From 83ce400928680a6c8123d492684b27857f5a2d95 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 26 Feb 2009 20:16:58 +0100 Subject: x86: set X86_FEATURE_TSC_RELIABLE If the TSC is constant and non-stop, also set it reliable. (We will turn this off in DMI quirks for multi-chassis systems) The performance number on a 16-way Nehalem system running 32 tasks that context-switch between each other is significant: sched_clock_stable=0 sched_clock_stable=1 .................... .................... 22.456925 million/sec 24.306972 million/sec [+8.2%] lmbench's "lat_ctx -s 0 2" goes from 0.63 microseconds to 0.59 microseconds - a 6.7% increase in context-switching performance. Perfstat of 1 million pipe context switches between two tasks: Performance counter stats for './pipe-test-1m': [before] [after] ............ ............ 37621.421089 36436.848378 task clock ticks (msecs) 0 0 CPU migrations (events) 2000274 2000189 context switches (events) 194 193 pagefaults (events) 8433799643 8171016416 CPU cycles (events) -3.21% 8370133368 8180999694 instructions (events) -2.31% 4158565 3895941 cache references (events) -6.74% 44312 46264 cache misses (events) 2349.287976 2279.362465 wall-time (msecs) -3.06% The speedup comes straight from the reduction in the instruction count. sched_clock_cpu() got simpler and the whole workload thus executes faster. Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/intel.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'arch/x86/kernel/cpu/intel.c') diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 24ff26a38ade..5fff00c70de0 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -4,6 +4,7 @@ #include #include #include +#include #include #include @@ -56,11 +57,16 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) /* * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate - * with P/T states and does not stop in deep C-states + * with P/T states and does not stop in deep C-states. + * + * It is also reliable across cores and sockets. (but not across + * cabinets - we turn it off in that case explicitly.) */ if (c->x86_power & (1 << 8)) { set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC); + set_cpu_cap(c, X86_FEATURE_TSC_RELIABLE); + sched_clock_stable = 1; } } -- cgit v1.2.3 From 02dde8b45c5460794b9052d7c12939fe3eb63c2c Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 12 Mar 2009 12:08:49 +0000 Subject: x86: move various CPU initialization objects into .cpuinit.rodata Impact: debuggability and micro-optimization Putting whatever is possible into the (final) .rodata section increases the likelihood of catching memory corruption bugs early, and reduces false cache line sharing. Signed-off-by: Jan Beulich LKML-Reference: <49B90961.76E4.0078.0@novell.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/intel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/kernel/cpu/intel.c') diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 191117f1ad51..968f15129ed8 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -410,7 +410,7 @@ static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 *c, unsigned i } #endif -static struct cpu_dev intel_cpu_dev __cpuinitdata = { +static const struct cpu_dev __cpuinitconst intel_cpu_dev = { .c_vendor = "Intel", .c_ident = { "GenuineIntel" }, #ifdef CONFIG_X86_32 -- cgit v1.2.3 From 13c6c53282d99c82e79b02477efd2c1e30a991ef Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 12 Mar 2009 12:37:34 +0000 Subject: x86, 32-bit: also use cpuinfo_x86's x86_{phys,virt}_bits members Impact: 32/64-bit consolidation In a first step, this allows fixing phys_addr_valid() for PAE (which until now reported all addresses to be valid). Subsequently, this will also allow simplifying some MTRR handling code. Signed-off-by: Jan Beulich LKML-Reference: <49B9101E.76E4.0078.0@novell.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/intel.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/x86/kernel/cpu/intel.c') diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 191117f1ad51..ae769471042e 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -54,6 +54,11 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) c->x86_cache_alignment = 128; #endif + /* CPUID workaround for 0F33/0F34 CPU */ + if (c->x86 == 0xF && c->x86_model == 0x3 + && (c->x86_mask == 0x3 || c->x86_mask == 0x4)) + c->x86_phys_bits = 36; + /* * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate * with P/T states and does not stop in deep C-states -- cgit v1.2.3