From b0797b60d0067fe437baa97a743c7d9de98fd769 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Tue, 13 Jul 2010 17:56:21 -0700 Subject: powerpc: Simplify update_vsyscall Currently powerpc's update_vsyscall calls an inline update_gtod. However, both are straightforward, and there are no other users, so this patch merges update_gtod into update_vsyscall. Signed-off-by: John Stultz Cc: Anton Blanchard Cc: Paul Mackerras LKML-Reference: <1279068988-21864-5-git-send-email-johnstul@us.ibm.com> Signed-off-by: Thomas Gleixner --- arch/powerpc/kernel/time.c | 55 +++++++++++++++++++++------------------------- 1 file changed, 25 insertions(+), 30 deletions(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 0441bbdadbd1..6fcd64886d1b 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -423,30 +423,6 @@ void udelay(unsigned long usecs) } EXPORT_SYMBOL(udelay); -static inline void update_gtod(u64 new_tb_stamp, u64 new_stamp_xsec, - u64 new_tb_to_xs) -{ - /* - * tb_update_count is used to allow the userspace gettimeofday code - * to assure itself that it sees a consistent view of the tb_to_xs and - * stamp_xsec variables. It reads the tb_update_count, then reads - * tb_to_xs and stamp_xsec and then reads tb_update_count again. If - * the two values of tb_update_count match and are even then the - * tb_to_xs and stamp_xsec values are consistent. If not, then it - * loops back and reads them again until this criteria is met. - * We expect the caller to have done the first increment of - * vdso_data->tb_update_count already. - */ - vdso_data->tb_orig_stamp = new_tb_stamp; - vdso_data->stamp_xsec = new_stamp_xsec; - vdso_data->tb_to_xs = new_tb_to_xs; - vdso_data->wtom_clock_sec = wall_to_monotonic.tv_sec; - vdso_data->wtom_clock_nsec = wall_to_monotonic.tv_nsec; - vdso_data->stamp_xtime = xtime; - smp_wmb(); - ++(vdso_data->tb_update_count); -} - #ifdef CONFIG_SMP unsigned long profile_pc(struct pt_regs *regs) { @@ -876,7 +852,7 @@ static cycle_t timebase_read(struct clocksource *cs) void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, u32 mult) { - u64 t2x, stamp_xsec; + u64 new_tb_to_xs, new_stamp_xsec; if (clock != &clocksource_timebase) return; @@ -887,11 +863,30 @@ void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, /* XXX this assumes clock->shift == 22 */ /* 4611686018 ~= 2^(20+64-22) / 1e9 */ - t2x = (u64) mult * 4611686018ULL; - stamp_xsec = (u64) xtime.tv_nsec * XSEC_PER_SEC; - do_div(stamp_xsec, 1000000000); - stamp_xsec += (u64) xtime.tv_sec * XSEC_PER_SEC; - update_gtod(clock->cycle_last, stamp_xsec, t2x); + new_tb_to_xs = (u64) mult * 4611686018ULL; + new_stamp_xsec = (u64) xtime.tv_nsec * XSEC_PER_SEC; + do_div(new_stamp_xsec, 1000000000); + new_stamp_xsec += (u64) xtime.tv_sec * XSEC_PER_SEC; + + /* + * tb_update_count is used to allow the userspace gettimeofday code + * to assure itself that it sees a consistent view of the tb_to_xs and + * stamp_xsec variables. It reads the tb_update_count, then reads + * tb_to_xs and stamp_xsec and then reads tb_update_count again. If + * the two values of tb_update_count match and are even then the + * tb_to_xs and stamp_xsec values are consistent. If not, then it + * loops back and reads them again until this criteria is met. + * We expect the caller to have done the first increment of + * vdso_data->tb_update_count already. + */ + vdso_data->tb_orig_stamp = clock->cycle_last; + vdso_data->stamp_xsec = new_stamp_xsec; + vdso_data->tb_to_xs = new_tb_to_xs; + vdso_data->wtom_clock_sec = wall_to_monotonic.tv_sec; + vdso_data->wtom_clock_nsec = wall_to_monotonic.tv_nsec; + vdso_data->stamp_xtime = xtime; + smp_wmb(); + ++(vdso_data->tb_update_count); } void update_vsyscall_tz(void) -- cgit v1.2.3 From 06d518e3dfb25334282c7e38b4d7a4eada215f6d Mon Sep 17 00:00:00 2001 From: John Stultz Date: Tue, 13 Jul 2010 17:56:22 -0700 Subject: powerpc: Cleanup xtime usage This removes powerpc's direct xtime usage, allowing for further generic timeekeping cleanups Signed-off-by: John Stultz Cc: Paul Mackerras Cc: Anton Blanchard LKML-Reference: <1279068988-21864-6-git-send-email-johnstul@us.ibm.com> Signed-off-by: Thomas Gleixner --- arch/powerpc/kernel/time.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 6fcd64886d1b..0711d60f40b0 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -864,9 +864,9 @@ void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, /* XXX this assumes clock->shift == 22 */ /* 4611686018 ~= 2^(20+64-22) / 1e9 */ new_tb_to_xs = (u64) mult * 4611686018ULL; - new_stamp_xsec = (u64) xtime.tv_nsec * XSEC_PER_SEC; + new_stamp_xsec = (u64) wall_time->tv_nsec * XSEC_PER_SEC; do_div(new_stamp_xsec, 1000000000); - new_stamp_xsec += (u64) xtime.tv_sec * XSEC_PER_SEC; + new_stamp_xsec += (u64) wall_time->tv_sec * XSEC_PER_SEC; /* * tb_update_count is used to allow the userspace gettimeofday code @@ -884,7 +884,7 @@ void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, vdso_data->tb_to_xs = new_tb_to_xs; vdso_data->wtom_clock_sec = wall_to_monotonic.tv_sec; vdso_data->wtom_clock_nsec = wall_to_monotonic.tv_nsec; - vdso_data->stamp_xtime = xtime; + vdso_data->stamp_xtime = *wall_time; smp_wmb(); ++(vdso_data->tb_update_count); } @@ -1093,7 +1093,7 @@ void __init time_init(void) vdso_data->tb_orig_stamp = tb_last_jiffy; vdso_data->tb_update_count = 0; vdso_data->tb_ticks_per_sec = tb_ticks_per_sec; - vdso_data->stamp_xsec = (u64) xtime.tv_sec * XSEC_PER_SEC; + vdso_data->stamp_xsec = (u64) get_seconds() * XSEC_PER_SEC; vdso_data->tb_to_xs = tb_to_xs; write_sequnlock_irqrestore(&xtime_lock, flags); -- cgit v1.2.3 From 7615856ebfee52b080c22d263ca4debbd0df0ac1 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Tue, 13 Jul 2010 17:56:23 -0700 Subject: timkeeping: Fix update_vsyscall to provide wall_to_monotonic offset update_vsyscall() did not provide the wall_to_monotoinc offset, so arch specific implementations tend to reference wall_to_monotonic directly. This limits future cleanups in the timekeeping core, so this patch fixes the update_vsyscall interface to provide wall_to_monotonic, allowing wall_to_monotonic to be made static as planned in Documentation/feature-removal-schedule.txt Signed-off-by: John Stultz Cc: Martin Schwidefsky Cc: Anton Blanchard Cc: Paul Mackerras Cc: Tony Luck LKML-Reference: <1279068988-21864-7-git-send-email-johnstul@us.ibm.com> Signed-off-by: Thomas Gleixner --- arch/ia64/kernel/time.c | 7 ++++--- arch/powerpc/kernel/time.c | 8 ++++---- arch/s390/kernel/time.c | 8 ++++---- arch/x86/kernel/vsyscall_64.c | 6 +++--- include/linux/clocksource.h | 6 ++++-- kernel/time/timekeeping.c | 9 ++++++--- 6 files changed, 25 insertions(+), 19 deletions(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c index 653b3c46ea82..ed6f22eb5b12 100644 --- a/arch/ia64/kernel/time.c +++ b/arch/ia64/kernel/time.c @@ -471,7 +471,8 @@ void update_vsyscall_tz(void) { } -void update_vsyscall(struct timespec *wall, struct clocksource *c, u32 mult) +void update_vsyscall(struct timespec *wall, struct timespec *wtm, + struct clocksource *c, u32 mult) { unsigned long flags; @@ -487,9 +488,9 @@ void update_vsyscall(struct timespec *wall, struct clocksource *c, u32 mult) /* copy kernel time structures */ fsyscall_gtod_data.wall_time.tv_sec = wall->tv_sec; fsyscall_gtod_data.wall_time.tv_nsec = wall->tv_nsec; - fsyscall_gtod_data.monotonic_time.tv_sec = wall_to_monotonic.tv_sec + fsyscall_gtod_data.monotonic_time.tv_sec = wtm->tv_sec + wall->tv_sec; - fsyscall_gtod_data.monotonic_time.tv_nsec = wall_to_monotonic.tv_nsec + fsyscall_gtod_data.monotonic_time.tv_nsec = wtm->tv_nsec + wall->tv_nsec; /* normalize */ diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 0711d60f40b0..e215f76bba1c 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -849,8 +849,8 @@ static cycle_t timebase_read(struct clocksource *cs) return (cycle_t)get_tb(); } -void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, - u32 mult) +void update_vsyscall(struct timespec *wall_time, struct timespec *wtm, + struct clocksource *clock, u32 mult) { u64 new_tb_to_xs, new_stamp_xsec; @@ -882,8 +882,8 @@ void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, vdso_data->tb_orig_stamp = clock->cycle_last; vdso_data->stamp_xsec = new_stamp_xsec; vdso_data->tb_to_xs = new_tb_to_xs; - vdso_data->wtom_clock_sec = wall_to_monotonic.tv_sec; - vdso_data->wtom_clock_nsec = wall_to_monotonic.tv_nsec; + vdso_data->wtom_clock_sec = wtm->tv_sec; + vdso_data->wtom_clock_nsec = wtm->tv_nsec; vdso_data->stamp_xtime = *wall_time; smp_wmb(); ++(vdso_data->tb_update_count); diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index a2163c95eb98..aeb30c6f279c 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -207,8 +207,8 @@ struct clocksource * __init clocksource_default_clock(void) return &clocksource_tod; } -void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, - u32 mult) +void update_vsyscall(struct timespec *wall_time, struct timespec *wtm, + struct clocksource *clock, u32 mult) { if (clock != &clocksource_tod) return; @@ -219,8 +219,8 @@ void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, vdso_data->xtime_tod_stamp = clock->cycle_last; vdso_data->xtime_clock_sec = wall_time->tv_sec; vdso_data->xtime_clock_nsec = wall_time->tv_nsec; - vdso_data->wtom_clock_sec = wall_to_monotonic.tv_sec; - vdso_data->wtom_clock_nsec = wall_to_monotonic.tv_nsec; + vdso_data->wtom_clock_sec = wtm->tv_sec; + vdso_data->wtom_clock_nsec = wtm->tv_nsec; vdso_data->ntp_mult = mult; smp_wmb(); ++vdso_data->tb_update_count; diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c index dce0c3c5a787..dcbb28c4b694 100644 --- a/arch/x86/kernel/vsyscall_64.c +++ b/arch/x86/kernel/vsyscall_64.c @@ -73,8 +73,8 @@ void update_vsyscall_tz(void) write_sequnlock_irqrestore(&vsyscall_gtod_data.lock, flags); } -void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, - u32 mult) +void update_vsyscall(struct timespec *wall_time, struct timespec *wtm, + struct clocksource *clock, u32 mult) { unsigned long flags; @@ -87,7 +87,7 @@ void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, vsyscall_gtod_data.clock.shift = clock->shift; vsyscall_gtod_data.wall_time_sec = wall_time->tv_sec; vsyscall_gtod_data.wall_time_nsec = wall_time->tv_nsec; - vsyscall_gtod_data.wall_to_monotonic = wall_to_monotonic; + vsyscall_gtod_data.wall_to_monotonic = *wtm; vsyscall_gtod_data.wall_time_coarse = __current_kernel_time(); write_sequnlock_irqrestore(&vsyscall_gtod_data.lock, flags); } diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h index 5ea3c60c160c..21677d99a161 100644 --- a/include/linux/clocksource.h +++ b/include/linux/clocksource.h @@ -313,11 +313,13 @@ clocksource_calc_mult_shift(struct clocksource *cs, u32 freq, u32 minsec) #ifdef CONFIG_GENERIC_TIME_VSYSCALL extern void -update_vsyscall(struct timespec *ts, struct clocksource *c, u32 mult); +update_vsyscall(struct timespec *ts, struct timespec *wtm, + struct clocksource *c, u32 mult); extern void update_vsyscall_tz(void); #else static inline void -update_vsyscall(struct timespec *ts, struct clocksource *c, u32 mult) +update_vsyscall(struct timespec *ts, struct timespec *wtm, + struct clocksource *c, u32 mult) { } diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 73edd4074b50..b15c3acafd5a 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -170,7 +170,8 @@ void timekeeping_leap_insert(int leapsecond) { xtime.tv_sec += leapsecond; wall_to_monotonic.tv_sec -= leapsecond; - update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult); + update_vsyscall(&xtime, &wall_to_monotonic, timekeeper.clock, + timekeeper.mult); } /** @@ -326,7 +327,8 @@ int do_settimeofday(struct timespec *tv) timekeeper.ntp_error = 0; ntp_clear(); - update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult); + update_vsyscall(&xtime, &wall_to_monotonic, timekeeper.clock, + timekeeper.mult); write_sequnlock_irqrestore(&xtime_lock, flags); @@ -809,7 +811,8 @@ void update_wall_time(void) } /* check to see if there is a new clocksource to use */ - update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult); + update_vsyscall(&xtime, &wall_to_monotonic, timekeeper.clock, + timekeeper.mult); } /** -- cgit v1.2.3 From 0e469db8f70c2645acdc90981c0480a3e19d5e68 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sun, 20 Jun 2010 19:03:08 +0000 Subject: powerpc: Rework VDSO gettimeofday to prevent time going backwards Currently it is possible for userspace to see the result of gettimeofday() going backwards by 1 microsecond, assuming that userspace is using the gettimeofday() in the VDSO. The VDSO gettimeofday() algorithm computes the time in "xsecs", which are units of 2^-20 seconds, or approximately 0.954 microseconds, using the algorithm now = (timebase - tb_orig_stamp) * tb_to_xs + stamp_xsec and then converts the time in xsecs to seconds and microseconds. The kernel updates the tb_orig_stamp and stamp_xsec values every tick in update_vsyscall(). If the length of the tick is not an integer number of xsecs, then some precision is lost in converting the current time to xsecs. For example, with CONFIG_HZ=1000, the tick is 1ms long, which is 1048.576 xsecs. That means that stamp_xsec will advance by either 1048 or 1049 on each tick. With the right conditions, it is possible for userspace to get (timebase - tb_orig_stamp) * tb_to_xs being 1049 if the kernel is slightly late in updating the vdso_datapage, and then for stamp_xsec to advance by 1048 when the kernel does update it, and for userspace to then see (timebase - tb_orig_stamp) * tb_to_xs being zero due to integer truncation. The result is that time appears to go backwards by 1 microsecond. To fix this we change the VDSO gettimeofday to use a new field in the VDSO datapage which stores the nanoseconds part of the time as a fractional number of seconds in a 0.32 binary fraction format. (Or put another way, as a 32-bit number in units of 0.23283 ns.) This is convenient because we can use the mulhwu instruction to convert it to either microseconds or nanoseconds. Since it turns out that computing the time of day using this new field is simpler than either using stamp_xsec (as gettimeofday does) or stamp_xtime.tv_nsec (as clock_gettime does), this converts both gettimeofday and clock_gettime to use the new field. The existing __do_get_tspec function is converted to use the new field and take a parameter in r7 that indicates the desired resolution, 1,000,000 for microseconds or 1,000,000,000 for nanoseconds. The __do_get_xsec function is then unused and is deleted. The new algorithm is now = ((timebase - tb_orig_stamp) << 12) * tb_to_xs + (stamp_xtime_seconds << 32) + stamp_sec_fraction with 'now' in units of 2^-32 seconds. That is then converted to seconds and either microseconds or nanoseconds with seconds = now >> 32 partseconds = ((now & 0xffffffff) * resolution) >> 32 The 32-bit VDSO code also makes a further simplification: it ignores the bottom 32 bits of the tb_to_xs value, which is a 0.64 format binary fraction. Doing so gets rid of 4 multiply instructions. Assuming a timebase frequency of 1GHz or less and an update interval of no more than 10ms, the upper 32 bits of tb_to_xs will be at least 4503599, so the error from ignoring the low 32 bits will be at most 2.2ns, which is more than an order of magnitude less than the time taken to do gettimeofday or clock_gettime on our fastest processors, so there is no possibility of seeing inconsistent values due to this. This also moves update_gtod() down next to its only caller, and makes update_vsyscall use the time passed in via the wall_time argument rather than accessing xtime directly. At present, wall_time always points to xtime, but that could change in future. Signed-off-by: Paul Mackerras Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/vdso_datapage.h | 2 + arch/powerpc/kernel/asm-offsets.c | 1 + arch/powerpc/kernel/time.c | 61 +++++----- arch/powerpc/kernel/vdso32/gettimeofday.S | 184 +++++++----------------------- arch/powerpc/kernel/vdso64/gettimeofday.S | 88 ++++---------- 5 files changed, 99 insertions(+), 237 deletions(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/powerpc/include/asm/vdso_datapage.h b/arch/powerpc/include/asm/vdso_datapage.h index 13c2c283e178..08679c5319b8 100644 --- a/arch/powerpc/include/asm/vdso_datapage.h +++ b/arch/powerpc/include/asm/vdso_datapage.h @@ -85,6 +85,7 @@ struct vdso_data { __s32 wtom_clock_sec; /* Wall to monotonic clock */ __s32 wtom_clock_nsec; struct timespec stamp_xtime; /* xtime as at tb_orig_stamp */ + __u32 stamp_sec_fraction; /* fractional seconds of stamp_xtime */ __u32 syscall_map_64[SYSCALL_MAP_SIZE]; /* map of syscalls */ __u32 syscall_map_32[SYSCALL_MAP_SIZE]; /* map of syscalls */ }; @@ -105,6 +106,7 @@ struct vdso_data { __s32 wtom_clock_sec; /* Wall to monotonic clock */ __s32 wtom_clock_nsec; struct timespec stamp_xtime; /* xtime as at tb_orig_stamp */ + __u32 stamp_sec_fraction; /* fractional seconds of stamp_xtime */ __u32 syscall_map_32[SYSCALL_MAP_SIZE]; /* map of syscalls */ __u32 dcache_block_size; /* L1 d-cache block size */ __u32 icache_block_size; /* L1 i-cache block size */ diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c index 496cc5b3984f..acbbac6aaa22 100644 --- a/arch/powerpc/kernel/asm-offsets.c +++ b/arch/powerpc/kernel/asm-offsets.c @@ -342,6 +342,7 @@ int main(void) DEFINE(WTOM_CLOCK_SEC, offsetof(struct vdso_data, wtom_clock_sec)); DEFINE(WTOM_CLOCK_NSEC, offsetof(struct vdso_data, wtom_clock_nsec)); DEFINE(STAMP_XTIME, offsetof(struct vdso_data, stamp_xtime)); + DEFINE(STAMP_SEC_FRAC, offsetof(struct vdso_data, stamp_sec_fraction)); DEFINE(CFG_ICACHE_BLOCKSZ, offsetof(struct vdso_data, icache_block_size)); DEFINE(CFG_DCACHE_BLOCKSZ, offsetof(struct vdso_data, dcache_block_size)); DEFINE(CFG_ICACHE_LOGBLOCKSZ, offsetof(struct vdso_data, icache_log_block_size)); diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 0441bbdadbd1..5adebaf47f13 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -423,30 +423,6 @@ void udelay(unsigned long usecs) } EXPORT_SYMBOL(udelay); -static inline void update_gtod(u64 new_tb_stamp, u64 new_stamp_xsec, - u64 new_tb_to_xs) -{ - /* - * tb_update_count is used to allow the userspace gettimeofday code - * to assure itself that it sees a consistent view of the tb_to_xs and - * stamp_xsec variables. It reads the tb_update_count, then reads - * tb_to_xs and stamp_xsec and then reads tb_update_count again. If - * the two values of tb_update_count match and are even then the - * tb_to_xs and stamp_xsec values are consistent. If not, then it - * loops back and reads them again until this criteria is met. - * We expect the caller to have done the first increment of - * vdso_data->tb_update_count already. - */ - vdso_data->tb_orig_stamp = new_tb_stamp; - vdso_data->stamp_xsec = new_stamp_xsec; - vdso_data->tb_to_xs = new_tb_to_xs; - vdso_data->wtom_clock_sec = wall_to_monotonic.tv_sec; - vdso_data->wtom_clock_nsec = wall_to_monotonic.tv_nsec; - vdso_data->stamp_xtime = xtime; - smp_wmb(); - ++(vdso_data->tb_update_count); -} - #ifdef CONFIG_SMP unsigned long profile_pc(struct pt_regs *regs) { @@ -873,10 +849,37 @@ static cycle_t timebase_read(struct clocksource *cs) return (cycle_t)get_tb(); } +static inline void update_gtod(u64 new_tb_stamp, u64 new_stamp_xsec, + u64 new_tb_to_xs, struct timespec *now, + u32 frac_sec) +{ + /* + * tb_update_count is used to allow the userspace gettimeofday code + * to assure itself that it sees a consistent view of the tb_to_xs and + * stamp_xsec variables. It reads the tb_update_count, then reads + * tb_to_xs and stamp_xsec and then reads tb_update_count again. If + * the two values of tb_update_count match and are even then the + * tb_to_xs and stamp_xsec values are consistent. If not, then it + * loops back and reads them again until this criteria is met. + * We expect the caller to have done the first increment of + * vdso_data->tb_update_count already. + */ + vdso_data->tb_orig_stamp = new_tb_stamp; + vdso_data->stamp_xsec = new_stamp_xsec; + vdso_data->tb_to_xs = new_tb_to_xs; + vdso_data->wtom_clock_sec = wall_to_monotonic.tv_sec; + vdso_data->wtom_clock_nsec = wall_to_monotonic.tv_nsec; + vdso_data->stamp_xtime = *now; + vdso_data->stamp_sec_fraction = frac_sec; + smp_wmb(); + ++(vdso_data->tb_update_count); +} + void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, u32 mult) { u64 t2x, stamp_xsec; + u32 frac_sec; if (clock != &clocksource_timebase) return; @@ -888,10 +891,14 @@ void update_vsyscall(struct timespec *wall_time, struct clocksource *clock, /* XXX this assumes clock->shift == 22 */ /* 4611686018 ~= 2^(20+64-22) / 1e9 */ t2x = (u64) mult * 4611686018ULL; - stamp_xsec = (u64) xtime.tv_nsec * XSEC_PER_SEC; + stamp_xsec = (u64) wall_time->tv_nsec * XSEC_PER_SEC; do_div(stamp_xsec, 1000000000); - stamp_xsec += (u64) xtime.tv_sec * XSEC_PER_SEC; - update_gtod(clock->cycle_last, stamp_xsec, t2x); + stamp_xsec += (u64) wall_time->tv_sec * XSEC_PER_SEC; + + BUG_ON(wall_time->tv_nsec >= NSEC_PER_SEC); + /* this is tv_nsec / 1e9 as a 0.32 fraction */ + frac_sec = ((u64) wall_time->tv_nsec * 18446744073ULL) >> 32; + update_gtod(clock->cycle_last, stamp_xsec, t2x, wall_time, frac_sec); } void update_vsyscall_tz(void) diff --git a/arch/powerpc/kernel/vdso32/gettimeofday.S b/arch/powerpc/kernel/vdso32/gettimeofday.S index ee038d4bf252..4ee09ee2e836 100644 --- a/arch/powerpc/kernel/vdso32/gettimeofday.S +++ b/arch/powerpc/kernel/vdso32/gettimeofday.S @@ -19,8 +19,10 @@ /* Offset for the low 32-bit part of a field of long type */ #ifdef CONFIG_PPC64 #define LOPART 4 +#define TSPEC_TV_SEC TSPC64_TV_SEC+LOPART #else #define LOPART 0 +#define TSPEC_TV_SEC TSPC32_TV_SEC #endif .text @@ -41,23 +43,11 @@ V_FUNCTION_BEGIN(__kernel_gettimeofday) mr r9, r3 /* datapage ptr in r9 */ cmplwi r10,0 /* check if tv is NULL */ beq 3f - bl __do_get_xsec@local /* get xsec from tb & kernel */ - bne- 2f /* out of line -> do syscall */ - - /* seconds are xsec >> 20 */ - rlwinm r5,r4,12,20,31 - rlwimi r5,r3,12,0,19 - stw r5,TVAL32_TV_SEC(r10) - - /* get remaining xsec and convert to usec. we scale - * up remaining xsec by 12 bits and get the top 32 bits - * of the multiplication - */ - rlwinm r5,r4,12,0,19 - lis r6,1000000@h - ori r6,r6,1000000@l - mulhwu r5,r5,r6 - stw r5,TVAL32_TV_USEC(r10) + lis r7,1000000@ha /* load up USEC_PER_SEC */ + addi r7,r7,1000000@l /* so we get microseconds in r4 */ + bl __do_get_tspec@local /* get sec/usec from tb & kernel */ + stw r3,TVAL32_TV_SEC(r10) + stw r4,TVAL32_TV_USEC(r10) 3: cmplwi r11,0 /* check if tz is NULL */ beq 1f @@ -70,14 +60,6 @@ V_FUNCTION_BEGIN(__kernel_gettimeofday) crclr cr0*4+so li r3,0 blr - -2: - mtlr r12 - mr r3,r10 - mr r4,r11 - li r0,__NR_gettimeofday - sc - blr .cfi_endproc V_FUNCTION_END(__kernel_gettimeofday) @@ -100,7 +82,8 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime) mr r11,r4 /* r11 saves tp */ bl __get_datapage@local /* get data page */ mr r9,r3 /* datapage ptr in r9 */ - + lis r7,NSEC_PER_SEC@h /* want nanoseconds */ + ori r7,r7,NSEC_PER_SEC@l 50: bl __do_get_tspec@local /* get sec/nsec from tb & kernel */ bne cr1,80f /* not monotonic -> all done */ @@ -198,83 +181,12 @@ V_FUNCTION_END(__kernel_clock_getres) /* - * This is the core of gettimeofday() & friends, it returns the xsec - * value in r3 & r4 and expects the datapage ptr (non clobbered) - * in r9. clobbers r0,r4,r5,r6,r7,r8. - * When returning, r8 contains the counter value that can be reused - * by the monotonic clock implementation - */ -__do_get_xsec: - .cfi_startproc - /* Check for update count & load values. We use the low - * order 32 bits of the update count - */ -1: lwz r8,(CFG_TB_UPDATE_COUNT+LOPART)(r9) - andi. r0,r8,1 /* pending update ? loop */ - bne- 1b - xor r0,r8,r8 /* create dependency */ - add r9,r9,r0 - - /* Load orig stamp (offset to TB) */ - lwz r5,CFG_TB_ORIG_STAMP(r9) - lwz r6,(CFG_TB_ORIG_STAMP+4)(r9) - - /* Get a stable TB value */ -2: mftbu r3 - mftbl r4 - mftbu r0 - cmpl cr0,r3,r0 - bne- 2b - - /* Substract tb orig stamp. If the high part is non-zero, we jump to - * the slow path which call the syscall. - * If it's ok, then we have our 32 bits tb_ticks value in r7 - */ - subfc r7,r6,r4 - subfe. r0,r5,r3 - bne- 3f - - /* Load scale factor & do multiplication */ - lwz r5,CFG_TB_TO_XS(r9) /* load values */ - lwz r6,(CFG_TB_TO_XS+4)(r9) - mulhwu r4,r7,r5 - mulhwu r6,r7,r6 - mullw r0,r7,r5 - addc r6,r6,r0 - - /* At this point, we have the scaled xsec value in r4 + XER:CA - * we load & add the stamp since epoch - */ - lwz r5,CFG_STAMP_XSEC(r9) - lwz r6,(CFG_STAMP_XSEC+4)(r9) - adde r4,r4,r6 - addze r3,r5 - - /* We now have our result in r3,r4. We create a fake dependency - * on that result and re-check the counter - */ - or r6,r4,r3 - xor r0,r6,r6 - add r9,r9,r0 - lwz r0,(CFG_TB_UPDATE_COUNT+LOPART)(r9) - cmpl cr0,r8,r0 /* check if updated */ - bne- 1b - - /* Warning ! The caller expects CR:EQ to be set to indicate a - * successful calculation (so it won't fallback to the syscall - * method). We have overriden that CR bit in the counter check, - * but fortunately, the loop exit condition _is_ CR:EQ set, so - * we can exit safely here. If you change this code, be careful - * of that side effect. - */ -3: blr - .cfi_endproc - -/* - * This is the core of clock_gettime(), it returns the current - * time in seconds and nanoseconds in r3 and r4. + * This is the core of clock_gettime() and gettimeofday(), + * it returns the current time in r3 (seconds) and r4. + * On entry, r7 gives the resolution of r4, either USEC_PER_SEC + * or NSEC_PER_SEC, giving r4 in microseconds or nanoseconds. * It expects the datapage ptr in r9 and doesn't clobber it. - * It clobbers r0, r5, r6, r10 and returns NSEC_PER_SEC in r7. + * It clobbers r0, r5 and r6. * On return, r8 contains the counter value that can be reused. * This clobbers cr0 but not any other cr field. */ @@ -297,70 +209,58 @@ __do_get_tspec: 2: mftbu r3 mftbl r4 mftbu r0 - cmpl cr0,r3,r0 + cmplw cr0,r3,r0 bne- 2b /* Subtract tb orig stamp and shift left 12 bits. */ - subfc r7,r6,r4 + subfc r4,r6,r4 subfe r0,r5,r3 slwi r0,r0,12 - rlwimi. r0,r7,12,20,31 - slwi r7,r7,12 + rlwimi. r0,r4,12,20,31 + slwi r4,r4,12 - /* Load scale factor & do multiplication */ + /* + * Load scale factor & do multiplication. + * We only use the high 32 bits of the tb_to_xs value. + * Even with a 1GHz timebase clock, the high 32 bits of + * tb_to_xs will be at least 4 million, so the error from + * ignoring the low 32 bits will be no more than 0.25ppm. + * The error will just make the clock run very very slightly + * slow until the next time the kernel updates the VDSO data, + * at which point the clock will catch up to the kernel's value, + * so there is no long-term error accumulation. + */ lwz r5,CFG_TB_TO_XS(r9) /* load values */ - lwz r6,(CFG_TB_TO_XS+4)(r9) - mulhwu r3,r7,r6 - mullw r10,r7,r5 - mulhwu r4,r7,r5 - addc r10,r3,r10 + mulhwu r4,r4,r5 li r3,0 beq+ 4f /* skip high part computation if 0 */ mulhwu r3,r0,r5 - mullw r7,r0,r5 - mulhwu r5,r0,r6 - mullw r6,r0,r6 - adde r4,r4,r7 - addze r3,r3 + mullw r5,r0,r5 addc r4,r4,r5 addze r3,r3 - addc r10,r10,r6 - -4: addze r4,r4 /* add in carry */ - lis r7,NSEC_PER_SEC@h - ori r7,r7,NSEC_PER_SEC@l - mulhwu r4,r4,r7 /* convert to nanoseconds */ - - /* At this point, we have seconds & nanoseconds since the xtime - * stamp in r3+CA and r4. Load & add the xtime stamp. +4: + /* At this point, we have seconds since the xtime stamp + * as a 32.32 fixed-point number in r3 and r4. + * Load & add the xtime stamp. */ -#ifdef CONFIG_PPC64 - lwz r5,STAMP_XTIME+TSPC64_TV_SEC+LOPART(r9) - lwz r6,STAMP_XTIME+TSPC64_TV_NSEC+LOPART(r9) -#else - lwz r5,STAMP_XTIME+TSPC32_TV_SEC(r9) - lwz r6,STAMP_XTIME+TSPC32_TV_NSEC(r9) -#endif - add r4,r4,r6 + lwz r5,STAMP_XTIME+TSPEC_TV_SEC(r9) + lwz r6,STAMP_SEC_FRAC(r9) + addc r4,r4,r6 adde r3,r3,r5 - /* We now have our result in r3,r4. We create a fake dependency - * on that result and re-check the counter + /* We create a fake dependency on the result in r3/r4 + * and re-check the counter */ or r6,r4,r3 xor r0,r6,r6 add r9,r9,r0 lwz r0,(CFG_TB_UPDATE_COUNT+LOPART)(r9) - cmpl cr0,r8,r0 /* check if updated */ + cmplw cr0,r8,r0 /* check if updated */ bne- 1b - /* check for nanosecond overflow and adjust if necessary */ - cmpw r4,r7 - bltlr /* all done if no overflow */ - subf r4,r7,r4 /* adjust if overflow */ - addi r3,r3,1 + mulhwu r4,r4,r7 /* convert to micro or nanoseconds */ blr .cfi_endproc diff --git a/arch/powerpc/kernel/vdso64/gettimeofday.S b/arch/powerpc/kernel/vdso64/gettimeofday.S index 262cd5857a56..e97a9a0dc4ac 100644 --- a/arch/powerpc/kernel/vdso64/gettimeofday.S +++ b/arch/powerpc/kernel/vdso64/gettimeofday.S @@ -33,18 +33,11 @@ V_FUNCTION_BEGIN(__kernel_gettimeofday) bl V_LOCAL_FUNC(__get_datapage) /* get data page */ cmpldi r11,0 /* check if tv is NULL */ beq 2f - bl V_LOCAL_FUNC(__do_get_xsec) /* get xsec from tb & kernel */ - lis r7,15 /* r7 = 1000000 = USEC_PER_SEC */ - ori r7,r7,16960 - rldicl r5,r4,44,20 /* r5 = sec = xsec / XSEC_PER_SEC */ - rldicr r6,r5,20,43 /* r6 = sec * XSEC_PER_SEC */ - std r5,TVAL64_TV_SEC(r11) /* store sec in tv */ - subf r0,r6,r4 /* r0 = xsec = (xsec - r6) */ - mulld r0,r0,r7 /* usec = (xsec * USEC_PER_SEC) / - * XSEC_PER_SEC - */ - rldicl r0,r0,44,20 - std r0,TVAL64_TV_USEC(r11) /* store usec in tv */ + lis r7,1000000@ha /* load up USEC_PER_SEC */ + addi r7,r7,1000000@l + bl V_LOCAL_FUNC(__do_get_tspec) /* get sec/us from tb & kernel */ + std r4,TVAL64_TV_SEC(r11) /* store sec in tv */ + std r5,TVAL64_TV_USEC(r11) /* store usec in tv */ 2: cmpldi r10,0 /* check if tz is NULL */ beq 1f lwz r4,CFG_TZ_MINUTEWEST(r3)/* fill tz */ @@ -77,6 +70,8 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime) .cfi_register lr,r12 mr r11,r4 /* r11 saves tp */ bl V_LOCAL_FUNC(__get_datapage) /* get data page */ + lis r7,NSEC_PER_SEC@h /* want nanoseconds */ + ori r7,r7,NSEC_PER_SEC@l 50: bl V_LOCAL_FUNC(__do_get_tspec) /* get time from tb & kernel */ bne cr1,80f /* if not monotonic, all done */ @@ -171,49 +166,12 @@ V_FUNCTION_END(__kernel_clock_getres) /* - * This is the core of gettimeofday(), it returns the xsec - * value in r4 and expects the datapage ptr (non clobbered) - * in r3. clobbers r0,r4,r5,r6,r7,r8 - * When returning, r8 contains the counter value that can be reused - */ -V_FUNCTION_BEGIN(__do_get_xsec) - .cfi_startproc - /* check for update count & load values */ -1: ld r8,CFG_TB_UPDATE_COUNT(r3) - andi. r0,r8,1 /* pending update ? loop */ - bne- 1b - xor r0,r8,r8 /* create dependency */ - add r3,r3,r0 - - /* Get TB & offset it. We use the MFTB macro which will generate - * workaround code for Cell. - */ - MFTB(r7) - ld r9,CFG_TB_ORIG_STAMP(r3) - subf r7,r9,r7 - - /* Scale result */ - ld r5,CFG_TB_TO_XS(r3) - mulhdu r7,r7,r5 - - /* Add stamp since epoch */ - ld r6,CFG_STAMP_XSEC(r3) - add r4,r6,r7 - - xor r0,r4,r4 - add r3,r3,r0 - ld r0,CFG_TB_UPDATE_COUNT(r3) - cmpld cr0,r0,r8 /* check if updated */ - bne- 1b - blr - .cfi_endproc -V_FUNCTION_END(__do_get_xsec) - -/* - * This is the core of clock_gettime(), it returns the current - * time in seconds and nanoseconds in r4 and r5. + * This is the core of clock_gettime() and gettimeofday(), + * it returns the current time in r4 (seconds) and r5. + * On entry, r7 gives the resolution of r5, either USEC_PER_SEC + * or NSEC_PER_SEC, giving r5 in microseconds or nanoseconds. * It expects the datapage ptr in r3 and doesn't clobber it. - * It clobbers r0 and r6 and returns NSEC_PER_SEC in r7. + * It clobbers r0, r6 and r9. * On return, r8 contains the counter value that can be reused. * This clobbers cr0 but not any other cr field. */ @@ -229,18 +187,18 @@ V_FUNCTION_BEGIN(__do_get_tspec) /* Get TB & offset it. We use the MFTB macro which will generate * workaround code for Cell. */ - MFTB(r7) + MFTB(r6) ld r9,CFG_TB_ORIG_STAMP(r3) - subf r7,r9,r7 + subf r6,r9,r6 /* Scale result */ ld r5,CFG_TB_TO_XS(r3) - sldi r7,r7,12 /* compute time since stamp_xtime */ - mulhdu r6,r7,r5 /* in units of 2^-32 seconds */ + sldi r6,r6,12 /* compute time since stamp_xtime */ + mulhdu r6,r6,r5 /* in units of 2^-32 seconds */ /* Add stamp since epoch */ ld r4,STAMP_XTIME+TSPC64_TV_SEC(r3) - ld r5,STAMP_XTIME+TSPC64_TV_NSEC(r3) + lwz r5,STAMP_SEC_FRAC(r3) or r0,r4,r5 or r0,r0,r6 xor r0,r0,r0 @@ -250,17 +208,11 @@ V_FUNCTION_BEGIN(__do_get_tspec) bne- 1b /* reload if so */ /* convert to seconds & nanoseconds and add to stamp */ - lis r7,NSEC_PER_SEC@h - ori r7,r7,NSEC_PER_SEC@l - mulhwu r0,r6,r7 /* compute nanoseconds and */ + add r6,r6,r5 /* add on fractional seconds of xtime */ + mulhwu r5,r6,r7 /* compute micro or nanoseconds and */ srdi r6,r6,32 /* seconds since stamp_xtime */ - clrldi r0,r0,32 - add r5,r5,r0 /* add nanoseconds together */ - cmpd r5,r7 /* overflow? */ + clrldi r5,r5,32 add r4,r4,r6 - bltlr /* all done if no overflow */ - subf r5,r7,r5 /* if overflow, adjust */ - addi r4,r4,1 blr .cfi_endproc V_FUNCTION_END(__do_get_tspec) -- cgit v1.2.3 From d75d68cfef4936ddf38d2694ae2f7d1f7c45db05 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sun, 20 Jun 2010 19:04:14 +0000 Subject: powerpc: Clean up obsolete code relating to decrementer and timebase Since the decrementer and timekeeping code was moved over to using the generic clockevents and timekeeping infrastructure, several variables and functions have been obsolete and effectively unused. This deletes them. In particular, wakeup_decrementer() is no longer needed since the generic code reprograms the decrementer as part of the process of resuming the timekeeping code, which happens during sysdev resume. Thus the wakeup_decrementer calls in the suspend_enter methods for 52xx platforms have been removed. The call in the powermac cpu frequency change code has been replaced by set_dec(1), which will cause a timer interrupt as soon as interrupts are enabled, and the generic code will then reprogram the decrementer with the correct value. This also simplifies the generic_suspend_en/disable_irqs functions and makes them static since they are not referenced outside time.c. The preempt_enable/disable calls are removed because the generic code has disabled all but the boot cpu at the point where these functions are called, so we can't be moved to another cpu. Signed-off-by: Paul Mackerras Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/machdep.h | 3 - arch/powerpc/include/asm/time.h | 7 -- arch/powerpc/kernel/smp.c | 2 - arch/powerpc/kernel/time.c | 136 +-------------------------- arch/powerpc/platforms/52xx/lite5200_pm.c | 3 - arch/powerpc/platforms/52xx/mpc52xx_pm.c | 3 - arch/powerpc/platforms/powermac/cpufreq_32.c | 8 +- 7 files changed, 9 insertions(+), 153 deletions(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/powerpc/include/asm/machdep.h b/arch/powerpc/include/asm/machdep.h index 9f0fc9e6ce0d..7716e68f7b18 100644 --- a/arch/powerpc/include/asm/machdep.h +++ b/arch/powerpc/include/asm/machdep.h @@ -366,8 +366,5 @@ static inline void log_error(char *buf, unsigned int err_type, int fatal) #define machine_late_initcall(mach,fn) __define_machine_initcall(mach,"7",fn,7) #define machine_late_initcall_sync(mach,fn) __define_machine_initcall(mach,"7s",fn,7s) -void generic_suspend_disable_irqs(void); -void generic_suspend_enable_irqs(void); - #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_MACHDEP_H */ diff --git a/arch/powerpc/include/asm/time.h b/arch/powerpc/include/asm/time.h index 27ccb764fdab..dc779dfcf258 100644 --- a/arch/powerpc/include/asm/time.h +++ b/arch/powerpc/include/asm/time.h @@ -28,16 +28,12 @@ extern unsigned long tb_ticks_per_jiffy; extern unsigned long tb_ticks_per_usec; extern unsigned long tb_ticks_per_sec; -extern u64 tb_to_xs; -extern unsigned tb_to_us; struct rtc_time; extern void to_tm(int tim, struct rtc_time * tm); extern void GregorianDay(struct rtc_time *tm); -extern time_t last_rtc_update; extern void generic_calibrate_decr(void); -extern void wakeup_decrementer(void); extern void snapshot_timebase(void); extern void set_dec_cpu6(unsigned int val); @@ -204,9 +200,6 @@ static inline unsigned long tb_ticks_since(unsigned long tstamp) extern u64 mulhdu(u64, u64); #endif -extern void smp_space_timers(unsigned int); - -extern unsigned mulhwu_scale_factor(unsigned, unsigned); extern void div128_by_32(u64 dividend_high, u64 dividend_low, unsigned divisor, struct div_result *dr); diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index 5c196d1086d9..8764daad309b 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -288,8 +288,6 @@ void __init smp_prepare_cpus(unsigned int max_cpus) max_cpus = NR_CPUS; else max_cpus = 1; - - smp_space_timers(max_cpus); for_each_possible_cpu(cpu) if (cpu != boot_cpuid) diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 5adebaf47f13..ccb8759c8532 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -149,16 +149,6 @@ unsigned long tb_ticks_per_usec = 100; /* sane default */ EXPORT_SYMBOL(tb_ticks_per_usec); unsigned long tb_ticks_per_sec; EXPORT_SYMBOL(tb_ticks_per_sec); /* for cputime_t conversions */ -u64 tb_to_xs; -unsigned tb_to_us; - -#define TICKLEN_SCALE NTP_SCALE_SHIFT -static u64 last_tick_len; /* units are ns / 2^TICKLEN_SCALE */ -static u64 ticklen_to_xs; /* 0.64 fraction */ - -/* If last_tick_len corresponds to about 1/HZ seconds, then - last_tick_len << TICKLEN_SHIFT will be about 2^63. */ -#define TICKLEN_SHIFT (63 - 30 - TICKLEN_SCALE + SHIFT_HZ) DEFINE_SPINLOCK(rtc_lock); EXPORT_SYMBOL_GPL(rtc_lock); @@ -174,7 +164,6 @@ unsigned long ppc_proc_freq; EXPORT_SYMBOL(ppc_proc_freq); unsigned long ppc_tb_freq; -static u64 tb_last_jiffy __cacheline_aligned_in_smp; static DEFINE_PER_CPU(u64, last_jiffy); #ifdef CONFIG_VIRT_CPU_ACCOUNTING @@ -446,7 +435,6 @@ EXPORT_SYMBOL(profile_pc); static int __init iSeries_tb_recal(void) { - struct div_result divres; unsigned long titan, tb; /* Make sure we only run on iSeries */ @@ -477,10 +465,7 @@ static int __init iSeries_tb_recal(void) tb_ticks_per_jiffy = new_tb_ticks_per_jiffy; tb_ticks_per_sec = new_tb_ticks_per_sec; calc_cputime_factors(); - div128_by_32( XSEC_PER_SEC, 0, tb_ticks_per_sec, &divres ); - tb_to_xs = divres.result_low; vdso_data->tb_ticks_per_sec = tb_ticks_per_sec; - vdso_data->tb_to_xs = tb_to_xs; setup_cputime_one_jiffy(); } else { @@ -643,27 +628,9 @@ void timer_interrupt(struct pt_regs * regs) trace_timer_interrupt_exit(regs); } -void wakeup_decrementer(void) -{ - unsigned long ticks; - - /* - * The timebase gets saved on sleep and restored on wakeup, - * so all we need to do is to reset the decrementer. - */ - ticks = tb_ticks_since(__get_cpu_var(last_jiffy)); - if (ticks < tb_ticks_per_jiffy) - ticks = tb_ticks_per_jiffy - ticks; - else - ticks = 1; - set_dec(ticks); -} - #ifdef CONFIG_SUSPEND -void generic_suspend_disable_irqs(void) +static void generic_suspend_disable_irqs(void) { - preempt_disable(); - /* Disable the decrementer, so that it doesn't interfere * with suspending. */ @@ -673,12 +640,9 @@ void generic_suspend_disable_irqs(void) set_dec(0x7fffffff); } -void generic_suspend_enable_irqs(void) +static void generic_suspend_enable_irqs(void) { - wakeup_decrementer(); - local_irq_enable(); - preempt_enable(); } /* Overrides the weak version in kernel/power/main.c */ @@ -698,23 +662,6 @@ void arch_suspend_enable_irqs(void) } #endif -#ifdef CONFIG_SMP -void __init smp_space_timers(unsigned int max_cpus) -{ - int i; - u64 previous_tb = per_cpu(last_jiffy, boot_cpuid); - - /* make sure tb > per_cpu(last_jiffy, cpu) for all cpus always */ - previous_tb -= tb_ticks_per_jiffy; - - for_each_possible_cpu(i) { - if (i == boot_cpuid) - continue; - per_cpu(last_jiffy, i) = previous_tb; - } -} -#endif - /* * Scheduler clock - returns current time in nanosec units. * @@ -1014,15 +961,13 @@ void secondary_cpu_time_init(void) /* This function is only called on the boot processor */ void __init time_init(void) { - unsigned long flags; struct div_result res; - u64 scale, x; + u64 scale; unsigned shift; if (__USE_RTC()) { /* 601 processor: dec counts down by 128 every 128ns */ ppc_tb_freq = 1000000000; - tb_last_jiffy = get_rtcl(); } else { /* Normal PowerPC with timebase register */ ppc_md.calibrate_decr(); @@ -1030,49 +975,14 @@ void __init time_init(void) ppc_tb_freq / 1000000, ppc_tb_freq % 1000000); printk(KERN_DEBUG "time_init: processor frequency = %lu.%.6lu MHz\n", ppc_proc_freq / 1000000, ppc_proc_freq % 1000000); - tb_last_jiffy = get_tb(); } tb_ticks_per_jiffy = ppc_tb_freq / HZ; tb_ticks_per_sec = ppc_tb_freq; tb_ticks_per_usec = ppc_tb_freq / 1000000; - tb_to_us = mulhwu_scale_factor(ppc_tb_freq, 1000000); calc_cputime_factors(); setup_cputime_one_jiffy(); - /* - * Calculate the length of each tick in ns. It will not be - * exactly 1e9/HZ unless ppc_tb_freq is divisible by HZ. - * We compute 1e9 * tb_ticks_per_jiffy / ppc_tb_freq, - * rounded up. - */ - x = (u64) NSEC_PER_SEC * tb_ticks_per_jiffy + ppc_tb_freq - 1; - do_div(x, ppc_tb_freq); - tick_nsec = x; - last_tick_len = x << TICKLEN_SCALE; - - /* - * Compute ticklen_to_xs, which is a factor which gets multiplied - * by (last_tick_len << TICKLEN_SHIFT) to get a tb_to_xs value. - * It is computed as: - * ticklen_to_xs = 2^N / (tb_ticks_per_jiffy * 1e9) - * where N = 64 + 20 - TICKLEN_SCALE - TICKLEN_SHIFT - * which turns out to be N = 51 - SHIFT_HZ. - * This gives the result as a 0.64 fixed-point fraction. - * That value is reduced by an offset amounting to 1 xsec per - * 2^31 timebase ticks to avoid problems with time going backwards - * by 1 xsec when we do timer_recalc_offset due to losing the - * fractional xsec. That offset is equal to ppc_tb_freq/2^51 - * since there are 2^20 xsec in a second. - */ - div128_by_32((1ULL << 51) - ppc_tb_freq, 0, - tb_ticks_per_jiffy << SHIFT_HZ, &res); - div128_by_32(res.result_high, res.result_low, NSEC_PER_SEC, &res); - ticklen_to_xs = res.result_low; - - /* Compute tb_to_xs from tick_nsec */ - tb_to_xs = mulhdu(last_tick_len << TICKLEN_SHIFT, ticklen_to_xs); - /* * Compute scale factor for sched_clock. * The calibrate_decr() function has set tb_ticks_per_sec, @@ -1094,21 +1004,14 @@ void __init time_init(void) /* Save the current timebase to pretty up CONFIG_PRINTK_TIME */ boot_tb = get_tb_or_rtc(); - write_seqlock_irqsave(&xtime_lock, flags); - /* If platform provided a timezone (pmac), we correct the time */ if (timezone_offset) { sys_tz.tz_minuteswest = -timezone_offset / 60; sys_tz.tz_dsttime = 0; } - vdso_data->tb_orig_stamp = tb_last_jiffy; vdso_data->tb_update_count = 0; vdso_data->tb_ticks_per_sec = tb_ticks_per_sec; - vdso_data->stamp_xsec = (u64) xtime.tv_sec * XSEC_PER_SEC; - vdso_data->tb_to_xs = tb_to_xs; - - write_sequnlock_irqrestore(&xtime_lock, flags); /* Start the decrementer on CPUs that have manual control * such as BookE @@ -1202,39 +1105,6 @@ void to_tm(int tim, struct rtc_time * tm) GregorianDay(tm); } -/* Auxiliary function to compute scaling factors */ -/* Actually the choice of a timebase running at 1/4 the of the bus - * frequency giving resolution of a few tens of nanoseconds is quite nice. - * It makes this computation very precise (27-28 bits typically) which - * is optimistic considering the stability of most processor clock - * oscillators and the precision with which the timebase frequency - * is measured but does not harm. - */ -unsigned mulhwu_scale_factor(unsigned inscale, unsigned outscale) -{ - unsigned mlt=0, tmp, err; - /* No concern for performance, it's done once: use a stupid - * but safe and compact method to find the multiplier. - */ - - for (tmp = 1U<<31; tmp != 0; tmp >>= 1) { - if (mulhwu(inscale, mlt|tmp) < outscale) - mlt |= tmp; - } - - /* We might still be off by 1 for the best approximation. - * A side effect of this is that if outscale is too large - * the returned value will be zero. - * Many corner cases have been checked and seem to work, - * some might have been forgotten in the test however. - */ - - err = inscale * (mlt+1); - if (err <= inscale/2) - mlt++; - return mlt; -} - /* * Divide a 128-bit dividend by a 32-bit divisor, leaving a 128 bit * result. diff --git a/arch/powerpc/platforms/52xx/lite5200_pm.c b/arch/powerpc/platforms/52xx/lite5200_pm.c index b5c753db125e..80234e5921f5 100644 --- a/arch/powerpc/platforms/52xx/lite5200_pm.c +++ b/arch/powerpc/platforms/52xx/lite5200_pm.c @@ -216,9 +216,6 @@ static int lite5200_pm_enter(suspend_state_t state) lite5200_restore_regs(); - /* restart jiffies */ - wakeup_decrementer(); - iounmap(mbar); return 0; } diff --git a/arch/powerpc/platforms/52xx/mpc52xx_pm.c b/arch/powerpc/platforms/52xx/mpc52xx_pm.c index 76722532bd95..568cef636275 100644 --- a/arch/powerpc/platforms/52xx/mpc52xx_pm.c +++ b/arch/powerpc/platforms/52xx/mpc52xx_pm.c @@ -171,9 +171,6 @@ int mpc52xx_pm_enter(suspend_state_t state) /* restore SRAM */ memcpy(sram, saved_sram, sram_size); - /* restart jiffies */ - wakeup_decrementer(); - /* reenable interrupts in PIC */ out_be32(&intr->main_mask, intr_main_mask); diff --git a/arch/powerpc/platforms/powermac/cpufreq_32.c b/arch/powerpc/platforms/powermac/cpufreq_32.c index 1e9eba175ff0..415ca6d6b273 100644 --- a/arch/powerpc/platforms/powermac/cpufreq_32.c +++ b/arch/powerpc/platforms/powermac/cpufreq_32.c @@ -310,8 +310,12 @@ static int pmu_set_cpu_speed(int low_speed) /* Restore low level PMU operations */ pmu_unlock(); - /* Restore decrementer */ - wakeup_decrementer(); + /* + * Restore decrementer; we'll take a decrementer interrupt + * as soon as interrupts are re-enabled and the generic + * clockevents code will reprogram it with the right value. + */ + set_dec(1); /* Restore interrupts */ mpic_cpu_set_priority(pic_prio); -- cgit v1.2.3 From b0d278b7d3ae9115939ddcea93f516308cc367e2 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Tue, 10 Aug 2010 20:38:23 +0000 Subject: powerpc/perf_event: Reduce latency of calling perf_event_do_pending Commit 0fe1ac48 ("powerpc/perf_event: Fix oops due to perf_event_do_pending call") moved the call to perf_event_do_pending in timer_interrupt() down so that it was after the irq_enter() call. Unfortunately this moved it after the code that checks whether it is time for the next decrementer clock event. The result is that the call to perf_event_do_pending() won't happen until the next decrementer clock event is due. This was pointed out by Milton Miller. This fixes it by moving the check for whether it's time for the next decrementer clock event down to the point where we're about to call the event handler, after we've called perf_event_do_pending. This has the side effect that on old pre-Core99 Powermacs where we use the ppc_n_lost_interrupts mechanism to replay interrupts, a replayed interrupt will incur a little more latency since it will now do the code from the irq_enter down to the irq_exit, that it used to skip. However, these machines are now old and rare enough that this doesn't matter. To make it clear that ppc_n_lost_interrupts is only used on Powermacs, and to speed up the code slightly on non-Powermac ppc32 machines, the code that tests ppc_n_lost_interrupts is now conditional on CONFIG_PMAC as well as CONFIG_PPC32. Signed-off-by: Paul Mackerras Cc: stable@kernel.org Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/time.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index ce53dfa7130d..8533b3b83f5d 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -577,20 +577,11 @@ void timer_interrupt(struct pt_regs * regs) * some CPUs will continuue to take decrementer exceptions */ set_dec(DECREMENTER_MAX); -#ifdef CONFIG_PPC32 +#if defined(CONFIG_PPC32) && defined(CONFIG_PMAC) if (atomic_read(&ppc_n_lost_interrupts) != 0) do_IRQ(regs); #endif - now = get_tb_or_rtc(); - if (now < decrementer->next_tb) { - /* not time for this event yet */ - now = decrementer->next_tb - now; - if (now <= DECREMENTER_MAX) - set_dec((int)now); - trace_timer_interrupt_exit(regs); - return; - } old_regs = set_irq_regs(regs); irq_enter(); @@ -606,8 +597,16 @@ void timer_interrupt(struct pt_regs * regs) get_lppaca()->int_dword.fields.decr_int = 0; #endif - if (evt->event_handler) - evt->event_handler(evt); + now = get_tb_or_rtc(); + if (now >= decrementer->next_tb) { + decrementer->next_tb = ~(u64)0; + if (evt->event_handler) + evt->event_handler(evt); + } else { + now = decrementer->next_tb - now; + if (now <= DECREMENTER_MAX) + set_dec((int)now); + } #ifdef CONFIG_PPC_ISERIES if (firmware_has_feature(FW_FEATURE_ISERIES) && hvlpevent_is_pending()) -- cgit v1.2.3 From cf9efce0ce3136fa076f53e53154e98455229514 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Thu, 26 Aug 2010 19:56:43 +0000 Subject: powerpc: Account time using timebase rather than PURR Currently, when CONFIG_VIRT_CPU_ACCOUNTING is enabled, we use the PURR register for measuring the user and system time used by processes, as well as other related times such as hardirq and softirq times. This turns out to be quite confusing for users because it means that a program will often be measured as taking less time when run on a multi-threaded processor (SMT2 or SMT4 mode) than it does when run on a single-threaded processor (ST mode), even though the program takes longer to finish. The discrepancy is accounted for as stolen time, which is also confusing, particularly when there are no other partitions running. This changes the accounting to use the timebase instead, meaning that the reported user and system times are the actual number of real-time seconds that the program was executing on the processor thread, regardless of which SMT mode the processor is in. Thus a program will generally show greater user and system times when run on a multi-threaded processor than on a single-threaded processor. On pSeries systems on POWER5 or later processors, we measure the stolen time (time when this partition wasn't running) using the hypervisor dispatch trace log. We check for new entries in the log on every entry from user mode and on every transition from kernel process context to soft or hard IRQ context (i.e. when account_system_vtime() gets called). So that we can correctly distinguish time stolen from user time and time stolen from system time, without having to check the log on every exit to user mode, we store separate timestamps for exit to user mode and entry from user mode. On systems that have a SPURR (POWER6 and POWER7), we read the SPURR in account_system_vtime() (as before), and then apportion the SPURR ticks since the last time we read it between scaled user time and scaled system time according to the relative proportions of user time and system time over the same interval. This avoids having to read the SPURR on every kernel entry and exit. On systems that have PURR but not SPURR (i.e., POWER5), we do the same using the PURR rather than the SPURR. This disables the DTL user interface in /sys/debug/kernel/powerpc/dtl for now since it conflicts with the use of the dispatch trace log by the time accounting code. Signed-off-by: Paul Mackerras Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/exception-64s.h | 3 +- arch/powerpc/include/asm/lppaca.h | 19 +++ arch/powerpc/include/asm/paca.h | 10 +- arch/powerpc/include/asm/ppc_asm.h | 50 +++--- arch/powerpc/include/asm/time.h | 5 - arch/powerpc/kernel/asm-offsets.c | 8 +- arch/powerpc/kernel/entry_64.S | 18 +++ arch/powerpc/kernel/process.c | 1 - arch/powerpc/kernel/smp.c | 5 - arch/powerpc/kernel/time.c | 268 +++++++++++++++---------------- arch/powerpc/platforms/pseries/dtl.c | 24 +-- arch/powerpc/platforms/pseries/lpar.c | 21 +++ arch/powerpc/platforms/pseries/setup.c | 52 ++++++ 13 files changed, 290 insertions(+), 194 deletions(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/powerpc/include/asm/exception-64s.h b/arch/powerpc/include/asm/exception-64s.h index 57c400071995..7778d6f0c878 100644 --- a/arch/powerpc/include/asm/exception-64s.h +++ b/arch/powerpc/include/asm/exception-64s.h @@ -137,7 +137,8 @@ li r10,0; \ ld r11,exception_marker@toc(r2); \ std r10,RESULT(r1); /* clear regs->result */ \ - std r11,STACK_FRAME_OVERHEAD-16(r1); /* mark the frame */ + std r11,STACK_FRAME_OVERHEAD-16(r1); /* mark the frame */ \ + ACCOUNT_STOLEN_TIME /* * Exception vectors. diff --git a/arch/powerpc/include/asm/lppaca.h b/arch/powerpc/include/asm/lppaca.h index 6d02624b622c..cfb85ec85750 100644 --- a/arch/powerpc/include/asm/lppaca.h +++ b/arch/powerpc/include/asm/lppaca.h @@ -172,6 +172,25 @@ struct slb_shadow { extern struct slb_shadow slb_shadow[]; +/* + * Layout of entries in the hypervisor's dispatch trace log buffer. + */ +struct dtl_entry { + u8 dispatch_reason; + u8 preempt_reason; + u16 processor_id; + u32 enqueue_to_dispatch_time; + u32 ready_to_enqueue_time; + u32 waiting_to_ready_time; + u64 timebase; + u64 fault_addr; + u64 srr0; + u64 srr1; +}; + +#define DISPATCH_LOG_BYTES 4096 /* bytes per cpu */ +#define N_DISPATCH_LOG (DISPATCH_LOG_BYTES / sizeof(struct dtl_entry)) + #endif /* CONFIG_PPC_BOOK3S */ #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_LPPACA_H */ diff --git a/arch/powerpc/include/asm/paca.h b/arch/powerpc/include/asm/paca.h index 1ff6662f7faf..6af6c1613409 100644 --- a/arch/powerpc/include/asm/paca.h +++ b/arch/powerpc/include/asm/paca.h @@ -85,6 +85,8 @@ struct paca_struct { u8 kexec_state; /* set when kexec down has irqs off */ #ifdef CONFIG_PPC_STD_MMU_64 struct slb_shadow *slb_shadow_ptr; + struct dtl_entry *dispatch_log; + struct dtl_entry *dispatch_log_end; /* * Now, starting in cacheline 2, the exception save areas @@ -134,8 +136,14 @@ struct paca_struct { /* Stuff for accurate time accounting */ u64 user_time; /* accumulated usermode TB ticks */ u64 system_time; /* accumulated system TB ticks */ - u64 startpurr; /* PURR/TB value snapshot */ + u64 user_time_scaled; /* accumulated usermode SPURR ticks */ + u64 starttime; /* TB value snapshot */ + u64 starttime_user; /* TB value on exit to usermode */ u64 startspurr; /* SPURR value snapshot */ + u64 utime_sspurr; /* ->user_time when ->startspurr set */ + u64 stolen_time; /* TB ticks taken by hypervisor */ + u64 dtl_ridx; /* read index in dispatch log */ + struct dtl_entry *dtl_curr; /* pointer corresponding to dtl_ridx */ #ifdef CONFIG_KVM_BOOK3S_HANDLER /* We use this to store guest state in */ diff --git a/arch/powerpc/include/asm/ppc_asm.h b/arch/powerpc/include/asm/ppc_asm.h index 498fe09263d3..98210067c1cc 100644 --- a/arch/powerpc/include/asm/ppc_asm.h +++ b/arch/powerpc/include/asm/ppc_asm.h @@ -9,6 +9,7 @@ #include #include #include +#include #ifndef __ASSEMBLY__ #error __FILE__ should only be used in assembler files @@ -26,17 +27,13 @@ #ifndef CONFIG_VIRT_CPU_ACCOUNTING #define ACCOUNT_CPU_USER_ENTRY(ra, rb) #define ACCOUNT_CPU_USER_EXIT(ra, rb) +#define ACCOUNT_STOLEN_TIME #else #define ACCOUNT_CPU_USER_ENTRY(ra, rb) \ beq 2f; /* if from kernel mode */ \ -BEGIN_FTR_SECTION; \ - mfspr ra,SPRN_PURR; /* get processor util. reg */ \ -END_FTR_SECTION_IFSET(CPU_FTR_PURR); \ -BEGIN_FTR_SECTION; \ - MFTB(ra); /* or get TB if no PURR */ \ -END_FTR_SECTION_IFCLR(CPU_FTR_PURR); \ - ld rb,PACA_STARTPURR(r13); \ - std ra,PACA_STARTPURR(r13); \ + MFTB(ra); /* get timebase */ \ + ld rb,PACA_STARTTIME_USER(r13); \ + std ra,PACA_STARTTIME(r13); \ subf rb,rb,ra; /* subtract start value */ \ ld ra,PACA_USER_TIME(r13); \ add ra,ra,rb; /* add on to user time */ \ @@ -44,19 +41,34 @@ END_FTR_SECTION_IFCLR(CPU_FTR_PURR); \ 2: #define ACCOUNT_CPU_USER_EXIT(ra, rb) \ -BEGIN_FTR_SECTION; \ - mfspr ra,SPRN_PURR; /* get processor util. reg */ \ -END_FTR_SECTION_IFSET(CPU_FTR_PURR); \ -BEGIN_FTR_SECTION; \ - MFTB(ra); /* or get TB if no PURR */ \ -END_FTR_SECTION_IFCLR(CPU_FTR_PURR); \ - ld rb,PACA_STARTPURR(r13); \ - std ra,PACA_STARTPURR(r13); \ + MFTB(ra); /* get timebase */ \ + ld rb,PACA_STARTTIME(r13); \ + std ra,PACA_STARTTIME_USER(r13); \ subf rb,rb,ra; /* subtract start value */ \ ld ra,PACA_SYSTEM_TIME(r13); \ - add ra,ra,rb; /* add on to user time */ \ - std ra,PACA_SYSTEM_TIME(r13); -#endif + add ra,ra,rb; /* add on to system time */ \ + std ra,PACA_SYSTEM_TIME(r13) + +#ifdef CONFIG_PPC_SPLPAR +#define ACCOUNT_STOLEN_TIME \ +BEGIN_FW_FTR_SECTION; \ + beq 33f; \ + /* from user - see if there are any DTL entries to process */ \ + ld r10,PACALPPACAPTR(r13); /* get ptr to VPA */ \ + ld r11,PACA_DTL_RIDX(r13); /* get log read index */ \ + ld r10,LPPACA_DTLIDX(r10); /* get log write index */ \ + cmpd cr1,r11,r10; \ + beq+ cr1,33f; \ + bl .accumulate_stolen_time; \ +33: \ +END_FW_FTR_SECTION_IFSET(FW_FEATURE_SPLPAR) + +#else /* CONFIG_PPC_SPLPAR */ +#define ACCOUNT_STOLEN_TIME + +#endif /* CONFIG_PPC_SPLPAR */ + +#endif /* CONFIG_VIRT_CPU_ACCOUNTING */ /* * Macros for storing registers into and loading registers from diff --git a/arch/powerpc/include/asm/time.h b/arch/powerpc/include/asm/time.h index dc779dfcf258..fe6f7c2c9c68 100644 --- a/arch/powerpc/include/asm/time.h +++ b/arch/powerpc/include/asm/time.h @@ -34,7 +34,6 @@ extern void to_tm(int tim, struct rtc_time * tm); extern void GregorianDay(struct rtc_time *tm); extern void generic_calibrate_decr(void); -extern void snapshot_timebase(void); extern void set_dec_cpu6(unsigned int val); @@ -212,12 +211,8 @@ struct cpu_usage { DECLARE_PER_CPU(struct cpu_usage, cpu_usage_array); #if defined(CONFIG_VIRT_CPU_ACCOUNTING) -extern void calculate_steal_time(void); -extern void snapshot_timebases(void); #define account_process_vtime(tsk) account_process_tick(tsk, 0) #else -#define calculate_steal_time() do { } while (0) -#define snapshot_timebases() do { } while (0) #define account_process_vtime(tsk) do { } while (0) #endif diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c index 1c0607ddccc0..c63494090854 100644 --- a/arch/powerpc/kernel/asm-offsets.c +++ b/arch/powerpc/kernel/asm-offsets.c @@ -181,17 +181,19 @@ int main(void) offsetof(struct slb_shadow, save_area[SLB_NUM_BOLTED - 1].vsid)); DEFINE(SLBSHADOW_STACKESID, offsetof(struct slb_shadow, save_area[SLB_NUM_BOLTED - 1].esid)); + DEFINE(SLBSHADOW_SAVEAREA, offsetof(struct slb_shadow, save_area)); DEFINE(LPPACASRR0, offsetof(struct lppaca, saved_srr0)); DEFINE(LPPACASRR1, offsetof(struct lppaca, saved_srr1)); DEFINE(LPPACAANYINT, offsetof(struct lppaca, int_dword.any_int)); DEFINE(LPPACADECRINT, offsetof(struct lppaca, int_dword.fields.decr_int)); - DEFINE(SLBSHADOW_SAVEAREA, offsetof(struct slb_shadow, save_area)); + DEFINE(LPPACA_DTLIDX, offsetof(struct lppaca, dtl_idx)); + DEFINE(PACA_DTL_RIDX, offsetof(struct paca_struct, dtl_ridx)); #endif /* CONFIG_PPC_STD_MMU_64 */ DEFINE(PACAEMERGSP, offsetof(struct paca_struct, emergency_sp)); DEFINE(PACAHWCPUID, offsetof(struct paca_struct, hw_cpu_id)); DEFINE(PACAKEXECSTATE, offsetof(struct paca_struct, kexec_state)); - DEFINE(PACA_STARTPURR, offsetof(struct paca_struct, startpurr)); - DEFINE(PACA_STARTSPURR, offsetof(struct paca_struct, startspurr)); + DEFINE(PACA_STARTTIME, offsetof(struct paca_struct, starttime)); + DEFINE(PACA_STARTTIME_USER, offsetof(struct paca_struct, starttime_user)); DEFINE(PACA_USER_TIME, offsetof(struct paca_struct, user_time)); DEFINE(PACA_SYSTEM_TIME, offsetof(struct paca_struct, system_time)); DEFINE(PACA_TRAP_SAVE, offsetof(struct paca_struct, trap_save)); diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S index 4d5fa12ca6e8..d82878c4daa6 100644 --- a/arch/powerpc/kernel/entry_64.S +++ b/arch/powerpc/kernel/entry_64.S @@ -97,6 +97,24 @@ system_call_common: addi r9,r1,STACK_FRAME_OVERHEAD ld r11,exception_marker@toc(r2) std r11,-16(r9) /* "regshere" marker */ +#if defined(CONFIG_VIRT_CPU_ACCOUNTING) && defined(CONFIG_PPC_SPLPAR) +BEGIN_FW_FTR_SECTION + beq 33f + /* if from user, see if there are any DTL entries to process */ + ld r10,PACALPPACAPTR(r13) /* get ptr to VPA */ + ld r11,PACA_DTL_RIDX(r13) /* get log read index */ + ld r10,LPPACA_DTLIDX(r10) /* get log write index */ + cmpd cr1,r11,r10 + beq+ cr1,33f + bl .accumulate_stolen_time + REST_GPR(0,r1) + REST_4GPRS(3,r1) + REST_2GPRS(7,r1) + addi r9,r1,STACK_FRAME_OVERHEAD +33: +END_FW_FTR_SECTION_IFSET(FW_FEATURE_SPLPAR) +#endif /* CONFIG_VIRT_CPU_ACCOUNTING && CONFIG_PPC_SPLPAR */ + #ifdef CONFIG_TRACE_IRQFLAGS bl .trace_hardirqs_on REST_GPR(0,r1) diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 37bc8ff16cac..84906d3fc860 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -517,7 +517,6 @@ struct task_struct *__switch_to(struct task_struct *prev, account_system_vtime(current); account_process_vtime(current); - calculate_steal_time(); /* * We can't take a PMU exception inside _switch() since there is a diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index 9019f0f1bb5e..68034bbf2e4f 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -508,9 +508,6 @@ int __devinit start_secondary(void *unused) if (smp_ops->take_timebase) smp_ops->take_timebase(); - if (system_state > SYSTEM_BOOTING) - snapshot_timebase(); - secondary_cpu_time_init(); ipi_call_lock(); @@ -575,8 +572,6 @@ void __init smp_cpus_done(unsigned int max_cpus) free_cpumask_var(old_mask); - snapshot_timebases(); - dump_numa_cpu_topology(); } diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 8533b3b83f5d..fca20643c368 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -164,8 +164,6 @@ unsigned long ppc_proc_freq; EXPORT_SYMBOL(ppc_proc_freq); unsigned long ppc_tb_freq; -static DEFINE_PER_CPU(u64, last_jiffy); - #ifdef CONFIG_VIRT_CPU_ACCOUNTING /* * Factors for converting from cputime_t (timebase ticks) to @@ -200,62 +198,151 @@ static void calc_cputime_factors(void) } /* - * Read the PURR on systems that have it, otherwise the timebase. + * Read the SPURR on systems that have it, otherwise the PURR, + * or if that doesn't exist return the timebase value passed in. */ -static u64 read_purr(void) +static u64 read_spurr(u64 tb) { + if (cpu_has_feature(CPU_FTR_SPURR)) + return mfspr(SPRN_SPURR); if (cpu_has_feature(CPU_FTR_PURR)) return mfspr(SPRN_PURR); - return mftb(); + return tb; } +#ifdef CONFIG_PPC_SPLPAR + /* - * Read the SPURR on systems that have it, otherwise the purr + * Scan the dispatch trace log and count up the stolen time. + * Should be called with interrupts disabled. */ -static u64 read_spurr(u64 purr) +static u64 scan_dispatch_log(u64 stop_tb) { - /* - * cpus without PURR won't have a SPURR - * We already know the former when we use this, so tell gcc - */ - if (cpu_has_feature(CPU_FTR_PURR) && cpu_has_feature(CPU_FTR_SPURR)) - return mfspr(SPRN_SPURR); - return purr; + unsigned long i = local_paca->dtl_ridx; + struct dtl_entry *dtl = local_paca->dtl_curr; + struct dtl_entry *dtl_end = local_paca->dispatch_log_end; + struct lppaca *vpa = local_paca->lppaca_ptr; + u64 tb_delta; + u64 stolen = 0; + u64 dtb; + + if (i == vpa->dtl_idx) + return 0; + while (i < vpa->dtl_idx) { + dtb = dtl->timebase; + tb_delta = dtl->enqueue_to_dispatch_time + + dtl->ready_to_enqueue_time; + barrier(); + if (i + N_DISPATCH_LOG < vpa->dtl_idx) { + /* buffer has overflowed */ + i = vpa->dtl_idx - N_DISPATCH_LOG; + dtl = local_paca->dispatch_log + (i % N_DISPATCH_LOG); + continue; + } + if (dtb > stop_tb) + break; + stolen += tb_delta; + ++i; + ++dtl; + if (dtl == dtl_end) + dtl = local_paca->dispatch_log; + } + local_paca->dtl_ridx = i; + local_paca->dtl_curr = dtl; + return stolen; } +/* + * Accumulate stolen time by scanning the dispatch trace log. + * Called on entry from user mode. + */ +void accumulate_stolen_time(void) +{ + u64 sst, ust; + + sst = scan_dispatch_log(get_paca()->starttime_user); + ust = scan_dispatch_log(get_paca()->starttime); + get_paca()->system_time -= sst; + get_paca()->user_time -= ust; + get_paca()->stolen_time += ust + sst; +} + +static inline u64 calculate_stolen_time(u64 stop_tb) +{ + u64 stolen = 0; + + if (get_paca()->dtl_ridx != get_paca()->lppaca_ptr->dtl_idx) { + stolen = scan_dispatch_log(stop_tb); + get_paca()->system_time -= stolen; + } + + stolen += get_paca()->stolen_time; + get_paca()->stolen_time = 0; + return stolen; +} + +#else /* CONFIG_PPC_SPLPAR */ +static inline u64 calculate_stolen_time(u64 stop_tb) +{ + return 0; +} + +#endif /* CONFIG_PPC_SPLPAR */ + /* * Account time for a transition between system, hard irq * or soft irq state. */ void account_system_vtime(struct task_struct *tsk) { - u64 now, nowscaled, delta, deltascaled, sys_time; + u64 now, nowscaled, delta, deltascaled; unsigned long flags; + u64 stolen, udelta, sys_scaled, user_scaled; local_irq_save(flags); - now = read_purr(); + now = mftb(); nowscaled = read_spurr(now); - delta = now - get_paca()->startpurr; + get_paca()->system_time += now - get_paca()->starttime; + get_paca()->starttime = now; deltascaled = nowscaled - get_paca()->startspurr; - get_paca()->startpurr = now; get_paca()->startspurr = nowscaled; - if (!in_interrupt()) { - /* deltascaled includes both user and system time. - * Hence scale it based on the purr ratio to estimate - * the system time */ - sys_time = get_paca()->system_time; - if (get_paca()->user_time) - deltascaled = deltascaled * sys_time / - (sys_time + get_paca()->user_time); - delta += sys_time; - get_paca()->system_time = 0; + + stolen = calculate_stolen_time(now); + + delta = get_paca()->system_time; + get_paca()->system_time = 0; + udelta = get_paca()->user_time - get_paca()->utime_sspurr; + get_paca()->utime_sspurr = get_paca()->user_time; + + /* + * Because we don't read the SPURR on every kernel entry/exit, + * deltascaled includes both user and system SPURR ticks. + * Apportion these ticks to system SPURR ticks and user + * SPURR ticks in the same ratio as the system time (delta) + * and user time (udelta) values obtained from the timebase + * over the same interval. The system ticks get accounted here; + * the user ticks get saved up in paca->user_time_scaled to be + * used by account_process_tick. + */ + sys_scaled = delta; + user_scaled = udelta; + if (deltascaled != delta + udelta) { + if (udelta) { + sys_scaled = deltascaled * delta / (delta + udelta); + user_scaled = deltascaled - sys_scaled; + } else { + sys_scaled = deltascaled; + } + } + get_paca()->user_time_scaled += user_scaled; + + if (in_irq() || idle_task(smp_processor_id()) != tsk) { + account_system_time(tsk, 0, delta, sys_scaled); + if (stolen) + account_steal_time(stolen); + } else { + account_idle_time(delta + stolen); } - if (in_irq() || idle_task(smp_processor_id()) != tsk) - account_system_time(tsk, 0, delta, deltascaled); - else - account_idle_time(delta); - __get_cpu_var(cputime_last_delta) = delta; - __get_cpu_var(cputime_scaled_last_delta) = deltascaled; local_irq_restore(flags); } EXPORT_SYMBOL_GPL(account_system_vtime); @@ -265,125 +352,26 @@ EXPORT_SYMBOL_GPL(account_system_vtime); * by the exception entry and exit code to the generic process * user and system time records. * Must be called with interrupts disabled. + * Assumes that account_system_vtime() has been called recently + * (i.e. since the last entry from usermode) so that + * get_paca()->user_time_scaled is up to date. */ void account_process_tick(struct task_struct *tsk, int user_tick) { cputime_t utime, utimescaled; utime = get_paca()->user_time; + utimescaled = get_paca()->user_time_scaled; get_paca()->user_time = 0; - utimescaled = cputime_to_scaled(utime); + get_paca()->user_time_scaled = 0; + get_paca()->utime_sspurr = 0; account_user_time(tsk, utime, utimescaled); } -/* - * Stuff for accounting stolen time. - */ -struct cpu_purr_data { - int initialized; /* thread is running */ - u64 tb; /* last TB value read */ - u64 purr; /* last PURR value read */ - u64 spurr; /* last SPURR value read */ -}; - -/* - * Each entry in the cpu_purr_data array is manipulated only by its - * "owner" cpu -- usually in the timer interrupt but also occasionally - * in process context for cpu online. As long as cpus do not touch - * each others' cpu_purr_data, disabling local interrupts is - * sufficient to serialize accesses. - */ -static DEFINE_PER_CPU(struct cpu_purr_data, cpu_purr_data); - -static void snapshot_tb_and_purr(void *data) -{ - unsigned long flags; - struct cpu_purr_data *p = &__get_cpu_var(cpu_purr_data); - - local_irq_save(flags); - p->tb = get_tb_or_rtc(); - p->purr = mfspr(SPRN_PURR); - wmb(); - p->initialized = 1; - local_irq_restore(flags); -} - -/* - * Called during boot when all cpus have come up. - */ -void snapshot_timebases(void) -{ - if (!cpu_has_feature(CPU_FTR_PURR)) - return; - on_each_cpu(snapshot_tb_and_purr, NULL, 1); -} - -/* - * Must be called with interrupts disabled. - */ -void calculate_steal_time(void) -{ - u64 tb, purr; - s64 stolen; - struct cpu_purr_data *pme; - - pme = &__get_cpu_var(cpu_purr_data); - if (!pme->initialized) - return; /* !CPU_FTR_PURR or early in early boot */ - tb = mftb(); - purr = mfspr(SPRN_PURR); - stolen = (tb - pme->tb) - (purr - pme->purr); - if (stolen > 0) { - if (idle_task(smp_processor_id()) != current) - account_steal_time(stolen); - else - account_idle_time(stolen); - } - pme->tb = tb; - pme->purr = purr; -} - -#ifdef CONFIG_PPC_SPLPAR -/* - * Must be called before the cpu is added to the online map when - * a cpu is being brought up at runtime. - */ -static void snapshot_purr(void) -{ - struct cpu_purr_data *pme; - unsigned long flags; - - if (!cpu_has_feature(CPU_FTR_PURR)) - return; - local_irq_save(flags); - pme = &__get_cpu_var(cpu_purr_data); - pme->tb = mftb(); - pme->purr = mfspr(SPRN_PURR); - pme->initialized = 1; - local_irq_restore(flags); -} - -#endif /* CONFIG_PPC_SPLPAR */ - #else /* ! CONFIG_VIRT_CPU_ACCOUNTING */ #define calc_cputime_factors() -#define calculate_steal_time() do { } while (0) #endif -#if !(defined(CONFIG_VIRT_CPU_ACCOUNTING) && defined(CONFIG_PPC_SPLPAR)) -#define snapshot_purr() do { } while (0) -#endif - -/* - * Called when a cpu comes up after the system has finished booting, - * i.e. as a result of a hotplug cpu action. - */ -void snapshot_timebase(void) -{ - __get_cpu_var(last_jiffy) = get_tb_or_rtc(); - snapshot_purr(); -} - void __delay(unsigned long loops) { unsigned long start; @@ -585,8 +573,6 @@ void timer_interrupt(struct pt_regs * regs) old_regs = set_irq_regs(regs); irq_enter(); - calculate_steal_time(); - if (test_perf_event_pending()) { clear_perf_event_pending(); perf_event_do_pending(); diff --git a/arch/powerpc/platforms/pseries/dtl.c b/arch/powerpc/platforms/pseries/dtl.c index adfd5441b612..0357655db49d 100644 --- a/arch/powerpc/platforms/pseries/dtl.c +++ b/arch/powerpc/platforms/pseries/dtl.c @@ -27,27 +27,10 @@ #include #include #include +#include #include "plpar_wrappers.h" -/* - * Layout of entries in the hypervisor's DTL buffer. Although we don't - * actually access the internals of an entry (we only need to know the size), - * we might as well define it here for reference. - */ -struct dtl_entry { - u8 dispatch_reason; - u8 preempt_reason; - u16 processor_id; - u32 enqueue_to_dispatch_time; - u32 ready_to_enqueue_time; - u32 waiting_to_ready_time; - u64 timebase; - u64 fault_addr; - u64 srr0; - u64 srr1; -}; - struct dtl { struct dtl_entry *buf; struct dentry *file; @@ -237,6 +220,11 @@ static int dtl_init(void) struct dentry *event_mask_file, *buf_entries_file; int rc, i; +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + /* disable this for now */ + return -ENODEV; +#endif + if (!firmware_has_feature(FW_FEATURE_SPLPAR)) return -ENODEV; diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c index a17fe4a9059f..f129040d974c 100644 --- a/arch/powerpc/platforms/pseries/lpar.c +++ b/arch/powerpc/platforms/pseries/lpar.c @@ -248,6 +248,8 @@ void vpa_init(int cpu) int hwcpu = get_hard_smp_processor_id(cpu); unsigned long addr; long ret; + struct paca_struct *pp; + struct dtl_entry *dtl; if (cpu_has_feature(CPU_FTR_ALTIVEC)) lppaca_of(cpu).vmxregs_in_use = 1; @@ -274,6 +276,25 @@ void vpa_init(int cpu) "registration for cpu %d (hw %d) of area %lx " "returns %ld\n", cpu, hwcpu, addr, ret); } + + /* + * Register dispatch trace log, if one has been allocated. + */ + pp = &paca[cpu]; + dtl = pp->dispatch_log; + if (dtl) { + pp->dtl_ridx = 0; + pp->dtl_curr = dtl; + lppaca_of(cpu).dtl_idx = 0; + + /* hypervisor reads buffer length from this field */ + dtl->enqueue_to_dispatch_time = DISPATCH_LOG_BYTES; + ret = register_dtl(hwcpu, __pa(dtl)); + if (ret) + pr_warn("DTL registration failed for cpu %d (%ld)\n", + cpu, ret); + lppaca_of(cpu).dtl_enable_mask = 2; + } } static long pSeries_lpar_hpte_insert(unsigned long hpte_group, diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c index a6d19e3a505e..d345bfd56bbe 100644 --- a/arch/powerpc/platforms/pseries/setup.c +++ b/arch/powerpc/platforms/pseries/setup.c @@ -273,6 +273,58 @@ static struct notifier_block pci_dn_reconfig_nb = { .notifier_call = pci_dn_reconfig_notifier, }; +#ifdef CONFIG_VIRT_CPU_ACCOUNTING +/* + * Allocate space for the dispatch trace log for all possible cpus + * and register the buffers with the hypervisor. This is used for + * computing time stolen by the hypervisor. + */ +static int alloc_dispatch_logs(void) +{ + int cpu, ret; + struct paca_struct *pp; + struct dtl_entry *dtl; + + if (!firmware_has_feature(FW_FEATURE_SPLPAR)) + return 0; + + for_each_possible_cpu(cpu) { + pp = &paca[cpu]; + dtl = kmalloc_node(DISPATCH_LOG_BYTES, GFP_KERNEL, + cpu_to_node(cpu)); + if (!dtl) { + pr_warn("Failed to allocate dispatch trace log for cpu %d\n", + cpu); + pr_warn("Stolen time statistics will be unreliable\n"); + break; + } + + pp->dtl_ridx = 0; + pp->dispatch_log = dtl; + pp->dispatch_log_end = dtl + N_DISPATCH_LOG; + pp->dtl_curr = dtl; + } + + /* Register the DTL for the current (boot) cpu */ + dtl = get_paca()->dispatch_log; + get_paca()->dtl_ridx = 0; + get_paca()->dtl_curr = dtl; + get_paca()->lppaca_ptr->dtl_idx = 0; + + /* hypervisor reads buffer length from this field */ + dtl->enqueue_to_dispatch_time = DISPATCH_LOG_BYTES; + ret = register_dtl(hard_smp_processor_id(), __pa(dtl)); + if (ret) + pr_warn("DTL registration failed for boot cpu %d (%d)\n", + smp_processor_id(), ret); + get_paca()->lppaca_ptr->dtl_enable_mask = 2; + + return 0; +} + +early_initcall(alloc_dispatch_logs); +#endif /* CONFIG_VIRT_CPU_ACCOUNTING */ + static void __init pSeries_setup_arch(void) { /* Discover PIC type and setup ppc_md accordingly */ -- cgit v1.2.3 From 872e439a45ed4a4bd499bc55cb0dffa74027f749 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Tue, 31 Aug 2010 01:59:53 +0000 Subject: powerpc/pseries: Re-enable dispatch trace log userspace interface Since the cpu accounting code uses the hypervisor dispatch trace log now when CONFIG_VIRT_CPU_ACCOUNTING = y, the previous commit disabled access to it via files in the /sys/kernel/debug/powerpc/dtl/ directory in that case. This restores those files. To do this, we now have a hook that the cpu accounting code will call as it processes each entry from the hypervisor dispatch trace log. The code in dtl.c now uses that to fill up its ring buffer, rather than having the hypervisor fill the ring buffer directly. This also fixes dtl_file_read() to handle overflow conditions a bit better and adds a spinlock to ensure that race conditions (multiple processes opening or reading the file concurrently) are handled correctly. Signed-off-by: Paul Mackerras Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/lppaca.h | 8 ++ arch/powerpc/kernel/time.c | 6 +- arch/powerpc/platforms/pseries/dtl.c | 206 ++++++++++++++++++++++++++++------- 3 files changed, 179 insertions(+), 41 deletions(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/powerpc/include/asm/lppaca.h b/arch/powerpc/include/asm/lppaca.h index cfb85ec85750..7f5e0fefebb0 100644 --- a/arch/powerpc/include/asm/lppaca.h +++ b/arch/powerpc/include/asm/lppaca.h @@ -191,6 +191,14 @@ struct dtl_entry { #define DISPATCH_LOG_BYTES 4096 /* bytes per cpu */ #define N_DISPATCH_LOG (DISPATCH_LOG_BYTES / sizeof(struct dtl_entry)) +/* + * When CONFIG_VIRT_CPU_ACCOUNTING = y, the cpu accounting code controls + * reading from the dispatch trace log. If other code wants to consume + * DTL entries, it can set this pointer to a function that will get + * called once for each DTL entry that gets processed. + */ +extern void (*dtl_consumer)(struct dtl_entry *entry, u64 index); + #endif /* CONFIG_PPC_BOOK3S */ #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_LPPACA_H */ diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index fca20643c368..bcb738b9ff8c 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -183,6 +183,8 @@ DEFINE_PER_CPU(unsigned long, cputime_scaled_last_delta); cputime_t cputime_one_jiffy; +void (*dtl_consumer)(struct dtl_entry *, u64); + static void calc_cputime_factors(void) { struct div_result res; @@ -218,7 +220,7 @@ static u64 read_spurr(u64 tb) */ static u64 scan_dispatch_log(u64 stop_tb) { - unsigned long i = local_paca->dtl_ridx; + u64 i = local_paca->dtl_ridx; struct dtl_entry *dtl = local_paca->dtl_curr; struct dtl_entry *dtl_end = local_paca->dispatch_log_end; struct lppaca *vpa = local_paca->lppaca_ptr; @@ -229,6 +231,8 @@ static u64 scan_dispatch_log(u64 stop_tb) if (i == vpa->dtl_idx) return 0; while (i < vpa->dtl_idx) { + if (dtl_consumer) + dtl_consumer(dtl, i); dtb = dtl->timebase; tb_delta = dtl->enqueue_to_dispatch_time + dtl->ready_to_enqueue_time; diff --git a/arch/powerpc/platforms/pseries/dtl.c b/arch/powerpc/platforms/pseries/dtl.c index 0357655db49d..c371bc06434b 100644 --- a/arch/powerpc/platforms/pseries/dtl.c +++ b/arch/powerpc/platforms/pseries/dtl.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -37,6 +38,7 @@ struct dtl { int cpu; int buf_entries; u64 last_idx; + spinlock_t lock; }; static DEFINE_PER_CPU(struct dtl, cpu_dtl); @@ -55,25 +57,97 @@ static u8 dtl_event_mask = 0x7; static int dtl_buf_entries = (16 * 85); -static int dtl_enable(struct dtl *dtl) +#ifdef CONFIG_VIRT_CPU_ACCOUNTING +struct dtl_ring { + u64 write_index; + struct dtl_entry *write_ptr; + struct dtl_entry *buf; + struct dtl_entry *buf_end; + u8 saved_dtl_mask; +}; + +static DEFINE_PER_CPU(struct dtl_ring, dtl_rings); + +static atomic_t dtl_count; + +/* + * The cpu accounting code controls the DTL ring buffer, and we get + * given entries as they are processed. + */ +static void consume_dtle(struct dtl_entry *dtle, u64 index) { - unsigned long addr; - int ret, hwcpu; + struct dtl_ring *dtlr = &__get_cpu_var(dtl_rings); + struct dtl_entry *wp = dtlr->write_ptr; + struct lppaca *vpa = local_paca->lppaca_ptr; - /* only allow one reader */ - if (dtl->buf) - return -EBUSY; + if (!wp) + return; - /* we need to store the original allocation size for use during read */ - dtl->buf_entries = dtl_buf_entries; + *wp = *dtle; + barrier(); - dtl->buf = kmalloc_node(dtl->buf_entries * sizeof(struct dtl_entry), - GFP_KERNEL, cpu_to_node(dtl->cpu)); - if (!dtl->buf) { - printk(KERN_WARNING "%s: buffer alloc failed for cpu %d\n", - __func__, dtl->cpu); - return -ENOMEM; - } + /* check for hypervisor ring buffer overflow, ignore this entry if so */ + if (index + N_DISPATCH_LOG < vpa->dtl_idx) + return; + + ++wp; + if (wp == dtlr->buf_end) + wp = dtlr->buf; + dtlr->write_ptr = wp; + + /* incrementing write_index makes the new entry visible */ + smp_wmb(); + ++dtlr->write_index; +} + +static int dtl_start(struct dtl *dtl) +{ + struct dtl_ring *dtlr = &per_cpu(dtl_rings, dtl->cpu); + + dtlr->buf = dtl->buf; + dtlr->buf_end = dtl->buf + dtl->buf_entries; + dtlr->write_index = 0; + + /* setting write_ptr enables logging into our buffer */ + smp_wmb(); + dtlr->write_ptr = dtl->buf; + + /* enable event logging */ + dtlr->saved_dtl_mask = lppaca_of(dtl->cpu).dtl_enable_mask; + lppaca_of(dtl->cpu).dtl_enable_mask |= dtl_event_mask; + + dtl_consumer = consume_dtle; + atomic_inc(&dtl_count); + return 0; +} + +static void dtl_stop(struct dtl *dtl) +{ + struct dtl_ring *dtlr = &per_cpu(dtl_rings, dtl->cpu); + + dtlr->write_ptr = NULL; + smp_wmb(); + + dtlr->buf = NULL; + + /* restore dtl_enable_mask */ + lppaca_of(dtl->cpu).dtl_enable_mask = dtlr->saved_dtl_mask; + + if (atomic_dec_and_test(&dtl_count)) + dtl_consumer = NULL; +} + +static u64 dtl_current_index(struct dtl *dtl) +{ + return per_cpu(dtl_rings, dtl->cpu).write_index; +} + +#else /* CONFIG_VIRT_CPU_ACCOUNTING */ + +static int dtl_start(struct dtl *dtl) +{ + unsigned long addr; + int ret, hwcpu; /* Register our dtl buffer with the hypervisor. The HV expects the * buffer size to be passed in the second word of the buffer */ @@ -85,12 +159,11 @@ static int dtl_enable(struct dtl *dtl) if (ret) { printk(KERN_WARNING "%s: DTL registration for cpu %d (hw %d) " "failed with %d\n", __func__, dtl->cpu, hwcpu, ret); - kfree(dtl->buf); return -EIO; } /* set our initial buffer indices */ - dtl->last_idx = lppaca_of(dtl->cpu).dtl_idx = 0; + lppaca_of(dtl->cpu).dtl_idx = 0; /* ensure that our updates to the lppaca fields have occurred before * we actually enable the logging */ @@ -102,17 +175,66 @@ static int dtl_enable(struct dtl *dtl) return 0; } -static void dtl_disable(struct dtl *dtl) +static void dtl_stop(struct dtl *dtl) { int hwcpu = get_hard_smp_processor_id(dtl->cpu); lppaca_of(dtl->cpu).dtl_enable_mask = 0x0; unregister_dtl(hwcpu, __pa(dtl->buf)); +} + +static u64 dtl_current_index(struct dtl *dtl) +{ + return lppaca_of(dtl->cpu).dtl_idx; +} +#endif /* CONFIG_VIRT_CPU_ACCOUNTING */ +static int dtl_enable(struct dtl *dtl) +{ + long int n_entries; + long int rc; + struct dtl_entry *buf = NULL; + + /* only allow one reader */ + if (dtl->buf) + return -EBUSY; + + n_entries = dtl_buf_entries; + buf = kmalloc_node(n_entries * sizeof(struct dtl_entry), + GFP_KERNEL, cpu_to_node(dtl->cpu)); + if (!buf) { + printk(KERN_WARNING "%s: buffer alloc failed for cpu %d\n", + __func__, dtl->cpu); + return -ENOMEM; + } + + spin_lock(&dtl->lock); + rc = -EBUSY; + if (!dtl->buf) { + /* store the original allocation size for use during read */ + dtl->buf_entries = n_entries; + dtl->buf = buf; + dtl->last_idx = 0; + rc = dtl_start(dtl); + if (rc) + dtl->buf = NULL; + } + spin_unlock(&dtl->lock); + + if (rc) + kfree(buf); + return rc; +} + +static void dtl_disable(struct dtl *dtl) +{ + spin_lock(&dtl->lock); + dtl_stop(dtl); kfree(dtl->buf); dtl->buf = NULL; dtl->buf_entries = 0; + spin_unlock(&dtl->lock); } /* file interface */ @@ -140,8 +262,9 @@ static int dtl_file_release(struct inode *inode, struct file *filp) static ssize_t dtl_file_read(struct file *filp, char __user *buf, size_t len, loff_t *pos) { - int rc, cur_idx, last_idx, n_read, n_req, read_size; + long int rc, n_read, n_req, read_size; struct dtl *dtl; + u64 cur_idx, last_idx, i; if ((len % sizeof(struct dtl_entry)) != 0) return -EINVAL; @@ -154,41 +277,48 @@ static ssize_t dtl_file_read(struct file *filp, char __user *buf, size_t len, /* actual number of entries read */ n_read = 0; - cur_idx = lppaca_of(dtl->cpu).dtl_idx; + spin_lock(&dtl->lock); + + cur_idx = dtl_current_index(dtl); last_idx = dtl->last_idx; - if (cur_idx - last_idx > dtl->buf_entries) { - pr_debug("%s: hv buffer overflow for cpu %d, samples lost\n", - __func__, dtl->cpu); - } + if (last_idx + dtl->buf_entries <= cur_idx) + last_idx = cur_idx - dtl->buf_entries + 1; + + if (last_idx + n_req > cur_idx) + n_req = cur_idx - last_idx; - cur_idx %= dtl->buf_entries; - last_idx %= dtl->buf_entries; + if (n_req > 0) + dtl->last_idx = last_idx + n_req; + + spin_unlock(&dtl->lock); + + if (n_req <= 0) + return 0; + + i = last_idx % dtl->buf_entries; /* read the tail of the buffer if we've wrapped */ - if (last_idx > cur_idx) { - read_size = min(n_req, dtl->buf_entries - last_idx); + if (i + n_req > dtl->buf_entries) { + read_size = dtl->buf_entries - i; - rc = copy_to_user(buf, &dtl->buf[last_idx], + rc = copy_to_user(buf, &dtl->buf[i], read_size * sizeof(struct dtl_entry)); if (rc) return -EFAULT; - last_idx = 0; + i = 0; n_req -= read_size; n_read += read_size; buf += read_size * sizeof(struct dtl_entry); } /* .. and now the head */ - read_size = min(n_req, cur_idx - last_idx); - rc = copy_to_user(buf, &dtl->buf[last_idx], - read_size * sizeof(struct dtl_entry)); + rc = copy_to_user(buf, &dtl->buf[i], n_req * sizeof(struct dtl_entry)); if (rc) return -EFAULT; - n_read += read_size; - dtl->last_idx += n_read; + n_read += n_req; return n_read * sizeof(struct dtl_entry); } @@ -220,11 +350,6 @@ static int dtl_init(void) struct dentry *event_mask_file, *buf_entries_file; int rc, i; -#ifdef CONFIG_VIRT_CPU_ACCOUNTING - /* disable this for now */ - return -ENODEV; -#endif - if (!firmware_has_feature(FW_FEATURE_SPLPAR)) return -ENODEV; @@ -251,6 +376,7 @@ static int dtl_init(void) /* set up the per-cpu log structures */ for_each_possible_cpu(i) { struct dtl *dtl = &per_cpu(cpu_dtl, i); + spin_lock_init(&dtl->lock); dtl->cpu = i; rc = dtl_setup_file(dtl); -- cgit v1.2.3 From 55ec2fca3e99f83b5c674e9aba713d848392f6cc Mon Sep 17 00:00:00 2001 From: Timur Tabi Date: Mon, 20 Sep 2010 11:23:41 -0500 Subject: powerpc: export ppc_proc_freq and ppc_tb_freq as GPL symbols Export the global variable 'ppc_tb_freq', so that modules (like the Book-E watchdog driver) can use it. To maintain consistency, ppc_proc_freq is changed to a GPL-only export. This is okay, because any module that needs this symbol should be an actual Linux driver, which must be GPL-licensed. Signed-off-by: Timur Tabi Acked-by: Josh Boyer Signed-off-by: Kumar Gala --- arch/powerpc/kernel/time.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index bcb738b9ff8c..644f9188d8e7 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -161,8 +161,9 @@ extern struct timezone sys_tz; static long timezone_offset; unsigned long ppc_proc_freq; -EXPORT_SYMBOL(ppc_proc_freq); +EXPORT_SYMBOL_GPL(ppc_proc_freq); unsigned long ppc_tb_freq; +EXPORT_SYMBOL_GPL(ppc_tb_freq); #ifdef CONFIG_VIRT_CPU_ACCOUNTING /* -- cgit v1.2.3 From e360adbe29241a0194e10e20595360dd7b98a2b3 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 14 Oct 2010 14:01:34 +0800 Subject: irq_work: Add generic hardirq context callbacks Provide a mechanism that allows running code in IRQ context. It is most useful for NMI code that needs to interact with the rest of the system -- like wakeup a task to drain buffers. Perf currently has such a mechanism, so extract that and provide it as a generic feature, independent of perf so that others may also benefit. The IRQ context callback is generated through self-IPIs where possible, or on architectures like powerpc the decrementer (the built-in timer facility) is set to generate an interrupt immediately. Architectures that don't have anything like this get to do with a callback from the timer tick. These architectures can call irq_work_run() at the tail of any IRQ handlers that might enqueue such work (like the perf IRQ handler) to avoid undue latencies in processing the work. Signed-off-by: Peter Zijlstra Acked-by: Kyle McMartin Acked-by: Martin Schwidefsky [ various fixes ] Signed-off-by: Huang Ying LKML-Reference: <1287036094.7768.291.camel@yhuang-dev> Signed-off-by: Ingo Molnar --- arch/alpha/Kconfig | 1 + arch/alpha/include/asm/perf_event.h | 5 -- arch/alpha/kernel/time.c | 30 +++---- arch/arm/Kconfig | 1 + arch/arm/include/asm/perf_event.h | 12 --- arch/arm/kernel/perf_event.c | 8 +- arch/frv/Kconfig | 1 + arch/frv/lib/Makefile | 2 +- arch/frv/lib/perf_event.c | 19 ---- arch/parisc/Kconfig | 1 + arch/parisc/include/asm/perf_event.h | 3 +- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/paca.h | 2 +- arch/powerpc/kernel/time.c | 42 ++++----- arch/s390/Kconfig | 1 + arch/s390/include/asm/perf_event.h | 3 +- arch/sh/Kconfig | 1 + arch/sh/include/asm/perf_event.h | 7 -- arch/sparc/Kconfig | 2 + arch/sparc/include/asm/perf_event.h | 4 - arch/sparc/kernel/pcr.c | 8 +- arch/x86/Kconfig | 1 + arch/x86/include/asm/entry_arch.h | 4 +- arch/x86/include/asm/hardirq.h | 2 +- arch/x86/include/asm/hw_irq.h | 2 +- arch/x86/include/asm/irq_vectors.h | 4 +- arch/x86/kernel/Makefile | 1 + arch/x86/kernel/cpu/perf_event.c | 19 ---- arch/x86/kernel/entry_64.S | 6 +- arch/x86/kernel/irq.c | 8 +- arch/x86/kernel/irq_work.c | 30 +++++++ arch/x86/kernel/irqinit.c | 6 +- include/linux/irq_work.h | 20 +++++ include/linux/perf_event.h | 11 +-- init/Kconfig | 8 ++ kernel/Makefile | 2 + kernel/irq_work.c | 164 +++++++++++++++++++++++++++++++++++ kernel/perf_event.c | 104 ++-------------------- kernel/timer.c | 7 +- 39 files changed, 311 insertions(+), 242 deletions(-) delete mode 100644 arch/frv/lib/perf_event.c create mode 100644 arch/x86/kernel/irq_work.c create mode 100644 include/linux/irq_work.h create mode 100644 kernel/irq_work.c (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index b9647bb66d13..d04ccd73af45 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -9,6 +9,7 @@ config ALPHA select HAVE_IDE select HAVE_OPROFILE select HAVE_SYSCALL_WRAPPERS + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select HAVE_DMA_ATTRS help diff --git a/arch/alpha/include/asm/perf_event.h b/arch/alpha/include/asm/perf_event.h index 4157cd3c44a9..fe792ca818f6 100644 --- a/arch/alpha/include/asm/perf_event.h +++ b/arch/alpha/include/asm/perf_event.h @@ -1,11 +1,6 @@ #ifndef __ASM_ALPHA_PERF_EVENT_H #define __ASM_ALPHA_PERF_EVENT_H -/* Alpha only supports software events through this interface. */ -extern void set_perf_event_pending(void); - -#define PERF_EVENT_INDEX_OFFSET 0 - #ifdef CONFIG_PERF_EVENTS extern void init_hw_perf_events(void); #else diff --git a/arch/alpha/kernel/time.c b/arch/alpha/kernel/time.c index 396af1799ea4..0f1d8493cfca 100644 --- a/arch/alpha/kernel/time.c +++ b/arch/alpha/kernel/time.c @@ -41,7 +41,7 @@ #include #include #include -#include +#include #include #include @@ -83,25 +83,25 @@ static struct { unsigned long est_cycle_freq; -#ifdef CONFIG_PERF_EVENTS +#ifdef CONFIG_IRQ_WORK -DEFINE_PER_CPU(u8, perf_event_pending); +DEFINE_PER_CPU(u8, irq_work_pending); -#define set_perf_event_pending_flag() __get_cpu_var(perf_event_pending) = 1 -#define test_perf_event_pending() __get_cpu_var(perf_event_pending) -#define clear_perf_event_pending() __get_cpu_var(perf_event_pending) = 0 +#define set_irq_work_pending_flag() __get_cpu_var(irq_work_pending) = 1 +#define test_irq_work_pending() __get_cpu_var(irq_work_pending) +#define clear_irq_work_pending() __get_cpu_var(irq_work_pending) = 0 -void set_perf_event_pending(void) +void set_irq_work_pending(void) { - set_perf_event_pending_flag(); + set_irq_work_pending_flag(); } -#else /* CONFIG_PERF_EVENTS */ +#else /* CONFIG_IRQ_WORK */ -#define test_perf_event_pending() 0 -#define clear_perf_event_pending() +#define test_irq_work_pending() 0 +#define clear_irq_work_pending() -#endif /* CONFIG_PERF_EVENTS */ +#endif /* CONFIG_IRQ_WORK */ static inline __u32 rpcc(void) @@ -191,9 +191,9 @@ irqreturn_t timer_interrupt(int irq, void *dev) write_sequnlock(&xtime_lock); - if (test_perf_event_pending()) { - clear_perf_event_pending(); - perf_event_do_pending(); + if (test_irq_work_pending()) { + clear_irq_work_pending(); + irq_work_run(); } #ifndef CONFIG_SMP diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 88c97bc7a6f5..7c0dfccd05bd 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -23,6 +23,7 @@ config ARM select HAVE_KERNEL_GZIP select HAVE_KERNEL_LZO select HAVE_KERNEL_LZMA + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select PERF_USE_VMALLOC select HAVE_REGS_AND_STACK_ACCESS_API diff --git a/arch/arm/include/asm/perf_event.h b/arch/arm/include/asm/perf_event.h index b5799a3b7117..c4aa4e8c6af9 100644 --- a/arch/arm/include/asm/perf_event.h +++ b/arch/arm/include/asm/perf_event.h @@ -12,18 +12,6 @@ #ifndef __ARM_PERF_EVENT_H__ #define __ARM_PERF_EVENT_H__ -/* - * NOP: on *most* (read: all supported) ARM platforms, the performance - * counter interrupts are regular interrupts and not an NMI. This - * means that when we receive the interrupt we can call - * perf_event_do_pending() that handles all of the work with - * interrupts disabled. - */ -static inline void -set_perf_event_pending(void) -{ -} - /* ARM performance counters start from 1 (in the cp15 accesses) so use the * same indexes here for consistency. */ #define PERF_EVENT_INDEX_OFFSET 1 diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 6cc6521881aa..49643b1467e6 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -1092,7 +1092,7 @@ armv6pmu_handle_irq(int irq_num, * platforms that can have the PMU interrupts raised as an NMI, this * will not work. */ - perf_event_do_pending(); + irq_work_run(); return IRQ_HANDLED; } @@ -2068,7 +2068,7 @@ static irqreturn_t armv7pmu_handle_irq(int irq_num, void *dev) * platforms that can have the PMU interrupts raised as an NMI, this * will not work. */ - perf_event_do_pending(); + irq_work_run(); return IRQ_HANDLED; } @@ -2436,7 +2436,7 @@ xscale1pmu_handle_irq(int irq_num, void *dev) armpmu->disable(hwc, idx); } - perf_event_do_pending(); + irq_work_run(); /* * Re-enable the PMU. @@ -2763,7 +2763,7 @@ xscale2pmu_handle_irq(int irq_num, void *dev) armpmu->disable(hwc, idx); } - perf_event_do_pending(); + irq_work_run(); /* * Re-enable the PMU. diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index 16399bd24993..0f2417df6323 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -7,6 +7,7 @@ config FRV default y select HAVE_IDE select HAVE_ARCH_TRACEHOOK + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS config ZONE_DMA diff --git a/arch/frv/lib/Makefile b/arch/frv/lib/Makefile index f4709756d0d9..4ff2fb1e6b16 100644 --- a/arch/frv/lib/Makefile +++ b/arch/frv/lib/Makefile @@ -5,4 +5,4 @@ lib-y := \ __ashldi3.o __lshrdi3.o __muldi3.o __ashrdi3.o __negdi2.o __ucmpdi2.o \ checksum.o memcpy.o memset.o atomic-ops.o atomic64-ops.o \ - outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o perf_event.o + outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o diff --git a/arch/frv/lib/perf_event.c b/arch/frv/lib/perf_event.c deleted file mode 100644 index 9ac5acfd2e91..000000000000 --- a/arch/frv/lib/perf_event.c +++ /dev/null @@ -1,19 +0,0 @@ -/* Performance event handling - * - * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. - */ - -#include - -/* - * mark the performance event as pending - */ -void set_perf_event_pending(void) -{ -} diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index 907417d187e1..79a04a9394d5 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -16,6 +16,7 @@ config PARISC select RTC_DRV_GENERIC select INIT_ALL_POSSIBLE select BUG + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select GENERIC_ATOMIC64 if !64BIT help diff --git a/arch/parisc/include/asm/perf_event.h b/arch/parisc/include/asm/perf_event.h index cc146427d8f9..1e0fd8ba6c03 100644 --- a/arch/parisc/include/asm/perf_event.h +++ b/arch/parisc/include/asm/perf_event.h @@ -1,7 +1,6 @@ #ifndef __ASM_PARISC_PERF_EVENT_H #define __ASM_PARISC_PERF_EVENT_H -/* parisc only supports software events through this interface. */ -static inline void set_perf_event_pending(void) { } +/* Empty, just to avoid compiling error */ #endif /* __ASM_PARISC_PERF_EVENT_H */ diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 631e5a0fb6ab..4b1e521d966f 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -138,6 +138,7 @@ config PPC select HAVE_OPROFILE select HAVE_SYSCALL_WRAPPERS if PPC64 select GENERIC_ATOMIC64 if PPC32 + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select HAVE_REGS_AND_STACK_ACCESS_API select HAVE_HW_BREAKPOINT if PERF_EVENTS && PPC_BOOK3S_64 diff --git a/arch/powerpc/include/asm/paca.h b/arch/powerpc/include/asm/paca.h index 1ff6662f7faf..9b287fdd8ea3 100644 --- a/arch/powerpc/include/asm/paca.h +++ b/arch/powerpc/include/asm/paca.h @@ -129,7 +129,7 @@ struct paca_struct { u8 soft_enabled; /* irq soft-enable flag */ u8 hard_enabled; /* set if irqs are enabled in MSR */ u8 io_sync; /* writel() needs spin_unlock sync */ - u8 perf_event_pending; /* PM interrupt while soft-disabled */ + u8 irq_work_pending; /* IRQ_WORK interrupt while soft-disable */ /* Stuff for accurate time accounting */ u64 user_time; /* accumulated usermode TB ticks */ diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 8533b3b83f5d..54888eb10c3b 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -53,7 +53,7 @@ #include #include #include -#include +#include #include #include @@ -493,60 +493,60 @@ void __init iSeries_time_init_early(void) } #endif /* CONFIG_PPC_ISERIES */ -#ifdef CONFIG_PERF_EVENTS +#ifdef CONFIG_IRQ_WORK /* * 64-bit uses a byte in the PACA, 32-bit uses a per-cpu variable... */ #ifdef CONFIG_PPC64 -static inline unsigned long test_perf_event_pending(void) +static inline unsigned long test_irq_work_pending(void) { unsigned long x; asm volatile("lbz %0,%1(13)" : "=r" (x) - : "i" (offsetof(struct paca_struct, perf_event_pending))); + : "i" (offsetof(struct paca_struct, irq_work_pending))); return x; } -static inline void set_perf_event_pending_flag(void) +static inline void set_irq_work_pending_flag(void) { asm volatile("stb %0,%1(13)" : : "r" (1), - "i" (offsetof(struct paca_struct, perf_event_pending))); + "i" (offsetof(struct paca_struct, irq_work_pending))); } -static inline void clear_perf_event_pending(void) +static inline void clear_irq_work_pending(void) { asm volatile("stb %0,%1(13)" : : "r" (0), - "i" (offsetof(struct paca_struct, perf_event_pending))); + "i" (offsetof(struct paca_struct, irq_work_pending))); } #else /* 32-bit */ -DEFINE_PER_CPU(u8, perf_event_pending); +DEFINE_PER_CPU(u8, irq_work_pending); -#define set_perf_event_pending_flag() __get_cpu_var(perf_event_pending) = 1 -#define test_perf_event_pending() __get_cpu_var(perf_event_pending) -#define clear_perf_event_pending() __get_cpu_var(perf_event_pending) = 0 +#define set_irq_work_pending_flag() __get_cpu_var(irq_work_pending) = 1 +#define test_irq_work_pending() __get_cpu_var(irq_work_pending) +#define clear_irq_work_pending() __get_cpu_var(irq_work_pending) = 0 #endif /* 32 vs 64 bit */ -void set_perf_event_pending(void) +void set_irq_work_pending(void) { preempt_disable(); - set_perf_event_pending_flag(); + set_irq_work_pending_flag(); set_dec(1); preempt_enable(); } -#else /* CONFIG_PERF_EVENTS */ +#else /* CONFIG_IRQ_WORK */ -#define test_perf_event_pending() 0 -#define clear_perf_event_pending() +#define test_irq_work_pending() 0 +#define clear_irq_work_pending() -#endif /* CONFIG_PERF_EVENTS */ +#endif /* CONFIG_IRQ_WORK */ /* * For iSeries shared processors, we have to let the hypervisor @@ -587,9 +587,9 @@ void timer_interrupt(struct pt_regs * regs) calculate_steal_time(); - if (test_perf_event_pending()) { - clear_perf_event_pending(); - perf_event_do_pending(); + if (test_irq_work_pending()) { + clear_irq_work_pending(); + irq_work_run(); } #ifdef CONFIG_PPC_ISERIES diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index f0777a47e3a5..958f0dadeadf 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -95,6 +95,7 @@ config S390 select HAVE_KVM if 64BIT select HAVE_ARCH_TRACEHOOK select INIT_ALL_POSSIBLE + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select HAVE_KERNEL_GZIP select HAVE_KERNEL_BZIP2 diff --git a/arch/s390/include/asm/perf_event.h b/arch/s390/include/asm/perf_event.h index 3840cbe77637..a75f168d2718 100644 --- a/arch/s390/include/asm/perf_event.h +++ b/arch/s390/include/asm/perf_event.h @@ -4,7 +4,6 @@ * Copyright 2009 Martin Schwidefsky, IBM Corporation. */ -static inline void set_perf_event_pending(void) {} -static inline void clear_perf_event_pending(void) {} +/* Empty, just to avoid compiling error */ #define PERF_EVENT_INDEX_OFFSET 0 diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index 35b6c3f85173..35b6879628a0 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -16,6 +16,7 @@ config SUPERH select HAVE_ARCH_TRACEHOOK select HAVE_DMA_API_DEBUG select HAVE_DMA_ATTRS + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select PERF_USE_VMALLOC select HAVE_KERNEL_GZIP diff --git a/arch/sh/include/asm/perf_event.h b/arch/sh/include/asm/perf_event.h index 3d0c9f36d150..14308bed7ea5 100644 --- a/arch/sh/include/asm/perf_event.h +++ b/arch/sh/include/asm/perf_event.h @@ -26,11 +26,4 @@ extern int register_sh_pmu(struct sh_pmu *); extern int reserve_pmc_hardware(void); extern void release_pmc_hardware(void); -static inline void set_perf_event_pending(void) -{ - /* Nothing to see here, move along. */ -} - -#define PERF_EVENT_INDEX_OFFSET 0 - #endif /* __ASM_SH_PERF_EVENT_H */ diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 9212cd42a832..3e9d31401fb2 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -26,6 +26,7 @@ config SPARC select ARCH_WANT_OPTIONAL_GPIOLIB select RTC_CLASS select RTC_DRV_M48T59 + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select PERF_USE_VMALLOC select HAVE_DMA_ATTRS @@ -54,6 +55,7 @@ config SPARC64 select RTC_DRV_BQ4802 select RTC_DRV_SUN4V select RTC_DRV_STARFIRE + select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select PERF_USE_VMALLOC diff --git a/arch/sparc/include/asm/perf_event.h b/arch/sparc/include/asm/perf_event.h index 727af70646cb..6e8bfa1786da 100644 --- a/arch/sparc/include/asm/perf_event.h +++ b/arch/sparc/include/asm/perf_event.h @@ -1,10 +1,6 @@ #ifndef __ASM_SPARC_PERF_EVENT_H #define __ASM_SPARC_PERF_EVENT_H -extern void set_perf_event_pending(void); - -#define PERF_EVENT_INDEX_OFFSET 0 - #ifdef CONFIG_PERF_EVENTS #include diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c index c4a6a50b4849..b87873c0e8ea 100644 --- a/arch/sparc/kernel/pcr.c +++ b/arch/sparc/kernel/pcr.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include #include @@ -43,14 +43,14 @@ void __irq_entry deferred_pcr_work_irq(int irq, struct pt_regs *regs) old_regs = set_irq_regs(regs); irq_enter(); -#ifdef CONFIG_PERF_EVENTS - perf_event_do_pending(); +#ifdef CONFIG_IRQ_WORK + irq_work_run(); #endif irq_exit(); set_irq_regs(old_regs); } -void set_perf_event_pending(void) +void arch_irq_work_raise(void) { set_softint(1 << PIL_DEFERRED_PCR_WORK); } diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 9815221976a7..fd227d6b8d9c 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -25,6 +25,7 @@ config X86 select HAVE_IDE select HAVE_OPROFILE select HAVE_PERF_EVENTS if (!M386 && !M486) + select HAVE_IRQ_WORK select HAVE_IOREMAP_PROT select HAVE_KPROBES select ARCH_WANT_OPTIONAL_GPIOLIB diff --git a/arch/x86/include/asm/entry_arch.h b/arch/x86/include/asm/entry_arch.h index 8e8ec663a98f..b8e96a18676b 100644 --- a/arch/x86/include/asm/entry_arch.h +++ b/arch/x86/include/asm/entry_arch.h @@ -49,8 +49,8 @@ BUILD_INTERRUPT(apic_timer_interrupt,LOCAL_TIMER_VECTOR) BUILD_INTERRUPT(error_interrupt,ERROR_APIC_VECTOR) BUILD_INTERRUPT(spurious_interrupt,SPURIOUS_APIC_VECTOR) -#ifdef CONFIG_PERF_EVENTS -BUILD_INTERRUPT(perf_pending_interrupt, LOCAL_PENDING_VECTOR) +#ifdef CONFIG_IRQ_WORK +BUILD_INTERRUPT(irq_work_interrupt, IRQ_WORK_VECTOR) #endif #ifdef CONFIG_X86_THERMAL_VECTOR diff --git a/arch/x86/include/asm/hardirq.h b/arch/x86/include/asm/hardirq.h index aeab29aee617..55e4de613f0e 100644 --- a/arch/x86/include/asm/hardirq.h +++ b/arch/x86/include/asm/hardirq.h @@ -14,7 +14,7 @@ typedef struct { #endif unsigned int x86_platform_ipis; /* arch dependent */ unsigned int apic_perf_irqs; - unsigned int apic_pending_irqs; + unsigned int apic_irq_work_irqs; #ifdef CONFIG_SMP unsigned int irq_resched_count; unsigned int irq_call_count; diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h index 46c0fe05f230..3a54a1ca1a02 100644 --- a/arch/x86/include/asm/hw_irq.h +++ b/arch/x86/include/asm/hw_irq.h @@ -29,7 +29,7 @@ extern void apic_timer_interrupt(void); extern void x86_platform_ipi(void); extern void error_interrupt(void); -extern void perf_pending_interrupt(void); +extern void irq_work_interrupt(void); extern void spurious_interrupt(void); extern void thermal_interrupt(void); diff --git a/arch/x86/include/asm/irq_vectors.h b/arch/x86/include/asm/irq_vectors.h index e2ca30092557..6af0894dafb4 100644 --- a/arch/x86/include/asm/irq_vectors.h +++ b/arch/x86/include/asm/irq_vectors.h @@ -114,9 +114,9 @@ #define X86_PLATFORM_IPI_VECTOR 0xed /* - * Performance monitoring pending work vector: + * IRQ work vector: */ -#define LOCAL_PENDING_VECTOR 0xec +#define IRQ_WORK_VECTOR 0xec #define UV_BAU_MESSAGE 0xea diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 9d3f485e5dd0..7490bf8d1459 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -35,6 +35,7 @@ obj-y := process_$(BITS).o signal.o entry_$(BITS).o obj-y += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o obj-y += time.o ioport.o ldt.o dumpstack.o obj-y += setup.o x86_init.o i8259.o irqinit.o jump_label.o +obj-$(CONFIG_IRQ_WORK) += irq_work.o obj-$(CONFIG_X86_VISWS) += visws_quirks.o obj-$(CONFIG_X86_32) += probe_roms_32.o obj-$(CONFIG_X86_32) += sys_i386_32.o i386_ksyms_32.o diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index e2513f26ba8b..fe73c1844a9a 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1196,25 +1196,6 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) return handled; } -void smp_perf_pending_interrupt(struct pt_regs *regs) -{ - irq_enter(); - ack_APIC_irq(); - inc_irq_stat(apic_pending_irqs); - perf_event_do_pending(); - irq_exit(); -} - -void set_perf_event_pending(void) -{ -#ifdef CONFIG_X86_LOCAL_APIC - if (!x86_pmu.apic || !x86_pmu_initialized()) - return; - - apic->send_IPI_self(LOCAL_PENDING_VECTOR); -#endif -} - void perf_events_lapic_init(void) { if (!x86_pmu.apic || !x86_pmu_initialized()) diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 17be5ec7cbba..c375c79065f8 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -1023,9 +1023,9 @@ apicinterrupt ERROR_APIC_VECTOR \ apicinterrupt SPURIOUS_APIC_VECTOR \ spurious_interrupt smp_spurious_interrupt -#ifdef CONFIG_PERF_EVENTS -apicinterrupt LOCAL_PENDING_VECTOR \ - perf_pending_interrupt smp_perf_pending_interrupt +#ifdef CONFIG_IRQ_WORK +apicinterrupt IRQ_WORK_VECTOR \ + irq_work_interrupt smp_irq_work_interrupt #endif /* diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index 91fd0c70a18a..44edb03fc9ec 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -67,10 +67,10 @@ static int show_other_interrupts(struct seq_file *p, int prec) for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->apic_perf_irqs); seq_printf(p, " Performance monitoring interrupts\n"); - seq_printf(p, "%*s: ", prec, "PND"); + seq_printf(p, "%*s: ", prec, "IWI"); for_each_online_cpu(j) - seq_printf(p, "%10u ", irq_stats(j)->apic_pending_irqs); - seq_printf(p, " Performance pending work\n"); + seq_printf(p, "%10u ", irq_stats(j)->apic_irq_work_irqs); + seq_printf(p, " IRQ work interrupts\n"); #endif if (x86_platform_ipi_callback) { seq_printf(p, "%*s: ", prec, "PLT"); @@ -185,7 +185,7 @@ u64 arch_irq_stat_cpu(unsigned int cpu) sum += irq_stats(cpu)->apic_timer_irqs; sum += irq_stats(cpu)->irq_spurious_count; sum += irq_stats(cpu)->apic_perf_irqs; - sum += irq_stats(cpu)->apic_pending_irqs; + sum += irq_stats(cpu)->apic_irq_work_irqs; #endif if (x86_platform_ipi_callback) sum += irq_stats(cpu)->x86_platform_ipis; diff --git a/arch/x86/kernel/irq_work.c b/arch/x86/kernel/irq_work.c new file mode 100644 index 000000000000..ca8f703a1e70 --- /dev/null +++ b/arch/x86/kernel/irq_work.c @@ -0,0 +1,30 @@ +/* + * x86 specific code for irq_work + * + * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra + */ + +#include +#include +#include +#include + +void smp_irq_work_interrupt(struct pt_regs *regs) +{ + irq_enter(); + ack_APIC_irq(); + inc_irq_stat(apic_irq_work_irqs); + irq_work_run(); + irq_exit(); +} + +void arch_irq_work_raise(void) +{ +#ifdef CONFIG_X86_LOCAL_APIC + if (!cpu_has_apic) + return; + + apic->send_IPI_self(IRQ_WORK_VECTOR); + apic_wait_icr_idle(); +#endif +} diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c index 990ae7cfc578..713969b9266b 100644 --- a/arch/x86/kernel/irqinit.c +++ b/arch/x86/kernel/irqinit.c @@ -224,9 +224,9 @@ static void __init apic_intr_init(void) alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt); alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt); - /* Performance monitoring interrupts: */ -# ifdef CONFIG_PERF_EVENTS - alloc_intr_gate(LOCAL_PENDING_VECTOR, perf_pending_interrupt); + /* IRQ work interrupts: */ +# ifdef CONFIG_IRQ_WORK + alloc_intr_gate(IRQ_WORK_VECTOR, irq_work_interrupt); # endif #endif diff --git a/include/linux/irq_work.h b/include/linux/irq_work.h new file mode 100644 index 000000000000..4fa09d4d0b71 --- /dev/null +++ b/include/linux/irq_work.h @@ -0,0 +1,20 @@ +#ifndef _LINUX_IRQ_WORK_H +#define _LINUX_IRQ_WORK_H + +struct irq_work { + struct irq_work *next; + void (*func)(struct irq_work *); +}; + +static inline +void init_irq_work(struct irq_work *entry, void (*func)(struct irq_work *)) +{ + entry->next = NULL; + entry->func = func; +} + +bool irq_work_queue(struct irq_work *entry); +void irq_work_run(void); +void irq_work_sync(struct irq_work *entry); + +#endif /* _LINUX_IRQ_WORK_H */ diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index a9227e985207..2ebfc9ae4755 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -486,6 +486,7 @@ struct perf_guest_info_callbacks { #include #include #include +#include #include #include @@ -672,11 +673,6 @@ struct perf_buffer { void *data_pages[0]; }; -struct perf_pending_entry { - struct perf_pending_entry *next; - void (*func)(struct perf_pending_entry *); -}; - struct perf_sample_data; typedef void (*perf_overflow_handler_t)(struct perf_event *, int, @@ -784,7 +780,7 @@ struct perf_event { int pending_wakeup; int pending_kill; int pending_disable; - struct perf_pending_entry pending; + struct irq_work pending; atomic_t event_limit; @@ -898,8 +894,6 @@ extern int perf_event_init_task(struct task_struct *child); extern void perf_event_exit_task(struct task_struct *child); extern void perf_event_free_task(struct task_struct *task); extern void perf_event_delayed_put(struct task_struct *task); -extern void set_perf_event_pending(void); -extern void perf_event_do_pending(void); extern void perf_event_print_debug(void); extern void perf_pmu_disable(struct pmu *pmu); extern void perf_pmu_enable(struct pmu *pmu); @@ -1078,7 +1072,6 @@ static inline int perf_event_init_task(struct task_struct *child) { return 0; } static inline void perf_event_exit_task(struct task_struct *child) { } static inline void perf_event_free_task(struct task_struct *task) { } static inline void perf_event_delayed_put(struct task_struct *task) { } -static inline void perf_event_do_pending(void) { } static inline void perf_event_print_debug(void) { } static inline int perf_event_task_disable(void) { return -EINVAL; } static inline int perf_event_task_enable(void) { return -EINVAL; } diff --git a/init/Kconfig b/init/Kconfig index 2de5b1cbadd9..1ef0b439908e 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -21,6 +21,13 @@ config CONSTRUCTORS depends on !UML default y +config HAVE_IRQ_WORK + bool + +config IRQ_WORK + bool + depends on HAVE_IRQ_WORK + menu "General setup" config EXPERIMENTAL @@ -987,6 +994,7 @@ config PERF_EVENTS default y if (PROFILING || PERF_COUNTERS) depends on HAVE_PERF_EVENTS select ANON_INODES + select IRQ_WORK help Enable kernel support for various performance events provided by software and hardware. diff --git a/kernel/Makefile b/kernel/Makefile index d52b473c99a1..4d9bf5f8531f 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -23,6 +23,7 @@ CFLAGS_REMOVE_rtmutex-debug.o = -pg CFLAGS_REMOVE_cgroup-debug.o = -pg CFLAGS_REMOVE_sched_clock.o = -pg CFLAGS_REMOVE_perf_event.o = -pg +CFLAGS_REMOVE_irq_work.o = -pg endif obj-$(CONFIG_FREEZER) += freezer.o @@ -100,6 +101,7 @@ obj-$(CONFIG_TRACING) += trace/ obj-$(CONFIG_X86_DS) += trace/ obj-$(CONFIG_RING_BUFFER) += trace/ obj-$(CONFIG_SMP) += sched_cpupri.o +obj-$(CONFIG_IRQ_WORK) += irq_work.o obj-$(CONFIG_PERF_EVENTS) += perf_event.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o obj-$(CONFIG_USER_RETURN_NOTIFIER) += user-return-notifier.o diff --git a/kernel/irq_work.c b/kernel/irq_work.c new file mode 100644 index 000000000000..f16763ff8481 --- /dev/null +++ b/kernel/irq_work.c @@ -0,0 +1,164 @@ +/* + * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra + * + * Provides a framework for enqueueing and running callbacks from hardirq + * context. The enqueueing is NMI-safe. + */ + +#include +#include +#include +#include + +/* + * An entry can be in one of four states: + * + * free NULL, 0 -> {claimed} : free to be used + * claimed NULL, 3 -> {pending} : claimed to be enqueued + * pending next, 3 -> {busy} : queued, pending callback + * busy NULL, 2 -> {free, claimed} : callback in progress, can be claimed + * + * We use the lower two bits of the next pointer to keep PENDING and BUSY + * flags. + */ + +#define IRQ_WORK_PENDING 1UL +#define IRQ_WORK_BUSY 2UL +#define IRQ_WORK_FLAGS 3UL + +static inline bool irq_work_is_set(struct irq_work *entry, int flags) +{ + return (unsigned long)entry->next & flags; +} + +static inline struct irq_work *irq_work_next(struct irq_work *entry) +{ + unsigned long next = (unsigned long)entry->next; + next &= ~IRQ_WORK_FLAGS; + return (struct irq_work *)next; +} + +static inline struct irq_work *next_flags(struct irq_work *entry, int flags) +{ + unsigned long next = (unsigned long)entry; + next |= flags; + return (struct irq_work *)next; +} + +static DEFINE_PER_CPU(struct irq_work *, irq_work_list); + +/* + * Claim the entry so that no one else will poke at it. + */ +static bool irq_work_claim(struct irq_work *entry) +{ + struct irq_work *next, *nflags; + + do { + next = entry->next; + if ((unsigned long)next & IRQ_WORK_PENDING) + return false; + nflags = next_flags(next, IRQ_WORK_FLAGS); + } while (cmpxchg(&entry->next, next, nflags) != next); + + return true; +} + + +void __weak arch_irq_work_raise(void) +{ + /* + * Lame architectures will get the timer tick callback + */ +} + +/* + * Queue the entry and raise the IPI if needed. + */ +static void __irq_work_queue(struct irq_work *entry) +{ + struct irq_work **head, *next; + + head = &get_cpu_var(irq_work_list); + + do { + next = *head; + /* Can assign non-atomic because we keep the flags set. */ + entry->next = next_flags(next, IRQ_WORK_FLAGS); + } while (cmpxchg(head, next, entry) != next); + + /* The list was empty, raise self-interrupt to start processing. */ + if (!irq_work_next(entry)) + arch_irq_work_raise(); + + put_cpu_var(irq_work_list); +} + +/* + * Enqueue the irq_work @entry, returns true on success, failure when the + * @entry was already enqueued by someone else. + * + * Can be re-enqueued while the callback is still in progress. + */ +bool irq_work_queue(struct irq_work *entry) +{ + if (!irq_work_claim(entry)) { + /* + * Already enqueued, can't do! + */ + return false; + } + + __irq_work_queue(entry); + return true; +} +EXPORT_SYMBOL_GPL(irq_work_queue); + +/* + * Run the irq_work entries on this cpu. Requires to be ran from hardirq + * context with local IRQs disabled. + */ +void irq_work_run(void) +{ + struct irq_work *list, **head; + + head = &__get_cpu_var(irq_work_list); + if (*head == NULL) + return; + + BUG_ON(!in_irq()); + BUG_ON(!irqs_disabled()); + + list = xchg(head, NULL); + while (list != NULL) { + struct irq_work *entry = list; + + list = irq_work_next(list); + + /* + * Clear the PENDING bit, after this point the @entry + * can be re-used. + */ + entry->next = next_flags(NULL, IRQ_WORK_BUSY); + entry->func(entry); + /* + * Clear the BUSY bit and return to the free state if + * no-one else claimed it meanwhile. + */ + cmpxchg(&entry->next, next_flags(NULL, IRQ_WORK_BUSY), NULL); + } +} +EXPORT_SYMBOL_GPL(irq_work_run); + +/* + * Synchronize against the irq_work @entry, ensures the entry is not + * currently in use. + */ +void irq_work_sync(struct irq_work *entry) +{ + WARN_ON_ONCE(irqs_disabled()); + + while (irq_work_is_set(entry, IRQ_WORK_BUSY)) + cpu_relax(); +} +EXPORT_SYMBOL_GPL(irq_work_sync); diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 634f86a4b2f9..99b9700e74d0 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -2206,12 +2206,11 @@ static void free_event_rcu(struct rcu_head *head) kfree(event); } -static void perf_pending_sync(struct perf_event *event); static void perf_buffer_put(struct perf_buffer *buffer); static void free_event(struct perf_event *event) { - perf_pending_sync(event); + irq_work_sync(&event->pending); if (!event->parent) { atomic_dec(&nr_events); @@ -3162,16 +3161,7 @@ void perf_event_wakeup(struct perf_event *event) } } -/* - * Pending wakeups - * - * Handle the case where we need to wakeup up from NMI (or rq->lock) context. - * - * The NMI bit means we cannot possibly take locks. Therefore, maintain a - * single linked list and use cmpxchg() to add entries lockless. - */ - -static void perf_pending_event(struct perf_pending_entry *entry) +static void perf_pending_event(struct irq_work *entry) { struct perf_event *event = container_of(entry, struct perf_event, pending); @@ -3187,89 +3177,6 @@ static void perf_pending_event(struct perf_pending_entry *entry) } } -#define PENDING_TAIL ((struct perf_pending_entry *)-1UL) - -static DEFINE_PER_CPU(struct perf_pending_entry *, perf_pending_head) = { - PENDING_TAIL, -}; - -static void perf_pending_queue(struct perf_pending_entry *entry, - void (*func)(struct perf_pending_entry *)) -{ - struct perf_pending_entry **head; - - if (cmpxchg(&entry->next, NULL, PENDING_TAIL) != NULL) - return; - - entry->func = func; - - head = &get_cpu_var(perf_pending_head); - - do { - entry->next = *head; - } while (cmpxchg(head, entry->next, entry) != entry->next); - - set_perf_event_pending(); - - put_cpu_var(perf_pending_head); -} - -static int __perf_pending_run(void) -{ - struct perf_pending_entry *list; - int nr = 0; - - list = xchg(&__get_cpu_var(perf_pending_head), PENDING_TAIL); - while (list != PENDING_TAIL) { - void (*func)(struct perf_pending_entry *); - struct perf_pending_entry *entry = list; - - list = list->next; - - func = entry->func; - entry->next = NULL; - /* - * Ensure we observe the unqueue before we issue the wakeup, - * so that we won't be waiting forever. - * -- see perf_not_pending(). - */ - smp_wmb(); - - func(entry); - nr++; - } - - return nr; -} - -static inline int perf_not_pending(struct perf_event *event) -{ - /* - * If we flush on whatever cpu we run, there is a chance we don't - * need to wait. - */ - get_cpu(); - __perf_pending_run(); - put_cpu(); - - /* - * Ensure we see the proper queue state before going to sleep - * so that we do not miss the wakeup. -- see perf_pending_handle() - */ - smp_rmb(); - return event->pending.next == NULL; -} - -static void perf_pending_sync(struct perf_event *event) -{ - wait_event(event->waitq, perf_not_pending(event)); -} - -void perf_event_do_pending(void) -{ - __perf_pending_run(); -} - /* * We assume there is only KVM supporting the callbacks. * Later on, we might change it to a list if there is @@ -3319,8 +3226,7 @@ static void perf_output_wakeup(struct perf_output_handle *handle) if (handle->nmi) { handle->event->pending_wakeup = 1; - perf_pending_queue(&handle->event->pending, - perf_pending_event); + irq_work_queue(&handle->event->pending); } else perf_event_wakeup(handle->event); } @@ -4356,8 +4262,7 @@ static int __perf_event_overflow(struct perf_event *event, int nmi, event->pending_kill = POLL_HUP; if (nmi) { event->pending_disable = 1; - perf_pending_queue(&event->pending, - perf_pending_event); + irq_work_queue(&event->pending); } else perf_event_disable(event); } @@ -5374,6 +5279,7 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, INIT_LIST_HEAD(&event->event_entry); INIT_LIST_HEAD(&event->sibling_list); init_waitqueue_head(&event->waitq); + init_irq_work(&event->pending, perf_pending_event); mutex_init(&event->mmap_mutex); diff --git a/kernel/timer.c b/kernel/timer.c index 97bf05baade7..68a9ae7679b7 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -37,7 +37,7 @@ #include #include #include -#include +#include #include #include @@ -1279,7 +1279,10 @@ void update_process_times(int user_tick) run_local_timers(); rcu_check_callbacks(cpu, user_tick); printk_tick(); - perf_event_do_pending(); +#ifdef CONFIG_IRQ_WORK + if (in_irq()) + irq_work_run(); +#endif scheduler_tick(); run_posix_cpu_timers(p); } -- cgit v1.2.3 From 364a1246522f99cbe58040e99af007ada31034ed Mon Sep 17 00:00:00 2001 From: Heiko Schocher Date: Mon, 22 Nov 2010 21:30:33 +0000 Subject: powerpc/time: printk time stamp init not correct problem: I see sometimes on my mpc5200 based board such printk timing information: [ 0.000000] NR_IRQS:512 nr_irqs:512 16 [ 0.000000] MPC52xx PIC is up and running! [ 0.000000] clocksource: timebase mult[79364d9] shift[22] registered [ 0.000000] console [ttyPSC0] enabled [ 130.300633] pid_max: default: 32768 minimum: 301 [ 130.305647] Mount-cache hash table entries: 512 [ 130.315818] NET: Registered protocol family 16 reason: if the tbu not starts from 0 when linux boots, boot_tb maybe could not store the real 64 bit tbu value, because boot_tp is only a 32 bit unsigned long. solution: change boot_tb to u64 [BenH: Made it u64 instead of unsigned long long] Signed-off-by: Heiko Schocher cc: Wolfgang Denk Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/time.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 010406958d97..09e4dea4a85a 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -155,7 +155,7 @@ EXPORT_SYMBOL_GPL(rtc_lock); static u64 tb_to_ns_scale __read_mostly; static unsigned tb_to_ns_shift __read_mostly; -static unsigned long boot_tb __read_mostly; +static u64 boot_tb __read_mostly; extern struct timezone sys_tz; static long timezone_offset; -- cgit v1.2.3 From b18ae08deac23187e4a22a8c94a1a473be8e8c93 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 3 Jan 2011 03:49:25 +0000 Subject: powerpc/cell: Use system_wq in cpufreq_spudemand With cmwq, there's no reason to use a separate workqueue in cpufreq_spudemand. Use system_wq instead. The work items are already sync canceled on stop, so it's already guaranteed that no work is running when spu_gov_exit() is entered. Signed-off-by: Tejun Heo Cc: Arnd Bergmann Cc: linuxppc-dev@lists.ozlabs.org Cc: Dave Jones Cc: cpufreq@vger.kernel.org Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/time.c | 25 ++++++++++++++++++++----- arch/powerpc/platforms/cell/cpufreq_spudemand.c | 20 +++----------------- 2 files changed, 23 insertions(+), 22 deletions(-) (limited to 'arch/powerpc/kernel/time.c') diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 09e4dea4a85a..09d31dbf43f9 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -265,11 +265,26 @@ void accumulate_stolen_time(void) { u64 sst, ust; - sst = scan_dispatch_log(get_paca()->starttime_user); - ust = scan_dispatch_log(get_paca()->starttime); - get_paca()->system_time -= sst; - get_paca()->user_time -= ust; - get_paca()->stolen_time += ust + sst; + u8 save_soft_enabled = local_paca->soft_enabled; + u8 save_hard_enabled = local_paca->hard_enabled; + + /* We are called early in the exception entry, before + * soft/hard_enabled are sync'ed to the expected state + * for the exception. We are hard disabled but the PACA + * needs to reflect that so various debug stuff doesn't + * complain + */ + local_paca->soft_enabled = 0; + local_paca->hard_enabled = 0; + + sst = scan_dispatch_log(local_paca->starttime_user); + ust = scan_dispatch_log(local_paca->starttime); + local_paca->system_time -= sst; + local_paca->user_time -= ust; + local_paca->stolen_time += ust + sst; + + local_paca->soft_enabled = save_soft_enabled; + local_paca->hard_enabled = save_hard_enabled; } static inline u64 calculate_stolen_time(u64 stop_tb) diff --git a/arch/powerpc/platforms/cell/cpufreq_spudemand.c b/arch/powerpc/platforms/cell/cpufreq_spudemand.c index 968c1c0b4d5b..d809836bcf5f 100644 --- a/arch/powerpc/platforms/cell/cpufreq_spudemand.c +++ b/arch/powerpc/platforms/cell/cpufreq_spudemand.c @@ -39,8 +39,6 @@ struct spu_gov_info_struct { }; static DEFINE_PER_CPU(struct spu_gov_info_struct, spu_gov_info); -static struct workqueue_struct *kspugov_wq; - static int calc_freq(struct spu_gov_info_struct *info) { int cpu; @@ -71,14 +69,14 @@ static void spu_gov_work(struct work_struct *work) __cpufreq_driver_target(info->policy, target_freq, CPUFREQ_RELATION_H); delay = usecs_to_jiffies(info->poll_int); - queue_delayed_work_on(info->policy->cpu, kspugov_wq, &info->work, delay); + schedule_delayed_work_on(info->policy->cpu, &info->work, delay); } static void spu_gov_init_work(struct spu_gov_info_struct *info) { int delay = usecs_to_jiffies(info->poll_int); INIT_DELAYED_WORK_DEFERRABLE(&info->work, spu_gov_work); - queue_delayed_work_on(info->policy->cpu, kspugov_wq, &info->work, delay); + schedule_delayed_work_on(info->policy->cpu, &info->work, delay); } static void spu_gov_cancel_work(struct spu_gov_info_struct *info) @@ -152,27 +150,15 @@ static int __init spu_gov_init(void) { int ret; - kspugov_wq = create_workqueue("kspugov"); - if (!kspugov_wq) { - printk(KERN_ERR "creation of kspugov failed\n"); - ret = -EFAULT; - goto out; - } - ret = cpufreq_register_governor(&spu_governor); - if (ret) { + if (ret) printk(KERN_ERR "registration of governor failed\n"); - destroy_workqueue(kspugov_wq); - goto out; - } -out: return ret; } static void __exit spu_gov_exit(void) { cpufreq_unregister_governor(&spu_governor); - destroy_workqueue(kspugov_wq); } -- cgit v1.2.3