc);
unsigned long deadline);
static int mv_softreset(struct ata_link *link, unsigned int *class,
unsigned long deadline);
+static void mv_pmp_error_handler(struct ata_port *ap);
+static void mv_process_crpb_entries(struct ata_port *ap,
+ struct mv_port_priv *pp);
/* .sg_tablesize is (MV_MAX_SG_CT / 2) in the structures below
* because we have to allow room for worst case splitting of
static struct ata_port_operations mv5_ops = {
.inherits = &ata_sff_port_ops,
+ .qc_defer = mv_qc_defer,
.qc_prep = mv_qc_prep,
.qc_issue = mv_qc_issue,
static struct ata_port_operations mv6_ops = {
.inherits = &mv5_ops,
- .qc_defer = sata_pmp_qc_defer_cmd_switch,
.dev_config = mv6_dev_config,
.scr_read = mv_scr_read,
.scr_write = mv_scr_write,
.pmp_hardreset = mv_pmp_hardreset,
.pmp_softreset = mv_softreset,
.softreset = mv_softreset,
- .error_handler = sata_pmp_error_handler,
+ .error_handler = mv_pmp_error_handler,
};
static struct ata_port_operations mv_iie_ops = {
.inherits = &mv6_ops,
- .qc_defer = ata_std_qc_defer, /* FIS-based switching */
.dev_config = ATA_OP_NULL,
.qc_prep = mv_qc_prep_iie,
};
}
}
+static void mv_wait_for_edma_empty_idle(struct ata_port *ap)
+{
+ void __iomem *port_mmio = mv_ap_base(ap);
+ const u32 empty_idle = (EDMA_STATUS_CACHE_EMPTY | EDMA_STATUS_IDLE);
+ const int per_loop = 5, timeout = (15 * 1000 / per_loop);
+ int i;
+
+ /*
+ * Wait for the EDMA engine to finish transactions in progress.
+ * No idea what a good "timeout" value might be, but measurements
+ * indicate that it often requires hundreds of microseconds
+ * with two drives in-use. So we use the 15msec value above
+ * as a rough guess at what even more drives might require.
+ */
+ for (i = 0; i < timeout; ++i) {
+ u32 edma_stat = readl(port_mmio + EDMA_STATUS_OFS);
+ if ((edma_stat & empty_idle) == empty_idle)
+ break;
+ udelay(per_loop);
+ }
+ /* ata_port_printk(ap, KERN_INFO, "%s: %u+ usecs\n", __func__, i); */
+}
+
/**
* mv_stop_edma_engine - Disable eDMA engine
* @port_mmio: io base address
if (!(pp->pp_flags & MV_PP_FLAG_EDMA_EN))
return 0;
pp->pp_flags &= ~MV_PP_FLAG_EDMA_EN;
+ mv_wait_for_edma_empty_idle(ap);
if (mv_stop_edma_engine(port_mmio)) {
ata_port_printk(ap, KERN_ERR, "Unable to stop eDMA\n");
return -EIO;
}
}
-static void mv_config_fbs(void __iomem *port_mmio, int enable_fbs)
+static int mv_qc_defer(struct ata_queued_cmd *qc)
{
- u32 old_fcfg, new_fcfg, old_ltmode, new_ltmode;
+ struct ata_link *link = qc->dev->link;
+ struct ata_port *ap = link->ap;
+ struct mv_port_priv *pp = ap->private_data;
+
+ /*
+ * Don't allow new commands if we're in a delayed EH state
+ * for NCQ and/or FIS-based switching.
+ */
+ if (pp->pp_flags & MV_PP_FLAG_DELAYED_EH)
+ return ATA_DEFER_PORT;
/*
- * Various bit settings required for operation
- * in FIS-based switching (fbs) mode on GenIIe:
+ * If the port is completely idle, then allow the new qc.
*/
- old_fcfg = readl(port_mmio + FIS_CFG_OFS);
- old_ltmode = readl(port_mmio + LTMODE_OFS);
- if (enable_fbs) {
- new_fcfg = old_fcfg | FIS_CFG_SINGLE_SYNC;
- new_ltmode = old_ltmode | LTMODE_BIT8;
- } else { /* disable fbs */
- new_fcfg = old_fcfg & ~FIS_CFG_SINGLE_SYNC;
- new_ltmode = old_ltmode & ~LTMODE_BIT8;
- }
- if (new_fcfg != old_fcfg)
- writelfl(new_fcfg, port_mmio + FIS_CFG_OFS);
+ if (ap->nr_active_links == 0)
+ return 0;
+
+ if (pp->pp_flags & MV_PP_FLAG_EDMA_EN) {
+ /*
+ * The port is operating in host queuing mode (EDMA).
+ * It can accomodate a new qc if the qc protocol
+ * is compatible with the current host queue mode.
+ */
+ if (pp->pp_flags & MV_PP_FLAG_NCQ_EN) {
+ /*
+ * The host queue (EDMA) is in NCQ mode.
+ * If the new qc is also an NCQ command,
+ * then allow the new qc.
+ */
+ if (qc->tf.protocol == ATA_PROT_NCQ)
+ return 0;
+ } else {
+ /*
+ * The host queue (EDMA) is in non-NCQ, DMA mode.
+ * If the new qc is also a non-NCQ, DMA command,
+ * then allow the new qc.
+ */
+ if (qc->tf.protocol == ATA_PROT_DMA)
+ return 0;
+ }
+ }
+ return ATA_DEFER_PORT;
+}
+
+static void mv_config_fbs(void __iomem *port_mmio, int want_ncq, int want_fbs)
+{
+ u32 new_fiscfg, old_fiscfg;
+ u32 new_ltmode, old_ltmode;
+ u32 new_haltcond, old_haltcond;
+
+ old_fiscfg = readl(port_mmio + FISCFG_OFS);
+ old_ltmode = readl(port_mmio + LTMODE_OFS);
+ old_haltcond = readl(port_mmio + EDMA_HALTCOND_OFS);
+
+ new_fiscfg = old_fiscfg & ~(FISCFG_SINGLE_SYNC | FISCFG_WAIT_DEV_ERR);
+ new_ltmode = old_ltmode & ~LTMODE_BIT8;
+ new_haltcond = old_haltcond | EDMA_ERR_DEV;
+
+ if (want_fbs) {
+ new_fiscfg = old_fiscfg | FISCFG_SINGLE_SYNC;
+ new_ltmode = old_ltmode | LTMODE_BIT8;
+ if (want_ncq)
+ new_haltcond &= ~EDMA_ERR_DEV;
+ else
+ new_fiscfg |= FISCFG_WAIT_DEV_ERR;
+ }
+
+ if (new_fiscfg != old_fiscfg)
+ writelfl(new_fiscfg, port_mmio + FISCFG_OFS);
if (new_ltmode != old_ltmode)
writelfl(new_ltmode, port_mmio + LTMODE_OFS);
+ if (new_haltcond != old_haltcond)
+ writelfl(new_haltcond, port_mmio + EDMA_HALTCOND_OFS);
+}
+
+static void mv_60x1_errata_sata25(struct ata_port *ap, int want_ncq)
+{
+ struct mv_host_priv *hpriv = ap->host->private_data;
+ u32 old, new;
+
+ /* workaround for 88SX60x1 FEr SATA#25 (part 1) */
+ old = readl(hpriv->base + MV_GPIO_PORT_CTL_OFS);
+ if (want_ncq)
+ new = old | (1 << 22);
+ else
+ new = old & ~(1 << 22);
+ if (new != old)
+ writel(new, hpriv->base + MV_GPIO_PORT_CTL_OFS);
}
static void mv_edma_cfg(struct ata_port *ap, int want_ncq)
/* set up non-NCQ EDMA configuration */
cfg = EDMA_CFG_Q_DEPTH; /* always 0x1f for *all* chips */
+ pp->pp_flags &= ~MV_PP_FLAG_FBS_EN;
if (IS_GEN_I(hpriv))
cfg |= (1 << 8); /* enab config burst size mask */
- else if (IS_GEN_II(hpriv))
+ else if (IS_GEN_II(hpriv)) {
cfg |= EDMA_CFG_RD_BRST_EXT | EDMA_CFG_WR_BUFF_LEN;
+ mv_60x1_errata_sata25(ap, want_ncq);
- else if (IS_GEN_IIE(hpriv)) {
- cfg |= (1 << 23); /* do not mask PM field in rx'd FIS */
- cfg |= (1 << 22); /* enab 4-entry host queue cache */
- cfg |= (1 << 18); /* enab early completion */
- cfg |= (1 << 17); /* enab cut-through (dis stor&forwrd) */
+ } else if (IS_GEN_IIE(hpriv)) {
+ int want_fbs = sata_pmp_attached(ap);
+ /*
+ * Possible future enhancement:
+ *
+ * The chip can use FBS with non-NCQ, if we allow it,
+ * But first we need to have the error handling in place
+ * for this mode (datasheet section 7.3.15.4.2.3).
+ * So disallow non-NCQ FBS for now.
+ */
+ want_fbs &= want_ncq;
+
+ mv_config_fbs(port_mmio, want_ncq, want_fbs);
- if (want_ncq && sata_pmp_attached(ap)) {
+ if (want_fbs) {
+ pp->pp_flags |= MV_PP_FLAG_FBS_EN;
cfg |= EDMA_CFG_EDMA_FBS; /* FIS-based switching */
- mv_config_fbs(port_mmio, 1);
- } else {
- mv_config_fbs(port_mmio, 0);
}
+
+ cfg |= (1 << 23); /* do not mask PM field in rx'd FIS */
+ cfg |= (1 << 22); /* enab 4-entry host queue cache */
+ if (HAS_PCI(ap->host))
+ cfg |= (1 << 18); /* enab early completion */
+ if (hpriv->hp_flags & MV_HP_CUT_THROUGH)
+ cfg |= (1 << 17); /* enab cut-thru (dis stor&forwrd) */
}
if (want_ncq) {
return qc;
}
-static void mv_unexpected_intr(struct ata_port *ap)
+static void mv_pmp_error_handler(struct ata_port *ap)
{
+ unsigned int pmp, pmp_map;
struct mv_port_priv *pp = ap->private_data;
- struct ata_eh_info *ehi = &ap->link.eh_info;
- char *when = "";
+ if (pp->pp_flags & MV_PP_FLAG_DELAYED_EH) {
+ /*
+ * Perform NCQ error analysis on failed PMPs
+ * before we freeze the port entirely.
+ *
+ * The failed PMPs are marked earlier by mv_pmp_eh_prep().
+ */
+ pmp_map = pp->delayed_eh_pmp_map;
+ pp->pp_flags &= ~MV_PP_FLAG_DELAYED_EH;
+ for (pmp = 0; pmp_map != 0; pmp++) {
+ unsigned int this_pmp = (1 << pmp);
+ if (pmp_map & this_pmp) {
+ struct ata_link *link = &ap->pmp_link[pmp];
+ pmp_map &= ~this_pmp;
+ ata_eh_analyze_ncq_error(link);
+ }
+ }
+ ata_port_freeze(ap);
+ }
+ sata_pmp_error_handler(ap);
+}
+
+static unsigned int mv_get_err_pmp_map(struct ata_port *ap)
+{
+ void __iomem *port_mmio = mv_ap_base(ap);
+
+ return readl(port_mmio + SATA_TESTCTL_OFS) >> 16;
+}
+
+static void mv_pmp_eh_prep(struct ata_port *ap, unsigned int pmp_map)
+{
+ struct ata_eh_info *ehi;
+ unsigned int pmp;
+
+ /*
+ * Initialize EH info for PMPs which saw device errors
+ */
+ ehi = &ap->link.eh_info;
+ for (pmp = 0; pmp_map != 0; pmp++) {
+ unsigned int this_pmp = (1 << pmp);
+ if (pmp_map & this_pmp) {
+ struct ata_link *link = &ap->pmp_link[pmp];
+
+ pmp_map &= ~this_pmp;
+ ehi = &link->eh_info;
+ ata_ehi_clear_desc(ehi);
+ ata_ehi_push_desc(ehi, "dev err");
+ ehi->err_mask |= AC_ERR_DEV;
+ ehi->action |= ATA_EH_RESET;
+ ata_link_abort(link);
+ }
+ }
+}
+
+static int mv_handle_fbs_ncq_dev_err(struct ata_port *ap)
+{
+ struct mv_port_priv *pp = ap->private_data;
+ int failed_links;
+ unsigned int old_map, new_map;
+
+ /*
+ * Device error during FBS+NCQ operation:
+ *
+ * Set a port flag to prevent further I/O being enqueued.
+ * Leave the EDMA running to drain outstanding commands from this port.
+ * Perform the post-mortem/EH only when all responses are complete.
+ * Follow recovery sequence from 6042/7042 datasheet (7.3.15.4.2.2).
+ */
+ if (!(pp->pp_flags & MV_PP_FLAG_DELAYED_EH)) {
+ pp->pp_flags |= MV_PP_FLAG_DELAYED_EH;
+ pp->delayed_eh_pmp_map = 0;
+ }
+ old_map = pp->delayed_eh_pmp_map;
+ new_map = old_map | mv_get_err_pmp_map(ap);
+
+ if (old_map != new_map) {
+ pp->delayed_eh_pmp_map = new_map;
+ mv_pmp_eh_prep(ap, new_map & ~old_map);
+ }
+ failed_links = hweight16(new_map);
+
+ ata_port_printk(ap, KERN_INFO, "%s: pmp_map=%04x qc_map=%04x "
+ "failed_links=%d nr_active_links=%d\n",
+ __func__, pp->delayed_eh_pmp_map,
+ ap->qc_active, failed_links,
+ ap->nr_active_links);
+
+ if (ap->nr_active_links <= failed_links) {
+ mv_process_crpb_entries(ap, pp);
+ mv_stop_edma(ap);
+ mv_eh_freeze(ap);
+ ata_port_printk(ap, KERN_INFO, "%s: done\n", __func__);
+ return 1; /* handled */
+ }
+ ata_port_printk(ap, KERN_INFO, "%s: waiting\n", __func__);
+ return 1; /* handled */
+}
+
+static int mv_handle_fbs_non_ncq_dev_err(struct ata_port *ap)
+{
/*
- * We got a device interrupt from something that
- * was supposed to be using EDMA or polling.
+ * Possible future enhancement:
+ *
+ * FBS+non-NCQ operation is not yet implemented.
+ * See related notes in mv_edma_cfg().
+ *
+ * Device error during FBS+non-NCQ operation:
+ *
+ * We need to snapshot the shadow registers for each failed command.
+ * Follow recovery sequence from 6042/7042 datasheet (7.3.15.4.2.3).
*/
+ return 0; /* not handled */
+}
+
+static int mv_handle_dev_err(struct ata_port *ap, u32 edma_err_cause)
+{
+ struct mv_port_priv *pp = ap->private_data;
+
+ if (!(pp->pp_flags & MV_PP_FLAG_EDMA_EN))
+ return 0; /* EDMA was not active: not handled */
+ if (!(pp->pp_flags & MV_PP_FLAG_FBS_EN))
+ return 0; /* FBS was not active: not handled */
+
+ if (!(edma_err_cause & EDMA_ERR_DEV))
+ return 0; /* non DEV error: not handled */
+ edma_err_cause &= ~EDMA_ERR_IRQ_TRANSIENT;
+ if (edma_err_cause & ~(EDMA_ERR_DEV | EDMA_ERR_SELF_DIS))
+ return 0; /* other problems: not handled */
+
+ if (pp->pp_flags & MV_PP_FLAG_NCQ_EN) {
+ /*
+ * EDMA should NOT have self-disabled for this case.
+ * If it did, then something is wrong elsewhere,
+ * and we cannot handle it here.
+ */
+ if (edma_err_cause & EDMA_ERR_SELF_DIS) {
+ ata_port_printk(ap, KERN_WARNING,
+ "%s: err_cause=0x%x pp_flags=0x%x\n",
+ __func__, edma_err_cause, pp->pp_flags);
+ return 0; /* not handled */
+ }
+ return mv_handle_fbs_ncq_dev_err(ap);
+ } else {
+ /*
+ * EDMA should have self-disabled for this case.
+ * If it did not, then something is wrong elsewhere,
+ * and we cannot handle it here.
+ */
+ if (!(edma_err_cause & EDMA_ERR_SELF_DIS)) {
+ ata_port_printk(ap, KERN_WARNING,
+ "%s: err_cause=0x%x pp_flags=0x%x\n",
+ __func__, edma_err_cause, pp->pp_flags);
+ return 0; /* not handled */
+ }
+ return mv_handle_fbs_non_ncq_dev_err(ap);
+ }
+ return 0; /* not handled */
+}
+
+static void mv_unexpected_intr(struct ata_port *ap, int edma_was_enabled)
+{
+ struct ata_eh_info *ehi = &ap->link.eh_info;
+ char *when = "idle";
+
ata_ehi_clear_desc(ehi);
- if (pp->pp_flags & MV_PP_FLAG_EDMA_EN) {
- when = " while EDMA enabled";
+ if (!ap || (ap->flags & ATA_FLAG_DISABLED)) {
+ when = "disabled";
+ } else if (edma_was_enabled) {
+ when = "EDMA enabled";
} else {
struct ata_queued_cmd *qc = ata_qc_from_tag(ap, ap->link.active_tag);
if (qc && (qc->tf.flags & ATA_TFLAG_POLLING))
- when = " while polling";
+ when = "polling";
}
- ata_ehi_push_desc(ehi, "unexpected device interrupt%s", when);
+ ata_ehi_push_desc(ehi, "unexpected device interrupt while %s", when);
ehi->err_mask |= AC_ERR_OTHER;
ehi->action |= ATA_EH_RESET;
ata_port_freeze(ap);
* LOCKING:
* Inherited from caller.
*/
-static void mv_err_intr(struct ata_port *ap, struct ata_queued_cmd *qc)
+static void mv_err_intr(struct ata_port *ap)
{
void __iomem *port_mmio = mv_ap_base(ap);
u32 edma_err_cause, eh_freeze_mask, serr = 0;
struct mv_host_priv *hpriv = ap->host->private_data;
unsigned int action = 0, err_mask = 0;
struct ata_eh_info *ehi = &ap->link.eh_info;
-
- ata_ehi_clear_desc(ehi);
+ struct ata_queued_cmd *qc;
+ int abort = 0;
/*
- * Read and clear the err_cause bits. This won't actually
- * clear for some errors (eg. SError), but we will be doing
- * a hard reset in those cases regardless, which *will* clear it.
+ * Read and clear the SError and err_cause bits.
*/
+ sata_scr_read(&ap->link, SCR_ERROR, &serr);
+ sata_scr_write_flush(&ap->link, SCR_ERROR, serr);
+
edma_err_cause = readl(port_mmio + EDMA_ERR_IRQ_CAUSE_OFS);
writelfl(~edma_err_cause, port_mmio + EDMA_ERR_IRQ_CAUSE_OFS);
- ata_ehi_push_desc(ehi, "edma_err_cause=%08x", edma_err_cause);
+ ata_port_printk(ap, KERN_INFO, "%s: err_cause=%08x pp_flags=0x%x\n",
+ __func__, edma_err_cause, pp->pp_flags);
+
+ if (edma_err_cause & EDMA_ERR_DEV) {
+ /*
+ * Device errors during FIS-based switching operation
+ * require special handling.
+ */
+ if (mv_handle_dev_err(ap, edma_err_cause))
+ return;
+ }
+ qc = mv_get_active_qc(ap);
+ ata_ehi_clear_desc(ehi);
+ ata_ehi_push_desc(ehi, "edma_err_cause=%08x pp_flags=%08x",
+ edma_err_cause, pp->pp_flags);
/*
* All generations share these EDMA error cause bits:
*/
- if (edma_err_cause & EDMA_ERR_DEV)
+ if (edma_err_cause & EDMA_ERR_DEV) {
err_mask |= AC_ERR_DEV;
+ action |= ATA_EH_RESET;
+ ata_ehi_push_desc(ehi, "dev error");
+ }
if (edma_err_cause & (EDMA_ERR_D_PAR | EDMA_ERR_PRD_PAR |
EDMA_ERR_CRQB_PAR | EDMA_ERR_CRPB_PAR |
EDMA_ERR_INTRL_PAR)) {
ata_ehi_push_desc(ehi, "EDMA self-disable");
}
if (edma_err_cause & EDMA_ERR_SERR) {
- /*
- * Ensure that we read our own SCR, not a pmp link SCR:
- */
- ap->ops->scr_read(ap, SCR_ERROR, &serr);
- /*
- * Don't clear SError here; leave it for libata-eh:
- */
ata_ehi_push_desc(ehi, "SError=%08x", serr);
err_mask |= AC_ERR_ATA_BUS;
action |= ATA_EH_RESET;
else
ehi->err_mask |= err_mask;
- if (edma_err_cause & eh_freeze_mask)
+ if (err_mask == AC_ERR_DEV) {
+ /*
+ * Cannot do ata_port_freeze() here,
+ * because it would kill PIO access,
+ * which is needed for further diagnosis.
+ */
+ mv_eh_freeze(ap);
+ abort = 1;
+ } else if (edma_err_cause & eh_freeze_mask) {
+ /*
+ * Note to self: ata_port_freeze() calls ata_port_abort()
+ */
ata_port_freeze(ap);
- else
- ata_port_abort(ap);
+ } else {
+ abort = 1;
+ }
+
+ if (abort) {
+ if (qc)
+ ata_link_abort(qc->dev->link);
+ else
+ ata_port_abort(ap);
+ }
}
static void mv_process_crpb_response(struct ata_port *ap,
}
}
ata_status = edma_status >> CRPB_FLAG_STATUS_SHIFT;
- qc->err_mask |= ac_err_mask(ata_status);
- ata_qc_complete(qc);
+ if (!ac_err_mask(ata_status))
+ ata_qc_complete(qc);
+ /* else: leave it for mv_err_intr() */
} else {
ata_port_printk(ap, KERN_ERR, "%s: no qc for tag=%d\n",
__func__, tag);
port_mmio + EDMA_RSP_Q_OUT_PTR_OFS);
}
+static void mv_port_intr(struct ata_port *ap, u32 port_cause)
+{
+ struct mv_port_priv *pp;
+ int edma_was_enabled;
+
+ if (!ap || (ap->flags & ATA_FLAG_DISABLED)) {
+ mv_unexpected_intr(ap, 0);
+ return;
+ }
+ /*
+ * Grab a snapshot of the EDMA_EN flag setting,
+ * so that we have a consistent view for this port,
+ * even if something we call of our routines changes it.
+ */
+ pp = ap->private_data;
+ edma_was_enabled = (pp->pp_flags & MV_PP_FLAG_EDMA_EN);
+ /*
+ * Process completed CRPB response(s) before other events.
+ */
+ if (edma_was_enabled && (port_cause & DONE_IRQ)) {
+ mv_process_crpb_entries(ap, pp);
+ if (pp->pp_flags & MV_PP_FLAG_DELAYED_EH)
+ mv_handle_fbs_ncq_dev_err(ap);
+ }
+ /*
+ * Handle chip-reported errors, or continue on to handle PIO.
+ */
+ if (unlikely(port_cause & ERR_IRQ)) {
+ mv_err_intr(ap);
+ } else if (!edma_was_enabled) {
+ struct ata_queued_cmd *qc = mv_get_active_qc(ap);
+ if (qc)
+ ata_sff_host_intr(ap, qc);
+ else
+ mv_unexpected_intr(ap, edma_was_enabled);
+ }
+}
+
/**
* mv_host_intr - Handle all interrupts on the given host controller
* @host: host specific structure
static int mv_host_intr(struct ata_host *host, u32 main_irq_cause)
{
struct mv_host_priv *hpriv = host->private_data;
- void __iomem *mmio = hpriv->base, *hc_mmio = NULL;
- u32 hc_irq_cause = 0;
+ void __iomem *mmio = hpriv->base, *hc_mmio;
unsigned int handled = 0, port;
for (port = 0; port < hpriv->n_ports; port++) {
struct ata_port *ap = host->ports[port];
- struct mv_port_priv *pp;
- unsigned int shift, hardport, port_cause;
- /*
- * When we move to the second hc, flag our cached
- * copies of hc_mmio (and hc_irq_cause) as invalid again.
- */
- if (port == MV_PORTS_PER_HC)
- hc_mmio = NULL;
- /*
- * Do nothing if port is not interrupting or is disabled:
- */
+ unsigned int p, shift, hardport, port_cause;
+
MV_PORT_TO_SHIFT_AND_HARDPORT(port, shift, hardport);
- port_cause = (main_irq_cause >> shift) & (DONE_IRQ | ERR_IRQ);
- if (!port_cause || !ap || (ap->flags & ATA_FLAG_DISABLED))
- continue;
/*
- * Each hc within the host has its own hc_irq_cause register.
- * We defer reading it until we know we need it, right now:
- *
- * FIXME later: we don't really need to read this register
- * (some logic changes required below if we go that way),
- * because it doesn't tell us anything new. But we do need
- * to write to it, outside the top of this loop,
- * to reset the interrupt triggers for next time.
+ * Each hc within the host has its own hc_irq_cause register,
+ * where the interrupting ports bits get ack'd.
*/
- if (!hc_mmio) {
+ if (hardport == 0) { /* first port on this hc ? */
+ u32 hc_cause = (main_irq_cause >> shift) & HC0_IRQ_PEND;
+ u32 port_mask, ack_irqs;
+ /*
+ * Skip this entire hc if nothing pending for any ports
+ */
+ if (!hc_cause) {
+ port += MV_PORTS_PER_HC - 1;
+ continue;
+ }
+ /*
+ * We don't need/want to read the hc_irq_cause register,
+ * because doing so hurts performance, and
+ * main_irq_cause already gives us everything we need.
+ *
+ * But we do have to *write* to the hc_irq_cause to ack
+ * the ports that we are handling this time through.
+ *
+ * This requires that we create a bitmap for those
+ * ports which interrupted us, and use that bitmap
+ * to ack (only) those ports via hc_irq_cause.
+ */
+ ack_irqs = 0;
+ for (p = 0; p < MV_PORTS_PER_HC; ++p) {
+ if ((port + p) >= hpriv->n_ports)
+ break;
+ port_mask = (DONE_IRQ | ERR_IRQ) << (p * 2);
+ if (hc_cause & port_mask)
+ ack_irqs |= (DMA_IRQ | DEV_IRQ) << p;
+ }
hc_mmio = mv_hc_base_from_port(mmio, port);
- hc_irq_cause = readl(hc_mmio + HC_IRQ_CAUSE_OFS);
- writelfl(~hc_irq_cause, hc_mmio + HC_IRQ_CAUSE_OFS);
+ writelfl(~ack_irqs, hc_mmio + HC_IRQ_CAUSE_OFS);
handled = 1;
}
/*
- * Process completed CRPB response(s) before other events.
- */
- pp = ap->private_data;
- if (hc_irq_cause & (DMA_IRQ << hardport)) {
- if (pp->pp_flags & MV_PP_FLAG_EDMA_EN)
- mv_process_crpb_entries(ap, pp);
- }
- /*
- * Handle chip-reported errors, or continue on to handle PIO.
+ * Handle interrupts signalled for this port:
*/
- if (unlikely(port_cause & ERR_IRQ)) {
- mv_err_intr(ap, mv_get_active_qc(ap));
- } else if (hc_irq_cause & (DEV_IRQ << hardport)) {
- if (!(pp->pp_flags & MV_PP_FLAG_EDMA_EN)) {
- struct ata_queued_cmd *qc = mv_get_active_qc(ap);
- if (qc) {
- ata_sff_host_intr(ap, qc);
- continue;
- }
- }
- mv_unexpected_intr(ap);
- }
+ port_cause = (main_irq_cause >> shift) & (DONE_IRQ | ERR_IRQ);
+ if (port_cause)
+ mv_port_intr(ap, port_cause);
}
return handled;
}
static void mv5_reset_flash(struct mv_host_priv *hpriv, void __iomem *mmio)
{
- writel(0x0fcfffff, mmio + MV_FLASH_CTL);
+ writel(0x0fcfffff, mmio + MV_FLASH_CTL_OFS);
}
static void mv5_read_preamp(struct mv_host_priv *hpriv, int idx,
{
u32 tmp;
- writel(0, mmio + MV_GPIO_PORT_CTL);
+ writel(0, mmio + MV_GPIO_PORT_CTL_OFS);
/* FIXME: handle MV_HP_ERRATA_50XXB2 errata */
int fix_apm_sq = (hpriv->hp_flags & MV_HP_ERRATA_50XXB0);
if (fix_apm_sq) {
- tmp = readl(phy_mmio + MV5_LT_MODE);
+ tmp = readl(phy_mmio + MV5_LTMODE_OFS);
tmp |= (1 << 19);
- writel(tmp, phy_mmio + MV5_LT_MODE);
+ writel(tmp, phy_mmio + MV5_LTMODE_OFS);
- tmp = readl(phy_mmio + MV5_PHY_CTL);
+ tmp = readl(phy_mmio + MV5_PHY_CTL_OFS);
tmp &= ~0x3;
tmp |= 0x1;
- writel(tmp, phy_mmio + MV5_PHY_CTL);
+ writel(tmp, phy_mmio + MV5_PHY_CTL_OFS);
}
tmp = readl(phy_mmio + MV5_PHY_MODE);
{
void __iomem *port_mmio = mv_port_base(mmio, port);
- /*
- * The datasheet warns against setting ATA_RST when EDMA is active
- * (but doesn't say what the problem might be). So we first try
- * to disable the EDMA engine before doing the ATA_RST operation.
- */
mv_reset_channel(hpriv, mmio, port);
ZERO(0x028); /* command */
ZERO(0x024); /* respq outp */
ZERO(0x020); /* respq inp */
ZERO(0x02c); /* test control */
- writel(0xbc, port_mmio + EDMA_IORDY_TMOUT);
+ writel(0xbc, port_mmio + EDMA_IORDY_TMOUT_OFS);
}
#undef ZERO
struct mv_host_priv *hpriv = host->private_data;
u32 tmp;
- tmp = readl(mmio + MV_PCI_MODE);
+ tmp = readl(mmio + MV_PCI_MODE_OFS);
tmp &= 0xff00ffff;
- writel(tmp, mmio + MV_PCI_MODE);
+ writel(tmp, mmio + MV_PCI_MODE_OFS);
ZERO(MV_PCI_DISC_TIMER);
ZERO(MV_PCI_MSI_TRIGGER);
- writel(0x000100ff, mmio + MV_PCI_XBAR_TMOUT);
+ writel(0x000100ff, mmio + MV_PCI_XBAR_TMOUT_OFS);
ZERO(PCI_HC_MAIN_IRQ_MASK_OFS);
ZERO(MV_PCI_SERR_MASK);
ZERO(hpriv->irq_cause_ofs);
mv5_reset_flash(hpriv, mmio);
- tmp = readl(mmio + MV_GPIO_PORT_CTL);
+ tmp = readl(mmio + MV_GPIO_PORT_CTL_OFS);
tmp &= 0x3;
tmp |= (1 << 5) | (1 << 6);
- writel(tmp, mmio + MV_GPIO_PORT_CTL);
+ writel(tmp, mmio + MV_GPIO_PORT_CTL_OFS);
}
/**
void __iomem *port_mmio;
u32 tmp;
- tmp = readl(mmio + MV_RESET_CFG);
+ tmp = readl(mmio + MV_RESET_CFG_OFS);
if ((tmp & (1 << 0)) == 0) {
hpriv->signal[idx].amps = 0x7 << 8;
hpriv->signal[idx].pre = 0x1 << 5;
static void mv6_enable_leds(struct mv_host_priv *hpriv, void __iomem *mmio)
{
- writel(0x00000060, mmio + MV_GPIO_PORT_CTL);
+ writel(0x00000060, mmio + MV_GPIO_PORT_CTL_OFS);
}
static void mv6_phy_errata(struct mv_host_priv *hpriv, void __iomem *mmio,
{
void __iomem *port_mmio = mv_port_base(mmio, port);
- /*
- * The datasheet warns against setting ATA_RST when EDMA is active
- * (but doesn't say what the problem might be). So we first try
- * to disable the EDMA engine before doing the ATA_RST operation.
- */
mv_reset_channel(hpriv, mmio, port);
ZERO(0x028); /* command */
ZERO(0x024); /* respq outp */
ZERO(0x020); /* respq inp */
ZERO(0x02c); /* test control */
- writel(0xbc, port_mmio + EDMA_IORDY_TMOUT);
+ writel(0xbc, port_mmio + EDMA_IORDY_TMOUT_OFS);
}
#undef ZERO
return;
}
-static void mv_setup_ifctl(void __iomem *port_mmio, int want_gen2i)
+static void mv_setup_ifcfg(void __iomem *port_mmio, int want_gen2i)
{
- u32 ifctl = readl(port_mmio + SATA_INTERFACE_CFG);
+ u32 ifcfg = readl(port_mmio + SATA_INTERFACE_CFG_OFS);
- ifctl = (ifctl & 0xf7f) | 0x9b1000; /* from chip spec */
+ ifcfg = (ifcfg & 0xf7f) | 0x9b1000; /* from chip spec */
if (want_gen2i)
- ifctl |= (1 << 7); /* enable gen2i speed */
- writelfl(ifctl, port_mmio + SATA_INTERFACE_CFG);
+ ifcfg |= (1 << 7); /* enable gen2i speed */
+ writelfl(ifcfg, port_mmio + SATA_INTERFACE_CFG_OFS);
}
-/*
- * Caller must ensure that EDMA is not active,
- * by first doing mv_stop_edma() where needed.
- */
static void mv_reset_channel(struct mv_host_priv *hpriv, void __iomem *mmio,
unsigned int port_no)
{
void __iomem *port_mmio = mv_port_base(mmio, port_no);
+ /*
+ * The datasheet warns against setting EDMA_RESET when EDMA is active
+ * (but doesn't say what the problem might be). So we first try
+ * to disable the EDMA engine before doing the EDMA_RESET operation.
+ */
mv_stop_edma_engine(port_mmio);
- writelfl(ATA_RST, port_mmio + EDMA_CMD_OFS);
+ writelfl(EDMA_RESET, port_mmio + EDMA_CMD_OFS);
if (!IS_GEN_I(hpriv)) {
- /* Enable 3.0gb/s link speed */
- mv_setup_ifctl(port_mmio, 1);
+ /* Enable 3.0gb/s link speed: this survives EDMA_RESET */
+ mv_setup_ifcfg(port_mmio, 1);
}
/*
- * Strobing ATA_RST here causes a hard reset of the SATA transport,
+ * Strobing EDMA_RESET here causes a hard reset of the SATA transport,
* link, and physical layers. It resets all SATA interface registers
* (except for SATA_INTERFACE_CFG), and issues a COMRESET to the dev.
*/
- writelfl(ATA_RST, port_mmio + EDMA_CMD_OFS);
+ writelfl(EDMA_RESET, port_mmio + EDMA_CMD_OFS);
udelay(25); /* allow reset propagation */
writelfl(0, port_mmio + EDMA_CMD_OFS);
sata_scr_read(link, SCR_STATUS, &sstatus);
if (!IS_GEN_I(hpriv) && ++attempts >= 5 && sstatus == 0x121) {
/* Force 1.5gb/s link speed and try again */
- mv_setup_ifctl(mv_ap_base(ap), 0);
+ mv_setup_ifcfg(mv_ap_base(ap), 0);
if (time_after(jiffies + HZ, deadline))
extra = HZ; /* only extend it once, max */
}
readl(port_mmio + EDMA_ERR_IRQ_MASK_OFS));
}
+static unsigned int mv_in_pcix_mode(struct ata_host *host)
+{
+ struct mv_host_priv *hpriv = host->private_data;
+ void __iomem *mmio = hpriv->base;
+ u32 reg;
+
+ if (!HAS_PCI(host) || !IS_PCIE(hpriv))
+ return 0; /* not PCI-X capable */
+ reg = readl(mmio + MV_PCI_MODE_OFS);
+ if ((reg & MV_PCI_MODE_MASK) == 0)
+ return 0; /* conventional PCI mode */
+ return 1; /* chip is in PCI-X mode */
+}
+
+static int mv_pci_cut_through_okay(struct ata_host *host)
+{
+ struct mv_host_priv *hpriv = host->private_data;
+ void __iomem *mmio = hpriv->base;
+ u32 reg;
+
+ if (!mv_in_pcix_mode(host)) {
+ reg = readl(mmio + PCI_COMMAND_OFS);
+ if (reg & PCI_COMMAND_MRDTRIG)
+ return 0; /* not okay */
+ }
+ return 1; /* okay */
+}
+
static int mv_chip_id(struct ata_host *host, unsigned int board_idx)
{
struct pci_dev *pdev = to_pci_dev(host->dev);
break;
case chip_7042:
- hp_flags |= MV_HP_PCIE;
+ hp_flags |= MV_HP_PCIE | MV_HP_CUT_THROUGH;
if (pdev->vendor == PCI_VENDOR_ID_TTI &&
(pdev->device == 0x2300 || pdev->device == 0x2310))
{
" and avoid the final two gigabytes on"
" all RocketRAID BIOS initialized drives.\n");
}
+ /* drop through */
case chip_6042:
hpriv->ops = &mv6xxx_ops;
hp_flags |= MV_HP_GEN_IIE;
+ if (board_idx == chip_6042 && mv_pci_cut_through_okay(host))
+ hp_flags |= MV_HP_CUT_THROUGH;
switch (pdev->revision) {
case 0x0:
}
/* Check whether this driver has already been added to a class. */
- if ((drv->entry.next != drv->entry.prev) ||
- (drv->entry.next != NULL)) {
+ if (drv->entry.next && !list_empty(&drv->entry)) {
printk(KERN_WARNING "sysdev: class %s: driver (%p) has already"
" been registered to a class, something is wrong, but "
"will forge on!\n", cls->name, drv);
{
unsigned long n_sect = bio->bi_size >> 9;
const int rw = bio_data_dir(bio);
+ struct hd_struct *part;
- all_stat_inc(disk, ios[rw], sector);
- all_stat_add(disk, ticks[rw], duration, sector);
- all_stat_add(disk, sectors[rw], n_sect, sector);
- all_stat_add(disk, io_ticks, duration, sector);
+ part = get_part(disk, sector);
+ all_stat_inc(disk, part, ios[rw], sector);
+ all_stat_add(disk, part, ticks[rw], duration, sector);
+ all_stat_add(disk, part, sectors[rw], n_sect, sector);
+ all_stat_add(disk, part, io_ticks, duration, sector);
}
void
return 0;
if (!info->xmit_buf)
- return;
+ return 0;
local_irq_save(flags);
if (info->xmit_cnt >= PAGE_SIZE - 1) {
sx_write_channel_byte(port, hi_mask, 0x1f);
break;
default:
- printk(KERN_INFO "sx: Invalid wordsize: %u\n", CFLAG & CSIZE);
+ printk(KERN_INFO "sx: Invalid wordsize: %u\n",
+ (unsigned int)CFLAG & CSIZE);
break;
}
set_bit(TTY_HW_COOK_IN, &port->gs.tty->flags);
}
sx_dprintk(SX_DEBUG_TERMIOS, "iflags: %x(%d) ",
- port->gs.tty->termios->c_iflag, I_OTHER(port->gs.tty));
+ (unsigned int)port->gs.tty->termios->c_iflag,
+ I_OTHER(port->gs.tty));
/* Tell line discipline whether we will do output cooking.
* If OPOST is set and no other output flags are set then we can do output
clear_bit(TTY_HW_COOK_OUT, &port->gs.tty->flags);
}
sx_dprintk(SX_DEBUG_TERMIOS, "oflags: %x(%d)\n",
- port->gs.tty->termios->c_oflag, O_OTHER(port->gs.tty));
+ (unsigned int)port->gs.tty->termios->c_oflag,
+ O_OTHER(port->gs.tty));
/* port->c_dcd = sx_get_CD (port); */
func_exit();
return 0;
tty->winsize.ws_row = vc_cons[currcons].d->vc_rows;
tty->winsize.ws_col = vc_cons[currcons].d->vc_cols;
}
+ if (vc->vc_utf)
+ tty->termios->c_iflag |= IUTF8;
+ else
+ tty->termios->c_iflag &= ~IUTF8;
release_console_sem();
vcs_make_sysfs(tty);
return ret;
console_driver->minor_start = 1;
console_driver->type = TTY_DRIVER_TYPE_CONSOLE;
console_driver->init_termios = tty_std_termios;
+ if (default_utf8)
+ console_driver->init_termios.c_iflag |= IUTF8;
console_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_RESET_TERMIOS;
tty_set_operations(console_driver, &con_ops);
if (tty_register_driver(console_driver))
#define PCI_VEND_DEV(vend, dev) PCI_VENDOR_ID_ ## vend, \
PCI_DEVICE_ID_ ## vend ## _ ## dev
-#define dev_name(dev) (dev)->dev_name
+#define edac_dev_name(dev) (dev)->dev_name
/* memory devices */
enum dev_type {
fail0:
edac_printk(KERN_WARNING, EDAC_MC,
"%s (%s) %s %s already assigned %d\n",
- rover->dev->bus_id, dev_name(rover),
+ rover->dev->bus_id, edac_dev_name(rover),
rover->mod_name, rover->ctl_name, rover->dev_idx);
return 1;
"'%s': DEV '%s' (%s)\n",
edac_dev->mod_name,
edac_dev->ctl_name,
- dev_name(edac_dev),
+ edac_dev_name(edac_dev),
edac_op_state_to_string(edac_dev->op_state));
mutex_unlock(&device_ctls_mutex);
edac_printk(KERN_INFO, EDAC_MC,
"Removed device %d for %s %s: DEV %s\n",
edac_dev->dev_idx,
- edac_dev->mod_name, edac_dev->ctl_name, dev_name(edac_dev));
+ edac_dev->mod_name, edac_dev->ctl_name, edac_dev_name(edac_dev));
return edac_dev;
}
fail0:
edac_printk(KERN_WARNING, EDAC_MC,
"%s (%s) %s %s already assigned %d\n", p->dev->bus_id,
- dev_name(mci), p->mod_name, p->ctl_name, p->mc_idx);
+ edac_dev_name(mci), p->mod_name, p->ctl_name, p->mc_idx);
return 1;
fail1:
/* Report action taken */
edac_mc_printk(mci, KERN_INFO, "Giving out device to '%s' '%s':"
- " DEV %s\n", mci->mod_name, mci->ctl_name, dev_name(mci));
+ " DEV %s\n", mci->mod_name, mci->ctl_name, edac_dev_name(mci));
mutex_unlock(&mem_ctls_mutex);
return 0;
edac_printk(KERN_INFO, EDAC_MC,
"Removed device %d for %s %s: DEV %s\n", mci->mc_idx,
- mci->mod_name, mci->ctl_name, dev_name(mci));
+ mci->mod_name, mci->ctl_name, edac_dev_name(mci));
return mci;
}
fail0:
edac_printk(KERN_WARNING, EDAC_PCI,
"%s (%s) %s %s already assigned %d\n",
- rover->dev->bus_id, dev_name(rover),
+ rover->dev->bus_id, edac_dev_name(rover),
rover->mod_name, rover->ctl_name, rover->pci_idx);
return 1;
" DEV '%s' (%s)\n",
pci->mod_name,
pci->ctl_name,
- dev_name(pci), edac_op_state_to_string(pci->op_state));
+ edac_dev_name(pci), edac_op_state_to_string(pci->op_state));
mutex_unlock(&edac_pci_ctls_mutex);
return 0;
edac_printk(KERN_INFO, EDAC_PCI,
"Removed device %d for %s %s: DEV %s\n",
- pci->pci_idx, pci->mod_name, pci->ctl_name, dev_name(pci));
+ pci->pci_idx, pci->mod_name, pci->ctl_name, edac_dev_name(pci));
return pci;
}
unsigned long data_addr = drive->hwif->io_ports.data_addr;
if (drive->media == ide_disk && rq && rq->cmd_type == REQ_TYPE_FS)
- return outsw(data_adr, buf, (len + 1) / 2);
+ return outsw(data_addr, buf, (len + 1) / 2);
outsw_swapw(data_addr, buf, (len + 1) / 2);
}
cq->sw_wptr++;
}
-void cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count)
+int cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count)
{
u32 ptr;
+ int flushed = 0;
PDBG("%s wq %p cq %p\n", __func__, wq, cq);
PDBG("%s rq_rptr %u rq_wptr %u skip count %u\n", __func__,
wq->rq_rptr, wq->rq_wptr, count);
ptr = wq->rq_rptr + count;
- while (ptr++ != wq->rq_wptr)
+ while (ptr++ != wq->rq_wptr) {
insert_recv_cqe(wq, cq);
+ flushed++;
+ }
+ return flushed;
}
static void insert_sq_cqe(struct t3_wq *wq, struct t3_cq *cq,
cq->sw_wptr++;
}
-void cxio_flush_sq(struct t3_wq *wq, struct t3_cq *cq, int count)
+int cxio_flush_sq(struct t3_wq *wq, struct t3_cq *cq, int count)
{
__u32 ptr;
+ int flushed = 0;
struct t3_swsq *sqp = wq->sq + Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2);
ptr = wq->sq_rptr + count;
insert_sq_cqe(wq, cq, sqp);
sqp++;
ptr++;
+ flushed++;
}
+ return flushed;
}
/*
* caller aquires the ctrl_qp lock before the call
*/
static int cxio_hal_ctrl_qp_write_mem(struct cxio_rdev *rdev_p, u32 addr,
- u32 len, void *data, int completion)
+ u32 len, void *data)
{
u32 i, nr_wqe, copy_len;
u8 *copy_data;
flag = 0;
if (i == (nr_wqe - 1)) {
/* last WQE */
- flag = completion ? T3_COMPLETION_FLAG : 0;
+ flag = T3_COMPLETION_FLAG;
if (len % 32)
utx_len = len / 32 + 1;
else
return 0;
}
-/* IN: stag key, pdid, perm, zbva, to, len, page_size, pbl, and pbl_size
- * OUT: stag index, actual pbl_size, pbl_addr allocated.
+/* IN: stag key, pdid, perm, zbva, to, len, page_size, pbl_size and pbl_addr
+ * OUT: stag index
* TBD: shared memory region support
*/
static int __cxio_tpt_op(struct cxio_rdev *rdev_p, u32 reset_tpt_entry,
u32 *stag, u8 stag_state, u32 pdid,
enum tpt_mem_type type, enum tpt_mem_perm perm,
- u32 zbva, u64 to, u32 len, u8 page_size, __be64 *pbl,
- u32 *pbl_size, u32 *pbl_addr)
+ u32 zbva, u64 to, u32 len, u8 page_size,
+ u32 pbl_size, u32 pbl_addr)
{
int err;
struct tpt_entry tpt;
u32 stag_idx;
u32 wptr;
- int rereg = (*stag != T3_STAG_UNSET);
stag_state = stag_state > 0;
stag_idx = (*stag) >> 8;
PDBG("%s stag_state 0x%0x type 0x%0x pdid 0x%0x, stag_idx 0x%x\n",
__func__, stag_state, type, pdid, stag_idx);
- if (reset_tpt_entry)
- cxio_hal_pblpool_free(rdev_p, *pbl_addr, *pbl_size << 3);
- else if (!rereg) {
- *pbl_addr = cxio_hal_pblpool_alloc(rdev_p, *pbl_size << 3);
- if (!*pbl_addr) {
- return -ENOMEM;
- }
- }
-
mutex_lock(&rdev_p->ctrl_qp.lock);
- /* write PBL first if any - update pbl only if pbl list exist */
- if (pbl) {
-
- PDBG("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n",
- __func__, *pbl_addr, rdev_p->rnic_info.pbl_base,
- *pbl_size);
- err = cxio_hal_ctrl_qp_write_mem(rdev_p,
- (*pbl_addr >> 5),
- (*pbl_size << 3), pbl, 0);
- if (err)
- goto ret;
- }
-
/* write TPT entry */
if (reset_tpt_entry)
memset(&tpt, 0, sizeof(tpt));
V_TPT_ADDR_TYPE((zbva ? TPT_ZBTO : TPT_VATO)) |
V_TPT_PAGE_SIZE(page_size));
tpt.rsvd_pbl_addr = reset_tpt_entry ? 0 :
- cpu_to_be32(V_TPT_PBL_ADDR(PBL_OFF(rdev_p, *pbl_addr)>>3));
+ cpu_to_be32(V_TPT_PBL_ADDR(PBL_OFF(rdev_p, pbl_addr)>>3));
tpt.len = cpu_to_be32(len);
tpt.va_hi = cpu_to_be32((u32) (to >> 32));
tpt.va_low_or_fbo = cpu_to_be32((u32) (to & 0xFFFFFFFFULL));
tpt.rsvd_bind_cnt_or_pstag = 0;
tpt.rsvd_pbl_size = reset_tpt_entry ? 0 :
- cpu_to_be32(V_TPT_PBL_SIZE((*pbl_size) >> 2));
+ cpu_to_be32(V_TPT_PBL_SIZE(pbl_size >> 2));
}
err = cxio_hal_ctrl_qp_write_mem(rdev_p,
stag_idx +
(rdev_p->rnic_info.tpt_base >> 5),
- sizeof(tpt), &tpt, 1);
+ sizeof(tpt), &tpt);
/* release the stag index to free pool */
if (reset_tpt_entry)
cxio_hal_put_stag(rdev_p->rscp, stag_idx);
-ret:
+
wptr = rdev_p->ctrl_qp.wptr;
mutex_unlock(&rdev_p->ctrl_qp.lock);
if (!err)
return err;
}
+int cxio_write_pbl(struct cxio_rdev *rdev_p, __be64 *pbl,
+ u32 pbl_addr, u32 pbl_size)
+{
+ u32 wptr;
+ int err;
+
+ PDBG("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n",
+ __func__, pbl_addr, rdev_p->rnic_info.pbl_base,
+ pbl_size);
+
+ mutex_lock(&rdev_p->ctrl_qp.lock);
+ err = cxio_hal_ctrl_qp_write_mem(rdev_p, pbl_addr >> 5, pbl_size << 3,
+ pbl);
+ wptr = rdev_p->ctrl_qp.wptr;
+ mutex_unlock(&rdev_p->ctrl_qp.lock);
+ if (err)
+ return err;
+
+ if (wait_event_interruptible(rdev_p->ctrl_qp.waitq,
+ SEQ32_GE(rdev_p->ctrl_qp.rptr,
+ wptr)))
+ return -ERESTARTSYS;
+
+ return 0;
+}
+
int cxio_register_phys_mem(struct cxio_rdev *rdev_p, u32 *stag, u32 pdid,
enum tpt_mem_perm perm, u32 zbva, u64 to, u32 len,
- u8 page_size, __be64 *pbl, u32 *pbl_size,
- u32 *pbl_addr)
+ u8 page_size, u32 pbl_size, u32 pbl_addr)
{
*stag = T3_STAG_UNSET;
return __cxio_tpt_op(rdev_p, 0, stag, 1, pdid, TPT_NON_SHARED_MR, perm,
- zbva, to, len, page_size, pbl, pbl_size, pbl_addr);
+ zbva, to, len, page_size, pbl_size, pbl_addr);
}
int cxio_reregister_phys_mem(struct cxio_rdev *rdev_p, u32 *stag, u32 pdid,
enum tpt_mem_perm perm, u32 zbva, u64 to, u32 len,
- u8 page_size, __be64 *pbl, u32 *pbl_size,
- u32 *pbl_addr)
+ u8 page_size, u32 pbl_size, u32 pbl_addr)
{
return __cxio_tpt_op(rdev_p, 0, stag, 1, pdid, TPT_NON_SHARED_MR, perm,
- zbva, to, len, page_size, pbl, pbl_size, pbl_addr);
+ zbva, to, len, page_size, pbl_size, pbl_addr);
}
int cxio_dereg_mem(struct cxio_rdev *rdev_p, u32 stag, u32 pbl_size,
u32 pbl_addr)
{
- return __cxio_tpt_op(rdev_p, 1, &stag, 0, 0, 0, 0, 0, 0ULL, 0, 0, NULL,
- &pbl_size, &pbl_addr);
+ return __cxio_tpt_op(rdev_p, 1, &stag, 0, 0, 0, 0, 0, 0ULL, 0, 0,
+ pbl_size, pbl_addr);
}
int cxio_allocate_window(struct cxio_rdev *rdev_p, u32 * stag, u32 pdid)
{
- u32 pbl_size = 0;
*stag = T3_STAG_UNSET;
return __cxio_tpt_op(rdev_p, 0, stag, 0, pdid, TPT_MW, 0, 0, 0ULL, 0, 0,
- NULL, &pbl_size, NULL);
+ 0, 0);
}
int cxio_deallocate_window(struct cxio_rdev *rdev_p, u32 stag)
{
- return __cxio_tpt_op(rdev_p, 1, &stag, 0, 0, 0, 0, 0, 0ULL, 0, 0, NULL,
- NULL, NULL);
+ return __cxio_tpt_op(rdev_p, 1, &stag, 0, 0, 0, 0, 0, 0ULL, 0, 0,
+ 0, 0);
}
int cxio_rdma_init(struct cxio_rdev *rdev_p, struct t3_rdma_init_attr *attr)
int cxio_destroy_qp(struct cxio_rdev *rdev, struct t3_wq *wq,
struct cxio_ucontext *uctx);
int cxio_peek_cq(struct t3_wq *wr, struct t3_cq *cq, int opcode);
+int cxio_write_pbl(struct cxio_rdev *rdev_p, __be64 *pbl,
+ u32 pbl_addr, u32 pbl_size);
int cxio_register_phys_mem(struct cxio_rdev *rdev, u32 * stag, u32 pdid,
enum tpt_mem_perm perm, u32 zbva, u64 to, u32 len,
- u8 page_size, __be64 *pbl, u32 *pbl_size,
- u32 *pbl_addr);
+ u8 page_size, u32 pbl_size, u32 pbl_addr);
int cxio_reregister_phys_mem(struct cxio_rdev *rdev, u32 * stag, u32 pdid,
enum tpt_mem_perm perm, u32 zbva, u64 to, u32 len,
- u8 page_size, __be64 *pbl, u32 *pbl_size,
- u32 *pbl_addr);
+ u8 page_size, u32 pbl_size, u32 pbl_addr);
int cxio_dereg_mem(struct cxio_rdev *rdev, u32 stag, u32 pbl_size,
u32 pbl_addr);
int cxio_allocate_window(struct cxio_rdev *rdev, u32 * stag, u32 pdid);
void cxio_hal_put_pdid(struct cxio_hal_resource *rscp, u32 pdid);
int __init cxio_hal_init(void);
void __exit cxio_hal_exit(void);
-void cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count);
-void cxio_flush_sq(struct t3_wq *wq, struct t3_cq *cq, int count);
+int cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count);
+int cxio_flush_sq(struct t3_wq *wq, struct t3_cq *cq, int count);
void cxio_count_rcqes(struct t3_cq *cq, struct t3_wq *wq, int *count);
void cxio_count_scqes(struct t3_cq *cq, struct t3_wq *wq, int *count);
void cxio_flush_hw_cq(struct t3_cq *cq);
*/
#define MIN_PBL_SHIFT 8 /* 256B == min PBL size (32 entries) */
-#define PBL_CHUNK 2*1024*1024
u32 cxio_hal_pblpool_alloc(struct cxio_rdev *rdev_p, int size)
{
int cxio_hal_pblpool_create(struct cxio_rdev *rdev_p)
{
- unsigned long i;
+ unsigned pbl_start, pbl_chunk;
+
rdev_p->pbl_pool = gen_pool_create(MIN_PBL_SHIFT, -1);
- if (rdev_p->pbl_pool)
- for (i = rdev_p->rnic_info.pbl_base;
- i <= rdev_p->rnic_info.pbl_top - PBL_CHUNK + 1;
- i += PBL_CHUNK)
- gen_pool_add(rdev_p->pbl_pool, i, PBL_CHUNK, -1);
- return rdev_p->pbl_pool ? 0 : -ENOMEM;
+ if (!rdev_p->pbl_pool)
+ return -ENOMEM;
+
+ pbl_start = rdev_p->rnic_info.pbl_base;
+ pbl_chunk = rdev_p->rnic_info.pbl_top - pbl_start + 1;
+
+ while (pbl_start < rdev_p->rnic_info.pbl_top) {
+ pbl_chunk = min(rdev_p->rnic_info.pbl_top - pbl_start + 1,
+ pbl_chunk);
+ if (gen_pool_add(rdev_p->pbl_pool, pbl_start, pbl_chunk, -1)) {
+ PDBG("%s failed to add PBL chunk (%x/%x)\n",
+ __func__, pbl_start, pbl_chunk);
+ if (pbl_chunk <= 1024 << MIN_PBL_SHIFT) {
+ printk(KERN_WARNING MOD "%s: Failed to add all PBL chunks (%x/%x)\n",
+ __func__, pbl_start, rdev_p->rnic_info.pbl_top - pbl_start);
+ return 0;
+ }
+ pbl_chunk >>= 1;
+ } else {
+ PDBG("%s added PBL chunk (%x/%x)\n",
+ __func__, pbl_start, pbl_chunk);
+ pbl_start += pbl_chunk;
+ }
+ }
+
+ return 0;
}
void cxio_hal_pblpool_destroy(struct cxio_rdev *rdev_p)
module_param(peer2peer, int, 0644);
MODULE_PARM_DESC(peer2peer, "Support peer2peer ULPs (default=0)");
-static int ep_timeout_secs = 10;
+static int ep_timeout_secs = 60;
module_param(ep_timeout_secs, int, 0644);
MODULE_PARM_DESC(ep_timeout_secs, "CM Endpoint operation timeout "
- "in seconds (default=10)");
+ "in seconds (default=60)");
static int mpa_rev = 1;
module_param(mpa_rev, int, 0644);
release = 1;
break;
case ABORTING:
- break;
case DEAD:
+ break;
default:
BUG_ON(1);
break;
#include <rdma/ib_verbs.h>
#include "cxio_hal.h"
+#include "cxio_resource.h"
#include "iwch.h"
#include "iwch_provider.h"
-int iwch_register_mem(struct iwch_dev *rhp, struct iwch_pd *php,
- struct iwch_mr *mhp,
- int shift,
- __be64 *page_list)
+static void iwch_finish_mem_reg(struct iwch_mr *mhp, u32 stag)
{
- u32 stag;
u32 mmid;
+ mhp->attr.state = 1;
+ mhp->attr.stag = stag;
+ mmid = stag >> 8;
+ mhp->ibmr.rkey = mhp->ibmr.lkey = stag;
+ insert_handle(mhp->rhp, &mhp->rhp->mmidr, mhp, mmid);
+ PDBG("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp);
+}
+
+int iwch_register_mem(struct iwch_dev *rhp, struct iwch_pd *php,
+ struct iwch_mr *mhp, int shift)
+{
+ u32 stag;
if (cxio_register_phys_mem(&rhp->rdev,
&stag, mhp->attr.pdid,
mhp->attr.zbva,
mhp->attr.va_fbo,
mhp->attr.len,
- shift-12,
- page_list,
- &mhp->attr.pbl_size, &mhp->attr.pbl_addr))
+ shift - 12,
+ mhp->attr.pbl_size, mhp->attr.pbl_addr))
return -ENOMEM;
- mhp->attr.state = 1;
- mhp->attr.stag = stag;
- mmid = stag >> 8;
- mhp->ibmr.rkey = mhp->ibmr.lkey = stag;
- insert_handle(rhp, &rhp->mmidr, mhp, mmid);
- PDBG("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp);
+
+ iwch_finish_mem_reg(mhp, stag);
+
return 0;
}
int iwch_reregister_mem(struct iwch_dev *rhp, struct iwch_pd *php,
struct iwch_mr *mhp,
int shift,
- __be64 *page_list,
int npages)
{
u32 stag;
- u32 mmid;
-
/* We could support this... */
if (npages > mhp->attr.pbl_size)
mhp->attr.zbva,
mhp->attr.va_fbo,
mhp->attr.len,
- shift-12,
- page_list,
- &mhp->attr.pbl_size, &mhp->attr.pbl_addr))
+ shift - 12,
+ mhp->attr.pbl_size, mhp->attr.pbl_addr))
return -ENOMEM;
- mhp->attr.state = 1;
- mhp->attr.stag = stag;
- mmid = stag >> 8;
- mhp->ibmr.rkey = mhp->ibmr.lkey = stag;
- insert_handle(rhp, &rhp->mmidr, mhp, mmid);
- PDBG("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp);
+
+ iwch_finish_mem_reg(mhp, stag);
+
+ return 0;
+}
+
+int iwch_alloc_pbl(struct iwch_mr *mhp, int npages)
+{
+ mhp->attr.pbl_addr = cxio_hal_pblpool_alloc(&mhp->rhp->rdev,
+ npages << 3);
+
+ if (!mhp->attr.pbl_addr)
+ return -ENOMEM;
+
+ mhp->attr.pbl_size = npages;
+
return 0;
}
+void iwch_free_pbl(struct iwch_mr *mhp)
+{
+ cxio_hal_pblpool_free(&mhp->rhp->rdev, mhp->attr.pbl_addr,
+ mhp->attr.pbl_size << 3);
+}
+
+int iwch_write_pbl(struct iwch_mr *mhp, __be64 *pages, int npages, int offset)
+{
+ return cxio_write_pbl(&mhp->rhp->rdev, pages,
+ mhp->attr.pbl_addr + (offset << 3), npages);
+}
+
int build_phys_page_list(struct ib_phys_buf *buffer_list,
int num_phys_buf,
u64 *iova_start,
mmid = mhp->attr.stag >> 8;
cxio_dereg_mem(&rhp->rdev, mhp->attr.stag, mhp->attr.pbl_size,
mhp->attr.pbl_addr);
+ iwch_free_pbl(mhp);
remove_handle(rhp, &rhp->mmidr, mmid);
if (mhp->kva)
kfree((void *) (unsigned long) mhp->kva);
if (!mhp)
return ERR_PTR(-ENOMEM);
+ mhp->rhp = rhp;
+
/* First check that we have enough alignment */
if ((*iova_start & ~PAGE_MASK) != (buffer_list[0].addr & ~PAGE_MASK)) {
ret = -EINVAL;
if (ret)
goto err;
- mhp->rhp = rhp;
+ ret = iwch_alloc_pbl(mhp, npages);
+ if (ret) {
+ kfree(page_list);
+ goto err_pbl;
+ }
+
+ ret = iwch_write_pbl(mhp, page_list, npages, 0);
+ kfree(page_list);
+ if (ret)
+ goto err_pbl;
+
mhp->attr.pdid = php->pdid;
mhp->attr.zbva = 0;
mhp->attr.len = (u32) total_size;
mhp->attr.pbl_size = npages;
- ret = iwch_register_mem(rhp, php, mhp, shift, page_list);
- kfree(page_list);
- if (ret) {
- goto err;
- }
+ ret = iwch_register_mem(rhp, php, mhp, shift);
+ if (ret)
+ goto err_pbl;
+
return &mhp->ibmr;
+
+err_pbl:
+ iwch_free_pbl(mhp);
+
err:
kfree(mhp);
return ERR_PTR(ret);
return ret;
}
- ret = iwch_reregister_mem(rhp, php, &mh, shift, page_list, npages);
+ ret = iwch_reregister_mem(rhp, php, &mh, shift, npages);
kfree(page_list);
if (ret) {
return ret;
if (!mhp)
return ERR_PTR(-ENOMEM);
+ mhp->rhp = rhp;
+
mhp->umem = ib_umem_get(pd->uobject->context, start, length, acc, 0);
if (IS_ERR(mhp->umem)) {
err = PTR_ERR(mhp->umem);
list_for_each_entry(chunk, &mhp->umem->chunk_list, list)
n += chunk->nents;
- pages = kmalloc(n * sizeof(u64), GFP_KERNEL);
+ err = iwch_alloc_pbl(mhp, n);
+ if (err)
+ goto err;
+
+ pages = (__be64 *) __get_free_page(GFP_KERNEL);
if (!pages) {
err = -ENOMEM;
- goto err;
+ goto err_pbl;
}
i = n = 0;
pages[i++] = cpu_to_be64(sg_dma_address(
&chunk->page_list[j]) +
mhp->umem->page_size * k);
+ if (i == PAGE_SIZE / sizeof *pages) {
+ err = iwch_write_pbl(mhp, pages, i, n);
+ if (err)
+ goto pbl_done;
+ n += i;
+ i = 0;
+ }
}
}
- mhp->rhp = rhp;
+ if (i)
+ err = iwch_write_pbl(mhp, pages, i, n);
+
+pbl_done:
+ free_page((unsigned long) pages);
+ if (err)
+ goto err_pbl;
+
mhp->attr.pdid = php->pdid;
mhp->attr.zbva = 0;
mhp->attr.perms = iwch_ib_to_tpt_access(acc);
mhp->attr.va_fbo = virt;
mhp->attr.page_size = shift - 12;
mhp->attr.len = (u32) length;
- mhp->attr.pbl_size = i;
- err = iwch_register_mem(rhp, php, mhp, shift, pages);
- kfree(pages);
+
+ err = iwch_register_mem(rhp, php, mhp, shift);
if (err)
- goto err;
+ goto err_pbl;
if (udata && !t3a_device(rhp)) {
uresp.pbl_addr = (mhp->attr.pbl_addr -
- rhp->rdev.rnic_info.pbl_base) >> 3;
+ rhp->rdev.rnic_info.pbl_base) >> 3;
PDBG("%s user resp pbl_addr 0x%x\n", __func__,
uresp.pbl_addr);
return &mhp->ibmr;
+err_pbl:
+ iwch_free_pbl(mhp);
+
err:
ib_umem_release(mhp->umem);
kfree(mhp);
int iwch_resume_qps(struct iwch_cq *chp);
void stop_read_rep_timer(struct iwch_qp *qhp);
int iwch_register_mem(struct iwch_dev *rhp, struct iwch_pd *php,
- struct iwch_mr *mhp,
- int shift,
- __be64 *page_list);
+ struct iwch_mr *mhp, int shift);
int iwch_reregister_mem(struct iwch_dev *rhp, struct iwch_pd *php,
struct iwch_mr *mhp,
int shift,
- __be64 *page_list,
int npages);
+int iwch_alloc_pbl(struct iwch_mr *mhp, int npages);
+void iwch_free_pbl(struct iwch_mr *mhp);
+int iwch_write_pbl(struct iwch_mr *mhp, __be64 *pages, int npages, int offset);
int build_phys_page_list(struct ib_phys_buf *buffer_list,
int num_phys_buf,
u64 *iova_start,
{
struct iwch_cq *rchp, *schp;
int count;
+ int flushed;
rchp = get_chp(qhp->rhp, qhp->attr.rcq);
schp = get_chp(qhp->rhp, qhp->attr.scq);
spin_lock(&qhp->lock);
cxio_flush_hw_cq(&rchp->cq);
cxio_count_rcqes(&rchp->cq, &qhp->wq, &count);
- cxio_flush_rq(&qhp->wq, &rchp->cq, count);
+ flushed = cxio_flush_rq(&qhp->wq, &rchp->cq, count);
spin_unlock(&qhp->lock);
spin_unlock_irqrestore(&rchp->lock, *flag);
- (*rchp->ibcq.comp_handler)(&rchp->ibcq, rchp->ibcq.cq_context);
+ if (flushed)
+ (*rchp->ibcq.comp_handler)(&rchp->ibcq, rchp->ibcq.cq_context);
/* locking heirarchy: cq lock first, then qp lock. */
spin_lock_irqsave(&schp->lock, *flag);
spin_lock(&qhp->lock);
cxio_flush_hw_cq(&schp->cq);
cxio_count_scqes(&schp->cq, &qhp->wq, &count);
- cxio_flush_sq(&qhp->wq, &schp->cq, count);
+ flushed = cxio_flush_sq(&qhp->wq, &schp->cq, count);
spin_unlock(&qhp->lock);
spin_unlock_irqrestore(&schp->lock, *flag);
- (*schp->ibcq.comp_handler)(&schp->ibcq, schp->ibcq.cq_context);
+ if (flushed)
+ (*schp->ibcq.comp_handler)(&schp->ibcq, schp->ibcq.cq_context);
/* deref */
if (atomic_dec_and_test(&qhp->refcnt))
ep = qhp->ep;
get_ep(&ep->com);
}
- flush_qp(qhp, &flag);
break;
case IWCH_QP_STATE_TERMINATE:
qhp->attr.state = IWCH_QP_STATE_TERMINATE;
}
switch (attrs->next_state) {
case IWCH_QP_STATE_IDLE:
+ flush_qp(qhp, &flag);
qhp->attr.state = IWCH_QP_STATE_IDLE;
qhp->attr.llp_stream_handle = NULL;
put_ep(&qhp->ep->com);
int mtu_shift;
u32 message_count;
u32 packet_count;
+ atomic_t nr_events; /* events seen */
+ wait_queue_head_t wait_completion;
};
#define IS_SRQ(qp) (qp->ext_type == EQPT_SRQ)
props->max_ee = limit_uint(rblock->max_rd_ee_context);
props->max_rdd = limit_uint(rblock->max_rd_domain);
props->max_fmr = limit_uint(rblock->max_mr);
- props->local_ca_ack_delay = limit_uint(rblock->local_ca_ack_delay);
props->max_qp_rd_atom = limit_uint(rblock->max_rr_qp);
props->max_ee_rd_atom = limit_uint(rblock->max_rr_ee_context);
props->max_res_rd_atom = limit_uint(rblock->max_rr_hca);
}
props->max_pkeys = 16;
- props->local_ca_ack_delay = limit_uint(rblock->local_ca_ack_delay);
+ props->local_ca_ack_delay = min_t(u8, rblock->local_ca_ack_delay, 255);
props->max_raw_ipv6_qp = limit_uint(rblock->max_raw_ipv6_qp);
props->max_raw_ethy_qp = limit_uint(rblock->max_raw_ethy_qp);
props->max_mcast_grp = limit_uint(rblock->max_mcast_grp);
return ret;
}
-static int map_mtu(struct ehca_shca *shca, u32 fw_mtu)
+static enum ib_mtu map_mtu(struct ehca_shca *shca, u32 fw_mtu)
{
switch (fw_mtu) {
case 0x1:
}
}
-static int map_number_of_vls(struct ehca_shca *shca, u32 vl_cap)
+static u8 map_number_of_vls(struct ehca_shca *shca, u32 vl_cap)
{
switch (vl_cap) {
case 0x1:
read_lock(&ehca_qp_idr_lock);
qp = idr_find(&ehca_qp_idr, token);
+ if (qp)
+ atomic_inc(&qp->nr_events);
read_unlock(&ehca_qp_idr_lock);
if (!qp)
if (fatal && qp->ext_type == EQPT_SRQBASE)
dispatch_qp_event(shca, qp, IB_EVENT_QP_LAST_WQE_REACHED);
+ if (atomic_dec_and_test(&qp->nr_events))
+ wake_up(&qp->wait_completion);
return;
}
return ERR_PTR(-ENOMEM);
}
+ atomic_set(&my_qp->nr_events, 0);
+ init_waitqueue_head(&my_qp->wait_completion);
spin_lock_init(&my_qp->spinlock_s);
spin_lock_init(&my_qp->spinlock_r);
my_qp->qp_type = qp_type;
idr_remove(&ehca_qp_idr, my_qp->token);
write_unlock_irqrestore(&ehca_qp_idr_lock, flags);
+ /* now wait until all pending events have completed */
+ wait_event(my_qp->wait_completion, !atomic_read(&my_qp->nr_events));
+
h_ret = hipz_h_destroy_qp(shca->ipz_hca_handle, my_qp);
if (h_ret != H_SUCCESS) {
ehca_err(dev, "hipz_h_destroy_qp() failed h_ret=%li "
}
reloop:
- for (last = 0, i = 1; !last; i++) {
+ for (last = 0, i = 1; !last; i += !last) {
hdr = dd->ipath_f_get_msgheader(dd, rhf_addr);
eflags = ipath_hdrget_err_flags(rhf_addr);
etype = ipath_hdrget_rcv_type(rhf_addr);
spin_unlock_irqrestore(&ipath_pioavail_lock, flags);
}
+/*
+ * used to force update of pioavailshadow if we can't get a pio buffer.
+ * Needed primarily due to exitting freeze mode after recovering
+ * from errors. Done lazily, because it's safer (known to not
+ * be writing pio buffers).
+ */
+static void ipath_reset_availshadow(struct ipath_devdata *dd)
+{
+ int i, im;
+ unsigned long flags;
+
+ spin_lock_irqsave(&ipath_pioavail_lock, flags);
+ for (i = 0; i < dd->ipath_pioavregs; i++) {
+ u64 val, oldval;
+ /* deal with 6110 chip bug on high register #s */
+ im = (i > 3 && (dd->ipath_flags & IPATH_SWAP_PIOBUFS)) ?
+ i ^ 1 : i;
+ val = le64_to_cpu(dd->ipath_pioavailregs_dma[im]);
+ /*
+ * busy out the buffers not in the kernel avail list,
+ * without changing the generation bits.
+ */
+ oldval = dd->ipath_pioavailshadow[i];
+ dd->ipath_pioavailshadow[i] = val |
+ ((~dd->ipath_pioavailkernel[i] <<
+ INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT) &
+ 0xaaaaaaaaaaaaaaaaULL); /* All BUSY bits in qword */
+ if (oldval != dd->ipath_pioavailshadow[i])
+ ipath_dbg("shadow[%d] was %Lx, now %lx\n",
+ i, oldval, dd->ipath_pioavailshadow[i]);
+ }
+ spin_unlock_irqrestore(&ipath_pioavail_lock, flags);
+}
+
/**
* ipath_setrcvhdrsize - set the receive header size
* @dd: the infinipath device
*/
ipath_stats.sps_nopiobufs++;
if (!(++dd->ipath_consec_nopiobuf % 100000)) {
- ipath_dbg("%u pio sends with no bufavail; dmacopy: "
- "%llx %llx %llx %llx; shadow: %lx %lx %lx %lx\n",
+ ipath_force_pio_avail_update(dd); /* at start */
+ ipath_dbg("%u tries no piobufavail ts%lx; dmacopy: "
+ "%llx %llx %llx %llx\n"
+ "ipath shadow: %lx %lx %lx %lx\n",
dd->ipath_consec_nopiobuf,
+ (unsigned long)get_cycles(),
(unsigned long long) le64_to_cpu(dma[0]),
(unsigned long long) le64_to_cpu(dma[1]),
(unsigned long long) le64_to_cpu(dma[2]),
*/
if ((dd->ipath_piobcnt2k + dd->ipath_piobcnt4k) >
(sizeof(shadow[0]) * 4 * 4))
- ipath_dbg("2nd group: dmacopy: %llx %llx "
- "%llx %llx; shadow: %lx %lx %lx %lx\n",
+ ipath_dbg("2nd group: dmacopy: "
+ "%llx %llx %llx %llx\n"
+ "ipath shadow: %lx %lx %lx %lx\n",
(unsigned long long)le64_to_cpu(dma[4]),
(unsigned long long)le64_to_cpu(dma[5]),
(unsigned long long)le64_to_cpu(dma[6]),
(unsigned long long)le64_to_cpu(dma[7]),
- shadow[4], shadow[5], shadow[6],
- shadow[7]);
+ shadow[4], shadow[5], shadow[6], shadow[7]);
+
+ /* at end, so update likely happened */
+ ipath_reset_availshadow(dd);
}
}
unsigned len, int avail)
{
unsigned long flags;
- unsigned end;
+ unsigned end, cnt = 0, next;
/* There are two bits per send buffer (busy and generation) */
start *= 2;
- len *= 2;
- end = start + len;
+ end = start + len * 2;
- /* Set or clear the generation bits. */
spin_lock_irqsave(&ipath_pioavail_lock, flags);
+ /* Set or clear the busy bit in the shadow. */
while (start < end) {
if (avail) {
- __clear_bit(start + INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT,
- dd->ipath_pioavailshadow);
+ unsigned long dma;
+ int i, im;
+ /*
+ * the BUSY bit will never be set, because we disarm
+ * the user buffers before we hand them back to the
+ * kernel. We do have to make sure the generation
+ * bit is set correctly in shadow, since it could
+ * have changed many times while allocated to user.
+ * We can't use the bitmap functions on the full
+ * dma array because it is always little-endian, so
+ * we have to flip to host-order first.
+ * BITS_PER_LONG is slightly wrong, since it's
+ * always 64 bits per register in chip...
+ * We only work on 64 bit kernels, so that's OK.
+ */
+ /* deal with 6110 chip bug on high register #s */
+ i = start / BITS_PER_LONG;
+ im = (i > 3 && (dd->ipath_flags & IPATH_SWAP_PIOBUFS)) ?
+ i ^ 1 : i;
+ __clear_bit(INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT
+ + start, dd->ipath_pioavailshadow);
+ dma = (unsigned long) le64_to_cpu(
+ dd->ipath_pioavailregs_dma[im]);
+ if (test_bit((INFINIPATH_SENDPIOAVAIL_CHECK_SHIFT
+ + start) % BITS_PER_LONG, &dma))
+ __set_bit(INFINIPATH_SENDPIOAVAIL_CHECK_SHIFT
+ + start, dd->ipath_pioavailshadow);
+ else
+ __clear_bit(INFINIPATH_SENDPIOAVAIL_CHECK_SHIFT
+ + start, dd->ipath_pioavailshadow);
__set_bit(start, dd->ipath_pioavailkernel);
} else {
__set_bit(start + INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT,
}
start += 2;
}
+
+ if (dd->ipath_pioupd_thresh) {
+ end = 2 * (dd->ipath_piobcnt2k + dd->ipath_piobcnt4k);
+ next = find_first_bit(dd->ipath_pioavailkernel, end);
+ while (next < end) {
+ cnt++;
+ next = find_next_bit(dd->ipath_pioavailkernel, end,
+ next + 1);
+ }
+ }
spin_unlock_irqrestore(&ipath_pioavail_lock, flags);
+
+ /*
+ * When moving buffers from kernel to user, if number assigned to
+ * the user is less than the pio update threshold, and threshold
+ * is supported (cnt was computed > 0), drop the update threshold
+ * so we update at least once per allocated number of buffers.
+ * In any case, if the kernel buffers are less than the threshold,
+ * drop the threshold. We don't bother increasing it, having once
+ * decreased it, since it would typically just cycle back and forth.
+ * If we don't decrease below buffers in use, we can wait a long
+ * time for an update, until some other context uses PIO buffers.
+ */
+ if (!avail && len < cnt)
+ cnt = len;
+ if (cnt < dd->ipath_pioupd_thresh) {
+ dd->ipath_pioupd_thresh = cnt;
+ ipath_dbg("Decreased pio update threshold to %u\n",
+ dd->ipath_pioupd_thresh);
+ spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags);
+ dd->ipath_sendctrl &= ~(INFINIPATH_S_UPDTHRESH_MASK
+ << INFINIPATH_S_UPDTHRESH_SHIFT);
+ dd->ipath_sendctrl |= dd->ipath_pioupd_thresh
+ << INFINIPATH_S_UPDTHRESH_SHIFT;
+ ipath_write_kreg(dd, dd->ipath_kregs->kr_sendctrl,
+ dd->ipath_sendctrl);
+ spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags);
+ }
}
/**
spin_lock_irqsave(&dd->ipath_sdma_lock, flags);
skip_cancel =
- !test_bit(IPATH_SDMA_DISABLED, statp) &&
- test_and_set_bit(IPATH_SDMA_ABORTING, statp);
+ test_and_set_bit(IPATH_SDMA_ABORTING, statp)
+ && !test_bit(IPATH_SDMA_DISABLED, statp);
spin_unlock_irqrestore(&dd->ipath_sdma_lock, flags);
if (skip_cancel)
goto bail;
ipath_disarm_piobufs(dd, 0,
dd->ipath_piobcnt2k + dd->ipath_piobcnt4k);
+ if (dd->ipath_flags & IPATH_HAS_SEND_DMA)
+ set_bit(IPATH_SDMA_DISARMED, &dd->ipath_sdma_status);
+
if (restore_sendctrl) {
/* else done by caller later if needed */
spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags);
/* only wait so long for intr */
dd->ipath_sdma_abort_intr_timeout = jiffies + HZ;
dd->ipath_sdma_reset_wait = 200;
- __set_bit(IPATH_SDMA_DISARMED, &dd->ipath_sdma_status);
if (!test_bit(IPATH_SDMA_SHUTDOWN, &dd->ipath_sdma_status))
tasklet_hi_schedule(&dd->ipath_sdma_abort_task);
spin_unlock_irqrestore(&dd->ipath_sdma_lock, flags);
(void *) dd->ipath_statusp -
(void *) dd->ipath_pioavailregs_dma;
if (!shared) {
- kinfo->spi_piocnt = dd->ipath_pbufsport;
+ kinfo->spi_piocnt = pd->port_piocnt;
kinfo->spi_piobufbase = (u64) pd->port_piobufs;
kinfo->__spi_uregbase = (u64) dd->ipath_uregbase +
dd->ipath_ureg_align * pd->port_port;
} else if (master) {
- kinfo->spi_piocnt = (dd->ipath_pbufsport / subport_cnt) +
- (dd->ipath_pbufsport % subport_cnt);
+ kinfo->spi_piocnt = (pd->port_piocnt / subport_cnt) +
+ (pd->port_piocnt % subport_cnt);
/* Master's PIO buffers are after all the slave's */
kinfo->spi_piobufbase = (u64) pd->port_piobufs +
dd->ipath_palign *
- (dd->ipath_pbufsport - kinfo->spi_piocnt);
+ (pd->port_piocnt - kinfo->spi_piocnt);
} else {
unsigned slave = subport_fp(fp) - 1;
- kinfo->spi_piocnt = dd->ipath_pbufsport / subport_cnt;
+ kinfo->spi_piocnt = pd->port_piocnt / subport_cnt;
kinfo->spi_piobufbase = (u64) pd->port_piobufs +
dd->ipath_palign * kinfo->spi_piocnt * slave;
}
- /*
- * Set the PIO avail update threshold to no larger
- * than the number of buffers per process. Note that
- * we decrease it here, but won't ever increase it.
- */
- if (dd->ipath_pioupd_thresh &&
- kinfo->spi_piocnt < dd->ipath_pioupd_thresh) {
- unsigned long flags;
-
- dd->ipath_pioupd_thresh = kinfo->spi_piocnt;
- ipath_dbg("Decreased pio update threshold to %u\n",
- dd->ipath_pioupd_thresh);
- spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags);
- dd->ipath_sendctrl &= ~(INFINIPATH_S_UPDTHRESH_MASK
- << INFINIPATH_S_UPDTHRESH_SHIFT);
- dd->ipath_sendctrl |= dd->ipath_pioupd_thresh
- << INFINIPATH_S_UPDTHRESH_SHIFT;
- ipath_write_kreg(dd, dd->ipath_kregs->kr_sendctrl,
- dd->ipath_sendctrl);
- spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags);
- }
-
if (shared) {
kinfo->spi_port_uregbase = (u64) dd->ipath_uregbase +
dd->ipath_ureg_align * pd->port_port;
ureg = dd->ipath_uregbase + dd->ipath_ureg_align * pd->port_port;
if (!pd->port_subport_cnt) {
/* port is not shared */
- piocnt = dd->ipath_pbufsport;
+ piocnt = pd->port_piocnt;
piobufs = pd->port_piobufs;
} else if (!subport_fp(fp)) {
/* caller is the master */
- piocnt = (dd->ipath_pbufsport / pd->port_subport_cnt) +
- (dd->ipath_pbufsport % pd->port_subport_cnt);
+ piocnt = (pd->port_piocnt / pd->port_subport_cnt) +
+ (pd->port_piocnt % pd->port_subport_cnt);
piobufs = pd->port_piobufs +
- dd->ipath_palign * (dd->ipath_pbufsport - piocnt);
+ dd->ipath_palign * (pd->port_piocnt - piocnt);
} else {
unsigned slave = subport_fp(fp) - 1;
/* caller is a slave */
- piocnt = dd->ipath_pbufsport / pd->port_subport_cnt;
+ piocnt = pd->port_piocnt / pd->port_subport_cnt;
piobufs = pd->port_piobufs + dd->ipath_palign * piocnt * slave;
}
port_fp(fp) = pd;
pd->port_pid = current->pid;
strncpy(pd->port_comm, current->comm, sizeof(pd->port_comm));
- ipath_chg_pioavailkernel(dd,
- dd->ipath_pbufsport * (pd->port_port - 1),
- dd->ipath_pbufsport, 0);
ipath_stats.sps_ports++;
ret = 0;
} else
/* for now we do nothing with rcvhdrcnt: uinfo->spu_rcvhdrcnt */
+ /* some ports may get extra buffers, calculate that here */
+ if (pd->port_port <= dd->ipath_ports_extrabuf)
+ pd->port_piocnt = dd->ipath_pbufsport + 1;
+ else
+ pd->port_piocnt = dd->ipath_pbufsport;
+
/* for right now, kernel piobufs are at end, so port 1 is at 0 */
+ if (pd->port_port <= dd->ipath_ports_extrabuf)
+ pd->port_pio_base = (dd->ipath_pbufsport + 1)
+ * (pd->port_port - 1);
+ else
+ pd->port_pio_base = dd->ipath_ports_extrabuf +
+ dd->ipath_pbufsport * (pd->port_port - 1);
pd->port_piobufs = dd->ipath_piobufbase +
- dd->ipath_pbufsport * (pd->port_port - 1) * dd->ipath_palign;
- ipath_cdbg(VERBOSE, "Set base of piobufs for port %u to 0x%x\n",
- pd->port_port, pd->port_piobufs);
+ pd->port_pio_base * dd->ipath_palign;
+ ipath_cdbg(VERBOSE, "piobuf base for port %u is 0x%x, piocnt %u,"
+ " first pio %u\n", pd->port_port, pd->port_piobufs,
+ pd->port_piocnt, pd->port_pio_base);
+ ipath_chg_pioavailkernel(dd, pd->port_pio_base, pd->port_piocnt, 0);
/*
* Now allocate the rcvhdr Q and eager TIDs; skip the TID
}
if (dd->ipath_kregbase) {
- int i;
/* atomically clear receive enable port and intr avail. */
clear_bit(dd->ipath_r_portenable_shift + port,
&dd->ipath_rcvctrl);
ipath_write_kreg_port(dd, dd->ipath_kregs->kr_rcvhdraddr,
pd->port_port, dd->ipath_dummy_hdrq_phys);
- i = dd->ipath_pbufsport * (port - 1);
- ipath_disarm_piobufs(dd, i, dd->ipath_pbufsport);
- ipath_chg_pioavailkernel(dd, i, dd->ipath_pbufsport, 1);
+ ipath_disarm_piobufs(dd, pd->port_pio_base, pd->port_piocnt);
+ ipath_chg_pioavailkernel(dd, pd->port_pio_base,
+ pd->port_piocnt, 1);
dd->ipath_f_clear_tids(dd, pd->port_port);
dev_info(&dd->pcidev->dev,
"Recovering from TXE PIO parity error\n");
- ipath_disarm_senderrbufs(dd, 1);
+ ipath_disarm_senderrbufs(dd);
}
ctrl = ipath_read_kreg32(dd, dd->ipath_kregs->kr_control);
if ((ctrl & INFINIPATH_C_FREEZEMODE) && !ipath_diag_inuse) {
/*
- * Parity errors in send memory are recoverable,
- * just cancel the send (if indicated in * sendbuffererror),
- * count the occurrence, unfreeze (if no other handled
- * hardware error bits are set), and continue.
+ * Parity errors in send memory are recoverable by h/w
+ * just do housekeeping, exit freeze mode and continue.
*/
if (hwerrs & ((INFINIPATH_HWE_TXEMEMPARITYERR_PIOBUF |
INFINIPATH_HWE_TXEMEMPARITYERR_PIOPBC)
hwerrs &= ~((INFINIPATH_HWE_TXEMEMPARITYERR_PIOBUF |
INFINIPATH_HWE_TXEMEMPARITYERR_PIOPBC)
<< INFINIPATH_HWE_TXEMEMPARITYERR_SHIFT);
- if (!hwerrs) {
- /* else leave in freeze mode */
- ipath_write_kreg(dd,
- dd->ipath_kregs->kr_control,
- dd->ipath_control);
- goto bail;
- }
}
if (hwerrs) {
/*
*dd->ipath_statusp |= IPATH_STATUS_HWERROR;
dd->ipath_flags &= ~IPATH_INITTED;
} else {
- ipath_dbg("Clearing freezemode on ignored hardware "
- "error\n");
+ ipath_dbg("Clearing freezemode on ignored or "
+ "recovered hardware error\n");
ipath_clear_freeze(dd);
}
}
"revision %u.%u!\n",
dd->ipath_majrev, dd->ipath_minrev);
ret = 1;
- } else if (dd->ipath_minrev == 1) {
- /* Rev1 chips are prototype. Complain, but allow use */
+ } else if (dd->ipath_minrev == 1 &&
+ !(dd->ipath_flags & IPATH_INITTED)) {
+ /* Rev1 chips are prototype. Complain at init, but allow use */
ipath_dev_err(dd, "Unsupported hardware "
"revision %u.%u, Contact support@qlogic.com\n",
dd->ipath_majrev, dd->ipath_minrev);
dd->ipath_rcvctrl);
dd->ipath_p0_rcvegrcnt = 2048; /* always */
if (dd->ipath_flags & IPATH_HAS_SEND_DMA)
- dd->ipath_pioreserved = 1; /* reserve a buffer */
+ dd->ipath_pioreserved = 3; /* kpiobufs used for PIO */
}
/*
* min buffers we want to have per port, after driver
*/
-#define IPATH_MIN_USER_PORT_BUFCNT 8
+#define IPATH_MIN_USER_PORT_BUFCNT 7
/*
* Number of ports we are configured to use (to allow for more pio
/*
* Number of buffers reserved for driver (verbs and layered drivers.)
- * Reserved at end of buffer list. Initialized based on
- * number of PIO buffers if not set via module interface.
+ * Initialized based on number of PIO buffers if not set via module interface.
* The problem with this is that it's global, but we'll use different
- * numbers for different chip types. So the default value is not
- * very useful. I've redefined it for the 1.3 release so that it's
- * zero unless set by the user to something else, in which case we
- * try to respect it.
+ * numbers for different chip types.
*/
static ushort ipath_kpiobufs;
pioavail = dd->ipath_pioavailregs_dma[i ^ 1];
else
pioavail = dd->ipath_pioavailregs_dma[i];
- dd->ipath_pioavailshadow[i] = le64_to_cpu(pioavail) |
- (~dd->ipath_pioavailkernel[i] <<
- INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT);
+ /*
+ * don't need to worry about ipath_pioavailkernel here
+ * because we will call ipath_chg_pioavailkernel() later
+ * in initialization, to busy out buffers as needed
+ */
+ dd->ipath_pioavailshadow[i] = le64_to_cpu(pioavail);
}
/* can get counters, stats, etc. */
dd->ipath_flags |= IPATH_PRESENT;
int ipath_init_chip(struct ipath_devdata *dd, int reinit)
{
int ret = 0;
- u32 val32, kpiobufs;
+ u32 kpiobufs, defkbufs;
u32 piobufs, uports;
u64 val;
struct ipath_portdata *pd;
gfp_t gfp_flags = GFP_USER | __GFP_COMP;
- unsigned long flags;
ret = init_housekeeping(dd, reinit);
if (ret)
dd->ipath_pioavregs = ALIGN(piobufs, sizeof(u64) * BITS_PER_BYTE / 2)
/ (sizeof(u64) * BITS_PER_BYTE / 2);
uports = dd->ipath_cfgports ? dd->ipath_cfgports - 1 : 0;
- if (ipath_kpiobufs == 0) {
- /* not set by user (this is default) */
- if (piobufs > 144)
- kpiobufs = 32;
- else
- kpiobufs = 16;
- }
+ if (piobufs > 144)
+ defkbufs = 32 + dd->ipath_pioreserved;
else
- kpiobufs = ipath_kpiobufs;
+ defkbufs = 16 + dd->ipath_pioreserved;
- if (kpiobufs + (uports * IPATH_MIN_USER_PORT_BUFCNT) > piobufs) {
+ if (ipath_kpiobufs && (ipath_kpiobufs +
+ (uports * IPATH_MIN_USER_PORT_BUFCNT)) > piobufs) {
int i = (int) piobufs -
(int) (uports * IPATH_MIN_USER_PORT_BUFCNT);
if (i < 1)
i = 1;
dev_info(&dd->pcidev->dev, "Allocating %d PIO bufs of "
"%d for kernel leaves too few for %d user ports "
- "(%d each); using %u\n", kpiobufs,
+ "(%d each); using %u\n", ipath_kpiobufs,
piobufs, uports, IPATH_MIN_USER_PORT_BUFCNT, i);
/*
* shouldn't change ipath_kpiobufs, because could be
* different for different devices...
*/
kpiobufs = i;
- }
+ } else if (ipath_kpiobufs)
+ kpiobufs = ipath_kpiobufs;
+ else
+ kpiobufs = defkbufs;
dd->ipath_lastport_piobuf = piobufs - kpiobufs;
dd->ipath_pbufsport =
uports ? dd->ipath_lastport_piobuf / uports : 0;
- val32 = dd->ipath_lastport_piobuf - (dd->ipath_pbufsport * uports);
- if (val32 > 0) {
- ipath_dbg("allocating %u pbufs/port leaves %u unused, "
- "add to kernel\n", dd->ipath_pbufsport, val32);
- dd->ipath_lastport_piobuf -= val32;
- kpiobufs += val32;
- ipath_dbg("%u pbufs/port leaves %u unused, add to kernel\n",
- dd->ipath_pbufsport, val32);
- }
+ /* if not an even divisor, some user ports get extra buffers */
+ dd->ipath_ports_extrabuf = dd->ipath_lastport_piobuf -
+ (dd->ipath_pbufsport * uports);
+ if (dd->ipath_ports_extrabuf)
+ ipath_dbg("%u pbufs/port leaves some unused, add 1 buffer to "
+ "ports <= %u\n", dd->ipath_pbufsport,
+ dd->ipath_ports_extrabuf);
dd->ipath_lastpioindex = 0;
dd->ipath_lastpioindexl = dd->ipath_piobcnt2k;
- ipath_chg_pioavailkernel(dd, 0, piobufs, 1);
+ /* ipath_pioavailshadow initialized earlier */
ipath_cdbg(VERBOSE, "%d PIO bufs for kernel out of %d total %u "
"each for %u user ports\n", kpiobufs,
piobufs, dd->ipath_pbufsport, uports);
- if (dd->ipath_pioupd_thresh) {
- if (dd->ipath_pbufsport < dd->ipath_pioupd_thresh)
- dd->ipath_pioupd_thresh = dd->ipath_pbufsport;
- if (kpiobufs < dd->ipath_pioupd_thresh)
- dd->ipath_pioupd_thresh = kpiobufs;
- }
-
ret = dd->ipath_f_early_init(dd);
if (ret) {
ipath_dev_err(dd, "Early initialization failure\n");
goto done;
}
- /*
- * Cancel any possible active sends from early driver load.
- * Follows early_init because some chips have to initialize
- * PIO buffers in early_init to avoid false parity errors.
- */
- ipath_cancel_sends(dd, 0);
-
/*
* Early_init sets rcvhdrentsize and rcvhdrsize, so this must be
* done after early_init.
ipath_write_kreg(dd, dd->ipath_kregs->kr_sendpioavailaddr,
dd->ipath_pioavailregs_phys);
+
/*
* this is to detect s/w errors, which the h/w works around by
* ignoring the low 6 bits of address, if it wasn't aligned.
~0ULL&~INFINIPATH_HWE_MEMBISTFAILED);
ipath_write_kreg(dd, dd->ipath_kregs->kr_control, 0ULL);
- spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags);
- dd->ipath_sendctrl = INFINIPATH_S_PIOENABLE;
- ipath_write_kreg(dd, dd->ipath_kregs->kr_sendctrl, dd->ipath_sendctrl);
- ipath_read_kreg64(dd, dd->ipath_kregs->kr_scratch);
- spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags);
-
/*
* before error clears, since we expect serdes pll errors during
* this, the first time after reset
else
enable_chip(dd, reinit);
+ /* after enable_chip, so pioavailshadow setup */
+ ipath_chg_pioavailkernel(dd, 0, piobufs, 1);
+
+ /*
+ * Cancel any possible active sends from early driver load.
+ * Follows early_init because some chips have to initialize
+ * PIO buffers in early_init to avoid false parity errors.
+ * After enable and ipath_chg_pioavailkernel so we can safely
+ * enable pioavail updates and PIOENABLE; packets are now
+ * ready to go out.
+ */
+ ipath_cancel_sends(dd, 1);
+
if (!reinit) {
/*
* Used when we close a port, for DMA already in flight
#include "ipath_verbs.h"
#include "ipath_common.h"
-/*
- * clear (write) a pio buffer, to clear a parity error. This routine
- * should only be called when in freeze mode, and the buffer should be
- * canceled afterwards.
- */
-static void ipath_clrpiobuf(struct ipath_devdata *dd, u32 pnum)
-{
- u32 __iomem *pbuf;
- u32 dwcnt; /* dword count to write */
- if (pnum < dd->ipath_piobcnt2k) {
- pbuf = (u32 __iomem *) (dd->ipath_pio2kbase + pnum *
- dd->ipath_palign);
- dwcnt = dd->ipath_piosize2k >> 2;
- }
- else {
- pbuf = (u32 __iomem *) (dd->ipath_pio4kbase +
- (pnum - dd->ipath_piobcnt2k) * dd->ipath_4kalign);
- dwcnt = dd->ipath_piosize4k >> 2;
- }
- dev_info(&dd->pcidev->dev,
- "Rewrite PIO buffer %u, to recover from parity error\n",
- pnum);
-
- /* no flush required, since already in freeze */
- writel(dwcnt + 1, pbuf);
- while (--dwcnt)
- writel(0, pbuf++);
-}
/*
* Called when we might have an error that is specific to a particular
* PIO buffer, and may need to cancel that buffer, so it can be re-used.
- * If rewrite is true, and bits are set in the sendbufferror registers,
- * we'll write to the buffer, for error recovery on parity errors.
*/
-void ipath_disarm_senderrbufs(struct ipath_devdata *dd, int rewrite)
+void ipath_disarm_senderrbufs(struct ipath_devdata *dd)
{
u32 piobcnt;
unsigned long sbuf[4];
}
for (i = 0; i < piobcnt; i++)
- if (test_bit(i, sbuf)) {
- if (rewrite)
- ipath_clrpiobuf(dd, i);
+ if (test_bit(i, sbuf))
ipath_disarm_piobufs(dd, i, 1);
- }
/* ignore armlaunch errs for a bit */
dd->ipath_lastcancel = jiffies+3;
}
{
u64 ignore_this_time = 0;
- ipath_disarm_senderrbufs(dd, 0);
+ ipath_disarm_senderrbufs(dd);
if ((errs & E_SUM_LINK_PKTERRS) &&
!(dd->ipath_flags & IPATH_LINKACTIVE)) {
/*
* processes (causing armlaunch), send errors due to going into freeze mode,
* etc., and try to avoid causing extra interrupts while doing so.
* Forcibly update the in-memory pioavail register copies after cleanup
- * because the chip won't do it for anything changing while in freeze mode
- * (we don't want to wait for the next pio buffer state change).
+ * because the chip won't do it while in freeze mode (the register values
+ * themselves are kept correct).
* Make sure that we don't lose any important interrupts by using the chip
* feature that says that writing 0 to a bit in *clear that is set in
* *status will cause an interrupt to be generated again (if allowed by
*/
void ipath_clear_freeze(struct ipath_devdata *dd)
{
- int i, im;
- u64 val;
-
/* disable error interrupts, to avoid confusion */
ipath_write_kreg(dd, dd->ipath_kregs->kr_errormask, 0ULL);
/* also disable interrupts; errormask is sometimes overwriten */
ipath_write_kreg(dd, dd->ipath_kregs->kr_intmask, 0ULL);
- /*
- * clear all sends, because they have may been
- * completed by usercode while in freeze mode, and
- * therefore would not be sent, and eventually
- * might cause the process to run out of bufs
- */
- ipath_cancel_sends(dd, 0);
+ ipath_cancel_sends(dd, 1);
+
+ /* clear the freeze, and be sure chip saw it */
ipath_write_kreg(dd, dd->ipath_kregs->kr_control,
dd->ipath_control);
+ ipath_read_kreg64(dd, dd->ipath_kregs->kr_scratch);
- /* ensure pio avail updates continue */
+ /* force in-memory update now we are out of freeze */
ipath_force_pio_avail_update(dd);
- /*
- * We just enabled pioavailupdate, so dma copy is almost certainly
- * not yet right, so read the registers directly. Similar to init
- */
- for (i = 0; i < dd->ipath_pioavregs; i++) {
- /* deal with 6110 chip bug */
- im = (i > 3 && (dd->ipath_flags & IPATH_SWAP_PIOBUFS)) ?
- i ^ 1 : i;
- val = ipath_read_kreg64(dd, (0x1000 / sizeof(u64)) + im);
- dd->ipath_pioavailregs_dma[i] = cpu_to_le64(val);
- dd->ipath_pioavailshadow[i] = val |
- (~dd->ipath_pioavailkernel[i] <<
- INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT);
- }
-
/*
* force new interrupt if any hwerr, error or interrupt bits are
* still set, and clear "safe" send packet errors related to freeze
ipath_read_kreg64(dd, dd->ipath_kregs->kr_scratch);
spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags);
- if (!(dd->ipath_flags & IPATH_HAS_SEND_DMA))
- handle_layer_pioavail(dd);
- else
- ipath_dbg("unexpected BUFAVAIL intr\n");
+ /* always process; sdma verbs uses PIO for acks and VL15 */
+ handle_layer_pioavail(dd);
}
ret = IRQ_HANDLED;
u16 port_subport_cnt;
/* non-zero if port is being shared. */
u16 port_subport_id;
+ /* number of pio bufs for this port (all procs, if shared) */
+ u32 port_piocnt;
+ /* first pio buffer for this port */
+ u32 port_pio_base;
/* chip offset of PIO buffers for this port */
u32 port_piobufs;
/* how many alloc_pages() chunks in port_rcvegrbuf_pages */
u32 ipath_lastrpkts;
/* pio bufs allocated per port */
u32 ipath_pbufsport;
+ /* if remainder on bufs/port, ports < extrabuf get 1 extra */
+ u32 ipath_ports_extrabuf;
u32 ipath_pioupd_thresh; /* update threshold, some chips */
/*
* number of ports configured as max; zero is set to number chip
int ipath_update_eeprom_log(struct ipath_devdata *dd);
void ipath_inc_eeprom_err(struct ipath_devdata *dd, u32 eidx, u32 incr);
u64 ipath_snap_cntr(struct ipath_devdata *, ipath_creg);
-void ipath_disarm_senderrbufs(struct ipath_devdata *, int);
+void ipath_disarm_senderrbufs(struct ipath_devdata *);
void ipath_force_pio_avail_update(struct ipath_devdata *);
void signal_ib_event(struct ipath_devdata *dd, enum ib_event_type ev);
qp->r_wrid_valid = 0;
wc.wr_id = qp->r_wr_id;
wc.status = IB_WC_SUCCESS;
- wc.opcode = IB_WC_RECV;
+ if (opcode == OP(RDMA_WRITE_LAST_WITH_IMMEDIATE) ||
+ opcode == OP(RDMA_WRITE_ONLY_WITH_IMMEDIATE))
+ wc.opcode = IB_WC_RECV_RDMA_WITH_IMM;
+ else
+ wc.opcode = IB_WC_RECV;
wc.vendor_err = 0;
wc.qp = &qp->ibqp;
wc.src_qp = qp->remote_qpn;
wake_up(&qp->wait);
}
-static void want_buffer(struct ipath_devdata *dd)
+static void want_buffer(struct ipath_devdata *dd, struct ipath_qp *qp)
{
- if (!(dd->ipath_flags & IPATH_HAS_SEND_DMA)) {
+ if (!(dd->ipath_flags & IPATH_HAS_SEND_DMA) ||
+ qp->ibqp.qp_type == IB_QPT_SMI) {
unsigned long flags;
spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags);
spin_lock_irqsave(&dev->pending_lock, flags);
list_add_tail(&qp->piowait, &dev->piowait);
spin_unlock_irqrestore(&dev->pending_lock, flags);
- want_buffer(dev->dd);
+ want_buffer(dev->dd, qp);
dev->n_piowait++;
}
spin_unlock_irqrestore(&dd->ipath_sdma_lock, flags);
/*
- * Don't restart sdma here. Wait until link is up to ACTIVE.
- * VL15 MADs used to bring the link up use PIO, and multiple
- * link transitions otherwise cause the sdma engine to be
+ * Don't restart sdma here (with the exception
+ * below). Wait until link is up to ACTIVE. VL15 MADs
+ * used to bring the link up use PIO, and multiple link
+ * transitions otherwise cause the sdma engine to be
* stopped and started multiple times.
- * The disable is done here, including the shadow, so the
- * state is kept consistent.
- * See ipath_restart_sdma() for the actual starting of sdma.
+ * The disable is done here, including the shadow,
+ * so the state is kept consistent.
+ * See ipath_restart_sdma() for the actual starting
+ * of sdma.
*/
spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags);
dd->ipath_sendctrl &= ~INFINIPATH_S_SDMAENABLE;
/* make sure I see next message */
dd->ipath_sdma_abort_jiffies = 0;
+ /*
+ * Not everything that takes SDMA offline is a link
+ * status change. If the link was up, restart SDMA.
+ */
+ if (dd->ipath_flags & IPATH_LINKACTIVE)
+ ipath_restart_sdma(dd);
+
goto done;
}
goto done;
}
- dd->ipath_sdma_status = 0;
+ /*
+ * Set initial status as if we had been up, then gone down.
+ * This lets initial start on transition to ACTIVE be the
+ * same as restart after link flap.
+ */
+ dd->ipath_sdma_status = IPATH_SDMA_ABORT_ABORTED;
dd->ipath_sdma_abort_jiffies = 0;
dd->ipath_sdma_generation = 0;
dd->ipath_sdma_descq_tail = 0;
ipath_write_kreg(dd, dd->ipath_kregs->kr_senddmaheadaddr,
dd->ipath_sdma_head_phys);
- /* Reserve all the former "kernel" piobufs */
- n = dd->ipath_piobcnt2k + dd->ipath_piobcnt4k - dd->ipath_pioreserved;
- for (i = dd->ipath_lastport_piobuf; i < n; ++i) {
+ /*
+ * Reserve all the former "kernel" piobufs, using high number range
+ * so we get as many 4K buffers as possible
+ */
+ n = dd->ipath_piobcnt2k + dd->ipath_piobcnt4k;
+ i = dd->ipath_lastport_piobuf + dd->ipath_pioreserved;
+ ipath_chg_pioavailkernel(dd, i, n - i , 0);
+ for (; i < n; ++i) {
unsigned word = i / 64;
unsigned bit = i & 63;
BUG_ON(word >= 3);
senddmabufmask[word] |= 1ULL << bit;
}
- ipath_chg_pioavailkernel(dd, dd->ipath_lastport_piobuf,
- n - dd->ipath_lastport_piobuf, 0);
ipath_write_kreg(dd, dd->ipath_kregs->kr_senddmabufmask0,
senddmabufmask[0]);
ipath_write_kreg(dd, dd->ipath_kregs->kr_senddmabufmask1,
ipath_read_kreg64(dd, dd->ipath_kregs->kr_scratch);
spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags);
+ /* notify upper layers */
+ ipath_ib_piobufavail(dd->verbs_dev);
+
bail:
return;
}
wqe = get_swqe_ptr(qp, qp->s_head);
wqe->wr = *wr;
- wqe->ssn = qp->s_ssn++;
wqe->length = 0;
if (wr->num_sge) {
acc = wr->opcode >= IB_WR_RDMA_READ ?
goto bail_inval;
} else if (wqe->length > to_idev(qp->ibqp.device)->dd->ipath_ibmtu)
goto bail_inval;
+ wqe->ssn = qp->s_ssn++;
qp->s_head = next;
ret = 0;
if (context)
ib_umem_release(cq->umem);
else
- mlx4_ib_free_cq_buf(dev, &cq->buf, entries);
+ mlx4_ib_free_cq_buf(dev, &cq->buf, cq->ibcq.cqe);
err_db:
if (!context)
mlx4_ib_db_unmap_user(to_mucontext(cq->uobject->context), &mcq->db);
ib_umem_release(mcq->umem);
} else {
- mlx4_ib_free_cq_buf(dev, &mcq->buf, cq->cqe + 1);
+ mlx4_ib_free_cq_buf(dev, &mcq->buf, cq->cqe);
mlx4_db_free(dev->dev, &mcq->db);
}
#endif
int hca_caps;
struct ipoib_ethtool_st ethtool;
+ struct timer_list poll_timer;
};
struct ipoib_ah {
int ipoib_poll(struct napi_struct *napi, int budget);
void ipoib_ib_completion(struct ib_cq *cq, void *dev_ptr);
+void ipoib_send_comp_handler(struct ib_cq *cq, void *dev_ptr);
struct ipoib_ah *ipoib_create_ah(struct net_device *dev,
struct ib_pd *pd, struct ib_ah_attr *attr);
netif_rx_schedule(dev, &priv->napi);
}
+static void drain_tx_cq(struct net_device *dev)
+{
+ struct ipoib_dev_priv *priv = netdev_priv(dev);
+ unsigned long flags;
+
+ spin_lock_irqsave(&priv->tx_lock, flags);
+ while (poll_tx(priv))
+ ; /* nothing */
+
+ if (netif_queue_stopped(dev))
+ mod_timer(&priv->poll_timer, jiffies + 1);
+
+ spin_unlock_irqrestore(&priv->tx_lock, flags);
+}
+
+void ipoib_send_comp_handler(struct ib_cq *cq, void *dev_ptr)
+{
+ drain_tx_cq((struct net_device *)dev_ptr);
+}
+
static inline int post_send(struct ipoib_dev_priv *priv,
unsigned int wr_id,
struct ib_ah *address, u32 qpn,
else
priv->tx_wr.send_flags &= ~IB_SEND_IP_CSUM;
+ if (++priv->tx_outstanding == ipoib_sendq_size) {
+ ipoib_dbg(priv, "TX ring full, stopping kernel net queue\n");
+ if (ib_req_notify_cq(priv->send_cq, IB_CQ_NEXT_COMP))
+ ipoib_warn(priv, "request notify on send CQ failed\n");
+ netif_stop_queue(dev);
+ }
+
if (unlikely(post_send(priv, priv->tx_head & (ipoib_sendq_size - 1),
address->ah, qpn, tx_req, phead, hlen))) {
ipoib_warn(priv, "post_send failed\n");
++dev->stats.tx_errors;
+ --priv->tx_outstanding;
ipoib_dma_unmap_tx(priv->ca, tx_req);
dev_kfree_skb_any(skb);
+ if (netif_queue_stopped(dev))
+ netif_wake_queue(dev);
} else {
dev->trans_start = jiffies;
++priv->tx_head;
skb_orphan(skb);
- if (++priv->tx_outstanding == ipoib_sendq_size) {
- ipoib_dbg(priv, "TX ring full, stopping kernel net queue\n");
- netif_stop_queue(dev);
- }
}
if (unlikely(priv->tx_outstanding > MAX_SEND_CQE))
- poll_tx(priv);
+ while (poll_tx(priv))
+ ; /* nothing */
}
static void __ipoib_reap_ah(struct net_device *dev)
round_jiffies_relative(HZ));
}
+static void ipoib_ib_tx_timer_func(unsigned long ctx)
+{
+ drain_tx_cq((struct net_device *)ctx);
+}
+
int ipoib_ib_dev_open(struct net_device *dev)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task,
round_jiffies_relative(HZ));
+ init_timer(&priv->poll_timer);
+ priv->poll_timer.function = ipoib_ib_tx_timer_func;
+ priv->poll_timer.data = (unsigned long)dev;
+
set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags);
return 0;
ipoib_dbg(priv, "All sends and receives done.\n");
timeout:
+ del_timer_sync(&priv->poll_timer);
qp_attr.qp_state = IB_QPS_RESET;
if (ib_modify_qp(priv->qp, &qp_attr, IB_QP_STATE))
ipoib_warn(priv, "Failed to modify QP to RESET state\n");
goto out_free_mr;
}
- priv->send_cq = ib_create_cq(priv->ca, NULL, NULL, dev, ipoib_sendq_size, 0);
+ priv->send_cq = ib_create_cq(priv->ca, ipoib_send_comp_handler, NULL,
+ dev, ipoib_sendq_size, 0);
if (IS_ERR(priv->send_cq)) {
printk(KERN_WARNING "%s: failed to create send CQ\n", ca->name);
goto out_free_recv_cq;
config INPUT_PCSPKR
tristate "PC Speaker support"
- depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES
+ depends on PCSPKR_PLATFORM
depends on SND_PCSP=n
help
Say Y here if you want the standard PC Speaker to be used for
#include <linux/time.h>
#include <linux/slab.h>
#include <linux/hil.h>
+#include <linux/semaphore.h>
#include <asm/io.h>
#include <asm/system.h>
*/
raid10_find_phys(conf, r10_bio);
retry_write:
- blocked_rdev = 0;
+ blocked_rdev = NULL;
rcu_read_lock();
for (i = 0; i < conf->copies; i++) {
int d = r10_bio->devs[i].devnum;
# Makefile for the kernel multimedia device drivers.
#
+obj-y := common/
+
obj-$(CONFIG_VIDEO_MEDIA) += common/
# Since hybrid devices are here, should be compiled if DVB and/or V4L
}
cx = kzalloc(sizeof(struct cx18), GFP_ATOMIC);
- if (cx == 0) {
+ if (!cx) {
spin_unlock(&cx18_cards_lock);
return -ENOMEM;
}
struct saa7134_fh *fh = priv;
struct saa7134_dev *dev = fh->dev;
int err;
- unsigned int flags;
+ unsigned long flags;
if (saa7134_no_overlay > 0) {
printk(KERN_ERR "V4L2_BUF_TYPE_VIDEO_OVERLAY: no_overlay\n");
* to test the HW NMI watchdog
* F## = Break at do_fork for ## iterations
* S## = Break at sys_open for ## iterations
+ * I## = Run the single step test ## iterations
*
* NOTE: that the do_fork and sys_open tests are mutually exclusive.
*
break;
case 1:
/* set breakpoint */
- break_helper("Z0", 0, sstep_addr);
+ break_helper("Z0", NULL, sstep_addr);
break;
case 2:
/* Continue */
break;
case 3:
/* Clear breakpoint */
- break_helper("z0", 0, sstep_addr);
+ break_helper("z0", NULL, sstep_addr);
break;
default:
eprintk("kgdbts: ERROR failed sstep get emulation\n");
{ "?", "S0*" }, /* Clear break points */
{ "kgdbts_break_test", "OK", sw_break, }, /* set sw breakpoint */
{ "c", "T0*", }, /* Continue */
- { "g", "kgdbts_break_test", 0, check_and_rewind_pc },
+ { "g", "kgdbts_break_test", NULL, check_and_rewind_pc },
{ "write", "OK", write_regs },
{ "kgdbts_break_test", "OK", sw_rem_break }, /*remove breakpoint */
{ "D", "OK" }, /* Detach */
- { "D", "OK", 0, got_break }, /* If the test worked we made it here */
+ { "D", "OK", NULL, got_break }, /* On success we made it here */
{ "", "" },
};
{ "?", "S0*" }, /* Clear break points */
{ "kgdbts_break_test", "OK", sw_break, }, /* set sw breakpoint */
{ "c", "T0*", }, /* Continue */
- { "g", "kgdbts_break_test", 0, check_and_rewind_pc },
+ { "g", "kgdbts_break_test", NULL, check_and_rewind_pc },
{ "write", "OK", write_regs }, /* Write registers */
{ "kgdbts_break_test", "OK", sw_rem_break }, /*remove breakpoint */
{ "s", "T0*", emul_sstep_get, emul_sstep_put }, /* Single step */
- { "g", "kgdbts_break_test", 0, check_single_step },
+ { "g", "kgdbts_break_test", NULL, check_single_step },
{ "kgdbts_break_test", "OK", sw_break, }, /* set sw breakpoint */
{ "c", "T0*", }, /* Continue */
- { "g", "kgdbts_break_test", 0, check_and_rewind_pc },
+ { "g", "kgdbts_break_test", NULL, check_and_rewind_pc },
{ "write", "OK", write_regs }, /* Write registers */
{ "D", "OK" }, /* Remove all breakpoints and continues */
{ "", "" },
{ "?", "S0*" }, /* Clear break points */
{ "do_fork", "OK", sw_break, }, /* set sw breakpoint */
{ "c", "T0*", }, /* Continue */
- { "g", "do_fork", 0, check_and_rewind_pc }, /* check location */
+ { "g", "do_fork", NULL, check_and_rewind_pc }, /* check location */
{ "write", "OK", write_regs }, /* Write registers */
{ "do_fork", "OK", sw_rem_break }, /*remove breakpoint */
{ "s", "T0*", emul_sstep_get, emul_sstep_put }, /* Single step */
- { "g", "do_fork", 0, check_single_step },
+ { "g", "do_fork", NULL, check_single_step },
{ "do_fork", "OK", sw_break, }, /* set sw breakpoint */
{ "7", "T0*", skip_back_repeat_test }, /* Loop based on repeat_test */
- { "D", "OK", 0, final_ack_set }, /* detach and unregister I/O */
+ { "D", "OK", NULL, final_ack_set }, /* detach and unregister I/O */
{ "", "" },
};
{ "?", "S0*" }, /* Clear break points */
{ "sys_open", "OK", sw_break, }, /* set sw breakpoint */
{ "c", "T0*", }, /* Continue */
- { "g", "sys_open", 0, check_and_rewind_pc }, /* check location */
+ { "g", "sys_open", NULL, check_and_rewind_pc }, /* check location */
{ "write", "OK", write_regs }, /* Write registers */
{ "sys_open", "OK", sw_rem_break }, /*remove breakpoint */
{ "s", "T0*", emul_sstep_get, emul_sstep_put }, /* Single step */
- { "g", "sys_open", 0, check_single_step },
+ { "g", "sys_open", NULL, check_single_step },
{ "sys_open", "OK", sw_break, }, /* set sw breakpoint */
{ "7", "T0*", skip_back_repeat_test }, /* Loop based on repeat_test */
- { "D", "OK", 0, final_ack_set }, /* detach and unregister I/O */
+ { "D", "OK", NULL, final_ack_set }, /* detach and unregister I/O */
{ "", "" },
};
{ "?", "S0*" }, /* Clear break points */
{ "kgdbts_break_test", "OK", hw_break, }, /* set hw breakpoint */
{ "c", "T0*", }, /* Continue */
- { "g", "kgdbts_break_test", 0, check_and_rewind_pc },
+ { "g", "kgdbts_break_test", NULL, check_and_rewind_pc },
{ "write", "OK", write_regs },
{ "kgdbts_break_test", "OK", hw_rem_break }, /*remove breakpoint */
{ "D", "OK" }, /* Detach */
- { "D", "OK", 0, got_break }, /* If the test worked we made it here */
+ { "D", "OK", NULL, got_break }, /* On success we made it here */
{ "", "" },
};
static struct test_struct hw_write_break_test[] = {
{ "?", "S0*" }, /* Clear break points */
{ "hw_break_val", "OK", hw_write_break, }, /* set hw breakpoint */
- { "c", "T0*", 0, got_break }, /* Continue */
- { "g", "silent", 0, check_and_rewind_pc },
+ { "c", "T0*", NULL, got_break }, /* Continue */
+ { "g", "silent", NULL, check_and_rewind_pc },
{ "write", "OK", write_regs },
{ "hw_break_val", "OK", hw_rem_write_break }, /*remove breakpoint */
{ "D", "OK" }, /* Detach */
- { "D", "OK", 0, got_break }, /* If the test worked we made it here */
+ { "D", "OK", NULL, got_break }, /* On success we made it here */
{ "", "" },
};
static struct test_struct hw_access_break_test[] = {
{ "?", "S0*" }, /* Clear break points */
{ "hw_break_val", "OK", hw_access_break, }, /* set hw breakpoint */
- { "c", "T0*", 0, got_break }, /* Continue */
- { "g", "silent", 0, check_and_rewind_pc },
+ { "c", "T0*", NULL, got_break }, /* Continue */
+ { "g", "silent", NULL, check_and_rewind_pc },
{ "write", "OK", write_regs },
{ "hw_break_val", "OK", hw_rem_access_break }, /*remove breakpoint */
{ "D", "OK" }, /* Detach */
- { "D", "OK", 0, got_break }, /* If the test worked we made it here */
+ { "D", "OK", NULL, got_break }, /* On success we made it here */
{ "", "" },
};
*/
static struct test_struct nmi_sleep_test[] = {
{ "?", "S0*" }, /* Clear break points */
- { "c", "T0*", 0, got_break }, /* Continue */
+ { "c", "T0*", NULL, got_break }, /* Continue */
{ "D", "OK" }, /* Detach */
- { "D", "OK", 0, got_break }, /* If the test worked we made it here */
+ { "D", "OK", NULL, got_break }, /* On success we made it here */
{ "", "" },
};
{
char *ptr;
int fork_test = 0;
- int sys_open_test = 0;
+ int do_sys_open_test = 0;
+ int sstep_test = 1000;
int nmi_sleep = 0;
+ int i;
ptr = strstr(config, "F");
if (ptr)
- fork_test = simple_strtol(ptr+1, NULL, 10);
+ fork_test = simple_strtol(ptr + 1, NULL, 10);
ptr = strstr(config, "S");
if (ptr)
- sys_open_test = simple_strtol(ptr+1, NULL, 10);
+ do_sys_open_test = simple_strtol(ptr + 1, NULL, 10);
ptr = strstr(config, "N");
if (ptr)
nmi_sleep = simple_strtol(ptr+1, NULL, 10);
+ ptr = strstr(config, "I");
+ if (ptr)
+ sstep_test = simple_strtol(ptr+1, NULL, 10);
/* required internal KGDB tests */
v1printk("kgdbts:RUN plant and detach test\n");
run_breakpoint_test(0);
v1printk("kgdbts:RUN bad memory access test\n");
run_bad_read_test();
- v1printk("kgdbts:RUN singlestep breakpoint test\n");
- run_singlestep_break_test();
+ v1printk("kgdbts:RUN singlestep test %i iterations\n", sstep_test);
+ for (i = 0; i < sstep_test; i++) {
+ run_singlestep_break_test();
+ if (i % 100 == 0)
+ v1printk("kgdbts:RUN singlestep [%i/%i]\n",
+ i, sstep_test);
+ }
/* ===Optional tests=== */
repeat_test = fork_test;
printk(KERN_INFO "kgdbts:RUN do_fork for %i breakpoints\n",
repeat_test);
- kthread_run(kgdbts_unreg_thread, 0, "kgdbts_unreg");
+ kthread_run(kgdbts_unreg_thread, NULL, "kgdbts_unreg");
run_do_fork_test();
return;
}
* executed because a kernel thread will be spawned at the very
* end to unregister the debug hooks.
*/
- if (sys_open_test) {
- repeat_test = sys_open_test;
+ if (do_sys_open_test) {
+ repeat_test = do_sys_open_test;
printk(KERN_INFO "kgdbts:RUN sys_open for %i breakpoints\n",
repeat_test);
- kthread_run(kgdbts_unreg_thread, 0, "kgdbts_unreg");
+ kthread_run(kgdbts_unreg_thread, NULL, "kgdbts_unreg");
run_sys_open_test();
return;
}
{"3c920B-EMB-WNM (ATI Radeon 9100 IGP)",
PCI_USES_MASTER, IS_TORNADO|HAS_MII|HAS_HWCKSM, 128, },
{"3c980 Cyclone",
- PCI_USES_MASTER, IS_CYCLONE|HAS_HWCKSM, 128, },
+ PCI_USES_MASTER, IS_CYCLONE|HAS_HWCKSM|EXTRA_PREAMBLE, 128, },
{"3c980C Python-T",
PCI_USES_MASTER, IS_CYCLONE|HAS_NWAY|HAS_HWCKSM, 128, },
struct sk_buff* tx_skbuff[TX_RING_SIZE];
unsigned int cur_rx, cur_tx; /* The next free ring entry */
unsigned int dirty_rx, dirty_tx; /* The ring entries to be free()ed. */
- struct net_device_stats stats; /* Generic stats */
struct vortex_extra_stats xstats; /* NIC-specific extra stats */
struct sk_buff *tx_skb; /* Packet being eaten by bus master ctrl. */
dma_addr_t tx_skb_dma; /* Allocated DMA address for bus master ctrl DMA. */
issue_and_wait(dev, TxReset);
- vp->stats.tx_errors++;
+ dev->stats.tx_errors++;
if (vp->full_bus_master_tx) {
printk(KERN_DEBUG "%s: Resetting the Tx ring pointer.\n", dev->name);
if (vp->cur_tx - vp->dirty_tx > 0 && ioread32(ioaddr + DownListPtr) == 0)
iowrite8(PKT_BUF_SZ>>8, ioaddr + TxFreeThreshold);
iowrite16(DownUnstall, ioaddr + EL3_CMD);
} else {
- vp->stats.tx_dropped++;
+ dev->stats.tx_dropped++;
netif_wake_queue(dev);
}
}
dump_tx_ring(dev);
}
- if (tx_status & 0x14) vp->stats.tx_fifo_errors++;
- if (tx_status & 0x38) vp->stats.tx_aborted_errors++;
+ if (tx_status & 0x14) dev->stats.tx_fifo_errors++;
+ if (tx_status & 0x38) dev->stats.tx_aborted_errors++;
if (tx_status & 0x08) vp->xstats.tx_max_collisions++;
iowrite8(0, ioaddr + TxStatus);
if (tx_status & 0x30) { /* txJabber or txUnderrun */
if (vortex_debug > 2)
printk(KERN_DEBUG "%s: Tx error, status %2.2x.\n",
dev->name, tx_status);
- if (tx_status & 0x04) vp->stats.tx_fifo_errors++;
- if (tx_status & 0x38) vp->stats.tx_aborted_errors++;
+ if (tx_status & 0x04) dev->stats.tx_fifo_errors++;
+ if (tx_status & 0x38) dev->stats.tx_aborted_errors++;
if (tx_status & 0x30) {
issue_and_wait(dev, TxReset);
}
} else {
printk(KERN_DEBUG "boomerang_interrupt: no skb!\n");
}
- /* vp->stats.tx_packets++; Counted below. */
+ /* dev->stats.tx_packets++; Counted below. */
dirty_tx++;
}
vp->dirty_tx = dirty_tx;
unsigned char rx_error = ioread8(ioaddr + RxErrors);
if (vortex_debug > 2)
printk(KERN_DEBUG " Rx error: status %2.2x.\n", rx_error);
- vp->stats.rx_errors++;
- if (rx_error & 0x01) vp->stats.rx_over_errors++;
- if (rx_error & 0x02) vp->stats.rx_length_errors++;
- if (rx_error & 0x04) vp->stats.rx_frame_errors++;
- if (rx_error & 0x08) vp->stats.rx_crc_errors++;
- if (rx_error & 0x10) vp->stats.rx_length_errors++;
+ dev->stats.rx_errors++;
+ if (rx_error & 0x01) dev->stats.rx_over_errors++;
+ if (rx_error & 0x02) dev->stats.rx_length_errors++;
+ if (rx_error & 0x04) dev->stats.rx_frame_errors++;
+ if (rx_error & 0x08) dev->stats.rx_crc_errors++;
+ if (rx_error & 0x10) dev->stats.rx_length_errors++;
} else {
/* The packet length: up to 4.5K!. */
int pkt_len = rx_status & 0x1fff;
skb->protocol = eth_type_trans(skb, dev);
netif_rx(skb);
dev->last_rx = jiffies;
- vp->stats.rx_packets++;
+ dev->stats.rx_packets++;
/* Wait a limited time to go to next packet. */
for (i = 200; i >= 0; i--)
if ( ! (ioread16(ioaddr + EL3_STATUS) & CmdInProgress))
} else if (vortex_debug > 0)
printk(KERN_NOTICE "%s: No memory to allocate a sk_buff of "
"size %d.\n", dev->name, pkt_len);
- vp->stats.rx_dropped++;
+ dev->stats.rx_dropped++;
}
issue_and_wait(dev, RxDiscard);
}
unsigned char rx_error = rx_status >> 16;
if (vortex_debug > 2)
printk(KERN_DEBUG " Rx error: status %2.2x.\n", rx_error);
- vp->stats.rx_errors++;
- if (rx_error & 0x01) vp->stats.rx_over_errors++;
- if (rx_error & 0x02) vp->stats.rx_length_errors++;
- if (rx_error & 0x04) vp->stats.rx_frame_errors++;
- if (rx_error & 0x08) vp->stats.rx_crc_errors++;
- if (rx_error & 0x10) vp->stats.rx_length_errors++;
+ dev->stats.rx_errors++;
+ if (rx_error & 0x01) dev->stats.rx_over_errors++;
+ if (rx_error & 0x02) dev->stats.rx_length_errors++;
+ if (rx_error & 0x04) dev->stats.rx_frame_errors++;
+ if (rx_error & 0x08) dev->stats.rx_crc_errors++;
+ if (rx_error & 0x10) dev->stats.rx_length_errors++;
} else {
/* The packet length: up to 4.5K!. */
int pkt_len = rx_status & 0x1fff;
}
netif_rx(skb);
dev->last_rx = jiffies;
- vp->stats.rx_packets++;
+ dev->stats.rx_packets++;
}
entry = (++vp->cur_rx) % RX_RING_SIZE;
}
del_timer_sync(&vp->rx_oom_timer);
del_timer_sync(&vp->timer);
- /* Turn off statistics ASAP. We update vp->stats below. */
+ /* Turn off statistics ASAP. We update dev->stats below. */
iowrite16(StatsDisable, ioaddr + EL3_CMD);
/* Disable the receiver and transmitter. */
update_stats(ioaddr, dev);
spin_unlock_irqrestore (&vp->lock, flags);
}
- return &vp->stats;
+ return &dev->stats;
}
/* Update statistics.
/* Unlike the 3c5x9 we need not turn off stats updates while reading. */
/* Switch to the stats window, and read everything. */
EL3WINDOW(6);
- vp->stats.tx_carrier_errors += ioread8(ioaddr + 0);
- vp->stats.tx_heartbeat_errors += ioread8(ioaddr + 1);
- vp->stats.tx_window_errors += ioread8(ioaddr + 4);
- vp->stats.rx_fifo_errors += ioread8(ioaddr + 5);
- vp->stats.tx_packets += ioread8(ioaddr + 6);
- vp->stats.tx_packets += (ioread8(ioaddr + 9)&0x30) << 4;
+ dev->stats.tx_carrier_errors += ioread8(ioaddr + 0);
+ dev->stats.tx_heartbeat_errors += ioread8(ioaddr + 1);
+ dev->stats.tx_window_errors += ioread8(ioaddr + 4);
+ dev->stats.rx_fifo_errors += ioread8(ioaddr + 5);
+ dev->stats.tx_packets += ioread8(ioaddr + 6);
+ dev->stats.tx_packets += (ioread8(ioaddr + 9)&0x30) << 4;
/* Rx packets */ ioread8(ioaddr + 7); /* Must read to clear */
/* Don't bother with register 9, an extension of registers 6&7.
If we do use the 6&7 values the atomic update assumption above
is invalid. */
- vp->stats.rx_bytes += ioread16(ioaddr + 10);
- vp->stats.tx_bytes += ioread16(ioaddr + 12);
+ dev->stats.rx_bytes += ioread16(ioaddr + 10);
+ dev->stats.tx_bytes += ioread16(ioaddr + 12);
/* Extra stats for get_ethtool_stats() */
vp->xstats.tx_multiple_collisions += ioread8(ioaddr + 2);
vp->xstats.tx_single_collisions += ioread8(ioaddr + 3);
EL3WINDOW(4);
vp->xstats.rx_bad_ssd += ioread8(ioaddr + 12);
- vp->stats.collisions = vp->xstats.tx_multiple_collisions
+ dev->stats.collisions = vp->xstats.tx_multiple_collisions
+ vp->xstats.tx_single_collisions
+ vp->xstats.tx_max_collisions;
{
u8 up = ioread8(ioaddr + 13);
- vp->stats.rx_bytes += (up & 0x0f) << 16;
- vp->stats.tx_bytes += (up & 0xf0) << 12;
+ dev->stats.rx_bytes += (up & 0x0f) << 16;
+ dev->stats.tx_bytes += (up & 0xf0) << 12;
}
EL3WINDOW(old_window >> 13);
To compile this driver as a module, choose M here. The module
will be called pcnet32.
-config PCNET32_NAPI
- bool "Use RX polling (NAPI)"
- depends on PCNET32
- help
- NAPI is a new driver API designed to reduce CPU and interrupt load
- when the driver is receiving lots of packets from the card. It is
- still somewhat experimental and thus not yet enabled by default.
-
- If your estimated Rx load is 10kpps or more, or if the card will be
- deployed on potentially unfriendly networks (e.g. in a firewall),
- then say Y here.
-
- If in doubt, say N.
-
config AMD8111_ETH
tristate "AMD 8111 (new PCI lance) support"
depends on NET_PCI && PCI
{
outb(0, ioaddr+DAYNA_RESET); /* Assert the reset port */
inb(ioaddr+DAYNA_RESET); /* Clear the reset */
- if(sleep)
- {
- long snap=jiffies;
-
- /* Let card finish initializing, about 1/3 second */
- while (time_before(jiffies, snap + HZ/3))
- schedule();
- }
- else
- mdelay(333);
+ if (sleep)
+ msleep(333);
+ else
+ mdelay(333);
}
+
netif_wake_queue(dev);
- return;
}
static void cops_load (struct net_device *dev)
res = netdev_set_master(slave_dev, bond_dev);
if (res) {
dprintk("Error %d calling netdev_set_master\n", res);
- goto err_close;
+ goto err_restore_mac;
}
/* open the slave since the application closed it */
res = dev_open(slave_dev);
if (res) {
dprintk("Openning slave %s failed\n", slave_dev->name);
- goto err_restore_mac;
+ goto err_unset_master;
}
new_slave->dev = slave_dev;
*/
res = bond_alb_init_slave(bond, new_slave);
if (res) {
- goto err_unset_master;
+ goto err_close;
}
}
res = bond_create_slave_symlinks(bond_dev, slave_dev);
if (res)
- goto err_unset_master;
+ goto err_close;
printk(KERN_INFO DRV_NAME
": %s: enslaving %s as a%s interface with a%s link.\n",
return 0;
/* Undo stages on error */
-err_unset_master:
- netdev_set_master(slave_dev, NULL);
-
err_close:
dev_close(slave_dev);
+err_unset_master:
+ netdev_set_master(slave_dev, NULL);
+
err_restore_mac:
if (!bond->params.fail_over_mac) {
memcpy(addr.sa_data, new_slave->perm_hwaddr, ETH_ALEN);
if (res < 0) {
rtnl_lock();
down_write(&bonding_rwsem);
- goto out_bond;
+ bond_deinit(bond_dev);
+ unregister_netdevice(bond_dev);
+ goto out_rtnl;
}
return 0;
destroy_workqueue(bond->wq);
}
+ bond_destroy_sysfs();
+
rtnl_lock();
bond_free_all();
- bond_destroy_sysfs();
rtnl_unlock();
out:
return res;
unregister_netdevice_notifier(&bond_netdev_notifier);
unregister_inetaddr_notifier(&bond_inetaddr_notifier);
+ bond_destroy_sysfs();
+
rtnl_lock();
bond_free_all();
- bond_destroy_sysfs();
rtnl_unlock();
}
": Unable remove bond %s due to open references.\n",
ifname);
res = -EPERM;
- goto out;
+ goto out_unlock;
}
printk(KERN_INFO DRV_NAME
": %s is being deleted...\n",
bond->dev->name);
bond_destroy(bond);
- up_write(&bonding_rwsem);
- rtnl_unlock();
- goto out;
+ goto out_unlock;
}
printk(KERN_ERR DRV_NAME
": unable to delete non-existent bond %s\n", ifname);
res = -ENODEV;
- up_write(&bonding_rwsem);
- rtnl_unlock();
- goto out;
+ goto out_unlock;
}
err_no_cmd:
printk(KERN_ERR DRV_NAME
": no command found in bonding_masters. Use +ifname or -ifname.\n");
- res = -EPERM;
+ return -EPERM;
+
+out_unlock:
+ up_write(&bonding_rwsem);
+ rtnl_unlock();
/* Always return either count or an error. If you return 0, you'll
* get called forever, which is bad.
u8 *fw_data;
struct ch_mem_range t;
- if (!capable(CAP_NET_ADMIN))
+ if (!capable(CAP_SYS_RAWIO))
return -EPERM;
if (copy_from_user(&t, useraddr, sizeof(t)))
return -EFAULT;
-
+ /* Check t.len sanity ? */
fw_data = kmalloc(t.len, GFP_KERNEL);
if (!fw_data)
return -ENOMEM;
#define IFE_E_PHY_ID 0x02A80330
#define IFE_PLUS_E_PHY_ID 0x02A80320
#define IFE_C_E_PHY_ID 0x02A80310
+#define BME1000_E_PHY_ID 0x01410CB0
+#define BME1000_E_PHY_ID_R2 0x01410CB1
/* M88E1000 Specific Registers */
#define M88E1000_PHY_SPEC_CTRL 0x10 /* PHY Specific Control Register */
#define M88EC018_EPSCR_DOWNSHIFT_COUNTER_MASK 0x0E00
#define M88EC018_EPSCR_DOWNSHIFT_COUNTER_5X 0x0800
+/* BME1000 PHY Specific Control Register */
+#define BME1000_PSCR_ENABLE_DOWNSHIFT 0x0800 /* 1 = enable downshift */
+
+
+#define PHY_PAGE_SHIFT 5
+#define PHY_REG(page, reg) (((page) << PHY_PAGE_SHIFT) | \
+ ((reg) & MAX_PHY_REG_ADDRESS))
+
/*
* Bits...
* 15-5: page
/* arrays of page information for packet split */
struct e1000_ps_page *ps_pages;
};
-
+ struct page *page;
};
struct e1000_ring {
#define FLAG_HAS_CTRLEXT_ON_LOAD (1 << 5)
#define FLAG_HAS_SWSM_ON_LOAD (1 << 6)
#define FLAG_HAS_JUMBO_FRAMES (1 << 7)
+#define FLAG_IS_ICH (1 << 9)
#define FLAG_HAS_SMART_POWER_DOWN (1 << 11)
#define FLAG_IS_QUAD_PORT_A (1 << 12)
#define FLAG_IS_QUAD_PORT (1 << 13)
bool state);
extern void e1000e_igp3_phy_powerdown_workaround_ich8lan(struct e1000_hw *hw);
extern void e1000e_gig_downshift_workaround_ich8lan(struct e1000_hw *hw);
+extern void e1000e_disable_gig_wol_ich8lan(struct e1000_hw *hw);
extern s32 e1000e_check_for_copper_link(struct e1000_hw *hw);
extern s32 e1000e_check_for_fiber_link(struct e1000_hw *hw);
extern s32 e1000e_read_phy_reg_m88(struct e1000_hw *hw, u32 offset, u16 *data);
extern s32 e1000e_write_phy_reg_m88(struct e1000_hw *hw, u32 offset, u16 data);
extern enum e1000_phy_type e1000e_get_phy_type_from_id(u32 phy_id);
+extern s32 e1000e_determine_phy_address(struct e1000_hw *hw);
+extern s32 e1000e_write_phy_reg_bm(struct e1000_hw *hw, u32 offset, u16 data);
+extern s32 e1000e_read_phy_reg_bm(struct e1000_hw *hw, u32 offset, u16 *data);
extern void e1000e_phy_force_speed_duplex_setup(struct e1000_hw *hw, u16 *phy_ctrl);
extern s32 e1000e_write_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 data);
extern s32 e1000e_read_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 *data);
for (i = 0; i < last_word - first_word + 1; i++) {
ret_val = e1000_read_nvm(hw, first_word + i, 1,
&eeprom_buff[i]);
- if (ret_val)
+ if (ret_val) {
+ /* a read error occurred, throw away the
+ * result */
+ memset(eeprom_buff, 0xff, sizeof(eeprom_buff));
break;
+ }
}
}
/* restore previous status */
ew32(STATUS, before);
- if ((mac->type != e1000_ich8lan) &&
- (mac->type != e1000_ich9lan)) {
+ if (!(adapter->flags & FLAG_IS_ICH)) {
REG_PATTERN_TEST(E1000_FCAL, 0xFFFFFFFF, 0xFFFFFFFF);
REG_PATTERN_TEST(E1000_FCAH, 0x0000FFFF, 0xFFFFFFFF);
REG_PATTERN_TEST(E1000_FCT, 0x0000FFFF, 0xFFFFFFFF);
REG_SET_AND_CHECK(E1000_RCTL, 0xFFFFFFFF, 0x00000000);
- before = (((mac->type == e1000_ich8lan) ||
- (mac->type == e1000_ich9lan)) ? 0x06C3B33E : 0x06DFB3FE);
+ before = ((adapter->flags & FLAG_IS_ICH) ? 0x06C3B33E : 0x06DFB3FE);
REG_SET_AND_CHECK(E1000_RCTL, before, 0x003FFFFB);
REG_SET_AND_CHECK(E1000_TCTL, 0xFFFFFFFF, 0x00000000);
REG_SET_AND_CHECK(E1000_RCTL, before, 0xFFFFFFFF);
REG_PATTERN_TEST(E1000_RDBAL, 0xFFFFFFF0, 0xFFFFFFFF);
- if ((mac->type != e1000_ich8lan) &&
- (mac->type != e1000_ich9lan))
+ if (!(adapter->flags & FLAG_IS_ICH))
REG_PATTERN_TEST(E1000_TXCW, 0xC000FFFF, 0x0000FFFF);
REG_PATTERN_TEST(E1000_TDBAL, 0xFFFFFFF0, 0xFFFFFFFF);
REG_PATTERN_TEST(E1000_TIDV, 0x0000FFFF, 0x0000FFFF);
/* Test each interrupt */
for (i = 0; i < 10; i++) {
-
- if (((adapter->hw.mac.type == e1000_ich8lan) ||
- (adapter->hw.mac.type == e1000_ich9lan)) && i == 8)
+ if ((adapter->flags & FLAG_IS_ICH) && (i == 8))
continue;
/* Interrupt to test */
struct e1000_hw *hw = &adapter->hw;
u32 ctrl_reg = 0;
u32 stat_reg = 0;
+ u16 phy_reg = 0;
hw->mac.autoneg = 0;
E1000_CTRL_SPD_100 |/* Force Speed to 100 */
E1000_CTRL_FD); /* Force Duplex to FULL */
break;
+ case e1000_phy_bm:
+ /* Set Default MAC Interface speed to 1GB */
+ e1e_rphy(hw, PHY_REG(2, 21), &phy_reg);
+ phy_reg &= ~0x0007;
+ phy_reg |= 0x006;
+ e1e_wphy(hw, PHY_REG(2, 21), phy_reg);
+ /* Assert SW reset for above settings to take effect */
+ e1000e_commit_phy(hw);
+ mdelay(1);
+ /* Force Full Duplex */
+ e1e_rphy(hw, PHY_REG(769, 16), &phy_reg);
+ e1e_wphy(hw, PHY_REG(769, 16), phy_reg | 0x000C);
+ /* Set Link Up (in force link) */
+ e1e_rphy(hw, PHY_REG(776, 16), &phy_reg);
+ e1e_wphy(hw, PHY_REG(776, 16), phy_reg | 0x0040);
+ /* Force Link */
+ e1e_rphy(hw, PHY_REG(769, 16), &phy_reg);
+ e1e_wphy(hw, PHY_REG(769, 16), phy_reg | 0x0040);
+ /* Set Early Link Enable */
+ e1e_rphy(hw, PHY_REG(769, 20), &phy_reg);
+ e1e_wphy(hw, PHY_REG(769, 20), phy_reg | 0x0400);
+ /* fall through */
default:
/* force 1000, set loopback */
e1e_wphy(hw, PHY_CONTROL, 0x4140);
E1000_CTRL_SPD_1000 |/* Force Speed to 1000 */
E1000_CTRL_FD); /* Force Duplex to FULL */
- if ((adapter->hw.mac.type == e1000_ich8lan) ||
- (adapter->hw.mac.type == e1000_ich9lan))
+ if (adapter->flags & FLAG_IS_ICH)
ctrl_reg |= E1000_CTRL_SLU; /* Set Link Up */
}
#define IGP01E1000_PHY_LINK_HEALTH 0x13 /* PHY Link Health */
#define IGP02E1000_PHY_POWER_MGMT 0x19 /* Power Management */
#define IGP01E1000_PHY_PAGE_SELECT 0x1F /* Page Select */
+#define BM_PHY_PAGE_SELECT 22 /* Page Select for BM */
+#define IGP_PAGE_SHIFT 5
+#define PHY_REG_MASK 0x1F
+
+#define BM_WUC_PAGE 800
+#define BM_WUC_ADDRESS_OPCODE 0x11
+#define BM_WUC_DATA_OPCODE 0x12
+#define BM_WUC_ENABLE_PAGE 769
+#define BM_WUC_ENABLE_REG 17
+#define BM_WUC_ENABLE_BIT (1 << 2)
+#define BM_WUC_HOST_WU_BIT (1 << 4)
+
+#define BM_WUC PHY_REG(BM_WUC_PAGE, 1)
+#define BM_WUFC PHY_REG(BM_WUC_PAGE, 2)
+#define BM_WUS PHY_REG(BM_WUC_PAGE, 3)
#define IGP01E1000_PHY_PCS_INIT_REG 0x00B4
#define IGP01E1000_PHY_POLARITY_MASK 0x0078
#define E1000_DEV_ID_ICH8_IFE_G 0x10C5
#define E1000_DEV_ID_ICH8_IGP_M 0x104D
#define E1000_DEV_ID_ICH9_IGP_AMT 0x10BD
+#define E1000_DEV_ID_ICH9_IGP_M_AMT 0x10F5
+#define E1000_DEV_ID_ICH9_IGP_M 0x10BF
+#define E1000_DEV_ID_ICH9_IGP_M_V 0x10CB
#define E1000_DEV_ID_ICH9_IGP_C 0x294C
#define E1000_DEV_ID_ICH9_IFE 0x10C0
#define E1000_DEV_ID_ICH9_IFE_GT 0x10C3
#define E1000_DEV_ID_ICH9_IFE_G 0x10C2
+#define E1000_DEV_ID_ICH10_R_BM_LM 0x10CC
+#define E1000_DEV_ID_ICH10_R_BM_LF 0x10CD
+#define E1000_DEV_ID_ICH10_R_BM_V 0x10CE
#define E1000_FUNC_1 1
e1000_phy_gg82563,
e1000_phy_igp_3,
e1000_phy_ife,
+ e1000_phy_bm,
};
enum e1000_bus_width {
* 82566DM Gigabit Network Connection
* 82566MC Gigabit Network Connection
* 82566MM Gigabit Network Connection
+ * 82567LM Gigabit Network Connection
+ * 82567LF Gigabit Network Connection
+ * 82567LM-2 Gigabit Network Connection
+ * 82567LF-2 Gigabit Network Connection
+ * 82567V-2 Gigabit Network Connection
+ * 82562GT-3 10/100 Network Connection
*/
#include <linux/netdevice.h>
phy->addr = 1;
phy->reset_delay_us = 100;
+ /*
+ * We may need to do this twice - once for IGP and if that fails,
+ * we'll set BM func pointers and try again
+ */
+ ret_val = e1000e_determine_phy_address(hw);
+ if (ret_val) {
+ hw->phy.ops.write_phy_reg = e1000e_write_phy_reg_bm;
+ hw->phy.ops.read_phy_reg = e1000e_read_phy_reg_bm;
+ ret_val = e1000e_determine_phy_address(hw);
+ if (ret_val)
+ return ret_val;
+ }
+
phy->id = 0;
while ((e1000_phy_unknown == e1000e_get_phy_type_from_id(phy->id)) &&
(i++ < 100)) {
phy->type = e1000_phy_ife;
phy->autoneg_mask = E1000_ALL_NOT_GIG;
break;
+ case BME1000_E_PHY_ID:
+ phy->type = e1000_phy_bm;
+ phy->autoneg_mask = AUTONEG_ADVERTISE_SPEED_DEFAULT;
+ hw->phy.ops.read_phy_reg = e1000e_read_phy_reg_bm;
+ hw->phy.ops.write_phy_reg = e1000e_write_phy_reg_bm;
+ hw->phy.ops.commit_phy = e1000e_phy_sw_reset;
+ break;
default:
return -E1000_ERR_PHY;
break;
return e1000_get_phy_info_ife_ich8lan(hw);
break;
case e1000_phy_igp_3:
+ case e1000_phy_bm:
return e1000e_get_phy_info_igp(hw);
break;
default:
s32 ret_val = 0;
u16 data;
- if (phy->type != e1000_phy_igp_3)
+ if (phy->type == e1000_phy_ife)
return ret_val;
phy_ctrl = er32(PHY_CTRL);
ret_val = e1000e_copper_link_setup_igp(hw);
if (ret_val)
return ret_val;
+ } else if (hw->phy.type == e1000_phy_bm) {
+ ret_val = e1000e_copper_link_setup_m88(hw);
+ if (ret_val)
+ return ret_val;
}
+ if (hw->phy.type == e1000_phy_ife) {
+ ret_val = e1e_rphy(hw, IFE_PHY_MDIX_CONTROL, ®_data);
+ if (ret_val)
+ return ret_val;
+
+ reg_data &= ~IFE_PMC_AUTO_MDIX;
+
+ switch (hw->phy.mdix) {
+ case 1:
+ reg_data &= ~IFE_PMC_FORCE_MDIX;
+ break;
+ case 2:
+ reg_data |= IFE_PMC_FORCE_MDIX;
+ break;
+ case 0:
+ default:
+ reg_data |= IFE_PMC_AUTO_MDIX;
+ break;
+ }
+ ret_val = e1e_wphy(hw, IFE_PHY_MDIX_CONTROL, reg_data);
+ if (ret_val)
+ return ret_val;
+ }
return e1000e_setup_copper_link(hw);
}
reg_data);
}
+/**
+ * e1000e_disable_gig_wol_ich8lan - disable gig during WoL
+ * @hw: pointer to the HW structure
+ *
+ * During S0 to Sx transition, it is possible the link remains at gig
+ * instead of negotiating to a lower speed. Before going to Sx, set
+ * 'LPLU Enabled' and 'Gig Disable' to force link speed negotiation
+ * to a lower speed.
+ *
+ * Should only be called for ICH9 devices.
+ **/
+void e1000e_disable_gig_wol_ich8lan(struct e1000_hw *hw)
+{
+ u32 phy_ctrl;
+
+ if (hw->mac.type == e1000_ich9lan) {
+ phy_ctrl = er32(PHY_CTRL);
+ phy_ctrl |= E1000_PHY_CTRL_D0A_LPLU |
+ E1000_PHY_CTRL_GBE_DISABLE;
+ ew32(PHY_CTRL, phy_ctrl);
+ }
+
+ return;
+}
+
/**
* e1000_cleanup_led_ich8lan - Restore the default LED operation
* @hw: pointer to the HW structure
struct e1000_info e1000_ich8_info = {
.mac = e1000_ich8lan,
.flags = FLAG_HAS_WOL
+ | FLAG_IS_ICH
| FLAG_RX_CSUM_ENABLED
| FLAG_HAS_CTRLEXT_ON_LOAD
| FLAG_HAS_AMT
struct e1000_info e1000_ich9_info = {
.mac = e1000_ich9lan,
.flags = FLAG_HAS_JUMBO_FRAMES
+ | FLAG_IS_ICH
| FLAG_HAS_WOL
| FLAG_RX_CSUM_ENABLED
| FLAG_HAS_CTRLEXT_ON_LOAD
#include <linux/if_vlan.h>
#include <linux/cpu.h>
#include <linux/smp.h>
+#include <linux/pm_qos_params.h>
#include "e1000.h"
-#define DRV_VERSION "0.2.1"
+#define DRV_VERSION "0.3.3.3-k2"
char e1000e_driver_name[] = "e1000e";
const char e1000e_driver_version[] = DRV_VERSION;
}
}
+/**
+ * e1000_alloc_jumbo_rx_buffers - Replace used jumbo receive buffers
+ * @adapter: address of board private structure
+ * @rx_ring: pointer to receive ring structure
+ * @cleaned_count: number of buffers to allocate this pass
+ **/
+
+static void e1000_alloc_jumbo_rx_buffers(struct e1000_adapter *adapter,
+ int cleaned_count)
+{
+ struct net_device *netdev = adapter->netdev;
+ struct pci_dev *pdev = adapter->pdev;
+ struct e1000_rx_desc *rx_desc;
+ struct e1000_ring *rx_ring = adapter->rx_ring;
+ struct e1000_buffer *buffer_info;
+ struct sk_buff *skb;
+ unsigned int i;
+ unsigned int bufsz = 256 -
+ 16 /* for skb_reserve */ -
+ NET_IP_ALIGN;
+
+ i = rx_ring->next_to_use;
+ buffer_info = &rx_ring->buffer_info[i];
+
+ while (cleaned_count--) {
+ skb = buffer_info->skb;
+ if (skb) {
+ skb_trim(skb, 0);
+ goto check_page;
+ }
+
+ skb = netdev_alloc_skb(netdev, bufsz);
+ if (unlikely(!skb)) {
+ /* Better luck next round */
+ adapter->alloc_rx_buff_failed++;
+ break;
+ }
+
+ /* Make buffer alignment 2 beyond a 16 byte boundary
+ * this will result in a 16 byte aligned IP header after
+ * the 14 byte MAC header is removed
+ */
+ skb_reserve(skb, NET_IP_ALIGN);
+
+ buffer_info->skb = skb;
+check_page:
+ /* allocate a new page if necessary */
+ if (!buffer_info->page) {
+ buffer_info->page = alloc_page(GFP_ATOMIC);
+ if (unlikely(!buffer_info->page)) {
+ adapter->alloc_rx_buff_failed++;
+ break;
+ }
+ }
+
+ if (!buffer_info->dma)
+ buffer_info->dma = pci_map_page(pdev,
+ buffer_info->page, 0,
+ PAGE_SIZE,
+ PCI_DMA_FROMDEVICE);
+
+ rx_desc = E1000_RX_DESC(*rx_ring, i);
+ rx_desc->buffer_addr = cpu_to_le64(buffer_info->dma);
+
+ if (unlikely(++i == rx_ring->count))
+ i = 0;
+ buffer_info = &rx_ring->buffer_info[i];
+ }
+
+ if (likely(rx_ring->next_to_use != i)) {
+ rx_ring->next_to_use = i;
+ if (unlikely(i-- == 0))
+ i = (rx_ring->count - 1);
+
+ /* Force memory writes to complete before letting h/w
+ * know there are new descriptors to fetch. (Only
+ * applicable for weak-ordered memory model archs,
+ * such as IA-64). */
+ wmb();
+ writel(i, adapter->hw.hw_addr + rx_ring->tail);
+ }
+}
+
/**
* e1000_clean_rx_irq - Send received data up the network stack; legacy
* @adapter: board private structure
return cleaned;
}
+/**
+ * e1000_consume_page - helper function
+ **/
+static void e1000_consume_page(struct e1000_buffer *bi, struct sk_buff *skb,
+ u16 length)
+{
+ bi->page = NULL;
+ skb->len += length;
+ skb->data_len += length;
+ skb->truesize += length;
+}
+
+/**
+ * e1000_clean_jumbo_rx_irq - Send received data up the network stack; legacy
+ * @adapter: board private structure
+ *
+ * the return value indicates whether actual cleaning was done, there
+ * is no guarantee that everything was cleaned
+ **/
+
+static bool e1000_clean_jumbo_rx_irq(struct e1000_adapter *adapter,
+ int *work_done, int work_to_do)
+{
+ struct net_device *netdev = adapter->netdev;
+ struct pci_dev *pdev = adapter->pdev;
+ struct e1000_ring *rx_ring = adapter->rx_ring;
+ struct e1000_rx_desc *rx_desc, *next_rxd;
+ struct e1000_buffer *buffer_info, *next_buffer;
+ u32 length;
+ unsigned int i;
+ int cleaned_count = 0;
+ bool cleaned = false;
+ unsigned int total_rx_bytes=0, total_rx_packets=0;
+
+ i = rx_ring->next_to_clean;
+ rx_desc = E1000_RX_DESC(*rx_ring, i);
+ buffer_info = &rx_ring->buffer_info[i];
+
+ while (rx_desc->status & E1000_RXD_STAT_DD) {
+ struct sk_buff *skb;
+ u8 status;
+
+ if (*work_done >= work_to_do)
+ break;
+ (*work_done)++;
+
+ status = rx_desc->status;
+ skb = buffer_info->skb;
+ buffer_info->skb = NULL;
+
+ ++i;
+ if (i == rx_ring->count)
+ i = 0;
+ next_rxd = E1000_RX_DESC(*rx_ring, i);
+ prefetch(next_rxd);
+
+ next_buffer = &rx_ring->buffer_info[i];
+
+ cleaned = true;
+ cleaned_count++;
+ pci_unmap_page(pdev, buffer_info->dma, PAGE_SIZE,
+ PCI_DMA_FROMDEVICE);
+ buffer_info->dma = 0;
+
+ length = le16_to_cpu(rx_desc->length);
+
+ /* errors is only valid for DD + EOP descriptors */
+ if (unlikely((status & E1000_RXD_STAT_EOP) &&
+ (rx_desc->errors & E1000_RXD_ERR_FRAME_ERR_MASK))) {
+ /* recycle both page and skb */
+ buffer_info->skb = skb;
+ /* an error means any chain goes out the window
+ * too */
+ if (rx_ring->rx_skb_top)
+ dev_kfree_skb(rx_ring->rx_skb_top);
+ rx_ring->rx_skb_top = NULL;
+ goto next_desc;
+ }
+
+#define rxtop rx_ring->rx_skb_top
+ if (!(status & E1000_RXD_STAT_EOP)) {
+ /* this descriptor is only the beginning (or middle) */
+ if (!rxtop) {
+ /* this is the beginning of a chain */
+ rxtop = skb;
+ skb_fill_page_desc(rxtop, 0, buffer_info->page,
+ 0, length);
+ } else {
+ /* this is the middle of a chain */
+ skb_fill_page_desc(rxtop,
+ skb_shinfo(rxtop)->nr_frags,
+ buffer_info->page, 0, length);
+ /* re-use the skb, only consumed the page */
+ buffer_info->skb = skb;
+ }
+ e1000_consume_page(buffer_info, rxtop, length);
+ goto next_desc;
+ } else {
+ if (rxtop) {
+ /* end of the chain */
+ skb_fill_page_desc(rxtop,
+ skb_shinfo(rxtop)->nr_frags,
+ buffer_info->page, 0, length);
+ /* re-use the current skb, we only consumed the
+ * page */
+ buffer_info->skb = skb;
+ skb = rxtop;
+ rxtop = NULL;
+ e1000_consume_page(buffer_info, skb, length);
+ } else {
+ /* no chain, got EOP, this buf is the packet
+ * copybreak to save the put_page/alloc_page */
+ if (length <= copybreak &&
+ skb_tailroom(skb) >= length) {
+ u8 *vaddr;
+ vaddr = kmap_atomic(buffer_info->page,
+ KM_SKB_DATA_SOFTIRQ);
+ memcpy(skb_tail_pointer(skb), vaddr,
+ length);
+ kunmap_atomic(vaddr,
+ KM_SKB_DATA_SOFTIRQ);
+ /* re-use the page, so don't erase
+ * buffer_info->page */
+ skb_put(skb, length);
+ } else {
+ skb_fill_page_desc(skb, 0,
+ buffer_info->page, 0,
+ length);
+ e1000_consume_page(buffer_info, skb,
+ length);
+ }
+ }
+ }
+
+ /* Receive Checksum Offload XXX recompute due to CRC strip? */
+ e1000_rx_checksum(adapter,
+ (u32)(status) |
+ ((u32)(rx_desc->errors) << 24),
+ le16_to_cpu(rx_desc->csum), skb);
+
+ /* probably a little skewed due to removing CRC */
+ total_rx_bytes += skb->len;
+ total_rx_packets++;
+
+ /* eth type trans needs skb->data to point to something */
+ if (!pskb_may_pull(skb, ETH_HLEN)) {
+ ndev_err(netdev, "pskb_may_pull failed.\n");
+ dev_kfree_skb(skb);
+ goto next_desc;
+ }
+
+ e1000_receive_skb(adapter, netdev, skb, status,
+ rx_desc->special);
+
+next_desc:
+ rx_desc->status = 0;
+
+ /* return some buffers to hardware, one at a time is too slow */
+ if (unlikely(cleaned_count >= E1000_RX_BUFFER_WRITE)) {
+ adapter->alloc_rx_buf(adapter, cleaned_count);
+ cleaned_count = 0;
+ }
+
+ /* use prefetched values */
+ rx_desc = next_rxd;
+ buffer_info = next_buffer;
+ }
+ rx_ring->next_to_clean = i;
+
+ cleaned_count = e1000_desc_unused(rx_ring);
+ if (cleaned_count)
+ adapter->alloc_rx_buf(adapter, cleaned_count);
+
+ adapter->total_rx_bytes += total_rx_bytes;
+ adapter->total_rx_packets += total_rx_packets;
+ adapter->net_stats.rx_bytes += total_rx_bytes;
+ adapter->net_stats.rx_packets += total_rx_packets;
+ return cleaned;
+}
+
/**
* e1000_clean_rx_ring - Free Rx Buffers per Queue
* @adapter: board private structure
pci_unmap_single(pdev, buffer_info->dma,
adapter->rx_buffer_len,
PCI_DMA_FROMDEVICE);
+ else if (adapter->clean_rx == e1000_clean_jumbo_rx_irq)
+ pci_unmap_page(pdev, buffer_info->dma,
+ PAGE_SIZE,
+ PCI_DMA_FROMDEVICE);
else if (adapter->clean_rx == e1000_clean_rx_irq_ps)
pci_unmap_single(pdev, buffer_info->dma,
adapter->rx_ps_bsize0,
buffer_info->dma = 0;
}
+ if (buffer_info->page) {
+ put_page(buffer_info->page);
+ buffer_info->page = NULL;
+ }
+
if (buffer_info->skb) {
dev_kfree_skb(buffer_info->skb);
buffer_info->skb = NULL;
* a lot of memory, since we allocate 3 pages at all times
* per packet.
*/
- adapter->rx_ps_pages = 0;
pages = PAGE_USE_COUNT(adapter->netdev->mtu);
- if ((pages <= 3) && (PAGE_SIZE <= 16384) && (rctl & E1000_RCTL_LPE))
+ if (!(adapter->flags & FLAG_IS_ICH) && (pages <= 3) &&
+ (PAGE_SIZE <= 16384) && (rctl & E1000_RCTL_LPE))
adapter->rx_ps_pages = pages;
+ else
+ adapter->rx_ps_pages = 0;
if (adapter->rx_ps_pages) {
/* Configure extra packet-split registers */
sizeof(union e1000_rx_desc_packet_split);
adapter->clean_rx = e1000_clean_rx_irq_ps;
adapter->alloc_rx_buf = e1000_alloc_rx_buffers_ps;
+ } else if (adapter->netdev->mtu > ETH_FRAME_LEN + ETH_FCS_LEN) {
+ rdlen = rx_ring->count * sizeof(struct e1000_rx_desc);
+ adapter->clean_rx = e1000_clean_jumbo_rx_irq;
+ adapter->alloc_rx_buf = e1000_alloc_jumbo_rx_buffers;
} else {
- rdlen = rx_ring->count *
- sizeof(struct e1000_rx_desc);
+ rdlen = rx_ring->count * sizeof(struct e1000_rx_desc);
adapter->clean_rx = e1000_clean_rx_irq;
adapter->alloc_rx_buf = e1000_alloc_rx_buffers;
}
* units), e.g. using jumbo frames when setting to E1000_ERT_2048
*/
if ((adapter->flags & FLAG_HAS_ERT) &&
- (adapter->netdev->mtu > ETH_DATA_LEN))
- ew32(ERT, E1000_ERT_2048);
+ (adapter->netdev->mtu > ETH_DATA_LEN)) {
+ u32 rxdctl = er32(RXDCTL(0));
+ ew32(RXDCTL(0), rxdctl | 0x3);
+ ew32(ERT, E1000_ERT_2048 | (1 << 13));
+ /*
+ * With jumbo frames and early-receive enabled, excessive
+ * C4->C2 latencies result in dropped transactions.
+ */
+ pm_qos_update_requirement(PM_QOS_CPU_DMA_LATENCY,
+ e1000e_driver_name, 55);
+ } else {
+ pm_qos_update_requirement(PM_QOS_CPU_DMA_LATENCY,
+ e1000e_driver_name,
+ PM_QOS_DEFAULT_VALUE);
+ }
/* Enable Receives */
ew32(RCTL, rctl);
/* Allow time for pending master requests to run */
mac->ops.reset_hw(hw);
+
+ /*
+ * For parts with AMT enabled, let the firmware know
+ * that the network interface is in control
+ */
+ if ((adapter->flags & FLAG_HAS_AMT) && e1000e_check_mng_mode(hw))
+ e1000_get_hw_control(adapter);
+
ew32(WUC, 0);
if (mac->ops.init_hw(hw))
* means we reserve 2 more, this pushes us to allocate from the next
* larger slab size.
* i.e. RXBUFFER_2048 --> size-4096 slab
+ * However with the new *_jumbo_rx* routines, jumbo receives will use
+ * fragmented skbs
*/
if (max_frame <= 256)
ew32(CTRL_EXT, ctrl_ext);
}
+ if (adapter->flags & FLAG_IS_ICH)
+ e1000e_disable_gig_wol_ich8lan(&adapter->hw);
+
/* Allow time for pending master requests to run */
e1000e_disable_pcie_master(&adapter->hw);
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IFE_GT), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_AMT), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_C), board_ich9lan },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M), board_ich9lan },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M_AMT), board_ich9lan },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M_V), board_ich9lan },
+
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_LM), board_ich9lan },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_LF), board_ich9lan },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_V), board_ich9lan },
{ } /* terminate list */
};
printk(KERN_INFO "%s: Copyright (c) 1999-2008 Intel Corporation.\n",
e1000e_driver_name);
ret = pci_register_driver(&e1000_driver);
-
+ pm_qos_add_requirement(PM_QOS_CPU_DMA_LATENCY, e1000e_driver_name,
+ PM_QOS_DEFAULT_VALUE);
+
return ret;
}
module_init(e1000_init_module);
static void __exit e1000_exit_module(void)
{
pci_unregister_driver(&e1000_driver);
+ pm_qos_remove_requirement(PM_QOS_CPU_DMA_LATENCY, e1000e_driver_name);
}
module_exit(e1000_exit_module);
static s32 e1000_phy_force_speed_duplex(struct e1000_hw *hw);
static s32 e1000_set_d0_lplu_state(struct e1000_hw *hw, bool active);
static s32 e1000_wait_autoneg(struct e1000_hw *hw);
+static u32 e1000_get_phy_addr_for_bm_page(u32 page, u32 reg);
+static s32 e1000_access_phy_wakeup_reg_bm(struct e1000_hw *hw, u32 offset,
+ u16 *data, bool read);
/* Cable length tables */
static const u16 e1000_m88_cable_length_table[] =
if (phy->disable_polarity_correction == 1)
phy_data |= M88E1000_PSCR_POLARITY_REVERSAL;
+ /* Enable downshift on BM (disabled by default) */
+ if (phy->type == e1000_phy_bm)
+ phy_data |= BME1000_PSCR_ENABLE_DOWNSHIFT;
+
ret_val = e1e_wphy(hw, M88E1000_PHY_SPEC_CTRL, phy_data);
if (ret_val)
return ret_val;
case IFE_C_E_PHY_ID:
phy_type = e1000_phy_ife;
break;
+ case BME1000_E_PHY_ID:
+ case BME1000_E_PHY_ID_R2:
+ phy_type = e1000_phy_bm;
+ break;
default:
phy_type = e1000_phy_unknown;
break;
return phy_type;
}
+/**
+ * e1000e_determine_phy_address - Determines PHY address.
+ * @hw: pointer to the HW structure
+ *
+ * This uses a trial and error method to loop through possible PHY
+ * addresses. It tests each by reading the PHY ID registers and
+ * checking for a match.
+ **/
+s32 e1000e_determine_phy_address(struct e1000_hw *hw)
+{
+ s32 ret_val = -E1000_ERR_PHY_TYPE;
+ u32 phy_addr= 0;
+ u32 i = 0;
+ enum e1000_phy_type phy_type = e1000_phy_unknown;
+
+ do {
+ for (phy_addr = 0; phy_addr < 4; phy_addr++) {
+ hw->phy.addr = phy_addr;
+ e1000e_get_phy_id(hw);
+ phy_type = e1000e_get_phy_type_from_id(hw->phy.id);
+
+ /*
+ * If phy_type is valid, break - we found our
+ * PHY address
+ */
+ if (phy_type != e1000_phy_unknown) {
+ ret_val = 0;
+ break;
+ }
+ }
+ i++;
+ } while ((ret_val != 0) && (i < 100));
+
+ return ret_val;
+}
+
+/**
+ * e1000_get_phy_addr_for_bm_page - Retrieve PHY page address
+ * @page: page to access
+ *
+ * Returns the phy address for the page requested.
+ **/
+static u32 e1000_get_phy_addr_for_bm_page(u32 page, u32 reg)
+{
+ u32 phy_addr = 2;
+
+ if ((page >= 768) || (page == 0 && reg == 25) || (reg == 31))
+ phy_addr = 1;
+
+ return phy_addr;
+}
+
+/**
+ * e1000e_write_phy_reg_bm - Write BM PHY register
+ * @hw: pointer to the HW structure
+ * @offset: register offset to write to
+ * @data: data to write at register offset
+ *
+ * Acquires semaphore, if necessary, then writes the data to PHY register
+ * at the offset. Release any acquired semaphores before exiting.
+ **/
+s32 e1000e_write_phy_reg_bm(struct e1000_hw *hw, u32 offset, u16 data)
+{
+ s32 ret_val;
+ u32 page_select = 0;
+ u32 page = offset >> IGP_PAGE_SHIFT;
+ u32 page_shift = 0;
+
+ /* Page 800 works differently than the rest so it has its own func */
+ if (page == BM_WUC_PAGE) {
+ ret_val = e1000_access_phy_wakeup_reg_bm(hw, offset, &data,
+ false);
+ goto out;
+ }
+
+ ret_val = hw->phy.ops.acquire_phy(hw);
+ if (ret_val)
+ goto out;
+
+ hw->phy.addr = e1000_get_phy_addr_for_bm_page(page, offset);
+
+ if (offset > MAX_PHY_MULTI_PAGE_REG) {
+ /*
+ * Page select is register 31 for phy address 1 and 22 for
+ * phy address 2 and 3. Page select is shifted only for
+ * phy address 1.
+ */
+ if (hw->phy.addr == 1) {
+ page_shift = IGP_PAGE_SHIFT;
+ page_select = IGP01E1000_PHY_PAGE_SELECT;
+ } else {
+ page_shift = 0;
+ page_select = BM_PHY_PAGE_SELECT;
+ }
+
+ /* Page is shifted left, PHY expects (page x 32) */
+ ret_val = e1000e_write_phy_reg_mdic(hw, page_select,
+ (page << page_shift));
+ if (ret_val) {
+ hw->phy.ops.release_phy(hw);
+ goto out;
+ }
+ }
+
+ ret_val = e1000e_write_phy_reg_mdic(hw, MAX_PHY_REG_ADDRESS & offset,
+ data);
+
+ hw->phy.ops.release_phy(hw);
+
+out:
+ return ret_val;
+}
+
+/**
+ * e1000e_read_phy_reg_bm - Read BM PHY register
+ * @hw: pointer to the HW structure
+ * @offset: register offset to be read
+ * @data: pointer to the read data
+ *
+ * Acquires semaphore, if necessary, then reads the PHY register at offset
+ * and storing the retrieved information in data. Release any acquired
+ * semaphores before exiting.
+ **/
+s32 e1000e_read_phy_reg_bm(struct e1000_hw *hw, u32 offset, u16 *data)
+{
+ s32 ret_val;
+ u32 page_select = 0;
+ u32 page = offset >> IGP_PAGE_SHIFT;
+ u32 page_shift = 0;
+
+ /* Page 800 works differently than the rest so it has its own func */
+ if (page == BM_WUC_PAGE) {
+ ret_val = e1000_access_phy_wakeup_reg_bm(hw, offset, data,
+ true);
+ goto out;
+ }
+
+ ret_val = hw->phy.ops.acquire_phy(hw);
+ if (ret_val)
+ goto out;
+
+ hw->phy.addr = e1000_get_phy_addr_for_bm_page(page, offset);
+
+ if (offset > MAX_PHY_MULTI_PAGE_REG) {
+ /*
+ * Page select is register 31 for phy address 1 and 22 for
+ * phy address 2 and 3. Page select is shifted only for
+ * phy address 1.
+ */
+ if (hw->phy.addr == 1) {
+ page_shift = IGP_PAGE_SHIFT;
+ page_select = IGP01E1000_PHY_PAGE_SELECT;
+ } else {
+ page_shift = 0;
+ page_select = BM_PHY_PAGE_SELECT;
+ }
+
+ /* Page is shifted left, PHY expects (page x 32) */
+ ret_val = e1000e_write_phy_reg_mdic(hw, page_select,
+ (page << page_shift));
+ if (ret_val) {
+ hw->phy.ops.release_phy(hw);
+ goto out;
+ }
+ }
+
+ ret_val = e1000e_read_phy_reg_mdic(hw, MAX_PHY_REG_ADDRESS & offset,
+ data);
+ hw->phy.ops.release_phy(hw);
+
+out:
+ return ret_val;
+}
+
+/**
+ * e1000_access_phy_wakeup_reg_bm - Read BM PHY wakeup register
+ * @hw: pointer to the HW structure
+ * @offset: register offset to be read or written
+ * @data: pointer to the data to read or write
+ * @read: determines if operation is read or write
+ *
+ * Acquires semaphore, if necessary, then reads the PHY register at offset
+ * and storing the retrieved information in data. Release any acquired
+ * semaphores before exiting. Note that procedure to read the wakeup
+ * registers are different. It works as such:
+ * 1) Set page 769, register 17, bit 2 = 1
+ * 2) Set page to 800 for host (801 if we were manageability)
+ * 3) Write the address using the address opcode (0x11)
+ * 4) Read or write the data using the data opcode (0x12)
+ * 5) Restore 769_17.2 to its original value
+ **/
+static s32 e1000_access_phy_wakeup_reg_bm(struct e1000_hw *hw, u32 offset,
+ u16 *data, bool read)
+{
+ s32 ret_val;
+ u16 reg = ((u16)offset) & PHY_REG_MASK;
+ u16 phy_reg = 0;
+ u8 phy_acquired = 1;
+
+
+ ret_val = hw->phy.ops.acquire_phy(hw);
+ if (ret_val) {
+ phy_acquired = 0;
+ goto out;
+ }
+
+ /* All operations in this function are phy address 1 */
+ hw->phy.addr = 1;
+
+ /* Set page 769 */
+ e1000e_write_phy_reg_mdic(hw, IGP01E1000_PHY_PAGE_SELECT,
+ (BM_WUC_ENABLE_PAGE << IGP_PAGE_SHIFT));
+
+ ret_val = e1000e_read_phy_reg_mdic(hw, BM_WUC_ENABLE_REG, &phy_reg);
+ if (ret_val)
+ goto out;
+
+ /* First clear bit 4 to avoid a power state change */
+ phy_reg &= ~(BM_WUC_HOST_WU_BIT);
+ ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_ENABLE_REG, phy_reg);
+ if (ret_val)
+ goto out;
+
+ /* Write bit 2 = 1, and clear bit 4 to 769_17 */
+ ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_ENABLE_REG,
+ phy_reg | BM_WUC_ENABLE_BIT);
+ if (ret_val)
+ goto out;
+
+ /* Select page 800 */
+ ret_val = e1000e_write_phy_reg_mdic(hw, IGP01E1000_PHY_PAGE_SELECT,
+ (BM_WUC_PAGE << IGP_PAGE_SHIFT));
+
+ /* Write the page 800 offset value using opcode 0x11 */
+ ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_ADDRESS_OPCODE, reg);
+ if (ret_val)
+ goto out;
+
+ if (read) {
+ /* Read the page 800 value using opcode 0x12 */
+ ret_val = e1000e_read_phy_reg_mdic(hw, BM_WUC_DATA_OPCODE,
+ data);
+ } else {
+ /* Read the page 800 value using opcode 0x12 */
+ ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_DATA_OPCODE,
+ *data);
+ }
+
+ if (ret_val)
+ goto out;
+
+ /*
+ * Restore 769_17.2 to its original value
+ * Set page 769
+ */
+ e1000e_write_phy_reg_mdic(hw, IGP01E1000_PHY_PAGE_SELECT,
+ (BM_WUC_ENABLE_PAGE << IGP_PAGE_SHIFT));
+
+ /* Clear 769_17.2 */
+ ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_ENABLE_REG, phy_reg);
+
+out:
+ if (phy_acquired == 1)
+ hw->phy.ops.release_phy(hw);
+ return ret_val;
+}
+
/**
* e1000e_commit_phy - Soft PHY reset
* @hw: pointer to the HW structure
0x0000,Cmd_MCast,
0x0076, /* link to next command */
#define CONF_NR_MULTICAST 0x44
- 0x0000, /* number of multicast addresses */
+ 0x0000, /* number of bytes in multicast address(es) */
#define CONF_MULTICAST 0x46
0x0000, 0x0000, 0x0000, /* some addresses */
0x0000, 0x0000, 0x0000,
static void eexp_setup_filter(struct net_device *dev)
{
- struct dev_mc_list *dmi = dev->mc_list;
+ struct dev_mc_list *dmi;
unsigned short ioaddr = dev->base_addr;
int count = dev->mc_count;
int i;
}
outw(CONF_NR_MULTICAST & ~31, ioaddr+SM_PTR);
- outw(count, ioaddr+SHADOW(CONF_NR_MULTICAST));
- for (i = 0; i < count; i++) {
- unsigned short *data = (unsigned short *)dmi->dmi_addr;
+ outw(6*count, ioaddr+SHADOW(CONF_NR_MULTICAST));
+ for (i = 0, dmi = dev->mc_list; i < count; i++, dmi = dmi->next) {
+ unsigned short *data;
if (!dmi) {
printk(KERN_INFO "%s: too few multicast addresses\n", dev->name);
break;
printk(KERN_INFO "%s: invalid multicast address length given.\n", dev->name);
continue;
}
+ data = (unsigned short *)dmi->dmi_addr;
outw((CONF_MULTICAST+(6*i)) & ~31, ioaddr+SM_PTR);
outw(data[0], ioaddr+SHADOW(CONF_MULTICAST+(6*i)));
outw((CONF_MULTICAST+(6*i)+2) & ~31, ioaddr+SM_PTR);
ret = of_address_to_resource(ofdev->node, 0, &res);
if (ret)
- return ret;
+ goto out_res;
snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", res.start);
kfree(new_bus->irq);
out_unmap_regs:
iounmap(fec->fecp);
+out_res:
out_fec:
kfree(fec);
out_mii:
static void gfar_netpoll(struct net_device *dev);
#endif
int gfar_clean_rx_ring(struct net_device *dev, int rx_work_limit);
+static int gfar_clean_tx_ring(struct net_device *dev);
static int gfar_process_frame(struct net_device *dev, struct sk_buff *skb, int length);
static void gfar_vlan_rx_register(struct net_device *netdev,
struct vlan_group *grp);
}
/* Changes the mac address if the controller is not running. */
-int gfar_set_mac_address(struct net_device *dev)
+static int gfar_set_mac_address(struct net_device *dev)
{
gfar_set_mac_for_addr(dev, 0, dev->dev_addr);
}
/* Interrupt Handler for Transmit complete */
-int gfar_clean_tx_ring(struct net_device *dev)
+static int gfar_clean_tx_ring(struct net_device *dev)
{
struct txbd8 *bdp;
struct gfar_private *priv = netdev_priv(dev);
extern void gfar_phy_test(struct mii_bus *bus, struct phy_device *phydev,
int enable, u32 regnum, u32 read);
void gfar_init_sysfs(struct net_device *dev);
+int gfar_local_mdio_write(struct gfar_mii __iomem *regs, int mii_id,
+ int regnum, u16 value);
+int gfar_local_mdio_read(struct gfar_mii __iomem *regs, int mii_id, int regnum);
#endif /* __GIANFAR_H */
spin_lock_irqsave(&priv->rxlock, flags);
if (length > priv->rx_buffer_size)
- return count;
+ goto out;
if (length == priv->rx_stash_size)
- return count;
+ goto out;
priv->rx_stash_size = length;
gfar_write(&priv->regs->attr, temp);
+out:
spin_unlock_irqrestore(&priv->rxlock, flags);
return count;
spin_lock_irqsave(&priv->rxlock, flags);
if (index > priv->rx_stash_size)
- return count;
+ goto out;
if (index == priv->rx_stash_index)
- return count;
+ goto out;
priv->rx_stash_index = index;
temp |= ATTRELI_EI(index);
gfar_write(&priv->regs->attreli, flags);
+out:
spin_unlock_irqrestore(&priv->rxlock, flags);
return count;
static int nsc_ircc_init_108(nsc_chip_t *chip, chipio_t *info);
static int nsc_ircc_init_338(nsc_chip_t *chip, chipio_t *info);
static int nsc_ircc_init_39x(nsc_chip_t *chip, chipio_t *info);
+#ifdef CONFIG_PNP
static int nsc_ircc_pnp_probe(struct pnp_dev *dev, const struct pnp_device_id *id);
+#endif
/* These are the known NSC chips */
static nsc_chip_t chips[] = {
MODULE_DEVICE_TABLE(pnp, nsc_ircc_pnp_table);
static struct pnp_driver nsc_ircc_pnp_driver = {
+#ifdef CONFIG_PNP
.name = "nsc-ircc",
.id_table = nsc_ircc_pnp_table,
.probe = nsc_ircc_pnp_probe,
+#endif
};
/* Some prototypes */
return 0;
}
+#ifdef CONFIG_PNP
/* PNP probing */
static int nsc_ircc_pnp_probe(struct pnp_dev *dev, const struct pnp_device_id *id)
{
return 0;
}
+#endif
/*
* Function nsc_ircc_setup (info)
static int pnp_driver_registered;
+#ifdef CONFIG_PNP
static int __init smsc_ircc_pnp_probe(struct pnp_dev *dev,
const struct pnp_device_id *dev_id)
{
.id_table = smsc_ircc_pnp_table,
.probe = smsc_ircc_pnp_probe,
};
-
+#else /* CONFIG_PNP */
+static struct pnp_driver smsc_ircc_pnp_driver;
+#endif
/*******************************************************************************
*
unregister_netdevice(dev);
if (list_empty(&port->vlans))
- macvlan_port_destroy(dev);
+ macvlan_port_destroy(port->dev);
}
static struct rtnl_link_ops macvlan_link_ops __read_mostly = {
u64 mtt_seg;
int err = -ENOMEM;
- if (page_shift < 12 || page_shift >= 32)
+ if (page_shift < (ffs(dev->caps.page_size_cap) - 1) || page_shift >= 32)
return -EINVAL;
/* All MTTs must fit in the same page */
*/
#define PHY_ADDR_REG 0x0000
#define SMI_REG 0x0004
+#define WINDOW_BASE(i) (0x0200 + ((i) << 3))
+#define WINDOW_SIZE(i) (0x0204 + ((i) << 3))
+#define WINDOW_REMAP_HIGH(i) (0x0280 + ((i) << 2))
+#define WINDOW_BAR_ENABLE 0x0290
+#define WINDOW_PROTECT(i) (0x0294 + ((i) << 4))
/*
* Per-port registers.
u32 late_collision;
};
+struct mv643xx_shared_private {
+ void __iomem *eth_base;
+
+ /* used to protect SMI_REG, which is shared across ports */
+ spinlock_t phy_lock;
+
+ u32 win_protect;
+
+ unsigned int t_clk;
+};
+
struct mv643xx_private {
+ struct mv643xx_shared_private *shared;
int port_num; /* User Ethernet port number */
+ struct mv643xx_shared_private *shared_smi;
+
u32 rx_sram_addr; /* Base address of rx sram area */
u32 rx_sram_size; /* Size of rx sram area */
u32 tx_sram_addr; /* Base address of tx sram area */
static char mv643xx_driver_name[] = "mv643xx_eth";
static char mv643xx_driver_version[] = "1.0";
-static void __iomem *mv643xx_eth_base;
-
-/* used to protect SMI_REG, which is shared across ports */
-static DEFINE_SPINLOCK(mv643xx_eth_phy_lock);
-
static inline u32 rdl(struct mv643xx_private *mp, int offset)
{
- return readl(mv643xx_eth_base + offset);
+ return readl(mp->shared->eth_base + offset);
}
static inline void wrl(struct mv643xx_private *mp, int offset, u32 data)
{
- writel(data, mv643xx_eth_base + offset);
+ writel(data, mp->shared->eth_base + offset);
}
/*
*
* INPUT:
* struct mv643xx_private *mp Ethernet port
- * unsigned int t_clk t_clk of the MV-643xx chip in HZ units
* unsigned int delay Delay in usec
*
* OUTPUT:
*
*/
static unsigned int eth_port_set_rx_coal(struct mv643xx_private *mp,
- unsigned int t_clk, unsigned int delay)
+ unsigned int delay)
{
unsigned int port_num = mp->port_num;
- unsigned int coal = ((t_clk / 1000000) * delay) / 64;
+ unsigned int coal = ((mp->shared->t_clk / 1000000) * delay) / 64;
/* Set RX Coalescing mechanism */
wrl(mp, SDMA_CONFIG_REG(port_num),
*
* INPUT:
* struct mv643xx_private *mp Ethernet port
- * unsigned int t_clk t_clk of the MV-643xx chip in HZ units
* unsigned int delay Delay in uSeconds
*
* OUTPUT:
*
*/
static unsigned int eth_port_set_tx_coal(struct mv643xx_private *mp,
- unsigned int t_clk, unsigned int delay)
+ unsigned int delay)
{
- unsigned int coal = ((t_clk / 1000000) * delay) / 64;
+ unsigned int coal = ((mp->shared->t_clk / 1000000) * delay) / 64;
/* Set TX Coalescing mechanism */
wrl(mp, TX_FIFO_URGENT_THRESHOLD_REG(mp->port_num), coal << 4);
#ifdef MV643XX_COAL
mp->rx_int_coal =
- eth_port_set_rx_coal(mp, 133000000, MV643XX_RX_COAL);
+ eth_port_set_rx_coal(mp, MV643XX_RX_COAL);
#endif
mp->tx_int_coal =
- eth_port_set_tx_coal(mp, 133000000, MV643XX_TX_COAL);
+ eth_port_set_tx_coal(mp, MV643XX_TX_COAL);
/* Unmask phy and link status changes interrupts */
wrl(mp, INTERRUPT_EXTEND_MASK_REG(port_num), ETH_INT_UNMASK_ALL_EXT);
return -ENODEV;
}
+ if (pd->shared == NULL) {
+ printk(KERN_ERR "No mv643xx_eth_platform_data->shared\n");
+ return -ENODEV;
+ }
+
dev = alloc_etherdev(sizeof(struct mv643xx_private));
if (!dev)
return -ENOMEM;
spin_lock_init(&mp->lock);
+ mp->shared = platform_get_drvdata(pd->shared);
port_num = mp->port_num = pd->port_number;
+ if (mp->shared->win_protect)
+ wrl(mp, WINDOW_PROTECT(port_num), mp->shared->win_protect);
+
+ mp->shared_smi = mp->shared;
+ if (pd->shared_smi != NULL)
+ mp->shared_smi = platform_get_drvdata(pd->shared_smi);
+
/* set default config values */
eth_port_uc_addr_get(mp, dev->dev_addr);
mp->rx_ring_size = PORT_DEFAULT_RECEIVE_QUEUE_SIZE;
return 0;
}
+static void mv643xx_eth_conf_mbus_windows(struct mv643xx_shared_private *msp,
+ struct mbus_dram_target_info *dram)
+{
+ void __iomem *base = msp->eth_base;
+ u32 win_enable;
+ u32 win_protect;
+ int i;
+
+ for (i = 0; i < 6; i++) {
+ writel(0, base + WINDOW_BASE(i));
+ writel(0, base + WINDOW_SIZE(i));
+ if (i < 4)
+ writel(0, base + WINDOW_REMAP_HIGH(i));
+ }
+
+ win_enable = 0x3f;
+ win_protect = 0;
+
+ for (i = 0; i < dram->num_cs; i++) {
+ struct mbus_dram_window *cs = dram->cs + i;
+
+ writel((cs->base & 0xffff0000) |
+ (cs->mbus_attr << 8) |
+ dram->mbus_dram_target_id, base + WINDOW_BASE(i));
+ writel((cs->size - 1) & 0xffff0000, base + WINDOW_SIZE(i));
+
+ win_enable &= ~(1 << i);
+ win_protect |= 3 << (2 * i);
+ }
+
+ writel(win_enable, base + WINDOW_BAR_ENABLE);
+ msp->win_protect = win_protect;
+}
+
static int mv643xx_eth_shared_probe(struct platform_device *pdev)
{
static int mv643xx_version_printed = 0;
+ struct mv643xx_eth_shared_platform_data *pd = pdev->dev.platform_data;
+ struct mv643xx_shared_private *msp;
struct resource *res;
+ int ret;
if (!mv643xx_version_printed++)
printk(KERN_NOTICE "MV-643xx 10/100/1000 Ethernet Driver\n");
+ ret = -EINVAL;
res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
if (res == NULL)
- return -ENODEV;
+ goto out;
- mv643xx_eth_base = ioremap(res->start, res->end - res->start + 1);
- if (mv643xx_eth_base == NULL)
- return -ENOMEM;
+ ret = -ENOMEM;
+ msp = kmalloc(sizeof(*msp), GFP_KERNEL);
+ if (msp == NULL)
+ goto out;
+ memset(msp, 0, sizeof(*msp));
+
+ msp->eth_base = ioremap(res->start, res->end - res->start + 1);
+ if (msp->eth_base == NULL)
+ goto out_free;
+
+ spin_lock_init(&msp->phy_lock);
+ msp->t_clk = (pd != NULL && pd->t_clk != 0) ? pd->t_clk : 133000000;
+
+ platform_set_drvdata(pdev, msp);
+
+ /*
+ * (Re-)program MBUS remapping windows if we are asked to.
+ */
+ if (pd != NULL && pd->dram != NULL)
+ mv643xx_eth_conf_mbus_windows(msp, pd->dram);
return 0;
+out_free:
+ kfree(msp);
+out:
+ return ret;
}
static int mv643xx_eth_shared_remove(struct platform_device *pdev)
{
- iounmap(mv643xx_eth_base);
- mv643xx_eth_base = NULL;
+ struct mv643xx_shared_private *msp = platform_get_drvdata(pdev);
+
+ iounmap(msp->eth_base);
+ kfree(msp);
return 0;
}
static void eth_port_read_smi_reg(struct mv643xx_private *mp,
unsigned int phy_reg, unsigned int *value)
{
+ void __iomem *smi_reg = mp->shared_smi->eth_base + SMI_REG;
int phy_addr = ethernet_phy_get(mp);
unsigned long flags;
int i;
/* the SMI register is a shared resource */
- spin_lock_irqsave(&mv643xx_eth_phy_lock, flags);
+ spin_lock_irqsave(&mp->shared_smi->phy_lock, flags);
/* wait for the SMI register to become available */
- for (i = 0; rdl(mp, SMI_REG) & ETH_SMI_BUSY; i++) {
+ for (i = 0; readl(smi_reg) & ETH_SMI_BUSY; i++) {
if (i == PHY_WAIT_ITERATIONS) {
printk("%s: PHY busy timeout\n", mp->dev->name);
goto out;
udelay(PHY_WAIT_MICRO_SECONDS);
}
- wrl(mp, SMI_REG,
- (phy_addr << 16) | (phy_reg << 21) | ETH_SMI_OPCODE_READ);
+ writel((phy_addr << 16) | (phy_reg << 21) | ETH_SMI_OPCODE_READ,
+ smi_reg);
/* now wait for the data to be valid */
- for (i = 0; !(rdl(mp, SMI_REG) & ETH_SMI_READ_VALID); i++) {
+ for (i = 0; !(readl(smi_reg) & ETH_SMI_READ_VALID); i++) {
if (i == PHY_WAIT_ITERATIONS) {
printk("%s: PHY read timeout\n", mp->dev->name);
goto out;
udelay(PHY_WAIT_MICRO_SECONDS);
}
- *value = rdl(mp, SMI_REG) & 0xffff;
+ *value = readl(smi_reg) & 0xffff;
out:
- spin_unlock_irqrestore(&mv643xx_eth_phy_lock, flags);
+ spin_unlock_irqrestore(&mp->shared_smi->phy_lock, flags);
}
/*
static void eth_port_write_smi_reg(struct mv643xx_private *mp,
unsigned int phy_reg, unsigned int value)
{
- int phy_addr;
- int i;
+ void __iomem *smi_reg = mp->shared_smi->eth_base + SMI_REG;
+ int phy_addr = ethernet_phy_get(mp);
unsigned long flags;
-
- phy_addr = ethernet_phy_get(mp);
+ int i;
/* the SMI register is a shared resource */
- spin_lock_irqsave(&mv643xx_eth_phy_lock, flags);
+ spin_lock_irqsave(&mp->shared_smi->phy_lock, flags);
/* wait for the SMI register to become available */
- for (i = 0; rdl(mp, SMI_REG) & ETH_SMI_BUSY; i++) {
+ for (i = 0; readl(smi_reg) & ETH_SMI_BUSY; i++) {
if (i == PHY_WAIT_ITERATIONS) {
printk("%s: PHY busy timeout\n", mp->dev->name);
goto out;
udelay(PHY_WAIT_MICRO_SECONDS);
}
- wrl(mp, SMI_REG, (phy_addr << 16) | (phy_reg << 21) |
- ETH_SMI_OPCODE_WRITE | (value & 0xffff));
+ writel((phy_addr << 16) | (phy_reg << 21) |
+ ETH_SMI_OPCODE_WRITE | (value & 0xffff), smi_reg);
out:
- spin_unlock_irqrestore(&mv643xx_eth_phy_lock, flags);
+ spin_unlock_irqrestore(&mp->shared_smi->phy_lock, flags);
}
/*
/* niu.c: Neptune ethernet driver.
*
- * Copyright (C) 2007 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 2007, 2008 David S. Miller (davem@davemloft.net)
*/
#include <linux/module.h>
#define DRV_MODULE_NAME "niu"
#define PFX DRV_MODULE_NAME ": "
-#define DRV_MODULE_VERSION "0.8"
-#define DRV_MODULE_RELDATE "April 24, 2008"
+#define DRV_MODULE_VERSION "0.9"
+#define DRV_MODULE_RELDATE "May 4, 2008"
static char version[] __devinitdata =
DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n";
parent->num_ports = nr64(ESPC_NUM_PORTS_MACS) &
ESPC_NUM_PORTS_MACS_VAL;
+ /* All of the current probing methods fail on
+ * Maramba on-board parts.
+ */
if (!parent->num_ports)
- return -ENODEV;
+ parent->num_ports = 4;
}
}
}
*************************************************************************/
#define DRV_NAME "pcnet32"
-#ifdef CONFIG_PCNET32_NAPI
-#define DRV_VERSION "1.34-NAPI"
-#else
-#define DRV_VERSION "1.34"
-#endif
-#define DRV_RELDATE "14.Aug.2007"
+#define DRV_VERSION "1.35"
+#define DRV_RELDATE "21.Apr.2008"
#define PFX DRV_NAME ": "
static const char *const version =
static void pcnet32_netif_stop(struct net_device *dev)
{
-#ifdef CONFIG_PCNET32_NAPI
struct pcnet32_private *lp = netdev_priv(dev);
-#endif
+
dev->trans_start = jiffies;
-#ifdef CONFIG_PCNET32_NAPI
napi_disable(&lp->napi);
-#endif
netif_tx_disable(dev);
}
static void pcnet32_netif_start(struct net_device *dev)
{
-#ifdef CONFIG_PCNET32_NAPI
struct pcnet32_private *lp = netdev_priv(dev);
ulong ioaddr = dev->base_addr;
u16 val;
-#endif
+
netif_wake_queue(dev);
-#ifdef CONFIG_PCNET32_NAPI
val = lp->a.read_csr(ioaddr, CSR3);
val &= 0x00ff;
lp->a.write_csr(ioaddr, CSR3, val);
napi_enable(&lp->napi);
-#endif
}
/*
rc = 1; /* default to fail */
if (netif_running(dev))
-#ifdef CONFIG_PCNET32_NAPI
pcnet32_netif_stop(dev);
-#else
- pcnet32_close(dev);
-#endif
spin_lock_irqsave(&lp->lock, flags);
lp->a.write_csr(ioaddr, CSR0, CSR0_STOP); /* stop the chip */
x = a->read_bcr(ioaddr, 32); /* reset internal loopback */
a->write_bcr(ioaddr, 32, (x & ~0x0002));
-#ifdef CONFIG_PCNET32_NAPI
if (netif_running(dev)) {
pcnet32_netif_start(dev);
pcnet32_restart(dev, CSR0_NORMAL);
lp->a.write_bcr(ioaddr, 20, 4); /* return to 16bit mode */
}
spin_unlock_irqrestore(&lp->lock, flags);
-#else
- if (netif_running(dev)) {
- spin_unlock_irqrestore(&lp->lock, flags);
- pcnet32_open(dev);
- } else {
- pcnet32_purge_rx_ring(dev);
- lp->a.write_bcr(ioaddr, 20, 4); /* return to 16bit mode */
- spin_unlock_irqrestore(&lp->lock, flags);
- }
-#endif
return (rc);
} /* end pcnet32_loopback_test */
}
dev->stats.rx_bytes += skb->len;
skb->protocol = eth_type_trans(skb, dev);
-#ifdef CONFIG_PCNET32_NAPI
netif_receive_skb(skb);
-#else
- netif_rx(skb);
-#endif
dev->last_rx = jiffies;
dev->stats.rx_packets++;
return;
return must_restart;
}
-#ifdef CONFIG_PCNET32_NAPI
static int pcnet32_poll(struct napi_struct *napi, int budget)
{
struct pcnet32_private *lp = container_of(napi, struct pcnet32_private, napi);
}
return work_done;
}
-#endif
#define PCNET32_REGS_PER_PHY 32
#define PCNET32_MAX_PHYS 32
/* napi.weight is used in both the napi and non-napi cases */
lp->napi.weight = lp->rx_ring_size / 2;
-#ifdef CONFIG_PCNET32_NAPI
netif_napi_add(dev, &lp->napi, pcnet32_poll, lp->rx_ring_size / 2);
-#endif
if (fdx && !(lp->options & PCNET32_PORT_ASEL) &&
((cards_found >= MAX_UNITS) || full_duplex[cards_found]))
goto err_free_ring;
}
-#ifdef CONFIG_PCNET32_NAPI
napi_enable(&lp->napi);
-#endif
/* Re-initialize the PCNET32, and start it when done. */
lp->a.write_csr(ioaddr, 1, (lp->init_dma_addr & 0xffff));
dev->name, csr0);
/* unlike for the lance, there is no restart needed */
}
-#ifdef CONFIG_PCNET32_NAPI
if (netif_rx_schedule_prep(dev, &lp->napi)) {
u16 val;
/* set interrupt masks */
__netif_rx_schedule(dev, &lp->napi);
break;
}
-#else
- pcnet32_rx(dev, lp->napi.weight);
- if (pcnet32_tx(dev)) {
- /* reset the chip to clear the error condition, then restart */
- lp->a.reset(ioaddr);
- lp->a.write_csr(ioaddr, CSR4, 0x0915); /* auto tx pad */
- pcnet32_restart(dev, CSR0_START);
- netif_wake_queue(dev);
- }
-#endif
csr0 = lp->a.read_csr(ioaddr, CSR0);
}
-#ifndef CONFIG_PCNET32_NAPI
- /* Set interrupt enable. */
- lp->a.write_csr(ioaddr, CSR0, CSR0_INTEN);
-#endif
-
if (netif_msg_intr(lp))
printk(KERN_DEBUG "%s: exiting interrupt, csr0=%#4.4x.\n",
dev->name, lp->a.read_csr(ioaddr, CSR0));
del_timer_sync(&lp->watchdog_timer);
netif_stop_queue(dev);
-#ifdef CONFIG_PCNET32_NAPI
napi_disable(&lp->napi);
-#endif
spin_lock_irqsave(&lp->lock, flags);
* Must not be called from interrupt context, or while the
* phydev->lock is held.
*/
-void phy_error(struct phy_device *phydev)
+static void phy_error(struct phy_device *phydev)
{
mutex_lock(&phydev->lock);
phydev->state = PHY_HALTED;
ULI526X_DBUG(0, "uli526x_open", 0);
- ret = request_irq(dev->irq, &uli526x_interrupt, IRQF_SHARED, dev->name, dev);
- if (ret)
- return ret;
-
/* system variable init */
db->cr6_data = CR6_DEFAULT | uli526x_cr6_user_set;
db->tx_packet_cnt = 0;
/* Initialize ULI526X board */
uli526x_init(dev);
+ ret = request_irq(dev->irq, &uli526x_interrupt, IRQF_SHARED, dev->name, dev);
+ if (ret)
+ return ret;
+
/* Active System Interface */
netif_wake_queue(dev);
* This setup frame initialize ULI526X address filter mode
*/
+#ifdef __BIG_ENDIAN
+#define FLT_SHIFT 16
+#else
+#define FLT_SHIFT 0
+#endif
+
static void send_filter_frame(struct net_device *dev, int mc_cnt)
{
struct uli526x_board_info *db = netdev_priv(dev);
/* Node address */
addrptr = (u16 *) dev->dev_addr;
- *suptr++ = addrptr[0];
- *suptr++ = addrptr[1];
- *suptr++ = addrptr[2];
+ *suptr++ = addrptr[0] << FLT_SHIFT;
+ *suptr++ = addrptr[1] << FLT_SHIFT;
+ *suptr++ = addrptr[2] << FLT_SHIFT;
/* broadcast address */
- *suptr++ = 0xffff;
- *suptr++ = 0xffff;
- *suptr++ = 0xffff;
+ *suptr++ = 0xffff << FLT_SHIFT;
+ *suptr++ = 0xffff << FLT_SHIFT;
+ *suptr++ = 0xffff << FLT_SHIFT;
/* fit the multicast address */
for (mcptr = dev->mc_list, i = 0; i < mc_cnt; i++, mcptr = mcptr->next) {
addrptr = (u16 *) mcptr->dmi_addr;
- *suptr++ = addrptr[0];
- *suptr++ = addrptr[1];
- *suptr++ = addrptr[2];
+ *suptr++ = addrptr[0] << FLT_SHIFT;
+ *suptr++ = addrptr[1] << FLT_SHIFT;
+ *suptr++ = addrptr[2] << FLT_SHIFT;
}
for (; i<14; i++) {
- *suptr++ = 0xffff;
- *suptr++ = 0xffff;
- *suptr++ = 0xffff;
+ *suptr++ = 0xffff << FLT_SHIFT;
+ *suptr++ = 0xffff << FLT_SHIFT;
+ *suptr++ = 0xffff << FLT_SHIFT;
}
/* prepare the setup frame */
#endif /* UGETH_VERBOSE_DEBUG */
#define UGETH_MSG_DEFAULT (NETIF_MSG_IFUP << 1 ) - 1
-void uec_set_ethtool_ops(struct net_device *netdev);
static DEFINE_SPINLOCK(ugeth_lock);
}
}
-static struct sk_buff *get_new_skb(struct ucc_geth_private *ugeth, u8 *bd)
+static struct sk_buff *get_new_skb(struct ucc_geth_private *ugeth,
+ u8 __iomem *bd)
{
struct sk_buff *skb = NULL;
skb->dev = ugeth->dev;
- out_be32(&((struct qe_bd *)bd)->buf,
+ out_be32(&((struct qe_bd __iomem *)bd)->buf,
dma_map_single(NULL,
skb->data,
ugeth->ug_info->uf_info.max_rx_buf_length +
UCC_GETH_RX_DATA_BUF_ALIGNMENT,
DMA_FROM_DEVICE));
- out_be32((u32 *)bd, (R_E | R_I | (in_be32((u32 *)bd) & R_W)));
+ out_be32((u32 __iomem *)bd,
+ (R_E | R_I | (in_be32((u32 __iomem*)bd) & R_W)));
return skb;
}
static int rx_bd_buffer_set(struct ucc_geth_private *ugeth, u8 rxQ)
{
- u8 *bd;
+ u8 __iomem *bd;
u32 bd_status;
struct sk_buff *skb;
int i;
i = 0;
do {
- bd_status = in_be32((u32*)bd);
+ bd_status = in_be32((u32 __iomem *)bd);
skb = get_new_skb(ugeth, bd);
if (!skb) /* If can not allocate data buffer,
}
static int fill_init_enet_entries(struct ucc_geth_private *ugeth,
- volatile u32 *p_start,
+ u32 *p_start,
u8 num_entries,
u32 thread_size,
u32 thread_alignment,
}
static int return_init_enet_entries(struct ucc_geth_private *ugeth,
- volatile u32 *p_start,
+ u32 *p_start,
u8 num_entries,
enum qe_risc_allocation risc,
int skip_page_for_first_entry)
int snum;
for (i = 0; i < num_entries; i++) {
+ u32 val = *p_start;
+
/* Check that this entry was actually valid --
needed in case failed in allocations */
- if ((*p_start & ENET_INIT_PARAM_RISC_MASK) == risc) {
+ if ((val & ENET_INIT_PARAM_RISC_MASK) == risc) {
snum =
- (u32) (*p_start & ENET_INIT_PARAM_SNUM_MASK) >>
+ (u32) (val & ENET_INIT_PARAM_SNUM_MASK) >>
ENET_INIT_PARAM_SNUM_SHIFT;
qe_put_snum((u8) snum);
if (!((i == 0) && skip_page_for_first_entry)) {
/* First entry of Rx does not have page */
init_enet_offset =
- (in_be32(p_start) &
- ENET_INIT_PARAM_PTR_MASK);
+ (val & ENET_INIT_PARAM_PTR_MASK);
qe_muram_free(init_enet_offset);
}
- *(p_start++) = 0; /* Just for cosmetics */
+ *p_start++ = 0;
}
}
#ifdef DEBUG
static int dump_init_enet_entries(struct ucc_geth_private *ugeth,
- volatile u32 *p_start,
+ u32 __iomem *p_start,
u8 num_entries,
u32 thread_size,
enum qe_risc_allocation risc,
int snum;
for (i = 0; i < num_entries; i++) {
+ u32 val = in_be32(p_start);
+
/* Check that this entry was actually valid --
needed in case failed in allocations */
- if ((*p_start & ENET_INIT_PARAM_RISC_MASK) == risc) {
+ if ((val & ENET_INIT_PARAM_RISC_MASK) == risc) {
snum =
- (u32) (*p_start & ENET_INIT_PARAM_SNUM_MASK) >>
+ (u32) (val & ENET_INIT_PARAM_SNUM_MASK) >>
ENET_INIT_PARAM_SNUM_SHIFT;
qe_put_snum((u8) snum);
if (!((i == 0) && skip_page_for_first_entry)) {
static int hw_clear_addr_in_paddr(struct ucc_geth_private *ugeth, u8 paddr_num)
{
- struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt;
+ struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt;
if (!(paddr_num < NUM_OF_PADDRS)) {
ugeth_warn("%s: Illagel paddr_num.", __FUNCTION__);
}
p_82xx_addr_filt =
- (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram->
+ (struct ucc_geth_82xx_address_filtering_pram __iomem *) ugeth->p_rx_glbl_pram->
addressfiltering;
/* Writing address ff.ff.ff.ff.ff.ff disables address
static void hw_add_addr_in_hash(struct ucc_geth_private *ugeth,
u8 *p_enet_addr)
{
- struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt;
+ struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt;
u32 cecr_subblock;
p_82xx_addr_filt =
- (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram->
+ (struct ucc_geth_82xx_address_filtering_pram __iomem *) ugeth->p_rx_glbl_pram->
addressfiltering;
cecr_subblock =
static void magic_packet_detection_enable(struct ucc_geth_private *ugeth)
{
struct ucc_fast_private *uccf;
- struct ucc_geth *ug_regs;
+ struct ucc_geth __iomem *ug_regs;
u32 maccfg2, uccm;
uccf = ugeth->uccf;
static void magic_packet_detection_disable(struct ucc_geth_private *ugeth)
{
struct ucc_fast_private *uccf;
- struct ucc_geth *ug_regs;
+ struct ucc_geth __iomem *ug_regs;
u32 maccfg2, uccm;
uccf = ugeth->uccf;
rx_firmware_statistics,
struct ucc_geth_hardware_statistics *hardware_statistics)
{
- struct ucc_fast *uf_regs;
- struct ucc_geth *ug_regs;
+ struct ucc_fast __iomem *uf_regs;
+ struct ucc_geth __iomem *ug_regs;
struct ucc_geth_tx_firmware_statistics_pram *p_tx_fw_statistics_pram;
struct ucc_geth_rx_firmware_statistics_pram *p_rx_fw_statistics_pram;
ug_regs = ugeth->ug_regs;
- uf_regs = (struct ucc_fast *) ug_regs;
+ uf_regs = (struct ucc_fast __iomem *) ug_regs;
p_tx_fw_statistics_pram = ugeth->p_tx_fw_statistics_pram;
p_rx_fw_statistics_pram = ugeth->p_rx_fw_statistics_pram;
}
#endif /* DEBUG */
-static void init_default_reg_vals(volatile u32 *upsmr_register,
- volatile u32 *maccfg1_register,
- volatile u32 *maccfg2_register)
+static void init_default_reg_vals(u32 __iomem *upsmr_register,
+ u32 __iomem *maccfg1_register,
+ u32 __iomem *maccfg2_register)
{
out_be32(upsmr_register, UCC_GETH_UPSMR_INIT);
out_be32(maccfg1_register, UCC_GETH_MACCFG1_INIT);
u8 alt_beb_truncation,
u8 max_retransmissions,
u8 collision_window,
- volatile u32 *hafdup_register)
+ u32 __iomem *hafdup_register)
{
u32 value = 0;
u8 non_btb_ipg,
u8 min_ifg,
u8 btb_ipg,
- volatile u32 *ipgifg_register)
+ u32 __iomem *ipgifg_register)
{
u32 value = 0;
int tx_flow_control_enable,
u16 pause_period,
u16 extension_field,
- volatile u32 *upsmr_register,
- volatile u32 *uempr_register,
- volatile u32 *maccfg1_register)
+ u32 __iomem *upsmr_register,
+ u32 __iomem *uempr_register,
+ u32 __iomem *maccfg1_register)
{
u32 value = 0;
static int init_hw_statistics_gathering_mode(int enable_hardware_statistics,
int auto_zero_hardware_statistics,
- volatile u32 *upsmr_register,
- volatile u16 *uescr_register)
+ u32 __iomem *upsmr_register,
+ u16 __iomem *uescr_register)
{
u32 upsmr_value = 0;
u16 uescr_value = 0;
static int init_firmware_statistics_gathering_mode(int
enable_tx_firmware_statistics,
int enable_rx_firmware_statistics,
- volatile u32 *tx_rmon_base_ptr,
+ u32 __iomem *tx_rmon_base_ptr,
u32 tx_firmware_statistics_structure_address,
- volatile u32 *rx_rmon_base_ptr,
+ u32 __iomem *rx_rmon_base_ptr,
u32 rx_firmware_statistics_structure_address,
- volatile u16 *temoder_register,
- volatile u32 *remoder_register)
+ u16 __iomem *temoder_register,
+ u32 __iomem *remoder_register)
{
/* Note: this function does not check if */
/* the parameters it receives are NULL */
u8 address_byte_3,
u8 address_byte_4,
u8 address_byte_5,
- volatile u32 *macstnaddr1_register,
- volatile u32 *macstnaddr2_register)
+ u32 __iomem *macstnaddr1_register,
+ u32 __iomem *macstnaddr2_register)
{
u32 value = 0;
}
static int init_check_frame_length_mode(int length_check,
- volatile u32 *maccfg2_register)
+ u32 __iomem *maccfg2_register)
{
u32 value = 0;
}
static int init_preamble_length(u8 preamble_length,
- volatile u32 *maccfg2_register)
+ u32 __iomem *maccfg2_register)
{
u32 value = 0;
static int init_rx_parameters(int reject_broadcast,
int receive_short_frames,
- int promiscuous, volatile u32 *upsmr_register)
+ int promiscuous, u32 __iomem *upsmr_register)
{
u32 value = 0;
}
static int init_max_rx_buff_len(u16 max_rx_buf_len,
- volatile u16 *mrblr_register)
+ u16 __iomem *mrblr_register)
{
/* max_rx_buf_len value must be a multiple of 128 */
if ((max_rx_buf_len == 0)
}
static int init_min_frame_len(u16 min_frame_length,
- volatile u16 *minflr_register,
- volatile u16 *mrblr_register)
+ u16 __iomem *minflr_register,
+ u16 __iomem *mrblr_register)
{
u16 mrblr_value = 0;
static int adjust_enet_interface(struct ucc_geth_private *ugeth)
{
struct ucc_geth_info *ug_info;
- struct ucc_geth *ug_regs;
- struct ucc_fast *uf_regs;
+ struct ucc_geth __iomem *ug_regs;
+ struct ucc_fast __iomem *uf_regs;
int ret_val;
u32 upsmr, maccfg2, tbiBaseAddress;
u16 value;
static void adjust_link(struct net_device *dev)
{
struct ucc_geth_private *ugeth = netdev_priv(dev);
- struct ucc_geth *ug_regs;
- struct ucc_fast *uf_regs;
+ struct ucc_geth __iomem *ug_regs;
+ struct ucc_fast __iomem *uf_regs;
struct phy_device *phydev = ugeth->phydev;
unsigned long flags;
int new_state = 0;
uccf = ugeth->uccf;
/* Clear acknowledge bit */
- temp = ugeth->p_rx_glbl_pram->rxgstpack;
+ temp = in_8(&ugeth->p_rx_glbl_pram->rxgstpack);
temp &= ~GRACEFUL_STOP_ACKNOWLEDGE_RX;
- ugeth->p_rx_glbl_pram->rxgstpack = temp;
+ out_8(&ugeth->p_rx_glbl_pram->rxgstpack, temp);
/* Keep issuing command and checking acknowledge bit until
it is asserted, according to spec */
qe_issue_cmd(QE_GRACEFUL_STOP_RX, cecr_subblock,
QE_CR_PROTOCOL_ETHERNET, 0);
- temp = ugeth->p_rx_glbl_pram->rxgstpack;
+ temp = in_8(&ugeth->p_rx_glbl_pram->rxgstpack);
} while (!(temp & GRACEFUL_STOP_ACKNOWLEDGE_RX));
uccf->stopped_rx = 1;
enum enet_addr_type
enet_addr_type)
{
- struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt;
+ struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt;
struct ucc_fast_private *uccf;
enum comm_dir comm_dir;
struct list_head *p_lh;
u16 i, num;
- u32 *addr_h, *addr_l;
+ u32 __iomem *addr_h;
+ u32 __iomem *addr_l;
u8 *p_counter;
uccf = ugeth->uccf;
p_82xx_addr_filt =
- (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram->
- addressfiltering;
+ (struct ucc_geth_82xx_address_filtering_pram __iomem *)
+ ugeth->p_rx_glbl_pram->addressfiltering;
if (enet_addr_type == ENET_ADDR_TYPE_GROUP) {
addr_h = &(p_82xx_addr_filt->gaddr_h);
static void ucc_geth_memclean(struct ucc_geth_private *ugeth)
{
u16 i, j;
- u8 *bd;
+ u8 __iomem *bd;
if (!ugeth)
return;
for (j = 0; j < ugeth->ug_info->bdRingLenTx[i]; j++) {
if (ugeth->tx_skbuff[i][j]) {
dma_unmap_single(NULL,
- ((struct qe_bd *)bd)->buf,
- (in_be32((u32 *)bd) &
+ in_be32(&((struct qe_bd __iomem *)bd)->buf),
+ (in_be32((u32 __iomem *)bd) &
BD_LENGTH_MASK),
DMA_TO_DEVICE);
dev_kfree_skb_any(ugeth->tx_skbuff[i][j]);
for (j = 0; j < ugeth->ug_info->bdRingLenRx[i]; j++) {
if (ugeth->rx_skbuff[i][j]) {
dma_unmap_single(NULL,
- ((struct qe_bd *)bd)->buf,
+ in_be32(&((struct qe_bd __iomem *)bd)->buf),
ugeth->ug_info->
uf_info.max_rx_buf_length +
UCC_GETH_RX_DATA_BUF_ALIGNMENT,
{
struct ucc_geth_private *ugeth;
struct dev_mc_list *dmi;
- struct ucc_fast *uf_regs;
- struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt;
+ struct ucc_fast __iomem *uf_regs;
+ struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt;
int i;
ugeth = netdev_priv(dev);
if (dev->flags & IFF_PROMISC) {
- uf_regs->upsmr |= UPSMR_PRO;
+ out_be32(&uf_regs->upsmr, in_be32(&uf_regs->upsmr) | UPSMR_PRO);
} else {
- uf_regs->upsmr &= ~UPSMR_PRO;
+ out_be32(&uf_regs->upsmr, in_be32(&uf_regs->upsmr)&~UPSMR_PRO);
p_82xx_addr_filt =
- (struct ucc_geth_82xx_address_filtering_pram *) ugeth->
+ (struct ucc_geth_82xx_address_filtering_pram __iomem *) ugeth->
p_rx_glbl_pram->addressfiltering;
if (dev->flags & IFF_ALLMULTI) {
static void ucc_geth_stop(struct ucc_geth_private *ugeth)
{
- struct ucc_geth *ug_regs = ugeth->ug_regs;
+ struct ucc_geth __iomem *ug_regs = ugeth->ug_regs;
struct phy_device *phydev = ugeth->phydev;
u32 tempval;
return -ENOMEM;
}
- ugeth->ug_regs = (struct ucc_geth *) ioremap(uf_info->regs, sizeof(struct ucc_geth));
+ ugeth->ug_regs = (struct ucc_geth __iomem *) ioremap(uf_info->regs, sizeof(struct ucc_geth));
return 0;
}
static int ucc_geth_startup(struct ucc_geth_private *ugeth)
{
- struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt;
- struct ucc_geth_init_pram *p_init_enet_pram;
+ struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt;
+ struct ucc_geth_init_pram __iomem *p_init_enet_pram;
struct ucc_fast_private *uccf;
struct ucc_geth_info *ug_info;
struct ucc_fast_info *uf_info;
- struct ucc_fast *uf_regs;
- struct ucc_geth *ug_regs;
+ struct ucc_fast __iomem *uf_regs;
+ struct ucc_geth __iomem *ug_regs;
int ret_val = -EINVAL;
u32 remoder = UCC_GETH_REMODER_INIT;
u32 init_enet_pram_offset, cecr_subblock, command, maccfg1;
u16 temoder = UCC_GETH_TEMODER_INIT;
u16 test;
u8 function_code = 0;
- u8 *bd, *endOfRing;
+ u8 __iomem *bd;
+ u8 __iomem *endOfRing;
u8 numThreadsRxNumerical, numThreadsTxNumerical;
ugeth_vdbg("%s: IN", __FUNCTION__);
if (UCC_GETH_TX_BD_RING_ALIGNMENT > 4)
align = UCC_GETH_TX_BD_RING_ALIGNMENT;
ugeth->tx_bd_ring_offset[j] =
- kmalloc((u32) (length + align), GFP_KERNEL);
+ (u32) kmalloc((u32) (length + align), GFP_KERNEL);
if (ugeth->tx_bd_ring_offset[j] != 0)
ugeth->p_tx_bd_ring[j] =
- (void*)((ugeth->tx_bd_ring_offset[j] +
+ (u8 __iomem *)((ugeth->tx_bd_ring_offset[j] +
align) & ~(align - 1));
} else if (uf_info->bd_mem_part == MEM_PART_MURAM) {
ugeth->tx_bd_ring_offset[j] =
UCC_GETH_TX_BD_RING_ALIGNMENT);
if (!IS_ERR_VALUE(ugeth->tx_bd_ring_offset[j]))
ugeth->p_tx_bd_ring[j] =
- (u8 *) qe_muram_addr(ugeth->
+ (u8 __iomem *) qe_muram_addr(ugeth->
tx_bd_ring_offset[j]);
}
if (!ugeth->p_tx_bd_ring[j]) {
return -ENOMEM;
}
/* Zero unused end of bd ring, according to spec */
- memset(ugeth->p_tx_bd_ring[j] +
- ug_info->bdRingLenTx[j] * sizeof(struct qe_bd), 0,
+ memset_io((void __iomem *)(ugeth->p_tx_bd_ring[j] +
+ ug_info->bdRingLenTx[j] * sizeof(struct qe_bd)), 0,
length - ug_info->bdRingLenTx[j] * sizeof(struct qe_bd));
}
if (UCC_GETH_RX_BD_RING_ALIGNMENT > 4)
align = UCC_GETH_RX_BD_RING_ALIGNMENT;
ugeth->rx_bd_ring_offset[j] =
- kmalloc((u32) (length + align), GFP_KERNEL);
+ (u32) kmalloc((u32) (length + align), GFP_KERNEL);
if (ugeth->rx_bd_ring_offset[j] != 0)
ugeth->p_rx_bd_ring[j] =
- (void*)((ugeth->rx_bd_ring_offset[j] +
+ (u8 __iomem *)((ugeth->rx_bd_ring_offset[j] +
align) & ~(align - 1));
} else if (uf_info->bd_mem_part == MEM_PART_MURAM) {
ugeth->rx_bd_ring_offset[j] =
UCC_GETH_RX_BD_RING_ALIGNMENT);
if (!IS_ERR_VALUE(ugeth->rx_bd_ring_offset[j]))
ugeth->p_rx_bd_ring[j] =
- (u8 *) qe_muram_addr(ugeth->
+ (u8 __iomem *) qe_muram_addr(ugeth->
rx_bd_ring_offset[j]);
}
if (!ugeth->p_rx_bd_ring[j]) {
bd = ugeth->confBd[j] = ugeth->txBd[j] = ugeth->p_tx_bd_ring[j];
for (i = 0; i < ug_info->bdRingLenTx[j]; i++) {
/* clear bd buffer */
- out_be32(&((struct qe_bd *)bd)->buf, 0);
+ out_be32(&((struct qe_bd __iomem *)bd)->buf, 0);
/* set bd status and length */
- out_be32((u32 *)bd, 0);
+ out_be32((u32 __iomem *)bd, 0);
bd += sizeof(struct qe_bd);
}
bd -= sizeof(struct qe_bd);
/* set bd status and length */
- out_be32((u32 *)bd, T_W); /* for last BD set Wrap bit */
+ out_be32((u32 __iomem *)bd, T_W); /* for last BD set Wrap bit */
}
/* Init Rx bds */
bd = ugeth->rxBd[j] = ugeth->p_rx_bd_ring[j];
for (i = 0; i < ug_info->bdRingLenRx[j]; i++) {
/* set bd status and length */
- out_be32((u32 *)bd, R_I);
+ out_be32((u32 __iomem *)bd, R_I);
/* clear bd buffer */
- out_be32(&((struct qe_bd *)bd)->buf, 0);
+ out_be32(&((struct qe_bd __iomem *)bd)->buf, 0);
bd += sizeof(struct qe_bd);
}
bd -= sizeof(struct qe_bd);
/* set bd status and length */
- out_be32((u32 *)bd, R_W); /* for last BD set Wrap bit */
+ out_be32((u32 __iomem *)bd, R_W); /* for last BD set Wrap bit */
}
/*
return -ENOMEM;
}
ugeth->p_tx_glbl_pram =
- (struct ucc_geth_tx_global_pram *) qe_muram_addr(ugeth->
+ (struct ucc_geth_tx_global_pram __iomem *) qe_muram_addr(ugeth->
tx_glbl_pram_offset);
/* Zero out p_tx_glbl_pram */
- memset(ugeth->p_tx_glbl_pram, 0, sizeof(struct ucc_geth_tx_global_pram));
+ memset_io((void __iomem *)ugeth->p_tx_glbl_pram, 0, sizeof(struct ucc_geth_tx_global_pram));
/* Fill global PRAM */
}
ugeth->p_thread_data_tx =
- (struct ucc_geth_thread_data_tx *) qe_muram_addr(ugeth->
+ (struct ucc_geth_thread_data_tx __iomem *) qe_muram_addr(ugeth->
thread_dat_tx_offset);
out_be32(&ugeth->p_tx_glbl_pram->tqptr, ugeth->thread_dat_tx_offset);
/* iphoffset */
for (i = 0; i < TX_IP_OFFSET_ENTRY_MAX; i++)
- ugeth->p_tx_glbl_pram->iphoffset[i] = ug_info->iphoffset[i];
+ out_8(&ugeth->p_tx_glbl_pram->iphoffset[i],
+ ug_info->iphoffset[i]);
/* SQPTR */
/* Size varies with number of Tx queues */
}
ugeth->p_send_q_mem_reg =
- (struct ucc_geth_send_queue_mem_region *) qe_muram_addr(ugeth->
+ (struct ucc_geth_send_queue_mem_region __iomem *) qe_muram_addr(ugeth->
send_q_mem_reg_offset);
out_be32(&ugeth->p_tx_glbl_pram->sqptr, ugeth->send_q_mem_reg_offset);
}
ugeth->p_scheduler =
- (struct ucc_geth_scheduler *) qe_muram_addr(ugeth->
+ (struct ucc_geth_scheduler __iomem *) qe_muram_addr(ugeth->
scheduler_offset);
out_be32(&ugeth->p_tx_glbl_pram->schedulerbasepointer,
ugeth->scheduler_offset);
/* Zero out p_scheduler */
- memset(ugeth->p_scheduler, 0, sizeof(struct ucc_geth_scheduler));
+ memset_io((void __iomem *)ugeth->p_scheduler, 0, sizeof(struct ucc_geth_scheduler));
/* Set values in scheduler */
out_be32(&ugeth->p_scheduler->mblinterval,
ug_info->mblinterval);
out_be16(&ugeth->p_scheduler->nortsrbytetime,
ug_info->nortsrbytetime);
- ugeth->p_scheduler->fracsiz = ug_info->fracsiz;
- ugeth->p_scheduler->strictpriorityq = ug_info->strictpriorityq;
- ugeth->p_scheduler->txasap = ug_info->txasap;
- ugeth->p_scheduler->extrabw = ug_info->extrabw;
+ out_8(&ugeth->p_scheduler->fracsiz, ug_info->fracsiz);
+ out_8(&ugeth->p_scheduler->strictpriorityq,
+ ug_info->strictpriorityq);
+ out_8(&ugeth->p_scheduler->txasap, ug_info->txasap);
+ out_8(&ugeth->p_scheduler->extrabw, ug_info->extrabw);
for (i = 0; i < NUM_TX_QUEUES; i++)
- ugeth->p_scheduler->weightfactor[i] =
- ug_info->weightfactor[i];
+ out_8(&ugeth->p_scheduler->weightfactor[i],
+ ug_info->weightfactor[i]);
/* Set pointers to cpucount registers in scheduler */
ugeth->p_cpucount[0] = &(ugeth->p_scheduler->cpucount0);
return -ENOMEM;
}
ugeth->p_tx_fw_statistics_pram =
- (struct ucc_geth_tx_firmware_statistics_pram *)
+ (struct ucc_geth_tx_firmware_statistics_pram __iomem *)
qe_muram_addr(ugeth->tx_fw_statistics_pram_offset);
/* Zero out p_tx_fw_statistics_pram */
- memset(ugeth->p_tx_fw_statistics_pram,
+ memset_io((void __iomem *)ugeth->p_tx_fw_statistics_pram,
0, sizeof(struct ucc_geth_tx_firmware_statistics_pram));
}
return -ENOMEM;
}
ugeth->p_rx_glbl_pram =
- (struct ucc_geth_rx_global_pram *) qe_muram_addr(ugeth->
+ (struct ucc_geth_rx_global_pram __iomem *) qe_muram_addr(ugeth->
rx_glbl_pram_offset);
/* Zero out p_rx_glbl_pram */
- memset(ugeth->p_rx_glbl_pram, 0, sizeof(struct ucc_geth_rx_global_pram));
+ memset_io((void __iomem *)ugeth->p_rx_glbl_pram, 0, sizeof(struct ucc_geth_rx_global_pram));
/* Fill global PRAM */
}
ugeth->p_thread_data_rx =
- (struct ucc_geth_thread_data_rx *) qe_muram_addr(ugeth->
+ (struct ucc_geth_thread_data_rx __iomem *) qe_muram_addr(ugeth->
thread_dat_rx_offset);
out_be32(&ugeth->p_rx_glbl_pram->rqptr, ugeth->thread_dat_rx_offset);
return -ENOMEM;
}
ugeth->p_rx_fw_statistics_pram =
- (struct ucc_geth_rx_firmware_statistics_pram *)
+ (struct ucc_geth_rx_firmware_statistics_pram __iomem *)
qe_muram_addr(ugeth->rx_fw_statistics_pram_offset);
/* Zero out p_rx_fw_statistics_pram */
- memset(ugeth->p_rx_fw_statistics_pram, 0,
+ memset_io((void __iomem *)ugeth->p_rx_fw_statistics_pram, 0,
sizeof(struct ucc_geth_rx_firmware_statistics_pram));
}
}
ugeth->p_rx_irq_coalescing_tbl =
- (struct ucc_geth_rx_interrupt_coalescing_table *)
+ (struct ucc_geth_rx_interrupt_coalescing_table __iomem *)
qe_muram_addr(ugeth->rx_irq_coalescing_tbl_offset);
out_be32(&ugeth->p_rx_glbl_pram->intcoalescingptr,
ugeth->rx_irq_coalescing_tbl_offset);
}
ugeth->p_rx_bd_qs_tbl =
- (struct ucc_geth_rx_bd_queues_entry *) qe_muram_addr(ugeth->
+ (struct ucc_geth_rx_bd_queues_entry __iomem *) qe_muram_addr(ugeth->
rx_bd_qs_tbl_offset);
out_be32(&ugeth->p_rx_glbl_pram->rbdqptr, ugeth->rx_bd_qs_tbl_offset);
/* Zero out p_rx_bd_qs_tbl */
- memset(ugeth->p_rx_bd_qs_tbl,
+ memset_io((void __iomem *)ugeth->p_rx_bd_qs_tbl,
0,
ug_info->numQueuesRx * (sizeof(struct ucc_geth_rx_bd_queues_entry) +
sizeof(struct ucc_geth_rx_prefetched_bds)));
&ugeth->p_rx_glbl_pram->remoder);
/* function code register */
- ugeth->p_rx_glbl_pram->rstate = function_code;
+ out_8(&ugeth->p_rx_glbl_pram->rstate, function_code);
/* initialize extended filtering */
if (ug_info->rxExtendedFiltering) {
}
ugeth->p_exf_glbl_param =
- (struct ucc_geth_exf_global_pram *) qe_muram_addr(ugeth->
+ (struct ucc_geth_exf_global_pram __iomem *) qe_muram_addr(ugeth->
exf_glbl_param_offset);
out_be32(&ugeth->p_rx_glbl_pram->exfGlobalParam,
ugeth->exf_glbl_param_offset);
ugeth_82xx_filtering_clear_addr_in_paddr(ugeth, (u8) j);
p_82xx_addr_filt =
- (struct ucc_geth_82xx_address_filtering_pram *) ugeth->
+ (struct ucc_geth_82xx_address_filtering_pram __iomem *) ugeth->
p_rx_glbl_pram->addressfiltering;
ugeth_82xx_filtering_clear_all_addr_in_hash(ugeth,
return -ENOMEM;
}
p_init_enet_pram =
- (struct ucc_geth_init_pram *) qe_muram_addr(init_enet_pram_offset);
+ (struct ucc_geth_init_pram __iomem *) qe_muram_addr(init_enet_pram_offset);
/* Copy shadow InitEnet command parameter structure into PRAM */
- p_init_enet_pram->resinit1 = ugeth->p_init_enet_param_shadow->resinit1;
- p_init_enet_pram->resinit2 = ugeth->p_init_enet_param_shadow->resinit2;
- p_init_enet_pram->resinit3 = ugeth->p_init_enet_param_shadow->resinit3;
- p_init_enet_pram->resinit4 = ugeth->p_init_enet_param_shadow->resinit4;
+ out_8(&p_init_enet_pram->resinit1,
+ ugeth->p_init_enet_param_shadow->resinit1);
+ out_8(&p_init_enet_pram->resinit2,
+ ugeth->p_init_enet_param_shadow->resinit2);
+ out_8(&p_init_enet_pram->resinit3,
+ ugeth->p_init_enet_param_shadow->resinit3);
+ out_8(&p_init_enet_pram->resinit4,
+ ugeth->p_init_enet_param_shadow->resinit4);
out_be16(&p_init_enet_pram->resinit5,
ugeth->p_init_enet_param_shadow->resinit5);
- p_init_enet_pram->largestexternallookupkeysize =
- ugeth->p_init_enet_param_shadow->largestexternallookupkeysize;
+ out_8(&p_init_enet_pram->largestexternallookupkeysize,
+ ugeth->p_init_enet_param_shadow->largestexternallookupkeysize);
out_be32(&p_init_enet_pram->rgftgfrxglobal,
ugeth->p_init_enet_param_shadow->rgftgfrxglobal);
for (i = 0; i < ENET_INIT_PARAM_MAX_ENTRIES_RX; i++)
#ifdef CONFIG_UGETH_TX_ON_DEMAND
struct ucc_fast_private *uccf;
#endif
- u8 *bd; /* BD pointer */
+ u8 __iomem *bd; /* BD pointer */
u32 bd_status;
u8 txQ = 0;
/* Start from the next BD that should be filled */
bd = ugeth->txBd[txQ];
- bd_status = in_be32((u32 *)bd);
+ bd_status = in_be32((u32 __iomem *)bd);
/* Save the skb pointer so we can free it later */
ugeth->tx_skbuff[txQ][ugeth->skb_curtx[txQ]] = skb;
1) & TX_RING_MOD_MASK(ugeth->ug_info->bdRingLenTx[txQ]);
/* set up the buffer descriptor */
- out_be32(&((struct qe_bd *)bd)->buf,
+ out_be32(&((struct qe_bd __iomem *)bd)->buf,
dma_map_single(NULL, skb->data, skb->len, DMA_TO_DEVICE));
/* printk(KERN_DEBUG"skb->data is 0x%x\n",skb->data); */
bd_status = (bd_status & T_W) | T_R | T_I | T_L | skb->len;
/* set bd status and length */
- out_be32((u32 *)bd, bd_status);
+ out_be32((u32 __iomem *)bd, bd_status);
dev->trans_start = jiffies;
static int ucc_geth_rx(struct ucc_geth_private *ugeth, u8 rxQ, int rx_work_limit)
{
struct sk_buff *skb;
- u8 *bd;
+ u8 __iomem *bd;
u16 length, howmany = 0;
u32 bd_status;
u8 *bdBuffer;
/* collect received buffers */
bd = ugeth->rxBd[rxQ];
- bd_status = in_be32((u32 *)bd);
+ bd_status = in_be32((u32 __iomem *)bd);
/* while there are received buffers and BD is full (~R_E) */
while (!((bd_status & (R_E)) || (--rx_work_limit < 0))) {
- bdBuffer = (u8 *) in_be32(&((struct qe_bd *)bd)->buf);
+ bdBuffer = (u8 *) in_be32(&((struct qe_bd __iomem *)bd)->buf);
length = (u16) ((bd_status & BD_LENGTH_MASK) - 4);
skb = ugeth->rx_skbuff[rxQ][ugeth->skb_currx[rxQ]];
else
bd += sizeof(struct qe_bd);
- bd_status = in_be32((u32 *)bd);
+ bd_status = in_be32((u32 __iomem *)bd);
}
ugeth->rxBd[rxQ] = bd;
{
/* Start from the next BD that should be filled */
struct ucc_geth_private *ugeth = netdev_priv(dev);
- u8 *bd; /* BD pointer */
+ u8 __iomem *bd; /* BD pointer */
u32 bd_status;
bd = ugeth->confBd[txQ];
- bd_status = in_be32((u32 *)bd);
+ bd_status = in_be32((u32 __iomem *)bd);
/* Normal processing. */
while ((bd_status & T_R) == 0) {
bd += sizeof(struct qe_bd);
else
bd = ugeth->p_tx_bd_ring[txQ];
- bd_status = in_be32((u32 *)bd);
+ bd_status = in_be32((u32 __iomem *)bd);
}
ugeth->confBd[txQ] = bd;
return 0;
return -EINVAL;
}
} else {
- prop = of_get_property(np, "rx-clock", NULL);
+ prop = of_get_property(np, "tx-clock", NULL);
if (!prop) {
printk(KERN_ERR
"ucc_geth: mising tx-clock-name property\n");
u32 iaddr_l; /* individual address filter, low */
u32 gaddr_h; /* group address filter, high */
u32 gaddr_l; /* group address filter, low */
- struct ucc_geth_82xx_enet_address taddr;
- struct ucc_geth_82xx_enet_address paddr[NUM_OF_PADDRS];
+ struct ucc_geth_82xx_enet_address __iomem taddr;
+ struct ucc_geth_82xx_enet_address __iomem paddr[NUM_OF_PADDRS];
u8 res0[0x40 - 0x38];
} __attribute__ ((packed));
struct ucc_fast_private *uccf;
struct net_device *dev;
struct napi_struct napi;
- struct ucc_geth *ug_regs;
+ struct ucc_geth __iomem *ug_regs;
struct ucc_geth_init_pram *p_init_enet_param_shadow;
- struct ucc_geth_exf_global_pram *p_exf_glbl_param;
+ struct ucc_geth_exf_global_pram __iomem *p_exf_glbl_param;
u32 exf_glbl_param_offset;
- struct ucc_geth_rx_global_pram *p_rx_glbl_pram;
+ struct ucc_geth_rx_global_pram __iomem *p_rx_glbl_pram;
u32 rx_glbl_pram_offset;
- struct ucc_geth_tx_global_pram *p_tx_glbl_pram;
+ struct ucc_geth_tx_global_pram __iomem *p_tx_glbl_pram;
u32 tx_glbl_pram_offset;
- struct ucc_geth_send_queue_mem_region *p_send_q_mem_reg;
+ struct ucc_geth_send_queue_mem_region __iomem *p_send_q_mem_reg;
u32 send_q_mem_reg_offset;
- struct ucc_geth_thread_data_tx *p_thread_data_tx;
+ struct ucc_geth_thread_data_tx __iomem *p_thread_data_tx;
u32 thread_dat_tx_offset;
- struct ucc_geth_thread_data_rx *p_thread_data_rx;
+ struct ucc_geth_thread_data_rx __iomem *p_thread_data_rx;
u32 thread_dat_rx_offset;
- struct ucc_geth_scheduler *p_scheduler;
+ struct ucc_geth_scheduler __iomem *p_scheduler;
u32 scheduler_offset;
- struct ucc_geth_tx_firmware_statistics_pram *p_tx_fw_statistics_pram;
+ struct ucc_geth_tx_firmware_statistics_pram __iomem *p_tx_fw_statistics_pram;
u32 tx_fw_statistics_pram_offset;
- struct ucc_geth_rx_firmware_statistics_pram *p_rx_fw_statistics_pram;
+ struct ucc_geth_rx_firmware_statistics_pram __iomem *p_rx_fw_statistics_pram;
u32 rx_fw_statistics_pram_offset;
- struct ucc_geth_rx_interrupt_coalescing_table *p_rx_irq_coalescing_tbl;
+ struct ucc_geth_rx_interrupt_coalescing_table __iomem *p_rx_irq_coalescing_tbl;
u32 rx_irq_coalescing_tbl_offset;
- struct ucc_geth_rx_bd_queues_entry *p_rx_bd_qs_tbl;
+ struct ucc_geth_rx_bd_queues_entry __iomem *p_rx_bd_qs_tbl;
u32 rx_bd_qs_tbl_offset;
- u8 *p_tx_bd_ring[NUM_TX_QUEUES];
+ u8 __iomem *p_tx_bd_ring[NUM_TX_QUEUES];
u32 tx_bd_ring_offset[NUM_TX_QUEUES];
- u8 *p_rx_bd_ring[NUM_RX_QUEUES];
+ u8 __iomem *p_rx_bd_ring[NUM_RX_QUEUES];
u32 rx_bd_ring_offset[NUM_RX_QUEUES];
- u8 *confBd[NUM_TX_QUEUES];
- u8 *txBd[NUM_TX_QUEUES];
- u8 *rxBd[NUM_RX_QUEUES];
+ u8 __iomem *confBd[NUM_TX_QUEUES];
+ u8 __iomem *txBd[NUM_TX_QUEUES];
+ u8 __iomem *rxBd[NUM_RX_QUEUES];
int badFrame[NUM_RX_QUEUES];
u16 cpucount[NUM_TX_QUEUES];
- volatile u16 *p_cpucount[NUM_TX_QUEUES];
+ u16 __iomem *p_cpucount[NUM_TX_QUEUES];
int indAddrRegUsed[NUM_OF_PADDRS];
u8 paddr[NUM_OF_PADDRS][ENET_NUM_OCTETS_PER_ADDRESS]; /* ethernet address */
u8 numGroupAddrInHash;
int oldlink;
};
+void uec_set_ethtool_ops(struct net_device *netdev);
+int init_flow_control_params(u32 automatic_flow_control_mode,
+ int rx_flow_control_enable, int tx_flow_control_enable,
+ u16 pause_period, u16 extension_field,
+ u32 __iomem *upsmr_register, u32 __iomem *uempr_register,
+ u32 __iomem *maccfg1_register);
+
+
#endif /* __UCC_GETH_H__ */
#define UEC_TX_FW_STATS_LEN ARRAY_SIZE(tx_fw_stat_gstrings)
#define UEC_RX_FW_STATS_LEN ARRAY_SIZE(rx_fw_stat_gstrings)
-extern int init_flow_control_params(u32 automatic_flow_control_mode,
- int rx_flow_control_enable,
- int tx_flow_control_enable, u16 pause_period,
- u16 extension_field, volatile u32 *upsmr_register,
- volatile u32 *uempr_register, volatile u32 *maccfg1_register);
-
static int
uec_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd)
{
}
/* Reset the MIIM registers, and wait for the bus to free */
-int uec_mdio_reset(struct mii_bus *bus)
+static int uec_mdio_reset(struct mii_bus *bus)
{
struct ucc_mii_mng __iomem *regs = (void __iomem *)bus->priv;
unsigned int timeout = PHY_INIT_TIMEOUT;
return err;
}
-int uec_mdio_remove(struct of_device *ofdev)
+static int uec_mdio_remove(struct of_device *ofdev)
{
struct device *device = &ofdev->dev;
struct mii_bus *bus = dev_get_drvdata(device);
// Buffalo LUA-U2-KTX
USB_DEVICE (0x0411, 0x003d),
.driver_info = (unsigned long) &ax8817x_info,
+}, {
+ // Buffalo LUA-U2-GT 10/100/1000
+ USB_DEVICE (0x0411, 0x006e),
+ .driver_info = (unsigned long) &ax88178_info,
}, {
// Sitecom LN-029 "USB 2.0 10/100 Ethernet adapter"
USB_DEVICE (0x6189, 0x182d),
list_for_each_safe(entry, tmp, &lapbeth_devices) {
lapbeth = list_entry(entry, struct lapbethdev, node);
+ dev_put(lapbeth->ethdev);
unregister_netdevice(lapbeth->axdev);
}
rtnl_unlock();
config IWLWIFI
- bool
- default n
+ tristate
config IWLCORE
tristate "Intel Wireless Wifi Core"
* reading the dword at 0x100 which must either be 0 or a valid extended
* capability header.
*/
-int pci_cfg_space_size_ext(struct pci_dev *dev, unsigned check_exp_pcix)
+int pci_cfg_space_size_ext(struct pci_dev *dev)
{
- int pos;
u32 status;
- if (!check_exp_pcix)
- goto skip;
+ if (pci_read_config_dword(dev, 256, &status) != PCIBIOS_SUCCESSFUL)
+ goto fail;
+ if (status == 0xffffffff)
+ goto fail;
+
+ return PCI_CFG_SPACE_EXP_SIZE;
+
+ fail:
+ return PCI_CFG_SPACE_SIZE;
+}
+
+int pci_cfg_space_size(struct pci_dev *dev)
+{
+ int pos;
+ u32 status;
pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
if (!pos) {
goto fail;
}
- skip:
- if (pci_read_config_dword(dev, 256, &status) != PCIBIOS_SUCCESSFUL)
- goto fail;
- if (status == 0xffffffff)
- goto fail;
-
- return PCI_CFG_SPACE_EXP_SIZE;
+ return pci_cfg_space_size_ext(dev);
fail:
return PCI_CFG_SPACE_SIZE;
}
-int pci_cfg_space_size(struct pci_dev *dev)
-{
- return pci_cfg_space_size_ext(dev, 1);
-}
-
static void pci_release_bus_bridge_dev(struct device *dev)
{
kfree(dev);
static int ds1511_rtc_set_time(struct device *dev, struct rtc_time *rtc_tm)
{
u8 mon, day, dow, hrs, min, sec, yrs, cen;
- unsigned int flags;
+ unsigned long flags;
/*
* won't have to change this for a while
static int ds1511_rtc_read_time(struct device *dev, struct rtc_time *rtc_tm)
{
unsigned int century;
- unsigned int flags;
+ unsigned long flags;
spin_lock_irqsave(&ds1511_lock, flags);
rtc_disable_update();
* Insert character into the screen at the current position with the
* current color and highlight. This function does NOT do cursor movement.
*/
-static int
-tty3270_put_character(struct tty3270 *tp, char ch)
+static void tty3270_put_character(struct tty3270 *tp, char ch)
{
struct tty3270_line *line;
struct tty3270_cell *cell;
cell->character = tp->view.ascebc[(unsigned int) ch];
cell->highlight = tp->highlight;
cell->f_color = tp->f_color;
- return 1;
}
/*
/*
* Put single characters to the ttys character buffer
*/
-static void
-tty3270_put_char(struct tty_struct *tty, unsigned char ch)
+static int tty3270_put_char(struct tty_struct *tty, unsigned char ch)
{
struct tty3270 *tp;
tp = tty->driver_data;
- if (!tp)
- return;
- if (tp->char_count < TTY3270_CHAR_BUF_SIZE)
- tp->char_buf[tp->char_count++] = ch;
+ if (!tp || tp->char_count >= TTY3270_CHAR_BUF_SIZE)
+ return 0;
+ tp->char_buf[tp->char_count++] = ch;
+ return 1;
}
/*
#include <asm/cio.h>
#include <asm/uaccess.h>
+#include <asm/cio.h>
#include "blacklist.h"
#include "cio.h"
* Function: blacklist_range
* (Un-)blacklist the devices from-to
*/
-static void
-blacklist_range (range_action action, unsigned int from, unsigned int to,
- unsigned int ssid)
+static int blacklist_range(range_action action, unsigned int from_ssid,
+ unsigned int to_ssid, unsigned int from,
+ unsigned int to, int msgtrigger)
{
- if (!to)
- to = from;
-
- if (from > to || to > __MAX_SUBCHANNEL || ssid > __MAX_SSID) {
- printk (KERN_WARNING "cio: Invalid blacklist range "
- "0.%x.%04x to 0.%x.%04x, skipping\n",
- ssid, from, ssid, to);
- return;
+ if ((from_ssid > to_ssid) || ((from_ssid == to_ssid) && (from > to))) {
+ if (msgtrigger)
+ printk(KERN_WARNING "cio: Invalid cio_ignore range "
+ "0.%x.%04x-0.%x.%04x\n", from_ssid, from,
+ to_ssid, to);
+ return 1;
}
- for (; from <= to; from++) {
+
+ while ((from_ssid < to_ssid) || ((from_ssid == to_ssid) &&
+ (from <= to))) {
if (action == add)
- set_bit (from, bl_dev[ssid]);
+ set_bit(from, bl_dev[from_ssid]);
else
- clear_bit (from, bl_dev[ssid]);
+ clear_bit(from, bl_dev[from_ssid]);
+ from++;
+ if (from > __MAX_SUBCHANNEL) {
+ from_ssid++;
+ from = 0;
+ }
}
+
+ return 0;
}
-/*
- * Function: blacklist_busid
- * Get devno/busid from given string.
- * Shamelessly grabbed from dasd_devmap.c.
- */
-static int
-blacklist_busid(char **str, int *id0, int *ssid, int *devno)
+static int pure_hex(char **cp, unsigned int *val, int min_digit,
+ int max_digit, int max_val)
{
- int val, old_style;
- char *sav;
+ int diff;
+ unsigned int value;
- sav = *str;
+ diff = 0;
+ *val = 0;
- /* check for leading '0x' */
- old_style = 0;
- if ((*str)[0] == '0' && (*str)[1] == 'x') {
- *str += 2;
- old_style = 1;
- }
- if (!isxdigit((*str)[0])) /* We require at least one hex digit */
- goto confused;
- val = simple_strtoul(*str, str, 16);
- if (old_style || (*str)[0] != '.') {
- *id0 = *ssid = 0;
- if (val < 0 || val > 0xffff)
- goto confused;
- *devno = val;
- if ((*str)[0] != ',' && (*str)[0] != '-' &&
- (*str)[0] != '\n' && (*str)[0] != '\0')
- goto confused;
- return 0;
+ while (isxdigit(**cp) && (diff <= max_digit)) {
+
+ if (isdigit(**cp))
+ value = **cp - '0';
+ else
+ value = tolower(**cp) - 'a' + 10;
+ *val = *val * 16 + value;
+ (*cp)++;
+ diff++;
}
- /* New style x.y.z busid */
- if (val < 0 || val > 0xff)
- goto confused;
- *id0 = val;
- (*str)++;
- if (!isxdigit((*str)[0])) /* We require at least one hex digit */
- goto confused;
- val = simple_strtoul(*str, str, 16);
- if (val < 0 || val > 0xff || (*str)++[0] != '.')
- goto confused;
- *ssid = val;
- if (!isxdigit((*str)[0])) /* We require at least one hex digit */
- goto confused;
- val = simple_strtoul(*str, str, 16);
- if (val < 0 || val > 0xffff)
- goto confused;
- *devno = val;
- if ((*str)[0] != ',' && (*str)[0] != '-' &&
- (*str)[0] != '\n' && (*str)[0] != '\0')
- goto confused;
+
+ if ((diff < min_digit) || (diff > max_digit) || (*val > max_val))
+ return 1;
+
return 0;
-confused:
- strsep(str, ",\n");
- printk(KERN_WARNING "cio: Invalid cio_ignore parameter '%s'\n", sav);
- return 1;
}
-static int
-blacklist_parse_parameters (char *str, range_action action)
+static int parse_busid(char *str, int *cssid, int *ssid, int *devno,
+ int msgtrigger)
{
- int from, to, from_id0, to_id0, from_ssid, to_ssid;
-
- while (*str != 0 && *str != '\n') {
- range_action ra = action;
- while(*str == ',')
- str++;
- if (*str == '!') {
- ra = !action;
- ++str;
+ char *str_work;
+ int val, rc, ret;
+
+ rc = 1;
+
+ if (*str == '\0')
+ goto out;
+
+ /* old style */
+ str_work = str;
+ val = simple_strtoul(str, &str_work, 16);
+
+ if (*str_work == '\0') {
+ if (val <= __MAX_SUBCHANNEL) {
+ *devno = val;
+ *ssid = 0;
+ *cssid = 0;
+ rc = 0;
}
+ goto out;
+ }
- /*
- * Since we have to parse the proc commands and the
- * kernel arguments we have to check four cases
- */
- if (strncmp(str,"all,",4) == 0 || strcmp(str,"all") == 0 ||
- strncmp(str,"all\n",4) == 0 || strncmp(str,"all ",4) == 0) {
- int j;
-
- str += 3;
- for (j=0; j <= __MAX_SSID; j++)
- blacklist_range(ra, 0, __MAX_SUBCHANNEL, j);
- } else {
- int rc;
+ /* new style */
+ str_work = str;
+ ret = pure_hex(&str_work, cssid, 1, 2, __MAX_CSSID);
+ if (ret || (str_work[0] != '.'))
+ goto out;
+ str_work++;
+ ret = pure_hex(&str_work, ssid, 1, 1, __MAX_SSID);
+ if (ret || (str_work[0] != '.'))
+ goto out;
+ str_work++;
+ ret = pure_hex(&str_work, devno, 4, 4, __MAX_SUBCHANNEL);
+ if (ret || (str_work[0] != '\0'))
+ goto out;
+
+ rc = 0;
+out:
+ if (rc && msgtrigger)
+ printk(KERN_WARNING "cio: Invalid cio_ignore device '%s'\n",
+ str);
+
+ return rc;
+}
- rc = blacklist_busid(&str, &from_id0,
- &from_ssid, &from);
- if (rc)
- continue;
- to = from;
- to_id0 = from_id0;
- to_ssid = from_ssid;
- if (*str == '-') {
- str++;
- rc = blacklist_busid(&str, &to_id0,
- &to_ssid, &to);
- if (rc)
- continue;
- }
- if (*str == '-') {
- printk(KERN_WARNING "cio: invalid cio_ignore "
- "parameter '%s'\n",
- strsep(&str, ",\n"));
- continue;
- }
- if ((from_id0 != to_id0) ||
- (from_ssid != to_ssid)) {
- printk(KERN_WARNING "cio: invalid cio_ignore "
- "range %x.%x.%04x-%x.%x.%04x\n",
- from_id0, from_ssid, from,
- to_id0, to_ssid, to);
- continue;
+static int blacklist_parse_parameters(char *str, range_action action,
+ int msgtrigger)
+{
+ int from_cssid, to_cssid, from_ssid, to_ssid, from, to;
+ int rc, totalrc;
+ char *parm;
+ range_action ra;
+
+ totalrc = 0;
+
+ while ((parm = strsep(&str, ","))) {
+ rc = 0;
+ ra = action;
+ if (*parm == '!') {
+ if (ra == add)
+ ra = free;
+ else
+ ra = add;
+ parm++;
+ }
+ if (strcmp(parm, "all") == 0) {
+ from_cssid = 0;
+ from_ssid = 0;
+ from = 0;
+ to_cssid = __MAX_CSSID;
+ to_ssid = __MAX_SSID;
+ to = __MAX_SUBCHANNEL;
+ } else {
+ rc = parse_busid(strsep(&parm, "-"), &from_cssid,
+ &from_ssid, &from, msgtrigger);
+ if (!rc) {
+ if (parm != NULL)
+ rc = parse_busid(parm, &to_cssid,
+ &to_ssid, &to,
+ msgtrigger);
+ else {
+ to_cssid = from_cssid;
+ to_ssid = from_ssid;
+ to = from;
+ }
}
- blacklist_range (ra, from, to, to_ssid);
}
+ if (!rc) {
+ rc = blacklist_range(ra, from_ssid, to_ssid, from, to,
+ msgtrigger);
+ if (rc)
+ totalrc = 1;
+ } else
+ totalrc = 1;
}
- return 1;
+
+ return totalrc;
}
-/* Parsing the commandline for blacklist parameters, e.g. to blacklist
- * bus ids 0.0.1234, 0.0.1235 and 0.0.1236, you could use any of:
- * - cio_ignore=1234-1236
- * - cio_ignore=0x1234-0x1235,1236
- * - cio_ignore=0x1234,1235-1236
- * - cio_ignore=1236 cio_ignore=1234-0x1236
- * - cio_ignore=1234 cio_ignore=1236 cio_ignore=0x1235
- * - cio_ignore=0.0.1234-0.0.1236
- * - cio_ignore=0.0.1234,0x1235,1236
- * - ...
- */
static int __init
blacklist_setup (char *str)
{
CIO_MSG_EVENT(6, "Reading blacklist parameters\n");
- return blacklist_parse_parameters (str, add);
+ if (blacklist_parse_parameters(str, add, 1))
+ return 0;
+ return 1;
}
__setup ("cio_ignore=", blacklist_setup);
* Function: blacklist_parse_proc_parameters
* parse the stuff which is piped to /proc/cio_ignore
*/
-static void
-blacklist_parse_proc_parameters (char *buf)
+static int blacklist_parse_proc_parameters(char *buf)
{
- if (strncmp (buf, "free ", 5) == 0) {
- blacklist_parse_parameters (buf + 5, free);
- } else if (strncmp (buf, "add ", 4) == 0) {
- /*
- * We don't need to check for known devices since
- * css_probe_device will handle this correctly.
- */
- blacklist_parse_parameters (buf + 4, add);
- } else {
- printk (KERN_WARNING "cio: cio_ignore: Parse error; \n"
- KERN_WARNING "try using 'free all|<devno-range>,"
- "<devno-range>,...'\n"
- KERN_WARNING "or 'add <devno-range>,"
- "<devno-range>,...'\n");
- return;
- }
+ int rc;
+ char *parm;
+
+ parm = strsep(&buf, " ");
+
+ if (strcmp("free", parm) == 0)
+ rc = blacklist_parse_parameters(buf, free, 0);
+ else if (strcmp("add", parm) == 0)
+ rc = blacklist_parse_parameters(buf, add, 0);
+ else
+ return 1;
css_schedule_reprobe();
+
+ return rc;
}
/* Iterator struct for all devices. */
size_t user_len, loff_t *offset)
{
char *buf;
+ size_t i;
+ ssize_t rc, ret;
if (*offset)
return -EINVAL;
buf = vmalloc (user_len + 1); /* maybe better use the stack? */
if (buf == NULL)
return -ENOMEM;
+ memset(buf, 0, user_len + 1);
+
if (strncpy_from_user (buf, user_buf, user_len) < 0) {
- vfree (buf);
- return -EFAULT;
+ rc = -EFAULT;
+ goto out_free;
}
- buf[user_len] = '\0';
- blacklist_parse_proc_parameters (buf);
+ i = user_len - 1;
+ while ((i >= 0) && (isspace(buf[i]) || (buf[i] == 0))) {
+ buf[i] = '\0';
+ i--;
+ }
+ ret = blacklist_parse_proc_parameters(buf);
+ if (ret)
+ rc = -EINVAL;
+ else
+ rc = user_len;
+out_free:
vfree (buf);
- return user_len;
+ return rc;
}
static const struct seq_operations cio_ignore_proc_seq_ops = {
debug_info_t *cio_debug_trace_id;
debug_info_t *cio_debug_crw_id;
-int cio_show_msg;
-
-static int __init
-cio_setup (char *parm)
-{
- if (!strcmp (parm, "yes"))
- cio_show_msg = 1;
- else if (!strcmp (parm, "no"))
- cio_show_msg = 0;
- else
- printk(KERN_ERR "cio: cio_setup: "
- "invalid cio_msg parameter '%s'", parm);
- return 1;
-}
-
-__setup ("cio_msg=", cio_setup);
-
/*
* Function: cio_debug_init
* Initializes three debug logs for common I/O:
stsch (sch->schid, &sch->schib);
- CIO_MSG_EVENT(0, "cio_start: 'not oper' status for "
+ CIO_MSG_EVENT(2, "cio_start: 'not oper' status for "
"subchannel 0.%x.%04x!\n", sch->schid.ssid,
sch->schid.sch_no);
sprintf(dbf_text, "no%s", sch->dev.bus_id);
* ... just being curious we check for non I/O subchannels
*/
if (sch->st != 0) {
- CIO_DEBUG(KERN_INFO, 0,
- "Subchannel 0.%x.%04x reports "
- "non-I/O subchannel type %04X\n",
- sch->schid.ssid, sch->schid.sch_no, sch->st);
+ CIO_MSG_EVENT(4, "Subchannel 0.%x.%04x reports "
+ "non-I/O subchannel type %04X\n",
+ sch->schid.ssid, sch->schid.sch_no, sch->st);
/* We stop here for non-io subchannels. */
err = sch->st;
goto out;
* This device must not be known to Linux. So we simply
* say that there is no device and return ENODEV.
*/
- CIO_MSG_EVENT(4, "Blacklisted device detected "
+ CIO_MSG_EVENT(6, "Blacklisted device detected "
"at devno %04X, subchannel set %x\n",
sch->schib.pmcw.dev, sch->schid.ssid);
err = -ENODEV;
sch->lpm = sch->schib.pmcw.pam & sch->opm;
sch->isc = 3;
- CIO_DEBUG(KERN_INFO, 0,
- "Detected device %04x on subchannel 0.%x.%04X"
- " - PIM = %02X, PAM = %02X, POM = %02X\n",
- sch->schib.pmcw.dev, sch->schid.ssid,
- sch->schid.sch_no, sch->schib.pmcw.pim,
- sch->schib.pmcw.pam, sch->schib.pmcw.pom);
+ CIO_MSG_EVENT(6, "Detected device %04x on subchannel 0.%x.%04X "
+ "- PIM = %02X, PAM = %02X, POM = %02X\n",
+ sch->schib.pmcw.dev, sch->schid.ssid,
+ sch->schid.sch_no, sch->schib.pmcw.pim,
+ sch->schib.pmcw.pam, sch->schib.pmcw.pom);
/*
* We now have to initially ...
#define cio_get_console_priv() NULL
#endif
-extern int cio_show_msg;
-
#endif
}
}
-#define CIO_DEBUG(printk_level, event_level, msg...) do { \
- if (cio_show_msg) \
- printk(printk_level "cio: " msg); \
- CIO_MSG_EVENT(event_level, msg); \
- } while (0)
-
#endif
{
int ret;
- CIO_MSG_EVENT(2, "reprobe start\n");
+ CIO_MSG_EVENT(4, "reprobe start\n");
need_reprobe = 0;
/* Make sure initial subchannel scan is done. */
atomic_read(&ccw_device_init_count) == 0);
ret = for_each_subchannel_staged(NULL, reprobe_subchannel, NULL);
- CIO_MSG_EVENT(2, "reprobe done (rc=%d, need_reprobe=%d)\n", ret,
+ CIO_MSG_EVENT(4, "reprobe done (rc=%d, need_reprobe=%d)\n", ret,
need_reprobe);
}
rc = device_schedule_callback(&cdev->dev,
ccw_device_remove_orphan_cb);
if (rc)
- CIO_MSG_EVENT(2, "Couldn't unregister orphan "
+ CIO_MSG_EVENT(0, "Couldn't unregister orphan "
"0.%x.%04x\n",
cdev->private->dev_id.ssid,
cdev->private->dev_id.devno);
rc = device_schedule_callback(cdev->dev.parent,
ccw_device_remove_sch_cb);
if (rc)
- CIO_MSG_EVENT(2, "Couldn't unregister disconnected device "
+ CIO_MSG_EVENT(0, "Couldn't unregister disconnected device "
"0.%x.%04x\n",
cdev->private->dev_id.ssid,
cdev->private->dev_id.devno);
if (ret == 0)
wait_event(cdev->private->wait_q, dev_fsm_final_state(cdev));
else {
- CIO_MSG_EVENT(2, "ccw_device_offline returned %d, "
+ CIO_MSG_EVENT(0, "ccw_device_offline returned %d, "
"device 0.%x.%04x\n",
ret, cdev->private->dev_id.ssid,
cdev->private->dev_id.devno);
if (ret == 0)
wait_event(cdev->private->wait_q, dev_fsm_final_state(cdev));
else {
- CIO_MSG_EVENT(2, "ccw_device_online returned %d, "
+ CIO_MSG_EVENT(0, "ccw_device_online returned %d, "
"device 0.%x.%04x\n",
ret, cdev->private->dev_id.ssid,
cdev->private->dev_id.devno);
if (ret == 0)
wait_event(cdev->private->wait_q, dev_fsm_final_state(cdev));
else
- CIO_MSG_EVENT(2, "ccw_device_offline returned %d, "
+ CIO_MSG_EVENT(0, "ccw_device_offline returned %d, "
"device 0.%x.%04x\n",
ret, cdev->private->dev_id.ssid,
cdev->private->dev_id.devno);
other_sch = to_subchannel(get_device(cdev->dev.parent));
ret = device_move(&cdev->dev, &sch->dev);
if (ret) {
- CIO_MSG_EVENT(2, "Moving disconnected device 0.%x.%04x failed "
+ CIO_MSG_EVENT(0, "Moving disconnected device 0.%x.%04x failed "
"(ret=%d)!\n", cdev->private->dev_id.ssid,
cdev->private->dev_id.devno, ret);
put_device(&other_sch->dev);
ret = device_reprobe(&cdev->dev);
if (ret)
/* We can't do much here. */
- CIO_MSG_EVENT(2, "device_reprobe() returned"
+ CIO_MSG_EVENT(0, "device_reprobe() returned"
" %d for 0.%x.%04x\n", ret,
cdev->private->dev_id.ssid,
cdev->private->dev_id.devno);
rc = device_move(&cdev->dev, &sch->dev);
mutex_unlock(&sch->reg_mutex);
if (rc) {
- CIO_MSG_EVENT(2, "Moving device 0.%x.%04x to subchannel "
+ CIO_MSG_EVENT(0, "Moving device 0.%x.%04x to subchannel "
"0.%x.%04x failed (ret=%d)!\n",
cdev->private->dev_id.ssid,
cdev->private->dev_id.devno, sch->schid.ssid,
wait_event(cdev->private->wait_q,
dev_fsm_final_state(cdev));
else
- //FIXME: we can't fail!
- CIO_MSG_EVENT(2, "ccw_device_offline returned %d, "
+ CIO_MSG_EVENT(0, "ccw_device_offline returned %d, "
"device 0.%x.%04x\n",
ret, cdev->private->dev_id.ssid,
cdev->private->dev_id.devno);
spin_lock_irq(cdev->ccwlock);
switch (cdev->private->state) {
case DEV_STATE_DISCONNECTED:
- CIO_MSG_EVENT(3, "recovery: trigger 0.%x.%04x\n",
+ CIO_MSG_EVENT(4, "recovery: trigger 0.%x.%04x\n",
cdev->private->dev_id.ssid,
cdev->private->dev_id.devno);
dev_fsm_event(cdev, DEV_EVENT_VERIFY);
}
spin_unlock_irq(&recovery_lock);
} else
- CIO_MSG_EVENT(2, "recovery: end\n");
+ CIO_MSG_EVENT(4, "recovery: end\n");
}
static DECLARE_WORK(recovery_work, recovery_work_func);
{
unsigned long flags;
- CIO_MSG_EVENT(2, "recovery: schedule\n");
+ CIO_MSG_EVENT(4, "recovery: schedule\n");
spin_lock_irqsave(&recovery_lock, flags);
if (!timer_pending(&recovery_timer) || (recovery_phase != 0)) {
recovery_phase = 0;
same_dev = 0; /* Keep the compiler quiet... */
switch (state) {
case DEV_STATE_NOT_OPER:
- CIO_DEBUG(KERN_WARNING, 2,
- "SenseID : unknown device %04x on subchannel "
- "0.%x.%04x\n", cdev->private->dev_id.devno,
- sch->schid.ssid, sch->schid.sch_no);
+ CIO_MSG_EVENT(2, "SenseID : unknown device %04x on "
+ "subchannel 0.%x.%04x\n",
+ cdev->private->dev_id.devno,
+ sch->schid.ssid, sch->schid.sch_no);
break;
case DEV_STATE_OFFLINE:
if (cdev->private->state == DEV_STATE_DISCONNECTED_SENSE_ID) {
return;
}
/* Issue device info message. */
- CIO_DEBUG(KERN_INFO, 2,
- "SenseID : device 0.%x.%04x reports: "
- "CU Type/Mod = %04X/%02X, Dev Type/Mod = "
- "%04X/%02X\n",
- cdev->private->dev_id.ssid,
- cdev->private->dev_id.devno,
- cdev->id.cu_type, cdev->id.cu_model,
- cdev->id.dev_type, cdev->id.dev_model);
+ CIO_MSG_EVENT(4, "SenseID : device 0.%x.%04x reports: "
+ "CU Type/Mod = %04X/%02X, Dev Type/Mod = "
+ "%04X/%02X\n",
+ cdev->private->dev_id.ssid,
+ cdev->private->dev_id.devno,
+ cdev->id.cu_type, cdev->id.cu_model,
+ cdev->id.dev_type, cdev->id.dev_model);
break;
case DEV_STATE_BOXED:
- CIO_DEBUG(KERN_WARNING, 2,
- "SenseID : boxed device %04x on subchannel "
- "0.%x.%04x\n", cdev->private->dev_id.devno,
- sch->schid.ssid, sch->schid.sch_no);
+ CIO_MSG_EVENT(0, "SenseID : boxed device %04x on "
+ " subchannel 0.%x.%04x\n",
+ cdev->private->dev_id.devno,
+ sch->schid.ssid, sch->schid.sch_no);
break;
}
cdev->private->state = state;
if (state == DEV_STATE_BOXED)
- CIO_DEBUG(KERN_WARNING, 2,
- "Boxed device %04x on subchannel %04x\n",
- cdev->private->dev_id.devno, sch->schid.sch_no);
+ CIO_MSG_EVENT(0, "Boxed device %04x on subchannel %04x\n",
+ cdev->private->dev_id.devno, sch->schid.sch_no);
if (cdev->private->flags.donotify) {
cdev->private->flags.donotify = 0;
/* Basic sense hasn't started. Try again. */
ccw_device_do_sense(cdev, irb);
else {
- CIO_MSG_EVENT(2, "Huh? 0.%x.%04x: unsolicited "
+ CIO_MSG_EVENT(0, "0.%x.%04x: unsolicited "
"interrupt during w4sense...\n",
cdev->private->dev_id.ssid,
cdev->private->dev_id.devno);
static void
ccw_device_bug(struct ccw_device *cdev, enum dev_event dev_event)
{
- CIO_MSG_EVENT(0, "dev_jumptable[%i][%i] == NULL\n",
- cdev->private->state, dev_event);
+ CIO_MSG_EVENT(0, "Internal state [%i][%i] not handled for device "
+ "0.%x.%04x\n", cdev->private->state, dev_event,
+ cdev->private->dev_id.ssid,
+ cdev->private->dev_id.devno);
BUG();
}
* sense id information. So, for intervention required,
* we use the "whack it until it talks" strategy...
*/
- CIO_MSG_EVENT(2, "SenseID : device %04x on Subchannel "
+ CIO_MSG_EVENT(0, "SenseID : device %04x on Subchannel "
"0.%x.%04x reports cmd reject\n",
cdev->private->dev_id.devno, sch->schid.ssid,
sch->schid.sch_no);
lpm = to_io_private(sch)->orb.lpm;
if ((lpm & sch->schib.pmcw.pim & sch->schib.pmcw.pam) != 0)
- CIO_MSG_EVENT(2, "SenseID : path %02X for device %04x "
+ CIO_MSG_EVENT(4, "SenseID : path %02X for device %04x "
"on subchannel 0.%x.%04x is "
"'not operational'\n", lpm,
cdev->private->dev_id.devno,
/* ret is 0, -EBUSY, -EACCES or -ENODEV */
if (ret != -EACCES)
return ret;
- CIO_MSG_EVENT(2, "SNID - Device %04x on Subchannel "
+ CIO_MSG_EVENT(3, "SNID - Device %04x on Subchannel "
"0.%x.%04x, lpm %02X, became 'not "
"operational'\n",
cdev->private->dev_id.devno,
u8 lpm;
lpm = to_io_private(sch)->orb.lpm;
- CIO_MSG_EVENT(2, "SNID - Device %04x on Subchannel 0.%x.%04x,"
+ CIO_MSG_EVENT(3, "SNID - Device %04x on Subchannel 0.%x.%04x,"
" lpm %02X, became 'not operational'\n",
cdev->private->dev_id.devno, sch->schid.ssid,
sch->schid.sch_no, lpm);
return ret;
}
/* PGID command failed on this path. */
- CIO_MSG_EVENT(2, "SPID - Device %04x on Subchannel "
+ CIO_MSG_EVENT(3, "SPID - Device %04x on Subchannel "
"0.%x.%04x, lpm %02X, became 'not operational'\n",
cdev->private->dev_id.devno, sch->schid.ssid,
sch->schid.sch_no, cdev->private->imask);
return ret;
}
/* nop command failed on this path. */
- CIO_MSG_EVENT(2, "NOP - Device %04x on Subchannel "
+ CIO_MSG_EVENT(3, "NOP - Device %04x on Subchannel "
"0.%x.%04x, lpm %02X, became 'not operational'\n",
cdev->private->dev_id.devno, sch->schid.ssid,
sch->schid.sch_no, cdev->private->imask);
return -EAGAIN;
}
if (irb->scsw.cc == 3) {
- CIO_MSG_EVENT(2, "SPID - Device %04x on Subchannel 0.%x.%04x,"
+ CIO_MSG_EVENT(3, "SPID - Device %04x on Subchannel 0.%x.%04x,"
" lpm %02X, became 'not operational'\n",
cdev->private->dev_id.devno, sch->schid.ssid,
sch->schid.sch_no, cdev->private->imask);
return -ETIME;
}
if (irb->scsw.cc == 3) {
- CIO_MSG_EVENT(2, "NOP - Device %04x on Subchannel 0.%x.%04x,"
+ CIO_MSG_EVENT(3, "NOP - Device %04x on Subchannel 0.%x.%04x,"
" lpm %02X, became 'not operational'\n",
cdev->private->dev_id.devno, sch->schid.ssid,
sch->schid.sch_no, cdev->private->imask);
int ccode;
struct semaphore *sem;
unsigned int chain;
+ int ignore;
sem = (struct semaphore *)param;
repeat:
- down_interruptible(sem);
+ ignore = down_interruptible(sem);
chain = 0;
while (1) {
if (unlikely(chain > 1)) {
#define BPP_DELAY 100
static const unsigned BPP_MAJOR = LP_MAJOR;
-static const char* dev_name = "bpp";
+static const char *bpp_dev_name = "bpp";
/* When switching from compatibility to a mode where I can read, try
the following mode first. */
config SCSI_MAC_ESP
tristate "Macintosh NCR53c9[46] SCSI"
depends on MAC && SCSI
+ select SCSI_SPI_ATTRS
help
This is the NCR 53c9x SCSI controller found on most of the 68040
based Macintoshes.
return rcode;
}
-
-/*
- * This routine returns information about the system. This does not effect
- * any logic and if the info is wrong - it doesn't matter.
- */
-
-/* Get all the info we can not get from kernel services */
-static int adpt_system_info(void __user *buffer)
-{
- sysInfo_S si;
-
- memset(&si, 0, sizeof(si));
-
- si.osType = OS_LINUX;
- si.osMajorVersion = 0;
- si.osMinorVersion = 0;
- si.osRevision = 0;
- si.busType = SI_PCI_BUS;
- si.processorFamily = DPTI_sig.dsProcessorFamily;
-
-#if defined __i386__
- adpt_i386_info(&si);
-#elif defined (__ia64__)
- adpt_ia64_info(&si);
-#elif defined(__sparc__)
- adpt_sparc_info(&si);
-#elif defined (__alpha__)
- adpt_alpha_info(&si);
-#else
- si.processorType = 0xff ;
-#endif
- if(copy_to_user(buffer, &si, sizeof(si))){
- printk(KERN_WARNING"dpti: Could not copy buffer TO user\n");
- return -EFAULT;
- }
-
- return 0;
-}
-
#if defined __ia64__
static void adpt_ia64_info(sysInfo_S* si)
{
}
#endif
-
#if defined __sparc__
static void adpt_sparc_info(sysInfo_S* si)
{
si->processorType = PROC_ULTRASPARC;
}
#endif
-
#if defined __alpha__
static void adpt_alpha_info(sysInfo_S* si)
{
#endif
#if defined __i386__
-
static void adpt_i386_info(sysInfo_S* si)
{
// This is all the info we need for now
break;
}
}
+#endif
+
+/*
+ * This routine returns information about the system. This does not effect
+ * any logic and if the info is wrong - it doesn't matter.
+ */
+/* Get all the info we can not get from kernel services */
+static int adpt_system_info(void __user *buffer)
+{
+ sysInfo_S si;
+
+ memset(&si, 0, sizeof(si));
+
+ si.osType = OS_LINUX;
+ si.osMajorVersion = 0;
+ si.osMinorVersion = 0;
+ si.osRevision = 0;
+ si.busType = SI_PCI_BUS;
+ si.processorFamily = DPTI_sig.dsProcessorFamily;
+
+#if defined __i386__
+ adpt_i386_info(&si);
+#elif defined (__ia64__)
+ adpt_ia64_info(&si);
+#elif defined(__sparc__)
+ adpt_sparc_info(&si);
+#elif defined (__alpha__)
+ adpt_alpha_info(&si);
+#else
+ si.processorType = 0xff ;
#endif
+ if (copy_to_user(buffer, &si, sizeof(si))){
+ printk(KERN_WARNING"dpti: Could not copy buffer TO user\n");
+ return -EFAULT;
+ }
+ return 0;
+}
static int adpt_ioctl(struct inode *inode, struct file *file, uint cmd,
ulong arg)
static void adpt_delay(int millisec);
#endif
-#if defined __ia64__
-static void adpt_ia64_info(sysInfo_S* si);
-#endif
-#if defined __sparc__
-static void adpt_sparc_info(sysInfo_S* si);
-#endif
-#if defined __alpha__
-static void adpt_sparc_info(sysInfo_S* si);
-#endif
-#if defined __i386__
-static void adpt_i386_info(sysInfo_S* si);
-#endif
-
#define PRINT_BUFFER_SIZE 512
#define HBA_FLAGS_DBG_FLAGS_MASK 0xffff0000 // Mask for debug flags
static void uart_flush_buffer(struct tty_struct *tty)
{
struct uart_state *state = tty->driver_data;
- struct uart_port *port = state->port;
+ struct uart_port *port;
unsigned long flags;
/*
return;
}
+ port = state->port;
pr_debug("uart_flush_buffer(%d) called\n", tty->index);
spin_lock_irqsave(&port->lock, flags);
config USB_ISP1760_OF
bool "Support for the OF platform bus"
- depends on USB_ISP1760_HCD && OF
+ depends on USB_ISP1760_HCD && PPC_OF
---help---
Enables support for the device present on the PowerPC
OpenFirmware platform bus.
static int iuu_bulk_write(struct usb_serial_port *port)
{
struct iuu_private *priv = usb_get_serial_port_data(port);
- unsigned int flags;
+ unsigned long flags;
int result;
int i;
char *buf_ptr = port->write_urb->transfer_buffer;
{
struct usb_serial_port *port = urb->context;
struct iuu_private *priv = usb_get_serial_port_data(port);
- unsigned int flags;
+ unsigned long flags;
int status;
int error = 0;
int len = 0;
int count)
{
struct iuu_private *priv = usb_get_serial_port_data(port);
- unsigned int flags;
+ unsigned long flags;
dbg("%s - enter", __func__);
if (count > 256)
if (!info->screen_base)
goto out_unmap_regs;
- bw2_blank(0, info);
+ bw2_blank(FB_BLANK_UNBLANK, info);
bw2_init_fix(info, linebytes);
if (!info->screen_base)
goto out_unmap_regs;
- cg3_blank(0, info);
+ cg3_blank(FB_BLANK_UNBLANK, info);
if (!of_find_property(dp, "width", NULL)) {
err = cg3_do_default_mode(par);
cg6_bt_init(par);
cg6_chip_init(info);
- cg6_blank(0, info);
+ cg6_blank(FB_BLANK_UNBLANK, info);
if (fb_alloc_cmap(&info->cmap, 256, 0))
goto out_unmap_regs;
* chosen console, it will have video outputs off in
* the DAC.
*/
- ffb_blank(0, info);
+ ffb_blank(FB_BLANK_UNBLANK, info);
if (fb_alloc_cmap(&info->cmap, 256, 0))
goto out_unmap_dac;
leo_init_wids(info);
leo_init_hw(info);
- leo_blank(0, info);
+ leo_blank(FB_BLANK_UNBLANK, info);
if (fb_alloc_cmap(&info->cmap, 256, 0))
goto out_unmap_regs;
if (!info->screen_base)
goto out_unmap_regs;
- p9100_blank(0, info);
+ p9100_blank(FB_BLANK_UNBLANK, info);
if (fb_alloc_cmap(&info->cmap, 256, 0))
goto out_unmap_screen;
struct tcx_thc {
u32 thc_rev;
- u32 thc_pad0[511];
+ u32 thc_pad0[511];
u32 thc_hs; /* hsync timing */
u32 thc_hsdvs;
u32 thc_hd;
};
/* Reset control plane so that WID is 8-bit plane. */
-static void __tcx_set_control_plane (struct tcx_par *par)
+static void __tcx_set_control_plane(struct tcx_par *par)
{
u32 __iomem *p, *pend;
-
+
if (par->lowdepth)
return;
sbus_writel(tmp, p);
}
}
-
-static void tcx_reset (struct fb_info *info)
+
+static void tcx_reset(struct fb_info *info)
{
struct tcx_par *par = (struct tcx_par *) info->par;
unsigned long flags;
info->screen_base, par->fbsize);
}
-static int __devinit tcx_init_one(struct of_device *op)
+static int __devinit tcx_probe(struct of_device *op,
+ const struct of_device_id *match)
{
struct device_node *dp = op->node;
struct fb_info *info;
return err;
}
-static int __devinit tcx_probe(struct of_device *dev, const struct of_device_id *match)
-{
- struct of_device *op = to_of_device(&dev->dev);
-
- return tcx_init_one(op);
-}
-
static int __devexit tcx_remove(struct of_device *op)
{
struct fb_info *info = dev_get_drvdata(&op->dev);
* affs fs inode data in memory
*/
struct affs_inode_info {
- u32 i_opencnt;
+ atomic_t i_opencnt;
struct semaphore i_link_lock; /* Protects internal inode access. */
struct semaphore i_ext_lock; /* Protects internal inode access. */
#define i_hash_lock i_ext_lock
extern unsigned long affs_parent_ino(struct inode *dir);
extern struct inode *affs_new_inode(struct inode *dir);
extern int affs_notify_change(struct dentry *dentry, struct iattr *attr);
-extern void affs_put_inode(struct inode *inode);
-extern void affs_drop_inode(struct inode *inode);
extern void affs_delete_inode(struct inode *inode);
extern void affs_clear_inode(struct inode *inode);
extern struct inode *affs_iget(struct super_block *sb,
{
if (atomic_read(&filp->f_count) != 1)
return 0;
- pr_debug("AFFS: open(%d)\n", AFFS_I(inode)->i_opencnt);
- AFFS_I(inode)->i_opencnt++;
+ pr_debug("AFFS: open(%lu,%d)\n",
+ inode->i_ino, atomic_read(&AFFS_I(inode)->i_opencnt));
+ atomic_inc(&AFFS_I(inode)->i_opencnt);
return 0;
}
{
if (atomic_read(&filp->f_count) != 0)
return 0;
- pr_debug("AFFS: release(%d)\n", AFFS_I(inode)->i_opencnt);
- AFFS_I(inode)->i_opencnt--;
- if (!AFFS_I(inode)->i_opencnt)
+ pr_debug("AFFS: release(%lu, %d)\n",
+ inode->i_ino, atomic_read(&AFFS_I(inode)->i_opencnt));
+
+ if (atomic_dec_and_test(&AFFS_I(inode)->i_opencnt)) {
+ mutex_lock(&inode->i_mutex);
+ if (inode->i_size != AFFS_I(inode)->mmu_private)
+ affs_truncate(inode);
affs_free_prealloc(inode);
+ mutex_unlock(&inode->i_mutex);
+ }
return 0;
}
/* inline the simplest case: same extended block as last time */
struct buffer_head *bh = AFFS_I(inode)->i_ext_bh;
if (ext == AFFS_I(inode)->i_ext_last)
- atomic_inc(&bh->b_count);
+ get_bh(bh);
else
/* we have to do more (not inlined) */
bh = affs_get_extblock_slow(inode, ext);
affs_brelse(AFFS_I(inode)->i_ext_bh);
AFFS_I(inode)->i_ext_last = ext;
AFFS_I(inode)->i_ext_bh = bh;
- atomic_inc(&bh->b_count);
+ get_bh(bh);
return bh;
pr_debug("AFFS: get_block(%u, %lu)\n", (u32)inode->i_ino, (unsigned long)block);
-
BUG_ON(block > (sector_t)0x7fffffffUL);
if (block >= AFFS_I(inode)->i_blkcnt) {
res = mapping->a_ops->write_begin(NULL, mapping, size, 0, 0, &page, &fsdata);
if (!res)
res = mapping->a_ops->write_end(NULL, mapping, size, 0, 0, page, fsdata);
+ else
+ inode->i_size = AFFS_I(inode)->mmu_private;
mark_inode_dirty(inode);
return;
} else if (inode->i_size == AFFS_I(inode)->mmu_private)
blk++;
} else
AFFS_HEAD(ext_bh)->first_data = 0;
+ AFFS_HEAD(ext_bh)->block_count = cpu_to_be32(i);
size = AFFS_SB(sb)->s_hashsize;
if (size > blkcnt - blk + i)
size = blkcnt - blk + i;
AFFS_I(inode)->i_extcnt = 1;
AFFS_I(inode)->i_ext_last = ~1;
AFFS_I(inode)->i_protect = prot;
- AFFS_I(inode)->i_opencnt = 0;
+ atomic_set(&AFFS_I(inode)->i_opencnt, 0);
AFFS_I(inode)->i_blkcnt = 0;
AFFS_I(inode)->i_lc = NULL;
AFFS_I(inode)->i_lc_size = 0;
inode->i_mode |= S_IFDIR;
} else
inode->i_mode = S_IRUGO | S_IXUGO | S_IWUSR | S_IFDIR;
- if (tail->link_chain)
- inode->i_nlink = 2;
/* Maybe it should be controlled by mount parameter? */
//inode->i_mode |= S_ISVTX;
inode->i_op = &affs_dir_inode_operations;
return error;
}
-void
-affs_put_inode(struct inode *inode)
-{
- pr_debug("AFFS: put_inode(ino=%lu, nlink=%u)\n", inode->i_ino, inode->i_nlink);
- affs_free_prealloc(inode);
-}
-
-void
-affs_drop_inode(struct inode *inode)
-{
- mutex_lock(&inode->i_mutex);
- if (inode->i_size != AFFS_I(inode)->mmu_private)
- affs_truncate(inode);
- mutex_unlock(&inode->i_mutex);
-
- generic_drop_inode(inode);
-}
-
void
affs_delete_inode(struct inode *inode)
{
pr_debug("AFFS: delete_inode(ino=%lu, nlink=%u)\n", inode->i_ino, inode->i_nlink);
truncate_inode_pages(&inode->i_data, 0);
inode->i_size = 0;
- if (S_ISREG(inode->i_mode))
- affs_truncate(inode);
+ affs_truncate(inode);
clear_inode(inode);
affs_free_block(inode->i_sb, inode->i_ino);
}
void
affs_clear_inode(struct inode *inode)
{
- unsigned long cache_page = (unsigned long) AFFS_I(inode)->i_lc;
+ unsigned long cache_page;
pr_debug("AFFS: clear_inode(ino=%lu, nlink=%u)\n", inode->i_ino, inode->i_nlink);
+
+ affs_free_prealloc(inode);
+ cache_page = (unsigned long)AFFS_I(inode)->i_lc;
if (cache_page) {
pr_debug("AFFS: freeing ext cache\n");
AFFS_I(inode)->i_lc = NULL;
inode->i_ino = block;
inode->i_nlink = 1;
inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
- AFFS_I(inode)->i_opencnt = 0;
+ atomic_set(&AFFS_I(inode)->i_opencnt, 0);
AFFS_I(inode)->i_blkcnt = 0;
AFFS_I(inode)->i_lc = NULL;
AFFS_I(inode)->i_lc_size = 0;
switch (type) {
case ST_LINKFILE:
case ST_LINKDIR:
- inode_bh = bh;
retval = -ENOSPC;
block = affs_alloc_block(dir, dir->i_ino);
if (!block)
goto err;
retval = -EIO;
+ inode_bh = bh;
bh = affs_getzeroblk(sb, block);
if (!bh)
goto err;
int
affs_unlink(struct inode *dir, struct dentry *dentry)
{
- pr_debug("AFFS: unlink(dir=%d, \"%.*s\")\n", (u32)dir->i_ino,
+ pr_debug("AFFS: unlink(dir=%d, %lu \"%.*s\")\n", (u32)dir->i_ino,
+ dentry->d_inode->i_ino,
(int)dentry->d_name.len, dentry->d_name.name);
return affs_remove_header(dentry);
int
affs_rmdir(struct inode *dir, struct dentry *dentry)
{
- pr_debug("AFFS: rmdir(dir=%u, \"%.*s\")\n", (u32)dir->i_ino,
+ pr_debug("AFFS: rmdir(dir=%u, %lu \"%.*s\")\n", (u32)dir->i_ino,
+ dentry->d_inode->i_ino,
(int)dentry->d_name.len, dentry->d_name.name);
return affs_remove_header(dentry);
static struct inode *affs_alloc_inode(struct super_block *sb)
{
- struct affs_inode_info *ei;
- ei = (struct affs_inode_info *)kmem_cache_alloc(affs_inode_cachep, GFP_KERNEL);
- if (!ei)
+ struct affs_inode_info *i;
+
+ i = kmem_cache_alloc(affs_inode_cachep, GFP_KERNEL);
+ if (!i)
return NULL;
- ei->vfs_inode.i_version = 1;
- return &ei->vfs_inode;
+
+ i->vfs_inode.i_version = 1;
+ i->i_lc = NULL;
+ i->i_ext_bh = NULL;
+ i->i_pa_cnt = 0;
+
+ return &i->vfs_inode;
}
static void affs_destroy_inode(struct inode *inode)
.alloc_inode = affs_alloc_inode,
.destroy_inode = affs_destroy_inode,
.write_inode = affs_write_inode,
- .put_inode = affs_put_inode,
- .drop_inode = affs_drop_inode,
.delete_inode = affs_delete_inode,
.clear_inode = affs_clear_inode,
.put_super = affs_put_super,
bio_init(bio);
if (likely(nr_iovecs)) {
- unsigned long idx = 0; /* shut up gcc */
+ unsigned long uninitialized_var(idx);
bvl = bvec_alloc_bs(gfp_mask, nr_iovecs, &idx, bs);
if (unlikely(!bvl)) {
* @data: pointer to buffer to copy
* @len: length in bytes
* @gfp_mask: allocation flags for bio and page allocation
+ * @reading: data direction is READ
*
* copy the kernel address into a bio suitable for io to a block
* device. Returns an error pointer in case of error.
void iput(struct inode *inode)
{
if (inode) {
- const struct super_operations *op = inode->i_sb->s_op;
-
BUG_ON(inode->i_state == I_CLEAR);
- if (op && op->put_inode)
- op->put_inode(inode);
-
if (atomic_dec_and_lock(&inode->i_count, &inode_lock))
iput_final(inode);
}
struct file_lock *file_lock = locks_alloc_lock();
struct flock flock;
struct inode *inode;
+ struct file *f;
int error;
if (file_lock == NULL)
* Attempt to detect a close/fcntl race and recover by
* releasing the lock that was just acquired.
*/
- if (!error && fcheck(fd) != filp && flock.l_type != F_UNLCK) {
+ /*
+ * we need that spin_lock here - it prevents reordering between
+ * update of inode->i_flock and check for it done in close().
+ * rcu_read_lock() wouldn't do.
+ */
+ spin_lock(¤t->files->file_lock);
+ f = fcheck(fd);
+ spin_unlock(¤t->files->file_lock);
+ if (!error && f != filp && flock.l_type != F_UNLCK) {
flock.l_type = F_UNLCK;
goto again;
}
struct file_lock *file_lock = locks_alloc_lock();
struct flock64 flock;
struct inode *inode;
+ struct file *f;
int error;
if (file_lock == NULL)
* Attempt to detect a close/fcntl race and recover by
* releasing the lock that was just acquired.
*/
- if (!error && fcheck(fd) != filp && flock.l_type != F_UNLCK) {
+ spin_lock(¤t->files->file_lock);
+ f = fcheck(fd);
+ spin_unlock(¤t->files->file_lock);
+ if (!error && f != filp && flock.l_type != F_UNLCK) {
flock.l_type = F_UNLCK;
goto again;
}
#include <linux/highmem.h>
#include <linux/pagemap.h>
#include <linux/audit.h>
+#include <linux/syscalls.h>
#include <asm/uaccess.h>
#include <asm/ioctls.h>
error = do_pipe(fd);
if (!error) {
- if (copy_to_user(fildes, fd, sizeof(fd)))
+ if (copy_to_user(fildes, fd, sizeof(fd))) {
+ sys_close(fd[0]);
+ sys_close(fd[1]);
error = -EFAULT;
+ }
}
return error;
}
#include <linux/highmem.h>
#include <linux/ptrace.h>
#include <linux/pagemap.h>
-#include <linux/ptrace.h>
#include <linux/mempolicy.h>
#include <linux/swap.h>
#include <linux/swapops.h>
-#include <linux/seq_file.h>
#include <asm/elf.h>
#include <asm/uaccess.h>
{
struct address_space *mapping = out->f_mapping;
struct inode *inode = mapping->host;
- int killsuid, killpriv;
+ struct splice_desc sd = {
+ .total_len = len,
+ .flags = flags,
+ .pos = *ppos,
+ .u.file = out,
+ };
ssize_t ret;
- int err = 0;
-
- killpriv = security_inode_need_killpriv(out->f_path.dentry);
- killsuid = should_remove_suid(out->f_path.dentry);
- if (unlikely(killsuid || killpriv)) {
- mutex_lock(&inode->i_mutex);
- if (killpriv)
- err = security_inode_killpriv(out->f_path.dentry);
- if (!err && killsuid)
- err = __remove_suid(out->f_path.dentry, killsuid);
- mutex_unlock(&inode->i_mutex);
- if (err)
- return err;
- }
- ret = splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_file);
+ inode_double_lock(inode, pipe->inode);
+ ret = remove_suid(out->f_path.dentry);
+ if (likely(!ret))
+ ret = __splice_from_pipe(pipe, &sd, pipe_to_file);
+ inode_double_unlock(inode, pipe->inode);
if (ret > 0) {
unsigned long nr_pages;
* sync it.
*/
if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) {
+ int err;
+
mutex_lock(&inode->i_mutex);
err = generic_osync_inode(inode, mapping,
OSYNC_METADATA|OSYNC_DATA);
ret = splice_direct_to_actor(in, &sd, direct_splice_actor);
if (ret > 0)
- *ppos = sd.pos;
+ *ppos += ret;
return ret;
}
#include <linux/buffer_head.h>
#include <linux/sched.h>
#include <linux/crc-itu-t.h>
+#include <linux/exportfs.h>
static inline int udf_match(int len1, const char *name1, int len2,
const char *name2)
sector_t offset;
struct extent_position epos = {};
struct udf_inode_info *dinfo = UDF_I(dir);
+ int isdotdot = dentry->d_name.len == 2 &&
+ dentry->d_name.name[0] == '.' && dentry->d_name.name[1] == '.';
size = udf_ext0_offset(dir) + dir->i_size;
f_pos = udf_ext0_offset(dir);
continue;
}
+ if ((cfi->fileCharacteristics & FID_FILE_CHAR_PARENT) &&
+ isdotdot) {
+ brelse(epos.bh);
+ return fi;
+ }
+
if (!lfi)
continue;
}
}
unlock_kernel();
- d_add(dentry, inode);
- return NULL;
+ return d_splice_alias(inode, dentry);
}
static struct fileIdentDesc *udf_add_entry(struct inode *dir,
uint16_t liu;
int block;
kernel_lb_addr eloc;
- uint32_t elen;
+ uint32_t elen = 0;
sector_t offset;
struct extent_position epos = {};
struct udf_inode_info *dinfo;
}
add:
- if (dinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB) {
+ /* Is there any extent whose size we need to round up? */
+ if (dinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB && elen) {
elen = (elen + sb->s_blocksize - 1) & ~(sb->s_blocksize - 1);
if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT)
epos.offset -= sizeof(short_ad);
return retval;
}
+static struct dentry *udf_get_parent(struct dentry *child)
+{
+ struct dentry *parent;
+ struct inode *inode = NULL;
+ struct dentry dotdot;
+ struct fileIdentDesc cfi;
+ struct udf_fileident_bh fibh;
+
+ dotdot.d_name.name = "..";
+ dotdot.d_name.len = 2;
+
+ lock_kernel();
+ if (!udf_find_entry(child->d_inode, &dotdot, &fibh, &cfi))
+ goto out_unlock;
+
+ if (fibh.sbh != fibh.ebh)
+ brelse(fibh.ebh);
+ brelse(fibh.sbh);
+
+ inode = udf_iget(child->d_inode->i_sb,
+ lelb_to_cpu(cfi.icb.extLocation));
+ if (!inode)
+ goto out_unlock;
+ unlock_kernel();
+
+ parent = d_alloc_anon(inode);
+ if (!parent) {
+ iput(inode);
+ parent = ERR_PTR(-ENOMEM);
+ }
+
+ return parent;
+out_unlock:
+ unlock_kernel();
+ return ERR_PTR(-EACCES);
+}
+
+
+static struct dentry *udf_nfs_get_inode(struct super_block *sb, u32 block,
+ u16 partref, __u32 generation)
+{
+ struct inode *inode;
+ struct dentry *result;
+ kernel_lb_addr loc;
+
+ if (block == 0)
+ return ERR_PTR(-ESTALE);
+
+ loc.logicalBlockNum = block;
+ loc.partitionReferenceNum = partref;
+ inode = udf_iget(sb, loc);
+
+ if (inode == NULL)
+ return ERR_PTR(-ENOMEM);
+
+ if (generation && inode->i_generation != generation) {
+ iput(inode);
+ return ERR_PTR(-ESTALE);
+ }
+ result = d_alloc_anon(inode);
+ if (!result) {
+ iput(inode);
+ return ERR_PTR(-ENOMEM);
+ }
+ return result;
+}
+
+static struct dentry *udf_fh_to_dentry(struct super_block *sb,
+ struct fid *fid, int fh_len, int fh_type)
+{
+ if ((fh_len != 3 && fh_len != 5) ||
+ (fh_type != FILEID_UDF_WITH_PARENT &&
+ fh_type != FILEID_UDF_WITHOUT_PARENT))
+ return NULL;
+
+ return udf_nfs_get_inode(sb, fid->udf.block, fid->udf.partref,
+ fid->udf.generation);
+}
+
+static struct dentry *udf_fh_to_parent(struct super_block *sb,
+ struct fid *fid, int fh_len, int fh_type)
+{
+ if (fh_len != 5 || fh_type != FILEID_UDF_WITH_PARENT)
+ return NULL;
+
+ return udf_nfs_get_inode(sb, fid->udf.parent_block,
+ fid->udf.parent_partref,
+ fid->udf.parent_generation);
+}
+static int udf_encode_fh(struct dentry *de, __u32 *fh, int *lenp,
+ int connectable)
+{
+ int len = *lenp;
+ struct inode *inode = de->d_inode;
+ kernel_lb_addr location = UDF_I(inode)->i_location;
+ struct fid *fid = (struct fid *)fh;
+ int type = FILEID_UDF_WITHOUT_PARENT;
+
+ if (len < 3 || (connectable && len < 5))
+ return 255;
+
+ *lenp = 3;
+ fid->udf.block = location.logicalBlockNum;
+ fid->udf.partref = location.partitionReferenceNum;
+ fid->udf.generation = inode->i_generation;
+
+ if (connectable && !S_ISDIR(inode->i_mode)) {
+ spin_lock(&de->d_lock);
+ inode = de->d_parent->d_inode;
+ location = UDF_I(inode)->i_location;
+ fid->udf.parent_block = location.logicalBlockNum;
+ fid->udf.parent_partref = location.partitionReferenceNum;
+ fid->udf.parent_generation = inode->i_generation;
+ spin_unlock(&de->d_lock);
+ *lenp = 5;
+ type = FILEID_UDF_WITH_PARENT;
+ }
+
+ return type;
+}
+
+const struct export_operations udf_export_ops = {
+ .encode_fh = udf_encode_fh,
+ .fh_to_dentry = udf_fh_to_dentry,
+ .fh_to_parent = udf_fh_to_parent,
+ .get_parent = udf_get_parent,
+};
+
const struct inode_operations udf_dir_inode_operations = {
.lookup = udf_lookup,
.create = udf_create,
#include <linux/slab.h>
#include <linux/buffer_head.h>
-inline uint32_t udf_get_pblock(struct super_block *sb, uint32_t block,
- uint16_t partition, uint32_t offset)
+uint32_t udf_get_pblock(struct super_block *sb, uint32_t block,
+ uint16_t partition, uint32_t offset)
{
struct udf_sb_info *sbi = UDF_SB(sb);
struct udf_part_map *map;
/* Fill in the rest of the superblock */
sb->s_op = &udf_sb_ops;
+ sb->s_export_op = &udf_export_ops;
sb->dq_op = NULL;
sb->s_dirt = 0;
sb->s_magic = UDF_SUPER_MAGIC;
struct buffer_head;
struct super_block;
+extern const struct export_operations udf_export_ops;
extern const struct inode_operations udf_dir_inode_operations;
extern const struct file_operations udf_dir_operations;
extern const struct inode_operations udf_file_inode_operations;
# define mac_scc ((*(volatile struct SCC*)MAC_SCC_BAS))
#endif
-/* hardware stuff */
-
-#define MACHW_DECLARE(name) unsigned name : 1
-#define MACHW_SET(name) (mac_hw_present.name = 1)
-#define MACHW_PRESENT(name) (mac_hw_present.name)
-
-struct mac_hw_present {
- /* video hardware */
- /* sound hardware */
- /* disk storage interfaces */
- MACHW_DECLARE(MAC_SCSI_80); /* Directly mapped NCR5380 */
- MACHW_DECLARE(MAC_SCSI_96); /* 53c9[46] */
- MACHW_DECLARE(MAC_SCSI_96_2); /* 2nd 53c9[46] Q900 and Q950 */
- MACHW_DECLARE(IDE); /* IDE Interface */
- /* other I/O hardware */
- MACHW_DECLARE(SCC); /* Serial Communications Contr. */
- /* DMA */
- MACHW_DECLARE(SCSI_DMA); /* DMA for the NCR5380 */
- /* real time clocks */
- MACHW_DECLARE(RTC_CLK); /* clock chip */
- /* supporting hardware */
- MACHW_DECLARE(VIA1); /* Versatile Interface Ad. 1 */
- MACHW_DECLARE(VIA2); /* Versatile Interface Ad. 2 */
- MACHW_DECLARE(RBV); /* Versatile Interface Ad. 2+ */
- /* NUBUS */
- MACHW_DECLARE(NUBUS); /* NUBUS */
-};
-
-extern struct mac_hw_present mac_hw_present;
-
#endif /* __ASSEMBLY__ */
#endif /* linux/machw.h */
extern void identify_cpu(struct mn10300_cpuinfo *);
extern void print_cpu_info(struct mn10300_cpuinfo *);
extern void dodgy_tsc(void);
-#define cpu_relax() do {} while (0)
+#define cpu_relax() barrier()
/*
* User space process size: 1.75GB (default).
struct kvm_vcpu_stat {
u32 exit_userspace;
+ u32 exit_null;
u32 exit_external_request;
u32 exit_external_interrupt;
u32 exit_stop_request;
return skey;
}
+#ifdef CONFIG_PAGE_STATES
+
+struct page;
+void arch_free_page(struct page *page, int order);
+void arch_alloc_page(struct page *page, int order);
+
+#define HAVE_ARCH_FREE_PAGE
+#define HAVE_ARCH_ALLOC_PAGE
+
+#endif
+
#endif /* !__ASSEMBLY__ */
/* to align the pointer to the (next) page boundary */
extern void user_enable_single_step(struct task_struct *);
extern void user_disable_single_step(struct task_struct *);
+#define __ARCH_WANT_COMPAT_SYS_PTRACE
+
#define user_mode(regs) (((regs)->psw.mask & PSW_MASK_PSTATE) != 0)
#define instruction_pointer(regs) ((regs)->psw.addr & PSW_ADDR_INSN)
#define regs_return_value(regs)((regs)->gprs[2])
#define pfault_fini() do { } while (0)
#endif /* CONFIG_PFAULT */
+#ifdef CONFIG_PAGE_STATES
+extern void cmma_init(void);
+#else
+static inline void cmma_init(void) { }
+#endif
+
#define finish_arch_switch(prev) do { \
set_fs(current->thread.mm_segment); \
account_vtime(prev); \
/* To avoid harmful races, pmd_none(x) should check only the lower when PAE */
#define pmd_none(x) (!(unsigned long)pmd_val((x)))
#define pmd_present(x) (pmd_val((x)) & _PAGE_PRESENT)
-
-extern int pmd_bad(pmd_t pmd);
-
-#define pmd_bad_v1(x) \
- (_KERNPG_TABLE != (pmd_val((x)) & ~(PAGE_MASK | _PAGE_USER)))
-#define pmd_bad_v2(x) \
- (_KERNPG_TABLE != (pmd_val((x)) & ~(PAGE_MASK | _PAGE_USER | \
- _PAGE_PSE | _PAGE_NX)))
+#define pmd_bad(x) ((pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))
static inline unsigned long pud_bad(pud_t pud)
{
- return pud_val(pud) &
- ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER | _PAGE_PSE | _PAGE_NX);
+ return pud_val(pud) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER);
}
static inline unsigned long pmd_bad(pmd_t pmd)
{
- return pmd_val(pmd) &
- ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER | _PAGE_PSE | _PAGE_NX);
+ return pmd_val(pmd) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER);
}
#define pte_none(x) (!pte_val((x)))
* 32 bit parent directory inode number.
*/
FILEID_INO32_GEN_PARENT = 2,
+
+ /*
+ * 32 bit block number, 16 bit partition reference,
+ * 16 bit unused, 32 bit generation number.
+ */
+ FILEID_UDF_WITHOUT_PARENT = 0x51,
+
+ /*
+ * 32 bit block number, 16 bit partition reference,
+ * 16 bit unused, 32 bit generation number,
+ * 32 bit parent block number, 32 bit parent generation number
+ */
+ FILEID_UDF_WITH_PARENT = 0x52,
};
struct fid {
u32 parent_ino;
u32 parent_gen;
} i32;
+ struct {
+ u32 block;
+ u16 partref;
+ u16 parent_partref;
+ u32 generation;
+ u32 parent_block;
+ u32 parent_generation;
+ } udf;
__u32 raw[0];
};
};
extern ssize_t vfs_writev(struct file *, const struct iovec __user *,
unsigned long, loff_t *);
-/*
- * NOTE: write_inode, delete_inode, clear_inode, put_inode can be called
- * without the big kernel lock held in all filesystems.
- */
struct super_operations {
struct inode *(*alloc_inode)(struct super_block *sb);
void (*destroy_inode)(struct inode *);
void (*dirty_inode) (struct inode *);
int (*write_inode) (struct inode *, int);
- void (*put_inode) (struct inode *);
void (*drop_inode) (struct inode *);
void (*delete_inode) (struct inode *);
void (*put_super) (struct super_block *);
extern void clear_inode(struct inode *);
extern void destroy_inode(struct inode *);
extern struct inode *new_inode(struct super_block *);
-extern int __remove_suid(struct dentry *, int);
extern int should_remove_suid(struct dentry *);
extern int remove_suid(struct dentry *);
static inline void disk_stat_set_all(struct gendisk *gendiskp, int value) {
int i;
+
for_each_possible_cpu(i)
memset(per_cpu_ptr(gendiskp->dkstats, i), value,
- sizeof (struct disk_stats));
+ sizeof(struct disk_stats));
}
#define __part_stat_add(part, field, addnd) \
(per_cpu_ptr(part->dkstats, smp_processor_id())->field += addnd)
-#define __all_stat_add(gendiskp, field, addnd, sector) \
+#define __all_stat_add(gendiskp, part, field, addnd, sector) \
({ \
- struct hd_struct *part = get_part(gendiskp, sector); \
if (part) \
__part_stat_add(part, field, addnd); \
__disk_stat_add(gendiskp, field, addnd); \
res; \
})
-static inline void part_stat_set_all(struct hd_struct *part, int value) {
+static inline void part_stat_set_all(struct hd_struct *part, int value)
+{
int i;
+
for_each_possible_cpu(i)
memset(per_cpu_ptr(part->dkstats, i), value,
- sizeof(struct disk_stats));
+ sizeof(struct disk_stats));
}
#else /* !CONFIG_SMP */
#define __part_stat_add(part, field, addnd) \
(part->dkstats.field += addnd)
-#define __all_stat_add(gendiskp, field, addnd, sector) \
+#define __all_stat_add(gendiskp, part, field, addnd, sector) \
({ \
- struct hd_struct *part = get_part(gendiskp, sector); \
if (part) \
part->dkstats.field += addnd; \
__disk_stat_add(gendiskp, field, addnd); \
#define part_stat_sub(gendiskp, field, subnd) \
part_stat_add(gendiskp, field, -subnd)
-#define all_stat_add(gendiskp, field, addnd, sector) \
+#define all_stat_add(gendiskp, part, field, addnd, sector) \
do { \
preempt_disable(); \
- __all_stat_add(gendiskp, field, addnd, sector); \
+ __all_stat_add(gendiskp, part, field, addnd, sector); \
preempt_enable(); \
} while (0)
#define all_stat_dec(gendiskp, field, sector) \
all_stat_add(gendiskp, field, -1, sector)
-#define __all_stat_inc(gendiskp, field, sector) \
- __all_stat_add(gendiskp, field, 1, sector)
-#define all_stat_inc(gendiskp, field, sector) \
- all_stat_add(gendiskp, field, 1, sector)
+#define __all_stat_inc(gendiskp, part, field, sector) \
+ __all_stat_add(gendiskp, part, field, 1, sector)
+#define all_stat_inc(gendiskp, part, field, sector) \
+ all_stat_add(gendiskp, part, field, 1, sector)
-#define __all_stat_sub(gendiskp, field, subnd, sector) \
- __all_stat_add(gendiskp, field, -subnd, sector)
-#define all_stat_sub(gendiskp, field, subnd, sector) \
- all_stat_add(gendiskp, field, -subnd, sector)
+#define __all_stat_sub(gendiskp, part, field, subnd, sector) \
+ __all_stat_add(gendiskp, part, field, -subnd, sector)
+#define all_stat_sub(gendiskp, part, field, subnd, sector) \
+ all_stat_add(gendiskp, part, field, -subnd, sector)
/* Inlines to alloc and free disk stats in struct gendisk */
#ifdef CONFIG_SMP
return (task_nice(task) + 20) / 5;
}
+/*
+ * This is for the case where the task hasn't asked for a specific IO class.
+ * Check for idle and rt task process, and return appropriate IO class.
+ */
+static inline int task_nice_ioclass(struct task_struct *task)
+{
+ if (task->policy == SCHED_IDLE)
+ return IOPRIO_CLASS_IDLE;
+ else if (task->policy == SCHED_FIFO || task->policy == SCHED_RR)
+ return IOPRIO_CLASS_RT;
+ else
+ return IOPRIO_CLASS_BE;
+}
+
/*
* For inheritance, return the highest of the two given priorities
*/
extern struct kgdb_arch arch_kgdb_ops;
+extern unsigned long __weak kgdb_arch_pc(int exception, struct pt_regs *regs);
+
extern int kgdb_register_io_module(struct kgdb_io *local_kgdb_io_ops);
extern void kgdb_unregister_io_module(struct kgdb_io *local_kgdb_io_ops);
-extern int kgdb_hex2long(char **ptr, long *long_val);
+extern int kgdb_hex2long(char **ptr, unsigned long *long_val);
extern int kgdb_mem2hex(char *mem, char *buf, int count);
extern int kgdb_hex2mem(char *buf, char *mem, int count);
extern void ata_eh_qc_complete(struct ata_queued_cmd *qc);
extern void ata_eh_qc_retry(struct ata_queued_cmd *qc);
+extern void ata_eh_analyze_ncq_error(struct ata_link *link);
extern void ata_do_eh(struct ata_port *ap, ata_prereset_fn_t prereset,
ata_reset_fn_t softreset, ata_reset_fn_t hardreset,
return *(struct ata_port **)&host->hostdata[0];
}
+static inline int ata_check_ready(u8 status)
+{
+ /* Some controllers report 0x77 or 0x7f during intermediate
+ * not-ready stages.
+ */
+ if (status == 0x77 || status == 0x7f)
+ return 0;
+
+ /* 0xff indicates either no device or device not ready */
+ if (status == 0xff)
+ return -ENODEV;
+
+ return !(status & ATA_BUSY);
+}
+
/**************************************************************************
* PMP - drivers/ata/libata-pmp.c
/*
* MV-643XX ethernet platform device data definition file.
*/
+
#ifndef __LINUX_MV643XX_ETH_H
#define __LINUX_MV643XX_ETH_H
-#define MV643XX_ETH_SHARED_NAME "mv643xx_eth_shared"
-#define MV643XX_ETH_NAME "mv643xx_eth"
+#include <linux/mbus.h>
+
+#define MV643XX_ETH_SHARED_NAME "mv643xx_eth"
+#define MV643XX_ETH_NAME "mv643xx_eth_port"
#define MV643XX_ETH_SHARED_REGS 0x2000
#define MV643XX_ETH_SHARED_REGS_SIZE 0x2000
#define MV643XX_ETH_BAR_4 0x2220
#define MV643XX_ETH_SIZE_REG_4 0x2224
#define MV643XX_ETH_BASE_ADDR_ENABLE_REG 0x2290
+struct mv643xx_eth_shared_platform_data {
+ struct mbus_dram_target_info *dram;
+ unsigned int t_clk;
+};
+
struct mv643xx_eth_platform_data {
+ struct platform_device *shared;
int port_number;
+
+ struct platform_device *shared_smi;
+
u16 force_phy_addr; /* force override if phy_addr == 0 */
u16 phy_addr;
struct nf_ct_sip_master {
unsigned int register_cseq;
+ unsigned int invite_cseq;
};
enum sip_expectation_classes {
#include <linux/mod_devicetable.h>
#include <linux/types.h>
+#include <linux/init.h>
#include <linux/ioport.h>
#include <linux/list.h>
#include <linux/compiler.h>
void pci_bus_add_devices(struct pci_bus *bus);
struct pci_bus *pci_scan_bus_parented(struct device *parent, int bus,
struct pci_ops *ops, void *sysdata);
-static inline struct pci_bus *pci_scan_bus(int bus, struct pci_ops *ops,
+static inline struct pci_bus * __devinit pci_scan_bus(int bus, struct pci_ops *ops,
void *sysdata)
{
struct pci_bus *root_bus;
void pci_walk_bus(struct pci_bus *top, void (*cb)(struct pci_dev *, void *),
void *userdata);
-int pci_cfg_space_size_ext(struct pci_dev *dev, unsigned check_exp_pcix);
+int pci_cfg_space_size_ext(struct pci_dev *dev);
int pci_cfg_space_size(struct pci_dev *dev);
unsigned char pci_bus_max_busnr(struct pci_bus *bus);
void mdiobus_unregister(struct mii_bus *bus);
void phy_sanitize_settings(struct phy_device *phydev);
int phy_stop_interrupts(struct phy_device *phydev);
+int phy_enable_interrupts(struct phy_device *phydev);
+int phy_disable_interrupts(struct phy_device *phydev);
static inline int phy_read_status(struct phy_device *phydev) {
return phydev->drv->read_status(phydev);
int (*run)(struct phy_device *));
int phy_scan_fixups(struct phy_device *phydev);
+int __init mdio_bus_init(void);
+void mdio_bus_exit(void);
+
extern struct bus_type mdio_bus_type;
#endif /* __PHY_H */
}
#endif
+extern unsigned long long time_sync_thresh;
+
/*
* Task state bitmask. NOTE! These bits are also
* encoded in fs/proc/array.c: get_task_state().
extern unsigned long long sched_clock(void);
+#ifndef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK
+static inline void sched_clock_init(void)
+{
+}
+
+static inline u64 sched_clock_cpu(int cpu)
+{
+ return sched_clock();
+}
+
+static inline void sched_clock_tick(void)
+{
+}
+
+static inline void sched_clock_idle_sleep_event(void)
+{
+}
+
+static inline void sched_clock_idle_wakeup_event(u64 delta_ns)
+{
+}
+#else
+extern void sched_clock_init(void);
+extern u64 sched_clock_cpu(int cpu);
+extern void sched_clock_tick(void);
+extern void sched_clock_idle_sleep_event(void);
+extern void sched_clock_idle_wakeup_event(u64 delta_ns);
+#endif
+
/*
* For kernel-internal use: high-speed (but slightly incorrect) per-cpu
* clock constructed from sched_clock():
clear_tsk_thread_flag(tsk,TIF_NEED_RESCHED);
}
+static inline int test_tsk_need_resched(struct task_struct *tsk)
+{
+ return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED));
+}
+
static inline int signal_pending(struct task_struct *p)
{
return unlikely(test_tsk_thread_flag(p,TIF_SIGPENDING));
static inline int need_resched(void)
{
- return unlikely(test_thread_flag(TIF_NEED_RESCHED));
+ return unlikely(test_tsk_need_resched(current));
}
/*
#else
#define MODULE_VERMAGIC_MODULE_UNLOAD ""
#endif
+#ifdef CONFIG_MODVERSIONS
+#define MODULE_VERMAGIC_MODVERSIONS "modversions "
+#else
+#define MODULE_VERMAGIC_MODVERSIONS ""
+#endif
#ifndef MODULE_ARCH_VERMAGIC
#define MODULE_ARCH_VERMAGIC ""
#endif
#define VERMAGIC_STRING \
UTS_RELEASE " " \
MODULE_VERMAGIC_SMP MODULE_VERMAGIC_PREEMPT \
- MODULE_VERMAGIC_MODULE_UNLOAD MODULE_ARCH_VERMAGIC
+ MODULE_VERMAGIC_MODULE_UNLOAD MODULE_VERMAGIC_MODVERSIONS \
+ MODULE_ARCH_VERMAGIC
{
return (inet_sk(sk)->pmtudisc == IP_PMTUDISC_DO ||
(inet_sk(sk)->pmtudisc == IP_PMTUDISC_WANT &&
- !(dst_metric(dst, RTAX_LOCK)&(1<<RTAX_MTU))));
+ !(dst_metric_locked(dst, RTAX_MTU))));
}
extern void __ip_select_ident(struct iphdr *iph, struct dst_entry *dst, int more);
extern void xfrm_audit_state_icvfail(struct xfrm_state *x,
struct sk_buff *skb, u8 proto);
#else
-#define xfrm_audit_policy_add(x, r, a, se, s) do { ; } while (0)
-#define xfrm_audit_policy_delete(x, r, a, se, s) do { ; } while (0)
-#define xfrm_audit_state_add(x, r, a, se, s) do { ; } while (0)
-#define xfrm_audit_state_delete(x, r, a, se, s) do { ; } while (0)
-#define xfrm_audit_state_replay_overflow(x, s) do { ; } while (0)
-#define xfrm_audit_state_notfound_simple(s, f) do { ; } while (0)
-#define xfrm_audit_state_notfound(s, f, sp, sq) do { ; } while (0)
-#define xfrm_audit_state_icvfail(x, s, p) do { ; } while (0)
+
+static inline void xfrm_audit_policy_add(struct xfrm_policy *xp, int result,
+ u32 auid, u32 ses, u32 secid)
+{
+}
+
+static inline void xfrm_audit_policy_delete(struct xfrm_policy *xp, int result,
+ u32 auid, u32 ses, u32 secid)
+{
+}
+
+static inline void xfrm_audit_state_add(struct xfrm_state *x, int result,
+ u32 auid, u32 ses, u32 secid)
+{
+}
+
+static inline void xfrm_audit_state_delete(struct xfrm_state *x, int result,
+ u32 auid, u32 ses, u32 secid)
+{
+}
+
+static inline void xfrm_audit_state_replay_overflow(struct xfrm_state *x,
+ struct sk_buff *skb)
+{
+}
+
+static inline void xfrm_audit_state_notfound_simple(struct sk_buff *skb,
+ u16 family)
+{
+}
+
+static inline void xfrm_audit_state_notfound(struct sk_buff *skb, u16 family,
+ __be32 net_spi, __be32 net_seq)
+{
+}
+
+static inline void xfrm_audit_state_icvfail(struct xfrm_state *x,
+ struct sk_buff *skb, u8 proto)
+{
+}
#endif /* CONFIG_AUDITSYSCALL */
static inline void xfrm_pol_hold(struct xfrm_policy *policy)
Say N if unsure.
+#
+# Architectures with an unreliable sched_clock() should select this:
+#
+config HAVE_UNSTABLE_SCHED_CLOCK
+ bool
+
config GROUP_SCHED
bool "Group CPU scheduler"
- default y
+ depends on EXPERIMENTAL
+ default n
help
This feature lets CPU scheduler recognize task groups and control CPU
bandwidth allocation to such task groups.
config FAIR_GROUP_SCHED
bool "Group scheduling for SCHED_OTHER"
depends on GROUP_SCHED
- default y
+ default GROUP_SCHED
config RT_GROUP_SCHED
bool "Group scheduling for SCHED_RR/FIFO"
help
Enable support for generating core dumps. Disabling saves about 4k.
+config PCSPKR_PLATFORM
+ bool "Enable PC-Speaker support" if EMBEDDED
+ depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES
+ default y
+ help
+ This option allows to disable the internal PC-Speaker
+ support, saving some memory.
+
config COMPAT_BRK
bool "Disable heap randomization"
default y
depends on MODULES
default n
help
- This option allows loading of modules even if that would set the
- 'F' (forced) taint, due to lack of version info. Which is
- usually a really bad idea.
+ Allow loading of modules without version information (ie. modprobe
+ --force). Forced module loading sets the 'F' (forced) taint flag and
+ is usually a really bad idea.
config MODULE_UNLOAD
bool "Module unloading"
softirq_init();
timekeeping_init();
time_init();
+ sched_clock_init();
profile_init();
if (!irqs_disabled())
printk("start_kernel(): bug: interrupts were enabled early\n");
rcupdate.o extable.o params.o posix-timers.o \
kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \
hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \
- notifier.o ksysfs.o pm_qos_params.o
+ notifier.o ksysfs.o pm_qos_params.o sched_clock.o
obj-$(CONFIG_SYSCTL_SYSCALL_CHECK) += sysctl_check.o
obj-$(CONFIG_STACKTRACE) += stacktrace.o
return task_cs(current) == cpuset_being_rebound;
}
-static int update_relax_domain_level(struct cpuset *cs, char *buf)
+static int update_relax_domain_level(struct cpuset *cs, s64 val)
{
- int val = simple_strtol(buf, NULL, 10);
-
- if (val < 0)
+ if ((int)val < 0)
val = -1;
if (val != cs->relax_domain_level) {
case FILE_MEMLIST:
retval = update_nodemask(cs, buffer);
break;
- case FILE_SCHED_RELAX_DOMAIN_LEVEL:
- retval = update_relax_domain_level(cs, buffer);
- break;
default:
retval = -EINVAL;
goto out2;
return retval;
}
+static int cpuset_write_s64(struct cgroup *cgrp, struct cftype *cft, s64 val)
+{
+ int retval = 0;
+ struct cpuset *cs = cgroup_cs(cgrp);
+ cpuset_filetype_t type = cft->private;
+
+ cgroup_lock();
+
+ if (cgroup_is_removed(cgrp)) {
+ cgroup_unlock();
+ return -ENODEV;
+ }
+ switch (type) {
+ case FILE_SCHED_RELAX_DOMAIN_LEVEL:
+ retval = update_relax_domain_level(cs, val);
+ break;
+ default:
+ retval = -EINVAL;
+ break;
+ }
+ cgroup_unlock();
+ return retval;
+}
+
/*
* These ascii lists should be read in a single call, by using a user
* buffer large enough to hold the entire map. If read in smaller
case FILE_MEMLIST:
s += cpuset_sprintf_memlist(s, cs);
break;
- case FILE_SCHED_RELAX_DOMAIN_LEVEL:
- s += sprintf(s, "%d", cs->relax_domain_level);
- break;
default:
retval = -EINVAL;
goto out;
}
}
+static s64 cpuset_read_s64(struct cgroup *cont, struct cftype *cft)
+{
+ struct cpuset *cs = cgroup_cs(cont);
+ cpuset_filetype_t type = cft->private;
+ switch (type) {
+ case FILE_SCHED_RELAX_DOMAIN_LEVEL:
+ return cs->relax_domain_level;
+ default:
+ BUG();
+ }
+}
+
/*
* for the common functions, 'private' gives the type of file
{
.name = "sched_relax_domain_level",
- .read_u64 = cpuset_read_u64,
- .write_u64 = cpuset_write_u64,
+ .read_s64 = cpuset_read_s64,
+ .write_s64 = cpuset_write_s64,
.private = FILE_SCHED_RELAX_DOMAIN_LEVEL,
},
/* Key which the futex is hashed on: */
union futex_key key;
- /* For fd, sigio sent using these: */
- int fd;
- struct file *filp;
-
/* Optional priority inheritance state: */
struct futex_pi_state *pi_state;
struct task_struct *task;
static struct futex_hash_bucket futex_queues[1<<FUTEX_HASHBITS];
-/* Futex-fs vfsmount entry: */
-static struct vfsmount *futex_mnt;
-
/*
* Take mm->mmap_sem, when futex is shared
*/
static void wake_futex(struct futex_q *q)
{
plist_del(&q->list, &q->list.plist);
- if (q->filp)
- send_sigio(&q->filp->f_owner, q->fd, POLL_IN);
/*
* The lock in wake_up_all() is a crucial memory barrier after the
* plist_del() and also before assigning to q->lock_ptr.
}
/* The key must be already stored in q->key. */
-static inline struct futex_hash_bucket *
-queue_lock(struct futex_q *q, int fd, struct file *filp)
+static inline struct futex_hash_bucket *queue_lock(struct futex_q *q)
{
struct futex_hash_bucket *hb;
- q->fd = fd;
- q->filp = filp;
-
init_waitqueue_head(&q->waiters);
get_futex_key_refs(&q->key);
return hb;
}
-static inline void __queue_me(struct futex_q *q, struct futex_hash_bucket *hb)
+static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb)
{
int prio;
* exactly once. They are called with the hashed spinlock held.
*/
-/* The key must be already stored in q->key. */
-static void queue_me(struct futex_q *q, int fd, struct file *filp)
-{
- struct futex_hash_bucket *hb;
-
- hb = queue_lock(q, fd, filp);
- __queue_me(q, hb);
-}
-
/* Return 1 if we were still queued (ie. 0 means we were woken) */
static int unqueue_me(struct futex_q *q)
{
if (unlikely(ret != 0))
goto out_release_sem;
- hb = queue_lock(&q, -1, NULL);
+ hb = queue_lock(&q);
/*
* Access the page AFTER the futex is queued.
goto out_unlock_release_sem;
/* Only actually queue if *uaddr contained val. */
- __queue_me(&q, hb);
+ queue_me(&q, hb);
/*
* Now the futex is queued and we have checked the data, we
goto out_release_sem;
retry_unlocked:
- hb = queue_lock(&q, -1, NULL);
+ hb = queue_lock(&q);
retry_locked:
ret = lock_taken = 0;
/*
* Only actually queue now that the atomic ops are done:
*/
- __queue_me(&q, hb);
+ queue_me(&q, hb);
/*
* Now the futex is queued and we have checked the data, we
return ret;
}
-static int futex_close(struct inode *inode, struct file *filp)
-{
- struct futex_q *q = filp->private_data;
-
- unqueue_me(q);
- kfree(q);
-
- return 0;
-}
-
-/* This is one-shot: once it's gone off you need a new fd */
-static unsigned int futex_poll(struct file *filp,
- struct poll_table_struct *wait)
-{
- struct futex_q *q = filp->private_data;
- int ret = 0;
-
- poll_wait(filp, &q->waiters, wait);
-
- /*
- * plist_node_empty() is safe here without any lock.
- * q->lock_ptr != 0 is not safe, because of ordering against wakeup.
- */
- if (plist_node_empty(&q->list))
- ret = POLLIN | POLLRDNORM;
-
- return ret;
-}
-
-static const struct file_operations futex_fops = {
- .release = futex_close,
- .poll = futex_poll,
-};
-
-/*
- * Signal allows caller to avoid the race which would occur if they
- * set the sigio stuff up afterwards.
- */
-static int futex_fd(u32 __user *uaddr, int signal)
-{
- struct futex_q *q;
- struct file *filp;
- int ret, err;
- struct rw_semaphore *fshared;
- static unsigned long printk_interval;
-
- if (printk_timed_ratelimit(&printk_interval, 60 * 60 * 1000)) {
- printk(KERN_WARNING "Process `%s' used FUTEX_FD, which "
- "will be removed from the kernel in June 2007\n",
- current->comm);
- }
-
- ret = -EINVAL;
- if (!valid_signal(signal))
- goto out;
-
- ret = get_unused_fd();
- if (ret < 0)
- goto out;
- filp = get_empty_filp();
- if (!filp) {
- put_unused_fd(ret);
- ret = -ENFILE;
- goto out;
- }
- filp->f_op = &futex_fops;
- filp->f_path.mnt = mntget(futex_mnt);
- filp->f_path.dentry = dget(futex_mnt->mnt_root);
- filp->f_mapping = filp->f_path.dentry->d_inode->i_mapping;
-
- if (signal) {
- err = __f_setown(filp, task_pid(current), PIDTYPE_PID, 1);
- if (err < 0) {
- goto error;
- }
- filp->f_owner.signum = signal;
- }
-
- q = kmalloc(sizeof(*q), GFP_KERNEL);
- if (!q) {
- err = -ENOMEM;
- goto error;
- }
- q->pi_state = NULL;
-
- fshared = ¤t->mm->mmap_sem;
- down_read(fshared);
- err = get_futex_key(uaddr, fshared, &q->key);
-
- if (unlikely(err != 0)) {
- up_read(fshared);
- kfree(q);
- goto error;
- }
-
- /*
- * queue_me() must be called before releasing mmap_sem, because
- * key->shared.inode needs to be referenced while holding it.
- */
- filp->private_data = q;
-
- queue_me(q, ret, filp);
- up_read(fshared);
-
- /* Now we map fd to filp, so userspace can access it */
- fd_install(ret, filp);
-out:
- return ret;
-error:
- put_unused_fd(ret);
- put_filp(filp);
- ret = err;
- goto out;
-}
-
/*
* Support for robust futexes: the kernel cleans up held futexes at
* thread exit time.
case FUTEX_WAKE_BITSET:
ret = futex_wake(uaddr, fshared, val, val3);
break;
- case FUTEX_FD:
- /* non-zero val means F_SETOWN(getpid()) & F_SETSIG(val) */
- ret = futex_fd(uaddr, val);
- break;
case FUTEX_REQUEUE:
ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, NULL);
break;
return do_futex(uaddr, op, val, tp, uaddr2, val2, val3);
}
-static int futexfs_get_sb(struct file_system_type *fs_type,
- int flags, const char *dev_name, void *data,
- struct vfsmount *mnt)
-{
- return get_sb_pseudo(fs_type, "futex", NULL, FUTEXFS_SUPER_MAGIC, mnt);
-}
-
-static struct file_system_type futex_fs_type = {
- .name = "futexfs",
- .get_sb = futexfs_get_sb,
- .kill_sb = kill_anon_super,
-};
-
static int __init futex_init(void)
{
u32 curval;
spin_lock_init(&futex_queues[i].lock);
}
- i = register_filesystem(&futex_fs_type);
- if (i)
- return i;
-
- futex_mnt = kern_mount(&futex_fs_type);
- if (IS_ERR(futex_mnt)) {
- unregister_filesystem(&futex_fs_type);
- return PTR_ERR(futex_mnt);
- }
-
return 0;
}
__initcall(futex_init);
int err_code;
int cpu;
int pass_exception;
- long threadid;
+ unsigned long threadid;
long kgdb_usethreadid;
struct pt_regs *linux_regs;
};
* the other CPUs might interfere with your debugging context, so
* use this with care:
*/
-int kgdb_do_roundup = 1;
+static int kgdb_do_roundup = 1;
static int __init opt_nokgdbroundup(char *str)
{
* While we find nice hex chars, build a long_val.
* Return number of chars processed.
*/
-int kgdb_hex2long(char **ptr, long *long_val)
+int kgdb_hex2long(char **ptr, unsigned long *long_val)
{
int hex_val;
int num = 0;
return 0;
}
-int remove_all_break(void)
+static int remove_all_break(void)
{
unsigned long addr;
int error;
if (!crc)
return 1;
+ /* No versions at all? modprobe --force does this. */
+ if (versindex == 0)
+ return try_to_force_load(mod, symname) == 0;
+
versions = (void *) sechdrs[versindex].sh_addr;
num_versions = sechdrs[versindex].sh_size
/ sizeof(struct modversion_info);
goto bad_version;
}
- if (!try_to_force_load(mod, symname))
- return 1;
+ printk(KERN_WARNING "%s: no symbol version for %s\n",
+ mod->name, symname);
+ return 0;
bad_version:
printk("%s: disagrees about version of symbol %s\n",
return check_version(sechdrs, versindex, "struct_module", mod, crc);
}
-/* First part is kernel version, which we ignore. */
-static inline int same_magic(const char *amagic, const char *bmagic)
+/* First part is kernel version, which we ignore if module has crcs. */
+static inline int same_magic(const char *amagic, const char *bmagic,
+ bool has_crcs)
{
- amagic += strcspn(amagic, " ");
- bmagic += strcspn(bmagic, " ");
+ if (has_crcs) {
+ amagic += strcspn(amagic, " ");
+ bmagic += strcspn(bmagic, " ");
+ }
return strcmp(amagic, bmagic) == 0;
}
#else
return 1;
}
-static inline int same_magic(const char *amagic, const char *bmagic)
+static inline int same_magic(const char *amagic, const char *bmagic,
+ bool has_crcs)
{
return strcmp(amagic, bmagic) == 0;
}
err = try_to_force_load(mod, "magic");
if (err)
goto free_hdr;
- } else if (!same_magic(modmagic, vermagic)) {
+ } else if (!same_magic(modmagic, vermagic, versindex)) {
printk(KERN_ERR "%s: version magic '%s' should be '%s'\n",
mod->name, modmagic, vermagic);
err = -ENOEXEC;
ret = 0;
spliced = 0;
- while (len && !spliced) {
+ while (len) {
ret = subbuf_splice_actor(in, ppos, pipe, len, flags, &nonpad_ret);
if (ret < 0)
break;
#include <asm/tlb.h>
#include <asm/irq_regs.h>
-/*
- * Scheduler clock - returns current time in nanosec units.
- * This is default implementation.
- * Architectures and sub-architectures can override this.
- */
-unsigned long long __attribute__((weak)) sched_clock(void)
-{
- return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ);
-}
-
/*
* Convert user-nice values [ -20 ... 0 ... 19 ]
* to static priority [ MAX_RT_PRIO..MAX_PRIO-1 ],
}
#endif
+/*
+ * sched_domains_mutex serializes calls to arch_init_sched_domains,
+ * detach_destroy_domains and partition_sched_domains.
+ */
+static DEFINE_MUTEX(sched_domains_mutex);
+
#ifdef CONFIG_GROUP_SCHED
#include <linux/cgroup.h>
*/
static DEFINE_SPINLOCK(task_group_lock);
-/* doms_cur_mutex serializes access to doms_cur[] array */
-static DEFINE_MUTEX(doms_cur_mutex);
-
#ifdef CONFIG_FAIR_GROUP_SCHED
#ifdef CONFIG_USER_SCHED
# define INIT_TASK_GROUP_LOAD (2*NICE_0_LOAD)
# define INIT_TASK_GROUP_LOAD NICE_0_LOAD
#endif
+/*
+ * A weight of 0, 1 or ULONG_MAX can cause arithmetics problems.
+ * (The default weight is 1024 - so there's no practical
+ * limitation from this.)
+ */
#define MIN_SHARES 2
+#define MAX_SHARES (ULONG_MAX - 1)
static int init_task_group_load = INIT_TASK_GROUP_LOAD;
#endif
#endif
}
-static inline void lock_doms_cur(void)
-{
- mutex_lock(&doms_cur_mutex);
-}
-
-static inline void unlock_doms_cur(void)
-{
- mutex_unlock(&doms_cur_mutex);
-}
-
#else
static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { }
-static inline void lock_doms_cur(void) { }
-static inline void unlock_doms_cur(void) { }
#endif /* CONFIG_GROUP_SCHED */
unsigned long next_balance;
struct mm_struct *prev_mm;
- u64 clock, prev_clock_raw;
- s64 clock_max_delta;
-
- unsigned int clock_warps, clock_overflows, clock_underflows;
- u64 idle_clock;
- unsigned int clock_deep_idle_events;
- u64 tick_timestamp;
+ u64 clock;
atomic_t nr_iowait;
#endif
}
-#ifdef CONFIG_NO_HZ
-static inline bool nohz_on(int cpu)
-{
- return tick_get_tick_sched(cpu)->nohz_mode != NOHZ_MODE_INACTIVE;
-}
-
-static inline u64 max_skipped_ticks(struct rq *rq)
-{
- return nohz_on(cpu_of(rq)) ? jiffies - rq->last_tick_seen + 2 : 1;
-}
-
-static inline void update_last_tick_seen(struct rq *rq)
-{
- rq->last_tick_seen = jiffies;
-}
-#else
-static inline u64 max_skipped_ticks(struct rq *rq)
-{
- return 1;
-}
-
-static inline void update_last_tick_seen(struct rq *rq)
-{
-}
-#endif
-
-/*
- * Update the per-runqueue clock, as finegrained as the platform can give
- * us, but without assuming monotonicity, etc.:
- */
-static void __update_rq_clock(struct rq *rq)
-{
- u64 prev_raw = rq->prev_clock_raw;
- u64 now = sched_clock();
- s64 delta = now - prev_raw;
- u64 clock = rq->clock;
-
-#ifdef CONFIG_SCHED_DEBUG
- WARN_ON_ONCE(cpu_of(rq) != smp_processor_id());
-#endif
- /*
- * Protect against sched_clock() occasionally going backwards:
- */
- if (unlikely(delta < 0)) {
- clock++;
- rq->clock_warps++;
- } else {
- /*
- * Catch too large forward jumps too:
- */
- u64 max_jump = max_skipped_ticks(rq) * TICK_NSEC;
- u64 max_time = rq->tick_timestamp + max_jump;
-
- if (unlikely(clock + delta > max_time)) {
- if (clock < max_time)
- clock = max_time;
- else
- clock++;
- rq->clock_overflows++;
- } else {
- if (unlikely(delta > rq->clock_max_delta))
- rq->clock_max_delta = delta;
- clock += delta;
- }
- }
-
- rq->prev_clock_raw = now;
- rq->clock = clock;
-}
-
-static void update_rq_clock(struct rq *rq)
-{
- if (likely(smp_processor_id() == cpu_of(rq)))
- __update_rq_clock(rq);
-}
-
/*
* The domain tree (rq->sd) is protected by RCU's quiescent state transition.
* See detach_destroy_domains: synchronize_sched for details.
#define task_rq(p) cpu_rq(task_cpu(p))
#define cpu_curr(cpu) (cpu_rq(cpu)->curr)
+static inline void update_rq_clock(struct rq *rq)
+{
+ rq->clock = sched_clock_cpu(cpu_of(rq));
+}
+
/*
* Tunables that become constants when CONFIG_SCHED_DEBUG is off:
*/
#define SCHED_FEAT(name, enabled) \
#name ,
-__read_mostly char *sched_feat_names[] = {
+static __read_mostly char *sched_feat_names[] = {
#include "sched_features.h"
NULL
};
#undef SCHED_FEAT
-int sched_feat_open(struct inode *inode, struct file *filp)
+static int sched_feat_open(struct inode *inode, struct file *filp)
{
filp->private_data = inode->i_private;
return 0;
return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC;
}
-static const unsigned long long time_sync_thresh = 100000;
+unsigned long long time_sync_thresh = 100000;
static DEFINE_PER_CPU(unsigned long long, time_offset);
static DEFINE_PER_CPU(unsigned long long, prev_cpu_time);
static DEFINE_SPINLOCK(time_sync_lock);
static unsigned long long prev_global_time;
-static unsigned long long __sync_cpu_clock(cycles_t time, int cpu)
+static unsigned long long __sync_cpu_clock(unsigned long long time, int cpu)
{
- unsigned long flags;
-
- spin_lock_irqsave(&time_sync_lock, flags);
+ /*
+ * We want this inlined, to not get tracer function calls
+ * in this critical section:
+ */
+ spin_acquire(&time_sync_lock.dep_map, 0, 0, _THIS_IP_);
+ __raw_spin_lock(&time_sync_lock.raw_lock);
if (time < prev_global_time) {
per_cpu(time_offset, cpu) += prev_global_time - time;
prev_global_time = time;
}
- spin_unlock_irqrestore(&time_sync_lock, flags);
+ __raw_spin_unlock(&time_sync_lock.raw_lock);
+ spin_release(&time_sync_lock.dep_map, 1, _THIS_IP_);
return time;
}
static unsigned long long __cpu_clock(int cpu)
{
unsigned long long now;
- unsigned long flags;
- struct rq *rq;
/*
* Only call sched_clock() if the scheduler has already been
if (unlikely(!scheduler_running))
return 0;
- local_irq_save(flags);
- rq = cpu_rq(cpu);
- update_rq_clock(rq);
- now = rq->clock;
- local_irq_restore(flags);
+ now = sched_clock_cpu(cpu);
return now;
}
unsigned long long cpu_clock(int cpu)
{
unsigned long long prev_cpu_time, time, delta_time;
+ unsigned long flags;
+ local_irq_save(flags);
prev_cpu_time = per_cpu(prev_cpu_time, cpu);
time = __cpu_clock(cpu) + per_cpu(time_offset, cpu);
delta_time = time-prev_cpu_time;
- if (unlikely(delta_time > time_sync_thresh))
+ if (unlikely(delta_time > time_sync_thresh)) {
time = __sync_cpu_clock(time, cpu);
+ per_cpu(prev_cpu_time, cpu) = time;
+ }
+ local_irq_restore(flags);
return time;
}
return rq;
}
-/*
- * We are going deep-idle (irqs are disabled):
- */
-void sched_clock_idle_sleep_event(void)
-{
- struct rq *rq = cpu_rq(smp_processor_id());
-
- spin_lock(&rq->lock);
- __update_rq_clock(rq);
- spin_unlock(&rq->lock);
- rq->clock_deep_idle_events++;
-}
-EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event);
-
-/*
- * We just idled delta nanoseconds (called with irqs disabled):
- */
-void sched_clock_idle_wakeup_event(u64 delta_ns)
-{
- struct rq *rq = cpu_rq(smp_processor_id());
- u64 now = sched_clock();
-
- rq->idle_clock += delta_ns;
- /*
- * Override the previous timestamp and ignore all
- * sched_clock() deltas that occured while we idled,
- * and use the PM-provided delta_ns to advance the
- * rq clock:
- */
- spin_lock(&rq->lock);
- rq->prev_clock_raw = now;
- rq->clock += delta_ns;
- spin_unlock(&rq->lock);
- touch_softlockup_watchdog();
-}
-EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event);
-
static void __resched_task(struct task_struct *p, int tif_bit);
static inline void resched_task(struct task_struct *p)
enum {
HRTICK_SET, /* re-programm hrtick_timer */
HRTICK_RESET, /* not a new slice */
+ HRTICK_BLOCK, /* stop hrtick operations */
};
/*
{
if (!sched_feat(HRTICK))
return 0;
+ if (unlikely(test_bit(HRTICK_BLOCK, &rq->hrtick_flags)))
+ return 0;
return hrtimer_is_hres_active(&rq->hrtick_timer);
}
WARN_ON_ONCE(cpu_of(rq) != smp_processor_id());
spin_lock(&rq->lock);
- __update_rq_clock(rq);
+ update_rq_clock(rq);
rq->curr->sched_class->task_tick(rq, rq->curr, 1);
spin_unlock(&rq->lock);
return HRTIMER_NORESTART;
}
-static inline void init_rq_hrtick(struct rq *rq)
+static void hotplug_hrtick_disable(int cpu)
+{
+ struct rq *rq = cpu_rq(cpu);
+ unsigned long flags;
+
+ spin_lock_irqsave(&rq->lock, flags);
+ rq->hrtick_flags = 0;
+ __set_bit(HRTICK_BLOCK, &rq->hrtick_flags);
+ spin_unlock_irqrestore(&rq->lock, flags);
+
+ hrtick_clear(rq);
+}
+
+static void hotplug_hrtick_enable(int cpu)
+{
+ struct rq *rq = cpu_rq(cpu);
+ unsigned long flags;
+
+ spin_lock_irqsave(&rq->lock, flags);
+ __clear_bit(HRTICK_BLOCK, &rq->hrtick_flags);
+ spin_unlock_irqrestore(&rq->lock, flags);
+}
+
+static int
+hotplug_hrtick(struct notifier_block *nfb, unsigned long action, void *hcpu)
+{
+ int cpu = (int)(long)hcpu;
+
+ switch (action) {
+ case CPU_UP_CANCELED:
+ case CPU_UP_CANCELED_FROZEN:
+ case CPU_DOWN_PREPARE:
+ case CPU_DOWN_PREPARE_FROZEN:
+ case CPU_DEAD:
+ case CPU_DEAD_FROZEN:
+ hotplug_hrtick_disable(cpu);
+ return NOTIFY_OK;
+
+ case CPU_UP_PREPARE:
+ case CPU_UP_PREPARE_FROZEN:
+ case CPU_DOWN_FAILED:
+ case CPU_DOWN_FAILED_FROZEN:
+ case CPU_ONLINE:
+ case CPU_ONLINE_FROZEN:
+ hotplug_hrtick_enable(cpu);
+ return NOTIFY_OK;
+ }
+
+ return NOTIFY_DONE;
+}
+
+static void init_hrtick(void)
+{
+ hotcpu_notifier(hotplug_hrtick, 0);
+}
+
+static void init_rq_hrtick(struct rq *rq)
{
rq->hrtick_flags = 0;
hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
void hrtick_resched(void)
{
}
+
+static inline void init_hrtick(void)
+{
+}
#endif
/*
{
u64 tmp;
- if (unlikely(!lw->inv_weight))
- lw->inv_weight = (WMULT_CONST-lw->weight/2) / (lw->weight+1);
+ if (!lw->inv_weight)
+ lw->inv_weight = 1 + (WMULT_CONST-lw->weight/2)/(lw->weight+1);
tmp = (u64)delta_exec * weight;
/*
if (shares < MIN_SHARES)
shares = MIN_SHARES;
+ else if (shares > MAX_SHARES)
+ shares = MAX_SHARES;
__set_se_shares(tg->se[tcpu], shares);
}
struct rq *rq = this_rq();
cputime64_t tmp;
- if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0))
- return account_guest_time(p, cputime);
+ if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) {
+ account_guest_time(p, cputime);
+ return;
+ }
p->stime = cputime_add(p->stime, cputime);
int cpu = smp_processor_id();
struct rq *rq = cpu_rq(cpu);
struct task_struct *curr = rq->curr;
- u64 next_tick = rq->tick_timestamp + TICK_NSEC;
+
+ sched_clock_tick();
spin_lock(&rq->lock);
- __update_rq_clock(rq);
- /*
- * Let rq->clock advance by at least TICK_NSEC:
- */
- if (unlikely(rq->clock < next_tick)) {
- rq->clock = next_tick;
- rq->clock_underflows++;
- }
- rq->tick_timestamp = rq->clock;
- update_last_tick_seen(rq);
+ update_rq_clock(rq);
update_cpu_load(rq);
curr->sched_class->task_tick(rq, curr, 0);
spin_unlock(&rq->lock);
* Do the rq-clock update outside the rq lock:
*/
local_irq_disable();
- __update_rq_clock(rq);
+ update_rq_clock(rq);
spin_lock(&rq->lock);
clear_tsk_need_resched(prev);
prev->sched_class->put_prev_task(rq, prev);
next = pick_next_task(rq, prev);
- sched_info_switch(prev, next);
-
if (likely(prev != next)) {
+ sched_info_switch(prev, next);
+
rq->nr_switches++;
rq->curr = next;
++*switch_count;
{
int i, j;
- lock_doms_cur();
+ mutex_lock(&sched_domains_mutex);
/* always unregister in case we don't destroy any domains */
unregister_sched_domain_sysctl();
register_sched_domain_sysctl();
- unlock_doms_cur();
+ mutex_unlock(&sched_domains_mutex);
}
#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
int err;
get_online_cpus();
+ mutex_lock(&sched_domains_mutex);
detach_destroy_domains(&cpu_online_map);
err = arch_init_sched_domains(&cpu_online_map);
+ mutex_unlock(&sched_domains_mutex);
put_online_cpus();
return err;
BUG_ON(sched_group_nodes_bycpu == NULL);
#endif
get_online_cpus();
+ mutex_lock(&sched_domains_mutex);
arch_init_sched_domains(&cpu_online_map);
cpus_andnot(non_isolated_cpus, cpu_possible_map, cpu_isolated_map);
if (cpus_empty(non_isolated_cpus))
cpu_set(smp_processor_id(), non_isolated_cpus);
+ mutex_unlock(&sched_domains_mutex);
put_online_cpus();
/* XXX: Theoretical race here - CPU may be hotplugged now */
hotcpu_notifier(update_sched_domains, 0);
+ init_hrtick();
/* Move init over to a non-isolated CPU */
if (set_cpus_allowed_ptr(current, &non_isolated_cpus) < 0)
se->my_q = cfs_rq;
se->load.weight = tg->shares;
- se->load.inv_weight = div64_u64(1ULL<<32, se->load.weight);
+ se->load.inv_weight = 0;
se->parent = parent;
}
#endif
spin_lock_init(&rq->lock);
lockdep_set_class(&rq->lock, &rq->rq_lock_key);
rq->nr_running = 0;
- rq->clock = 1;
- update_last_tick_seen(rq);
init_cfs_rq(&rq->cfs, rq);
init_rt_rq(&rq->rt, rq);
#ifdef CONFIG_FAIR_GROUP_SCHED
static void normalize_task(struct rq *rq, struct task_struct *p)
{
int on_rq;
+
update_rq_clock(rq);
on_rq = p->se.on_rq;
if (on_rq)
p->se.sleep_start = 0;
p->se.block_start = 0;
#endif
- task_rq(p)->clock = 0;
if (!rt_task(p)) {
/*
dequeue_entity(cfs_rq, se, 0);
se->load.weight = shares;
- se->load.inv_weight = div64_u64((1ULL<<32), shares);
+ se->load.inv_weight = 0;
if (on_rq)
enqueue_entity(cfs_rq, se, 0);
if (!tg->se[0])
return -EINVAL;
- /*
- * A weight of 0 or 1 can cause arithmetics problems.
- * (The default weight is 1024 - so there's no practical
- * limitation from this.)
- */
if (shares < MIN_SHARES)
shares = MIN_SHARES;
+ else if (shares > MAX_SHARES)
+ shares = MAX_SHARES;
mutex_lock(&shares_mutex);
if (tg->shares == shares)
* force a rebalance
*/
cfs_rq_set_shares(tg->cfs_rq[i], 0);
- set_se_shares(tg->se[i], shares/nr_cpu_ids);
+ set_se_shares(tg->se[i], shares);
}
/*
--- /dev/null
+/*
+ * sched_clock for unstable cpu clocks
+ *
+ * Copyright (C) 2008 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
+ *
+ * Based on code by:
+ * Ingo Molnar <mingo@redhat.com>
+ * Guillaume Chazarain <guichaz@gmail.com>
+ *
+ * Create a semi stable clock from a mixture of other events, including:
+ * - gtod
+ * - jiffies
+ * - sched_clock()
+ * - explicit idle events
+ *
+ * We use gtod as base and the unstable clock deltas. The deltas are filtered,
+ * making it monotonic and keeping it within an expected window. This window
+ * is set up using jiffies.
+ *
+ * Furthermore, explicit sleep and wakeup hooks allow us to account for time
+ * that is otherwise invisible (TSC gets stopped).
+ *
+ * The clock: sched_clock_cpu() is monotonic per cpu, and should be somewhat
+ * consistent between cpus (never more than 1 jiffies difference).
+ */
+#include <linux/sched.h>
+#include <linux/percpu.h>
+#include <linux/spinlock.h>
+#include <linux/ktime.h>
+#include <linux/module.h>
+
+
+#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK
+
+struct sched_clock_data {
+ /*
+ * Raw spinlock - this is a special case: this might be called
+ * from within instrumentation code so we dont want to do any
+ * instrumentation ourselves.
+ */
+ raw_spinlock_t lock;
+
+ unsigned long prev_jiffies;
+ u64 prev_raw;
+ u64 tick_raw;
+ u64 tick_gtod;
+ u64 clock;
+};
+
+static DEFINE_PER_CPU_SHARED_ALIGNED(struct sched_clock_data, sched_clock_data);
+
+static inline struct sched_clock_data *this_scd(void)
+{
+ return &__get_cpu_var(sched_clock_data);
+}
+
+static inline struct sched_clock_data *cpu_sdc(int cpu)
+{
+ return &per_cpu(sched_clock_data, cpu);
+}
+
+void sched_clock_init(void)
+{
+ u64 ktime_now = ktime_to_ns(ktime_get());
+ u64 now = 0;
+ int cpu;
+
+ for_each_possible_cpu(cpu) {
+ struct sched_clock_data *scd = cpu_sdc(cpu);
+
+ scd->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED;
+ scd->prev_jiffies = jiffies;
+ scd->prev_raw = now;
+ scd->tick_raw = now;
+ scd->tick_gtod = ktime_now;
+ scd->clock = ktime_now;
+ }
+}
+
+/*
+ * update the percpu scd from the raw @now value
+ *
+ * - filter out backward motion
+ * - use jiffies to generate a min,max window to clip the raw values
+ */
+static void __update_sched_clock(struct sched_clock_data *scd, u64 now)
+{
+ unsigned long now_jiffies = jiffies;
+ long delta_jiffies = now_jiffies - scd->prev_jiffies;
+ u64 clock = scd->clock;
+ u64 min_clock, max_clock;
+ s64 delta = now - scd->prev_raw;
+
+ WARN_ON_ONCE(!irqs_disabled());
+ min_clock = scd->tick_gtod + delta_jiffies * TICK_NSEC;
+
+ if (unlikely(delta < 0)) {
+ clock++;
+ goto out;
+ }
+
+ max_clock = min_clock + TICK_NSEC;
+
+ if (unlikely(clock + delta > max_clock)) {
+ if (clock < max_clock)
+ clock = max_clock;
+ else
+ clock++;
+ } else {
+ clock += delta;
+ }
+
+ out:
+ if (unlikely(clock < min_clock))
+ clock = min_clock;
+
+ scd->prev_raw = now;
+ scd->prev_jiffies = now_jiffies;
+ scd->clock = clock;
+}
+
+static void lock_double_clock(struct sched_clock_data *data1,
+ struct sched_clock_data *data2)
+{
+ if (data1 < data2) {
+ __raw_spin_lock(&data1->lock);
+ __raw_spin_lock(&data2->lock);
+ } else {
+ __raw_spin_lock(&data2->lock);
+ __raw_spin_lock(&data1->lock);
+ }
+}
+
+u64 sched_clock_cpu(int cpu)
+{
+ struct sched_clock_data *scd = cpu_sdc(cpu);
+ u64 now, clock;
+
+ WARN_ON_ONCE(!irqs_disabled());
+ now = sched_clock();
+
+ if (cpu != raw_smp_processor_id()) {
+ /*
+ * in order to update a remote cpu's clock based on our
+ * unstable raw time rebase it against:
+ * tick_raw (offset between raw counters)
+ * tick_gotd (tick offset between cpus)
+ */
+ struct sched_clock_data *my_scd = this_scd();
+
+ lock_double_clock(scd, my_scd);
+
+ now -= my_scd->tick_raw;
+ now += scd->tick_raw;
+
+ now -= my_scd->tick_gtod;
+ now += scd->tick_gtod;
+
+ __raw_spin_unlock(&my_scd->lock);
+ } else {
+ __raw_spin_lock(&scd->lock);
+ }
+
+ __update_sched_clock(scd, now);
+ clock = scd->clock;
+
+ __raw_spin_unlock(&scd->lock);
+
+ return clock;
+}
+
+void sched_clock_tick(void)
+{
+ struct sched_clock_data *scd = this_scd();
+ u64 now, now_gtod;
+
+ WARN_ON_ONCE(!irqs_disabled());
+
+ now = sched_clock();
+ now_gtod = ktime_to_ns(ktime_get());
+
+ __raw_spin_lock(&scd->lock);
+ __update_sched_clock(scd, now);
+ /*
+ * update tick_gtod after __update_sched_clock() because that will
+ * already observe 1 new jiffy; adding a new tick_gtod to that would
+ * increase the clock 2 jiffies.
+ */
+ scd->tick_raw = now;
+ scd->tick_gtod = now_gtod;
+ __raw_spin_unlock(&scd->lock);
+}
+
+/*
+ * We are going deep-idle (irqs are disabled):
+ */
+void sched_clock_idle_sleep_event(void)
+{
+ sched_clock_cpu(smp_processor_id());
+}
+EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event);
+
+/*
+ * We just idled delta nanoseconds (called with irqs disabled):
+ */
+void sched_clock_idle_wakeup_event(u64 delta_ns)
+{
+ struct sched_clock_data *scd = this_scd();
+ u64 now = sched_clock();
+
+ /*
+ * Override the previous timestamp and ignore all
+ * sched_clock() deltas that occured while we idled,
+ * and use the PM-provided delta_ns to advance the
+ * rq clock:
+ */
+ __raw_spin_lock(&scd->lock);
+ scd->prev_raw = now;
+ scd->clock += delta_ns;
+ __raw_spin_unlock(&scd->lock);
+
+ touch_softlockup_watchdog();
+}
+EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event);
+
+#endif
+
+/*
+ * Scheduler clock - returns current time in nanosec units.
+ * This is default implementation.
+ * Architectures and sub-architectures can override this.
+ */
+unsigned long long __attribute__((weak)) sched_clock(void)
+{
+ return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ);
+}
PN(next_balance);
P(curr->pid);
PN(clock);
- PN(idle_clock);
- PN(prev_clock_raw);
- P(clock_warps);
- P(clock_overflows);
- P(clock_underflows);
- P(clock_deep_idle_events);
- PN(clock_max_delta);
P(cpu_load[0]);
P(cpu_load[1]);
P(cpu_load[2]);
if (!initial) {
/* sleeps upto a single latency don't count. */
if (sched_feat(NEW_FAIR_SLEEPERS)) {
+ unsigned long thresh = sysctl_sched_latency;
+
+ /*
+ * convert the sleeper threshold into virtual time
+ */
if (sched_feat(NORMALIZED_SLEEPER))
- vruntime -= calc_delta_weight(sysctl_sched_latency, se);
- else
- vruntime -= sysctl_sched_latency;
+ thresh = calc_delta_fair(thresh, se);
+
+ vruntime -= thresh;
}
/* ensure we never gain time by being placed backwards. */
* Update run-time statistics of the 'current'.
*/
update_curr(cfs_rq);
+ account_entity_enqueue(cfs_rq, se);
if (wakeup) {
place_entity(cfs_rq, se, 0);
check_spread(cfs_rq, se);
if (se != cfs_rq->curr)
__enqueue_entity(cfs_rq, se);
- account_entity_enqueue(cfs_rq, se);
}
static void update_avg(u64 *avg, u64 sample)
* queued ticks are scheduled to match the slice, so don't bother
* validating it and just reschedule.
*/
- if (queued)
- return resched_task(rq_of(cfs_rq)->curr);
+ if (queued) {
+ resched_task(rq_of(cfs_rq)->curr);
+ return;
+ }
/*
* don't let the period tick interfere with the hrtick preemption
*/
return;
if (likely(!sysctl_sched_compat_yield) && curr->policy != SCHED_BATCH) {
- __update_rq_clock(rq);
+ update_rq_clock(rq);
/*
* Update run-time statistics of the 'current'.
*/
* sibling runqueue info. This will avoid the checks and cache miss
* penalities associated with that.
*/
- if (idle_cpu(cpu) || cpu_rq(cpu)->nr_running > 1)
+ if (idle_cpu(cpu) || cpu_rq(cpu)->cfs.nr_running > 1)
return cpu;
for_each_domain(cpu, sd) {
};
#ifdef CONFIG_SCHED_DEBUG
-static void
-print_cfs_rq_tasks(struct seq_file *m, struct cfs_rq *cfs_rq, int depth)
-{
- struct sched_entity *se;
-
- if (!cfs_rq)
- return;
-
- list_for_each_entry_rcu(se, &cfs_rq->tasks, group_node) {
- int i;
-
- for (i = depth; i; i--)
- seq_puts(m, " ");
-
- seq_printf(m, "%lu %s %lu\n",
- se->load.weight,
- entity_is_task(se) ? "T" : "G",
- calc_delta_weight(SCHED_LOAD_SCALE, se)
- );
- if (!entity_is_task(se))
- print_cfs_rq_tasks(m, group_cfs_rq(se), depth + 1);
- }
-}
-
static void print_cfs_stats(struct seq_file *m, int cpu)
{
struct cfs_rq *cfs_rq;
rcu_read_lock();
for_each_leaf_cfs_rq(cpu_rq(cpu), cfs_rq)
print_cfs_rq(m, cpu, cfs_rq);
-
- seq_printf(m, "\nWeight tree:\n");
- print_cfs_rq_tasks(m, &cpu_rq(cpu)->cfs, 1);
rcu_read_unlock();
}
#endif
/*
* Simple, special scheduling class for the per-CPU idle tasks:
*/
-const struct sched_class idle_sched_class = {
+static const struct sched_class idle_sched_class = {
/* .next is NULL */
/* no enqueue/yield_task for idle tasks */
}
}
-
+/*
+ * If we are not running and we are not going to reschedule soon, we should
+ * try to push tasks away now
+ */
static void task_wake_up_rt(struct rq *rq, struct task_struct *p)
{
if (!task_running(rq, p) &&
- (p->prio >= rq->rt.highest_prio) &&
+ !test_tsk_need_resched(rq->curr) &&
rq->rt.overloaded)
push_rt_tasks(rq);
}
p->se.exec_start = rq->clock;
}
-const struct sched_class rt_sched_class = {
+static const struct sched_class rt_sched_class = {
.next = &fair_sched_class,
.enqueue_task = enqueue_task_rt,
.dequeue_task = dequeue_task_rt,
unsigned long flags;
spin_lock_irqsave(&sem->lock, flags);
- if (likely(sem->count > 0))
- sem->count--;
- else
+ if (unlikely(!sem->count))
__down(sem);
+ sem->count--;
spin_unlock_irqrestore(&sem->lock, flags);
}
EXPORT_SYMBOL(down);
int result = 0;
spin_lock_irqsave(&sem->lock, flags);
- if (likely(sem->count > 0))
- sem->count--;
- else
+ if (unlikely(!sem->count))
result = __down_interruptible(sem);
+ if (!result)
+ sem->count--;
spin_unlock_irqrestore(&sem->lock, flags);
return result;
int result = 0;
spin_lock_irqsave(&sem->lock, flags);
- if (likely(sem->count > 0))
- sem->count--;
- else
+ if (unlikely(!sem->count))
result = __down_killable(sem);
+ if (!result)
+ sem->count--;
spin_unlock_irqrestore(&sem->lock, flags);
return result;
int result = 0;
spin_lock_irqsave(&sem->lock, flags);
- if (likely(sem->count > 0))
- sem->count--;
- else
+ if (unlikely(!sem->count))
result = __down_timeout(sem, jiffies);
+ if (!result)
+ sem->count--;
spin_unlock_irqrestore(&sem->lock, flags);
return result;
unsigned long flags;
spin_lock_irqsave(&sem->lock, flags);
- if (likely(list_empty(&sem->wait_list)))
- sem->count++;
- else
+ sem->count++;
+ if (unlikely(!list_empty(&sem->wait_list)))
__up(sem);
spin_unlock_irqrestore(&sem->lock, flags);
}
struct semaphore_waiter {
struct list_head list;
struct task_struct *task;
- int up;
};
/*
{
struct task_struct *task = current;
struct semaphore_waiter waiter;
+ int ret = 0;
- list_add_tail(&waiter.list, &sem->wait_list);
waiter.task = task;
- waiter.up = 0;
+ list_add_tail(&waiter.list, &sem->wait_list);
for (;;) {
- if (state == TASK_INTERRUPTIBLE && signal_pending(task))
- goto interrupted;
- if (state == TASK_KILLABLE && fatal_signal_pending(task))
- goto interrupted;
- if (timeout <= 0)
- goto timed_out;
+ if (state == TASK_INTERRUPTIBLE && signal_pending(task)) {
+ ret = -EINTR;
+ break;
+ }
+ if (state == TASK_KILLABLE && fatal_signal_pending(task)) {
+ ret = -EINTR;
+ break;
+ }
+ if (timeout <= 0) {
+ ret = -ETIME;
+ break;
+ }
__set_task_state(task, state);
spin_unlock_irq(&sem->lock);
timeout = schedule_timeout(timeout);
spin_lock_irq(&sem->lock);
- if (waiter.up)
- return 0;
+ if (sem->count > 0)
+ break;
}
- timed_out:
- list_del(&waiter.list);
- return -ETIME;
-
- interrupted:
list_del(&waiter.list);
- return -EINTR;
+ return ret;
}
static noinline void __sched __down(struct semaphore *sem)
{
struct semaphore_waiter *waiter = list_first_entry(&sem->wait_list,
struct semaphore_waiter, list);
- list_del(&waiter->list);
- waiter->up = 1;
wake_up_process(waiter->task);
}
+config HAVE_ARCH_KGDB_SHADOW_INFO
+ bool
+
+config HAVE_ARCH_KGDB
+ bool
+
menuconfig KGDB
bool "KGDB: kernel debugging with remote gdb"
select FRAME_POINTER
at http://kgdb.sourceforge.net as well as in DocBook form
in Documentation/DocBook/. If unsure, say N.
-config HAVE_ARCH_KGDB_SHADOW_INFO
- bool
-
-config HAVE_ARCH_KGDB
- bool
+if KGDB
config KGDB_SERIAL_CONSOLE
tristate "KGDB: use kgdb over the serial console"
- depends on KGDB
select CONSOLE_POLL
select MAGIC_SYSRQ
default y
config KGDB_TESTS
bool "KGDB: internal test suite"
- depends on KGDB
default n
help
This is a kgdb I/O module specifically designed to test
boot. See the drivers/misc/kgdbts.c for detailed
information about other strings you could use beyond the
default of V1F100.
+
+endif # KGDB
}
EXPORT_SYMBOL(should_remove_suid);
-int __remove_suid(struct dentry *dentry, int kill)
+static int __remove_suid(struct dentry *dentry, int kill)
{
struct iattr newattrs;
goto no_page_table;
pmd = pmd_offset(pud, address);
- if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd)))
+ if (pmd_none(*pmd))
goto no_page_table;
if (pmd_huge(*pmd)) {
goto out;
}
+ if (unlikely(pmd_bad(*pmd)))
+ goto no_page_table;
+
ptep = pte_offset_map_lock(mm, pmd, address, &ptl);
if (!ptep)
goto out;
if (!n)
continue;
- if (atomic_read(&n->total_objects))
+ if (atomic_long_read(&n->total_objects))
return 1;
}
return 0;
/* skb==NULL means VCC is being destroyed */
br2684_close_vcc(brvcc);
if (list_empty(&brdev->brvccs)) {
- read_lock(&devs_lock);
+ write_lock_irq(&devs_lock);
list_del(&brdev->br2684_devs);
- read_unlock(&devs_lock);
+ write_unlock_irq(&devs_lock);
unregister_netdev(net_dev);
free_netdev(net_dev);
}
rtnl_lock();
if (strchr(dev->name, '%')) {
ret = dev_alloc_name(dev, dev->name);
- if (ret < 0) {
- free_netdev(dev);
- goto out;
- }
+ if (ret < 0)
+ goto out_free;
}
ret = register_netdevice(dev);
if (ret)
- goto out;
+ goto out_free;
ret = br_sysfs_addbr(dev);
if (ret)
out:
rtnl_unlock();
return ret;
+
+out_free:
+ free_netdev(dev);
+ goto out;
}
int br_del_bridge(const char *name)
*/
int can_send(struct sk_buff *skb, int loop)
{
+ struct sk_buff *newskb = NULL;
int err;
if (skb->dev->type != ARPHRD_CAN) {
* If the interface is not capable to do loopback
* itself, we do it here.
*/
- struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC);
-
+ newskb = skb_clone(skb, GFP_ATOMIC);
if (!newskb) {
kfree_skb(skb);
return -ENOMEM;
newskb->sk = skb->sk;
newskb->ip_summed = CHECKSUM_UNNECESSARY;
newskb->pkt_type = PACKET_BROADCAST;
- netif_rx(newskb);
}
} else {
/* indication for the CAN driver: no loopback required */
if (err > 0)
err = net_xmit_errno(err);
+ if (err) {
+ if (newskb)
+ kfree_skb(newskb);
+ return err;
+ }
+
+ if (newskb)
+ netif_rx(newskb);
+
/* update statistics */
can_stats.tx_frames++;
can_stats.tx_frames_delta++;
- return err;
+ return 0;
}
EXPORT_SYMBOL(can_send);
{
int ret = 0;
+ ASSERT_RTNL();
+
/*
* Is it already up?
*/
*/
int dev_close(struct net_device *dev)
{
+ ASSERT_RTNL();
+
might_sleep();
if (!(dev->flags & IFF_UP))
rtnl_lock();
for_each_netdev_safe(net, dev, next) {
int err;
+ char fb_name[IFNAMSIZ];
/* Ignore unmoveable devices (i.e. loopback) */
if (dev->features & NETIF_F_NETNS_LOCAL)
continue;
/* Push remaing network devices to init_net */
- err = dev_change_net_namespace(dev, &init_net, "dev%d");
+ snprintf(fb_name, IFNAMSIZ, "dev%d", dev->ifindex);
+ err = dev_change_net_namespace(dev, &init_net, fb_name);
if (err) {
- printk(KERN_WARNING "%s: failed to move %s to init_net: %d\n",
+ printk(KERN_EMERG "%s: failed to move %s to init_net: %d\n",
__func__, dev->name, err);
- unregister_netdevice(dev);
+ BUG();
}
}
rtnl_unlock();
goto nodata;
/*
- * See comment in sk_buff definition, just before the 'tail' member
+ * Only clear those fields we need to clear, not those that we will
+ * actually initialise below. Hence, don't put any more fields after
+ * the tail pointer in struct sk_buff!
*/
memset(skb, 0, offsetof(struct sk_buff, tail));
skb->truesize = size + sizeof(struct sk_buff);
if (len > 3) {
DCCP_WARN("invalid length %d\n", len);
- return 1;
+ return -EINVAL;
}
/* XXX add further sanity checks */
else
min_mtu -= 21;
- if (dst->metrics[RTAX_MTU-1] > mtu && mtu >= min_mtu) {
+ if (dst_metric(dst, RTAX_MTU) > mtu && mtu >= min_mtu) {
if (!(dst_metric_locked(dst, RTAX_MTU))) {
dst->metrics[RTAX_MTU-1] = mtu;
dst_set_expires(dst, dn_rt_mtu_expires);
}
if (!(dst_metric_locked(dst, RTAX_ADVMSS))) {
u32 mss = mtu - DN_MAX_NSP_DATA_HEADER;
- if (dst->metrics[RTAX_ADVMSS-1] > mss)
+ if (dst_metric(dst, RTAX_ADVMSS) > mss)
dst->metrics[RTAX_ADVMSS-1] = mss;
}
}
rt->u.dst.neighbour = n;
}
- if (rt->u.dst.metrics[RTAX_MTU-1] == 0 ||
- rt->u.dst.metrics[RTAX_MTU-1] > rt->u.dst.dev->mtu)
+ if (dst_metric(&rt->u.dst, RTAX_MTU) == 0 ||
+ dst_metric(&rt->u.dst, RTAX_MTU) > rt->u.dst.dev->mtu)
rt->u.dst.metrics[RTAX_MTU-1] = rt->u.dst.dev->mtu;
mss = dn_mss_from_pmtu(dev, dst_mtu(&rt->u.dst));
- if (rt->u.dst.metrics[RTAX_ADVMSS-1] == 0 ||
- rt->u.dst.metrics[RTAX_ADVMSS-1] > mss)
+ if (dst_metric(&rt->u.dst, RTAX_ADVMSS) == 0 ||
+ dst_metric(&rt->u.dst, RTAX_ADVMSS) > mss)
rt->u.dst.metrics[RTAX_ADVMSS-1] = mss;
return 0;
}
iph = ip_hdr(skb);
/*
- * RFC1122: 3.1.2.2 MUST silently discard any IP frame that fails the checksum.
+ * RFC1122: 3.2.1.2 MUST silently discard any IP frame that fails the checksum.
*
* Is the datagram acceptable?
*
/* BSD 4.2 compatibility hack :-( */
if (mtu == 0 &&
- old_mtu >= rth->u.dst.metrics[RTAX_MTU-1] &&
+ old_mtu >= dst_metric(&rth->u.dst, RTAX_MTU) &&
old_mtu >= 68 + (iph->ihl << 2))
old_mtu -= iph->ihl << 2;
mtu = guess_mtu(old_mtu);
}
- if (mtu <= rth->u.dst.metrics[RTAX_MTU-1]) {
- if (mtu < rth->u.dst.metrics[RTAX_MTU-1]) {
+ if (mtu <= dst_metric(&rth->u.dst, RTAX_MTU)) {
+ if (mtu < dst_metric(&rth->u.dst, RTAX_MTU)) {
dst_confirm(&rth->u.dst);
if (mtu < ip_rt_min_pmtu) {
mtu = ip_rt_min_pmtu;
static void ip_rt_update_pmtu(struct dst_entry *dst, u32 mtu)
{
- if (dst->metrics[RTAX_MTU-1] > mtu && mtu >= 68 &&
+ if (dst_metric(dst, RTAX_MTU) > mtu && mtu >= 68 &&
!(dst_metric_locked(dst, RTAX_MTU))) {
if (mtu < ip_rt_min_pmtu) {
mtu = ip_rt_min_pmtu;
sizeof(rt->u.dst.metrics));
if (fi->fib_mtu == 0) {
rt->u.dst.metrics[RTAX_MTU-1] = rt->u.dst.dev->mtu;
- if (rt->u.dst.metrics[RTAX_LOCK-1] & (1 << RTAX_MTU) &&
+ if (dst_metric_locked(&rt->u.dst, RTAX_MTU) &&
rt->rt_gateway != rt->rt_dst &&
rt->u.dst.dev->mtu > 576)
rt->u.dst.metrics[RTAX_MTU-1] = 576;
} else
rt->u.dst.metrics[RTAX_MTU-1]= rt->u.dst.dev->mtu;
- if (rt->u.dst.metrics[RTAX_HOPLIMIT-1] == 0)
+ if (dst_metric(&rt->u.dst, RTAX_HOPLIMIT) == 0)
rt->u.dst.metrics[RTAX_HOPLIMIT-1] = sysctl_ip_default_ttl;
- if (rt->u.dst.metrics[RTAX_MTU-1] > IP_MAX_MTU)
+ if (dst_metric(&rt->u.dst, RTAX_MTU) > IP_MAX_MTU)
rt->u.dst.metrics[RTAX_MTU-1] = IP_MAX_MTU;
- if (rt->u.dst.metrics[RTAX_ADVMSS-1] == 0)
+ if (dst_metric(&rt->u.dst, RTAX_ADVMSS) == 0)
rt->u.dst.metrics[RTAX_ADVMSS-1] = max_t(unsigned int, rt->u.dst.dev->mtu - 40,
ip_rt_min_advmss);
- if (rt->u.dst.metrics[RTAX_ADVMSS-1] > 65535 - 40)
+ if (dst_metric(&rt->u.dst, RTAX_ADVMSS) > 65535 - 40)
rt->u.dst.metrics[RTAX_ADVMSS-1] = 65535 - 40;
#ifdef CONFIG_NET_CLS_ROUTE
#include <linux/mm.h>
#include <linux/module.h>
#include <linux/sysctl.h>
+#include <net/dst.h>
#include <net/tcp.h>
#include <net/inet_common.h>
#include <linux/ipsec.h>
#define FLAG_FORWARD_PROGRESS (FLAG_ACKED|FLAG_DATA_SACKED)
#define FLAG_ANY_PROGRESS (FLAG_FORWARD_PROGRESS|FLAG_SND_UNA_ADVANCED)
-#define IsSackFrto() (sysctl_tcp_frto == 0x2)
-
#define TCP_REMNANT (TCP_FLAG_FIN|TCP_FLAG_URG|TCP_FLAG_SYN|TCP_FLAG_PSH)
#define TCP_HP_BITS (~(TCP_RESERVED_BITS|TCP_FLAG_PSH))
u32 rto_min = TCP_RTO_MIN;
if (dst && dst_metric_locked(dst, RTAX_RTO_MIN))
- rto_min = dst->metrics[RTAX_RTO_MIN - 1];
+ rto_min = dst_metric(dst, RTAX_RTO_MIN);
return rto_min;
}
dst->metrics[RTAX_RTTVAR - 1] = m;
else
dst->metrics[RTAX_RTTVAR-1] -=
- (dst->metrics[RTAX_RTTVAR-1] - m)>>2;
+ (dst_metric(dst, RTAX_RTTVAR) - m)>>2;
}
if (tp->snd_ssthresh >= 0xFFFF) {
dst->metrics[RTAX_SSTHRESH-1] =
max(tp->snd_cwnd >> 1, tp->snd_ssthresh);
if (!dst_metric_locked(dst, RTAX_CWND))
- dst->metrics[RTAX_CWND-1] = (dst->metrics[RTAX_CWND-1] + tp->snd_cwnd) >> 1;
+ dst->metrics[RTAX_CWND-1] = (dst_metric(dst, RTAX_CWND) + tp->snd_cwnd) >> 1;
} else {
/* Else slow start did not finish, cwnd is non-sense,
ssthresh may be also invalid.
*/
if (!dst_metric_locked(dst, RTAX_CWND))
- dst->metrics[RTAX_CWND-1] = (dst->metrics[RTAX_CWND-1] + tp->snd_ssthresh) >> 1;
- if (dst->metrics[RTAX_SSTHRESH-1] &&
+ dst->metrics[RTAX_CWND-1] = (dst_metric(dst, RTAX_CWND) + tp->snd_ssthresh) >> 1;
+ if (dst_metric(dst, RTAX_SSTHRESH) &&
!dst_metric_locked(dst, RTAX_SSTHRESH) &&
- tp->snd_ssthresh > dst->metrics[RTAX_SSTHRESH-1])
+ tp->snd_ssthresh > dst_metric(dst, RTAX_SSTHRESH))
dst->metrics[RTAX_SSTHRESH-1] = tp->snd_ssthresh;
}
if (!dst_metric_locked(dst, RTAX_REORDERING)) {
- if (dst->metrics[RTAX_REORDERING-1] < tp->reordering &&
+ if (dst_metric(dst, RTAX_REORDERING) < tp->reordering &&
tp->reordering != sysctl_tcp_reordering)
dst->metrics[RTAX_REORDERING-1] = tp->reordering;
}
tp->sacked_out = 0;
}
+static int tcp_is_sackfrto(const struct tcp_sock *tp)
+{
+ return (sysctl_tcp_frto == 0x2) && !tcp_is_reno(tp);
+}
+
/* F-RTO can only be used if TCP has never retransmitted anything other than
* head (SACK enhanced variant from Appendix B of RFC4138 is more robust here)
*/
if (icsk->icsk_mtup.probe_size)
return 0;
- if (IsSackFrto())
+ if (tcp_is_sackfrto(tp))
return 1;
/* Avoid expensive walking of rexmit queue if possible */
/* Earlier loss recovery underway (see RFC4138; Appendix B).
* The last condition is necessary at least in tp->frto_counter case.
*/
- if (IsSackFrto() && (tp->frto_counter ||
+ if (tcp_is_sackfrto(tp) && (tp->frto_counter ||
((1 << icsk->icsk_ca_state) & (TCPF_CA_Recovery|TCPF_CA_Loss))) &&
after(tp->high_seq, tp->snd_una)) {
tp->frto_highmark = tp->high_seq;
return 1;
}
- if (!IsSackFrto() || tcp_is_reno(tp)) {
+ if (!tcp_is_sackfrto(tp)) {
/* RFC4138 shortcoming in step 2; should also have case c):
* ACK isn't duplicate nor advances window, e.g., opposite dir
* data, winupdate
}
}
- if (rt->u.dst.metrics[RTAX_HOPLIMIT-1] == 0)
+ if (dst_metric(&rt->u.dst, RTAX_HOPLIMIT) == 0)
rt->u.dst.metrics[RTAX_HOPLIMIT-1] = -1;
- if (!rt->u.dst.metrics[RTAX_MTU-1])
+ if (!dst_metric(&rt->u.dst, RTAX_MTU))
rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(dev);
- if (!rt->u.dst.metrics[RTAX_ADVMSS-1])
+ if (!dst_metric(&rt->u.dst, RTAX_ADVMSS))
rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst));
rt->u.dst.dev = dev;
rt->rt6i_idev = idev;
}
icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
- kfree_skb(skb);
read_unlock(&ipip6_lock);
out:
+ kfree_skb(skb);
return 0;
}
fail_rate:
ieee80211_debugfs_remove_netdev(IEEE80211_DEV_TO_SUB_IF(local->mdev));
unregister_netdevice(local->mdev);
+ local->mdev = NULL;
fail_dev:
rtnl_unlock();
sta_info_stop(local);
debugfs_hw_del(local);
destroy_workqueue(local->hw.workqueue);
fail_workqueue:
- ieee80211_if_free(local->mdev);
- local->mdev = NULL;
+ if (local->mdev != NULL) {
+ ieee80211_if_free(local->mdev);
+ local->mdev = NULL;
+ }
fail_mdev_alloc:
wiphy_unregister(local->hw.wiphy);
return result;
struct rc_pid_sta_info *sinfo = inode->i_private;
struct rc_pid_event_buffer *events = &sinfo->events;
struct rc_pid_events_file_info *file_info;
- unsigned int status;
+ unsigned long status;
/* Allocate a state struct */
file_info = kmalloc(sizeof(*file_info), GFP_KERNEL);
char pb[RC_PID_PRINT_BUF_SIZE];
int ret;
int p;
- unsigned int status;
+ unsigned long status;
/* Check if there is something to read. */
if (events->next_entry == file_info->next_entry) {
tristate 'DCCP protocol connection tracking support (EXPERIMENTAL)'
depends on EXPERIMENTAL && NF_CONNTRACK
depends on NETFILTER_ADVANCED
+ default IP_DCCP
help
With this option enabled, the layer 3 independent connection
tracking code will be able to do state tracking on DCCP connections.
tristate 'SCTP protocol connection tracking support (EXPERIMENTAL)'
depends on EXPERIMENTAL && NF_CONNTRACK
depends on NETFILTER_ADVANCED
+ default IP_SCTP
help
With this option enabled, the layer 3 independent connection
tracking code will be able to do state tracking on SCTP connections.
tristate '"dccp" protocol match support'
depends on NETFILTER_XTABLES
depends on NETFILTER_ADVANCED
+ default IP_DCCP
help
With this option enabled, you will be able to use the iptables
`dccp' match in order to match on DCCP source/destination ports
tristate '"sctp" protocol match support (EXPERIMENTAL)'
depends on NETFILTER_XTABLES && EXPERIMENTAL
depends on NETFILTER_ADVANCED
+ default IP_SCTP
help
With this option enabled, you will be able to use the
`sctp' match in order to match on SCTP source/destination ports
{
enum ip_conntrack_info ctinfo;
struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+ struct nf_conn_help *help = nfct_help(ct);
unsigned int matchoff, matchlen;
unsigned int mediaoff, medialen;
unsigned int sdpoff;
if (nf_nat_sdp_session && ct->status & IPS_NAT_MASK)
ret = nf_nat_sdp_session(skb, dptr, sdpoff, datalen, &rtp_addr);
+ if (ret == NF_ACCEPT && i > 0)
+ help->help.ct_sip_info.invite_cseq = cseq;
+
return ret;
}
static int process_invite_response(struct sk_buff *skb,
{
enum ip_conntrack_info ctinfo;
struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+ struct nf_conn_help *help = nfct_help(ct);
if ((code >= 100 && code <= 199) ||
(code >= 200 && code <= 299))
return process_sdp(skb, dptr, datalen, cseq);
- else {
+ else if (help->help.ct_sip_info.invite_cseq == cseq)
flush_expectations(ct, true);
- return NF_ACCEPT;
- }
+ return NF_ACCEPT;
}
static int process_update_response(struct sk_buff *skb,
{
enum ip_conntrack_info ctinfo;
struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+ struct nf_conn_help *help = nfct_help(ct);
if ((code >= 100 && code <= 199) ||
(code >= 200 && code <= 299))
return process_sdp(skb, dptr, datalen, cseq);
- else {
+ else if (help->help.ct_sip_info.invite_cseq == cseq)
flush_expectations(ct, true);
- return NF_ACCEPT;
- }
+ return NF_ACCEPT;
}
static int process_prack_response(struct sk_buff *skb,
{
enum ip_conntrack_info ctinfo;
struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+ struct nf_conn_help *help = nfct_help(ct);
if ((code >= 100 && code <= 199) ||
(code >= 200 && code <= 299))
return process_sdp(skb, dptr, datalen, cseq);
- else {
+ else if (help->help.ct_sip_info.invite_cseq == cseq)
flush_expectations(ct, true);
- return NF_ACCEPT;
- }
+ return NF_ACCEPT;
}
static int process_bye_request(struct sk_buff *skb,
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*
- * Authors: Jamal Hadi Salim (2005)
+ * Authors: Jamal Hadi Salim (2005-8)
*
*/
.lock = &simp_lock,
};
+#define SIMP_MAX_DATA 32
static int tcf_simp(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
{
struct tcf_defact *d = a->priv;
return ret;
}
-static int alloc_defdata(struct tcf_defact *d, u32 datalen, void *defdata)
+static int alloc_defdata(struct tcf_defact *d, char *defdata)
{
- d->tcfd_defdata = kmemdup(defdata, datalen, GFP_KERNEL);
+ d->tcfd_defdata = kstrndup(defdata, SIMP_MAX_DATA, GFP_KERNEL);
if (unlikely(!d->tcfd_defdata))
return -ENOMEM;
- d->tcfd_datalen = datalen;
+
return 0;
}
-static int realloc_defdata(struct tcf_defact *d, u32 datalen, void *defdata)
+static void reset_policy(struct tcf_defact *d, char *defdata,
+ struct tc_defact *p)
{
- kfree(d->tcfd_defdata);
- return alloc_defdata(d, datalen, defdata);
+ spin_lock_bh(&d->tcf_lock);
+ d->tcf_action = p->action;
+ memset(d->tcfd_defdata, 0, SIMP_MAX_DATA);
+ strlcpy(d->tcfd_defdata, defdata, SIMP_MAX_DATA);
+ spin_unlock_bh(&d->tcf_lock);
}
static const struct nla_policy simple_policy[TCA_DEF_MAX + 1] = {
[TCA_DEF_PARMS] = { .len = sizeof(struct tc_defact) },
+ [TCA_DEF_DATA] = { .type = NLA_STRING, .len = SIMP_MAX_DATA },
};
static int tcf_simp_init(struct nlattr *nla, struct nlattr *est,
struct tc_defact *parm;
struct tcf_defact *d;
struct tcf_common *pc;
- void *defdata;
- u32 datalen = 0;
+ char *defdata;
int ret = 0, err;
if (nla == NULL)
return -EINVAL;
- err = nla_parse_nested(tb, TCA_DEF_MAX, nla, NULL);
+ err = nla_parse_nested(tb, TCA_DEF_MAX, nla, simple_policy);
if (err < 0)
return err;
if (tb[TCA_DEF_PARMS] == NULL)
return -EINVAL;
- parm = nla_data(tb[TCA_DEF_PARMS]);
- defdata = nla_data(tb[TCA_DEF_DATA]);
- if (defdata == NULL)
+ if (tb[TCA_DEF_DATA] == NULL)
return -EINVAL;
- datalen = nla_len(tb[TCA_DEF_DATA]);
- if (datalen == 0)
- return -EINVAL;
+ parm = nla_data(tb[TCA_DEF_PARMS]);
+ defdata = nla_data(tb[TCA_DEF_DATA]);
pc = tcf_hash_check(parm->index, a, bind, &simp_hash_info);
if (!pc) {
return -ENOMEM;
d = to_defact(pc);
- ret = alloc_defdata(d, datalen, defdata);
+ ret = alloc_defdata(d, defdata);
if (ret < 0) {
kfree(pc);
return ret;
}
+ d->tcf_action = parm->action;
ret = ACT_P_CREATED;
} else {
d = to_defact(pc);
tcf_simp_release(d, bind);
return -EEXIST;
}
- realloc_defdata(d, datalen, defdata);
+ reset_policy(d, defdata, parm);
}
- spin_lock_bh(&d->tcf_lock);
- d->tcf_action = parm->action;
- spin_unlock_bh(&d->tcf_lock);
-
if (ret == ACT_P_CREATED)
tcf_hash_insert(pc, &simp_hash_info);
return ret;
opt.bindcnt = d->tcf_bindcnt - bind;
opt.action = d->tcf_action;
NLA_PUT(skb, TCA_DEF_PARMS, sizeof(opt), &opt);
- NLA_PUT(skb, TCA_DEF_DATA, d->tcfd_datalen, d->tcfd_defdata);
+ NLA_PUT_STRING(skb, TCA_DEF_DATA, d->tcfd_defdata);
t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install);
t.lastuse = jiffies_to_clock_t(jiffies - d->tcf_tm.lastuse);
t.expires = jiffies_to_clock_t(d->tcf_tm.expires);
return 1;
}
-static void htb_parent_to_leaf(struct htb_class *cl, struct Qdisc *new_q)
+static void htb_parent_to_leaf(struct htb_sched *q, struct htb_class *cl,
+ struct Qdisc *new_q)
{
struct htb_class *parent = cl->parent;
BUG_TRAP(!cl->level && cl->un.leaf.q && !cl->prio_activity);
+ if (parent->cmode != HTB_CAN_SEND)
+ htb_safe_rb_erase(&parent->pq_node, q->wait_pq + parent->level);
+
parent->level = 0;
memset(&parent->un.inner, 0, sizeof(parent->un.inner));
INIT_LIST_HEAD(&parent->un.leaf.drop_list);
htb_deactivate(q, cl);
if (last_child)
- htb_parent_to_leaf(cl, new_q);
+ htb_parent_to_leaf(q, cl, new_q);
if (--cl->refcnt == 0)
htb_destroy_class(sch, cl);
#include <linux/mm.h>
#include <linux/interrupt.h>
#include <linux/module.h>
-#include <linux/sched.h>
#include <linux/sunrpc/types.h>
#include <linux/sunrpc/xdr.h>
/*
* TIPC message buffer code
*
- * TIPC message buffer headroom reserves space for a link-level header
- * (in case the message is sent off-node),
- * while ensuring TIPC header is word aligned for quicker access
+ * TIPC message buffer headroom reserves space for the worst-case
+ * link-level device header (in case the message is sent off-node).
*
- * The largest header currently supported is 18 bytes, which is used when
- * the standard 14 byte Ethernet header has 4 added bytes for VLAN info
+ * Note: Headroom should be a multiple of 4 to ensure the TIPC header fields
+ * are word aligned for quicker access
*/
-#define BUF_HEADROOM 20u
+#define BUF_HEADROOM LL_MAX_HEADER
struct tipc_skb_cb {
void *handle;
config SND_PCSP
- tristate "Internal PC speaker support"
- depends on X86_PC && HIGH_RES_TIMERS
+ tristate "PC-Speaker support"
+ depends on PCSPKR_PLATFORM && X86_PC && HIGH_RES_TIMERS
depends on INPUT
depends on SND
select SND_PCM
spin_unlock_irq(&chip->substream_lock);
}
+#ifdef CONFIG_PM
static int pcsp_suspend(struct platform_device *dev, pm_message_t state)
{
struct snd_pcsp *chip = platform_get_drvdata(dev);
snd_pcm_suspend_all(chip->pcm);
return 0;
}
+#else
+#define pcsp_suspend NULL
+#endif /* CONFIG_PM */
static void pcsp_shutdown(struct platform_device *dev)
{
return 1;
mem = ioremap(base, 128);
- if(mem == 0UL)
+ if (!mem)
return 1;
map = readw(mem + 0x18); /* Read the SMI enables */
iounmap(mem);
config SND_FM801_TEA575X_BOOL
bool "ForteMedia FM801 + TEA5757 tuner"
depends on SND_FM801
+ depends on VIDEO_V4L1=y || VIDEO_V4L1=SND_FM801
help
Say Y here to include support for soundcards based on the ForteMedia
FM801 chip with a TEA5757 tuner connected to GPIO1-3 pins (Media
Forte SF256-PCS-02) into the snd-fm801 driver.
- This will enable support for the old V4L1 API.
-
config SND_FM801_TEA575X
tristate
depends on SND_FM801_TEA575X_BOOL
default SND_FM801
- select VIDEO_V4L1
- select VIDEO_DEV
config SND_HDA_INTEL
tristate "Intel HD Audio"
int patch_vt1617a(struct snd_ac97 * ac97)
{
int err = 0;
+ int val;
/* we choose to not fail out at this point, but we tell the
caller when we return */
/* bring analog power consumption to normal by turning off the
* headphone amplifier, like WinXP driver for EPIA SP
*/
- snd_ac97_write_cache(ac97, 0x5c, 0x20);
+ /* We need to check the bit before writing it.
+ * On some (many?) hardwares, setting bit actually clears it!
+ */
+ val = snd_ac97_read(ac97, 0x5c);
+ if (!(val & 0x20))
+ snd_ac97_write_cache(ac97, 0x5c, 0x20);
+
ac97->ext_id |= AC97_EI_SPDIF; /* force the detection of spdif */
ac97->rates[AC97_RATES_SPDIF] = SNDRV_PCM_RATE_44100 | SNDRV_PCM_RATE_48000;
ac97->build_ops = &patch_vt1616_ops;
hda_nid_t nid,
int pin_type, int dac_idx)
{
- alc_set_pin_output(codec, nid, pin_type);
+ snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL,
+ pin_type);
+ snd_hda_codec_write(codec, dac_idx, 0, AC_VERB_SET_AMP_GAIN_MUTE,
+ AMP_OUT_UNMUTE);
}
static void alc861_auto_init_multi_out(struct hda_codec *codec)
{ .id = 0x83847635, .name = "STAC9250D", .patch = patch_stac925x },
{ .id = 0x83847636, .name = "STAC9251", .patch = patch_stac925x },
{ .id = 0x83847637, .name = "STAC9250D", .patch = patch_stac925x },
+ { .id = 0x83847645, .name = "92HD206X", .patch = patch_stac927x },
+ { .id = 0x83847646, .name = "92HD206D", .patch = patch_stac927x },
/* The following does not take into account .id=0x83847661 when subsys =
* 104D0C00 which is STAC9225s. Because of this, some SZ Notebooks are
* currently not fully supported.
if (prtd->period_ptr >= prtd->dma_buffer_end) {
prtd->period_ptr = prtd->dma_buffer;
}
- at91_ssc_write(params->ssc_base + params->pdc->xnpr, prtd->period_ptr);
+ at91_ssc_write(params->ssc_base + params->pdc->xnpr,
+ prtd->period_ptr);
at91_ssc_write(params->ssc_base + params->pdc->xncr,
prtd->period_size / params->pdc_xfer_size);
}
at91_ssc_write(params->ssc_base + AT91_SSC_IER,
params->mask->ssc_endx | params->mask->ssc_endbuf);
- at91_ssc_write(params->ssc_base + ATMEL_PDC_PTCR, params->mask->pdc_enable);
+ at91_ssc_write(params->ssc_base + ATMEL_PDC_PTCR,
+ params->mask->pdc_enable);
- DBG("sr=%lx imr=%lx\n", at91_ssc_read(params->ssc_base + AT91_SSC_SR),
- at91_ssc_read(params->ssc_base + AT91_SSC_IER));
+ DBG("sr=%lx imr=%lx\n",
+ at91_ssc_read(params->ssc_base + AT91_SSC_SR),
+ at91_ssc_read(params->ssc_base + AT91_SSC_IMR));
break;
case SNDRV_PCM_TRIGGER_STOP:
printk(KERN_WARNING "at91-ssc: request_irq failure\n");
DBG("Stopping pid %d clock\n", ssc_p->ssc.pid);
- at91_sys_write(AT91_PMC_PCER, 1<<ssc_p->ssc.pid);
+ at91_sys_write(AT91_PMC_PCDR, 1<<ssc_p->ssc.pid);
return ret;
}
iismod |= S3C2410_IISMOD_SLAVE;
break;
case SND_SOC_DAIFMT_CBS_CFS:
+ iismod &= ~S3C2410_IISMOD_SLAVE;
break;
default:
return -EINVAL;
iismod |= S3C2410_IISMOD_MSB;
break;
case SND_SOC_DAIFMT_I2S:
+ iismod &= ~S3C2410_IISMOD_MSB;
break;
default:
return -EINVAL;
ret = s3c2410_dma_request(prtd->params->channel,
prtd->params->client, NULL);
- if (ret) {
+ if (ret < 0) {
DBG(KERN_ERR "failed to get dma channel\n");
return ret;
}
projects / linux-2.6-omap-h63xx.git / commitdiff