aboutsummaryrefslogtreecommitdiffstats
path: root/arch/sparc64
diff options
context:
space:
mode:
authorHugh Dickins <hugh@veritas.com>2005-11-07 14:09:01 -0800
committerDavid S. Miller <davem@davemloft.net>2005-11-07 14:09:01 -0800
commitdedeb0029b9c83420fc1337d4ee53daa7b2a0ad4 (patch)
treed87e66e1d6240cd412c20ecbc12f5b810c9807e4 /arch/sparc64
parentb8ae48656db860d4c83a29aa7b0588fc89361935 (diff)
downloadkernel_samsung_crespo-dedeb0029b9c83420fc1337d4ee53daa7b2a0ad4.zip
kernel_samsung_crespo-dedeb0029b9c83420fc1337d4ee53daa7b2a0ad4.tar.gz
kernel_samsung_crespo-dedeb0029b9c83420fc1337d4ee53daa7b2a0ad4.tar.bz2
[SPARC64] mm: context switch ptlock
sparc64 is unique among architectures in taking the page_table_lock in its context switch (well, cris does too, but erroneously, and it's not yet SMP anyway). This seems to be a private affair between switch_mm and activate_mm, using page_table_lock as a per-mm lock, without any relation to its uses elsewhere. That's fine, but comment it as such; and unlock sooner in switch_mm, more like in activate_mm (preemption is disabled here). There is a block of "if (0)"ed code in smp_flush_tlb_pending which would have liked to rely on the page_table_lock, in switch_mm and elsewhere; but its comment explains how dup_mmap's flush_tlb_mm defeated it. And though that could have been changed at any time over the past few years, now the chance vanishes as we push the page_table_lock downwards, and perhaps split it per page table page. Just delete that block of code. Which leaves the mysterious spin_unlock_wait(&oldmm->page_table_lock) in kernel/fork.c copy_mm. Textual analysis (supported by Nick Piggin) suggests that the comment was written by DaveM, and that it relates to the defeated approach in the sparc64 smp_flush_tlb_pending. Just delete this block too. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'arch/sparc64')
-rw-r--r--arch/sparc64/kernel/smp.c31
1 files changed, 5 insertions, 26 deletions
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index b137fd6..a9089e2 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -883,34 +883,13 @@ void smp_flush_tlb_pending(struct mm_struct *mm, unsigned long nr, unsigned long
u32 ctx = CTX_HWBITS(mm->context);
int cpu = get_cpu();
- if (mm == current->active_mm && atomic_read(&mm->mm_users) == 1) {
+ if (mm == current->active_mm && atomic_read(&mm->mm_users) == 1)
mm->cpu_vm_mask = cpumask_of_cpu(cpu);
- goto local_flush_and_out;
- } else {
- /* This optimization is not valid. Normally
- * we will be holding the page_table_lock, but
- * there is an exception which is copy_page_range()
- * when forking. The lock is held during the individual
- * page table updates in the parent, but not at the
- * top level, which is where we are invoked.
- */
- if (0) {
- cpumask_t this_cpu_mask = cpumask_of_cpu(cpu);
-
- /* By virtue of running under the mm->page_table_lock,
- * and mmu_context.h:switch_mm doing the same, the
- * following operation is safe.
- */
- if (cpus_equal(mm->cpu_vm_mask, this_cpu_mask))
- goto local_flush_and_out;
- }
- }
-
- smp_cross_call_masked(&xcall_flush_tlb_pending,
- ctx, nr, (unsigned long) vaddrs,
- mm->cpu_vm_mask);
+ else
+ smp_cross_call_masked(&xcall_flush_tlb_pending,
+ ctx, nr, (unsigned long) vaddrs,
+ mm->cpu_vm_mask);
-local_flush_and_out:
__flush_tlb_pending(ctx, nr, vaddrs);
put_cpu();