summaryrefslogtreecommitdiff
path: root/arch/ia64/mm
diff options
context:
space:
mode:
authorChen, Kenneth W <kenneth.w.chen@intel.com>2006-10-13 10:08:13 -0700
committerTony Luck <tony.luck@intel.com>2007-02-06 15:04:48 -0800
commit00b65985fb2fc542b855b03fcda0d0f2bab4f442 (patch)
treedc9372aced10184945862b9adf0848da3e0e946f /arch/ia64/mm
parenta0776ec8e97bf109e7d973d09fc3e1814eb32bfb (diff)
[IA64] relax per-cpu TLB requirement to DTC
Instead of pinning per-cpu TLB into a DTR, use DTC. This will free up one TLB entry for application, or even kernel if access pattern to per-cpu data area has high temporal locality. Since per-cpu is mapped at the top of region 7 address, we just need to add special case in alt_dtlb_miss. The physical address of per-cpu data is already conveniently stored in IA64_KR(PER_CPU_DATA). Latency for alt_dtlb_miss is not affected as we can hide all the latency. It was measured that alt_dtlb_miss handler has 23 cycles latency before and after the patch. The performance effect is massive for applications that put lots of tlb pressure on CPU. Workload environment like database online transaction processing or application uses tera-byte of memory would benefit the most. Measurement with industry standard database benchmark shown an upward of 1.6% gain. While smaller workloads like cpu, java also showing small improvement. Signed-off-by: Ken Chen <kenneth.w.chen@intel.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
Diffstat (limited to 'arch/ia64/mm')
-rw-r--r--arch/ia64/mm/init.c11
1 files changed, 1 insertions, 10 deletions
diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c
index 1373fae7657f..07d82cd7cbdd 100644
--- a/arch/ia64/mm/init.c
+++ b/arch/ia64/mm/init.c
@@ -337,7 +337,7 @@ setup_gate (void)
void __devinit
ia64_mmu_init (void *my_cpu_data)
{
- unsigned long psr, pta, impl_va_bits;
+ unsigned long pta, impl_va_bits;
extern void __devinit tlb_init (void);
#ifdef CONFIG_DISABLE_VHPT
@@ -346,15 +346,6 @@ ia64_mmu_init (void *my_cpu_data)
# define VHPT_ENABLE_BIT 1
#endif
- /* Pin mapping for percpu area into TLB */
- psr = ia64_clear_ic();
- ia64_itr(0x2, IA64_TR_PERCPU_DATA, PERCPU_ADDR,
- pte_val(pfn_pte(__pa(my_cpu_data) >> PAGE_SHIFT, PAGE_KERNEL)),
- PERCPU_PAGE_SHIFT);
-
- ia64_set_psr(psr);
- ia64_srlz_i();
-
/*
* Check if the virtually mapped linear page table (VMLPT) overlaps with a mapped
* address space. The IA-64 architecture guarantees that at least 50 bits of