From: Luming Yu <luyu@redhat.com> Subject: [RHEL 5.2 PATCH] bz 250825: fsys_gettimeofday leaps days if it runs with nojitter Date: Tue, 07 Aug 2007 11:09:16 +0800 Bugzilla: 250825 Message-Id: <46B7E25C.2020808@redhat.com> Changelog: [ia64] fsys_gettimeofday leaps days if it runs with nojitter bz 250825 Description of problem: The ".acq" semantics of the load only apply w.r.t. other data access. Reading the clock (ar.itc) isn't a data access so strange things can happen here. Specifically the read of ar.itc can be launched as soon as the read of xtime_lock.sequence is ISSUED. Since this may cache miss, and that might cause a thread switch, and there may be cache contention for the line containing xtime_lock, it may be a long time before the actual value is returned, so the ar.itc value may be very stale. Move the consumption of r28 up before the read of ar.itc to make sure that we really have got the current value of xtime_lock.sequence before look at ar.itc. The upstream patch link: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff_plain;h=829a9996259e4d0b20ce7b94c49b985d6ba6b760 The back port for rhel 4.6 is attached, and tested by me with a simple gettimeofday test case. It works Please ACK. Thanks, Luming X-Git-Tag: v2.6.23-rc1~521^2~1 X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=829a9996259e4d0b20ce7b94c49b985d6ba6b760 [IA64] ar.itc access must really be after xtime_lock.sequence has been read The ".acq" semantics of the load only apply w.r.t. other data access. Reading the clock (ar.itc) isn't a data access so strange things can happen here. Specifically the read of ar.itc can be launched as soon as the read of xtime_lock.sequence is ISSUED. Since this may cache miss, and that might cause a thread switch, and there may be cache contention for the line containing xtime_lock, it may be a long time before the actual value is returned, so the ar.itc value may be very stale. Move the consumption of r28 up before the read of ar.itc to make sure that we really have got the current value of xtime_lock.sequence before look at ar.itc. Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com> Signed-off-by: Tony Luck <tony.luck@intel.com> --- diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S index 8589e84..3f926c2 100644 --- a/arch/ia64/kernel/fsys.S +++ b/arch/ia64/kernel/fsys.S @@ -247,6 +247,9 @@ ENTRY(fsys_gettimeofday) .time_redo: .pred.rel.mutex p8,p9,p10 ld4.acq r28 = [r29] // xtime_lock.sequence. Must come first for locking purposes + ;; + and r28 = ~1,r28 // Make sequence even to force retry if odd + ;; (p8) mov r2 = ar.itc // CPU_TIMER. 36 clocks latency!!! add r22 = IA64_TIME_INTERPOLATOR_LAST_COUNTER_OFFSET,r20 (p9) ld8 r2 = [r30] // readq(ti->address). Could also have latency issues.. @@ -284,7 +287,6 @@ EX(.fail_efault, probe.w.fault r31, 3) // This takes 5 cycles and we have spare (p15) ld8 r17 = [r19],-IA64_TIMESPEC_TV_NSEC_OFFSET (p7) cmp.ne p7,p0 = r25,r3 // if cmpxchg not successful redo // simulate tbit.nz.or p7,p0 = r28,0 - and r28 = ~1,r28 // Make sequence even to force retry if odd getf.sig r2 = f8 mf add r8 = r8,r18 // Add time interpolator offset