Re: [PATCH 1/1] x86 TSC: set X86_FEATURE_TSC_RELIABLE, per CPUID

2015-06-09 Thread Thomas Gleixner
On Mon, 8 Jun 2015, Andy Lutomirski wrote: > On Wed, Jun 3, 2015 at 12:03 PM, Len Brown wrote: > > If you boot this machine with "tsc=reliable", to disable the cpu_up > > check_tsc_sync, > > what happens? Does the run-time clocksource code detect the bogus TSC > > values? > > If yes, how long do

Re: [PATCH 1/1] x86 TSC: set X86_FEATURE_TSC_RELIABLE, per CPUID

2015-06-08 Thread Andy Lutomirski
On Wed, Jun 3, 2015 at 12:03 PM, Len Brown wrote: > On Mon, Jun 1, 2015 at 9:12 PM, Andy Lutomirski wrote: >> On Mon, Jun 1, 2015 at 5:45 PM, Len Brown wrote: >>> On Mon, Jun 1, 2015 at 2:40 PM, Andy Lutomirski wrote: On 05/30/2015 10:44 PM, Len Brown wrote: > > From: Len Brown >>

Re: [PATCH 1/1] x86 TSC: set X86_FEATURE_TSC_RELIABLE, per CPUID

2015-06-06 Thread Andy Lutomirski
On Thu, Jun 4, 2015 at 12:17 AM, H. Peter Anvin wrote: > On 06/01/2015 05:45 PM, Len Brown wrote: >> >> BTW, it also begs the question if Linux could actually *repair* the BIOS >> damage? >> > > We may very well be able to by setting the TSC offset register to zero > on all CPUs. What TSC offset

Re: [PATCH 1/1] x86 TSC: set X86_FEATURE_TSC_RELIABLE, per CPUID

2015-06-04 Thread H. Peter Anvin
On 06/01/2015 05:45 PM, Len Brown wrote: > > BTW, it also begs the question if Linux could actually *repair* the BIOS > damage? > We may very well be able to by setting the TSC offset register to zero on all CPUs. -hpa -- To unsubscribe from this list: send the line "unsubscribe linu

Re: [PATCH 1/1] x86 TSC: set X86_FEATURE_TSC_RELIABLE, per CPUID

2015-06-03 Thread Len Brown
On Mon, Jun 1, 2015 at 9:12 PM, Andy Lutomirski wrote: > On Mon, Jun 1, 2015 at 5:45 PM, Len Brown wrote: >> On Mon, Jun 1, 2015 at 2:40 PM, Andy Lutomirski wrote: >>> On 05/30/2015 10:44 PM, Len Brown wrote: From: Len Brown Speed cpu_up() by believing CPUID's "invariant TSC

Re: [PATCH 1/1] x86 TSC: set X86_FEATURE_TSC_RELIABLE, per CPUID

2015-06-01 Thread Andy Lutomirski
On Mon, Jun 1, 2015 at 5:45 PM, Len Brown wrote: > On Mon, Jun 1, 2015 at 2:40 PM, Andy Lutomirski wrote: >> On 05/30/2015 10:44 PM, Len Brown wrote: >>> >>> From: Len Brown >>> >>> Speed cpu_up() by believing CPUID's "invariant TSC" flag, >>> and skipping the TSC warp test on single socket syst

Re: [PATCH 1/1] x86 TSC: set X86_FEATURE_TSC_RELIABLE, per CPUID

2015-06-01 Thread Len Brown
On Mon, Jun 1, 2015 at 2:40 PM, Andy Lutomirski wrote: > On 05/30/2015 10:44 PM, Len Brown wrote: >> >> From: Len Brown >> >> Speed cpu_up() by believing CPUID's "invariant TSC" flag, >> and skipping the TSC warp test on single socket systems. > > > I'm typing this email on a "Intel(R) Core(TM) i

Re: [PATCH 1/1] x86 TSC: set X86_FEATURE_TSC_RELIABLE, per CPUID

2015-06-01 Thread Andy Lutomirski
On 05/30/2015 10:44 PM, Len Brown wrote: From: Len Brown Speed cpu_up() by believing CPUID's "invariant TSC" flag, and skipping the TSC warp test on single socket systems. I'm typing this email on a "Intel(R) Core(TM) i7-3930K CPU @ 3.20GHz" with a "X79A-GD65 (8D) (MS-7760)" motherboard. (D

[PATCH 1/1] x86 TSC: set X86_FEATURE_TSC_RELIABLE, per CPUID

2015-05-30 Thread Len Brown
From: Len Brown Speed cpu_up() by believing CPUID's "invariant TSC" flag, and skipping the TSC warp test on single socket systems. On my desktop, cpu_up() duration drops from 4.4ms to 2.4ms. That savings is per-CPU, and occurs during boot, online and resume from S3. Originally, this was done fo