search for: sleep_start

..._cpu(cpu_time).calibration_timer); } /* Late init function (after all CPUs are booted). */ @@ -926,6 +1013,56 @@ unsigned long get_localtime(struct domai { return wc_sec + (wc_nsec + NOW()) / 1000000000ULL + d->time_offset_seconds; +} + +/* timer suspend/resume */ +static u64 sleep_start; +int time_suspend(void) +{ + sleep_start = get_cmos_time(); + + /* Better to cancel calibration timer for accuracy */ + destroy_percpu_time(); + + if (cur_timer->suspend) + cur_timer->suspend(); + return 0; +} + +int time_resume(void) +{ + u64 sleep_diff; + u64 no...

...: 0 [ 386.320037] .se->exec_start : 386317.815742 [ 386.320040] .se->vruntime : 23467.274189 [ 386.320044] .se->sum_exec_runtime : 13.339940 [ 386.320047] .se->statistics.wait_start : 0.000000 [ 386.320051] .se->statistics.sleep_start : 0.000000 [ 386.320054] .se->statistics.block_start : 0.000000 [ 386.320057] .se->statistics.sleep_max : 0.000000 [ 386.320060] .se->statistics.block_max : 0.000000 [ 386.320063] .se->statistics.exec_max : 1.952409 [ 386.320067] .se->statistics.s...

I''ve been working on running down intermittent ENOSPC issues. I can only seem to replicate ENOSPC errors when running zlib compression. However, I have been seeing similar ENOSPC errors to a lesser extent when playing with the LZ4HC patches. I apologize for not following up on this sooner, but I had drifted away from using zlib, and didn''t notice there was still an issue. My

...ync_cmos_timer, sync_cmos_clock, 0, 0); @@ -375,7 +376,8 @@ static void sync_cmos_clock(unsigned lon void notify_arch_cmos_timer(void) { - mod_timer(&sync_cmos_timer, jiffies + 1); + if (!no_sync_cmos_timer) + mod_timer(&sync_cmos_timer, jiffies + 1); } static long clock_cmos_diff, sleep_start; @@ -446,16 +448,19 @@ static int time_init_device(void) device_initcall(time_init_device); -#ifdef CONFIG_HPET_TIMER -extern void (*late_time_init)(void); -/* Duplicate of time_init() below, with hpet_enable part added */ -static void __init hpet_time_init(void) +void __init init_xtime_from_c...

...ync_cmos_timer, sync_cmos_clock, 0, 0); @@ -375,7 +376,8 @@ static void sync_cmos_clock(unsigned lon void notify_arch_cmos_timer(void) { - mod_timer(&sync_cmos_timer, jiffies + 1); + if (!no_sync_cmos_timer) + mod_timer(&sync_cmos_timer, jiffies + 1); } static long clock_cmos_diff, sleep_start; @@ -446,16 +448,19 @@ static int time_init_device(void) device_initcall(time_init_device); -#ifdef CONFIG_HPET_TIMER -extern void (*late_time_init)(void); -/* Duplicate of time_init() below, with hpet_enable part added */ -static void __init hpet_time_init(void) +void __init init_xtime_from_c...

We need an smb_mb() before waitqueue_active to avoid missing wakeups. Before Mitch was hitting a deadlock between the ordered flushers and the transaction commit because the ordered flushers were waiting for more refs and were never woken up, so those smp_mb()''s are the most important. Everything else I added for correctness sake and to avoid getting bitten by this again somewhere else.

Unlike full virtualization in which the virtual machine provides the same platform interface as running natively on the hardware, paravirtualization requires modification to the guest operating system to work with the platform interface provided by the hypervisor. Xen was designed with performance in mind. Calls to the hypervisor are minimized, batched if necessary, and non-critical codepaths

Unlike full virtualization in which the virtual machine provides the same platform interface as running natively on the hardware, paravirtualization requires modification to the guest operating system to work with the platform interface provided by the hypervisor. Xen was designed with performance in mind. Calls to the hypervisor are minimized, batched if necessary, and non-critical codepaths