- 06 Apr, 2009 40 commits
-
-
Peter Zijlstra authored
It just occured to me it is possible to have multiple contending updates of the userpage (mmap information vs overflow vs counter). This would break the seqlock logic. It appear the arch code uses this from NMI context, so we cannot possibly serialize its use, therefore separate the data_head update from it and let it return to its original use. The arch code needs to make sure there are no contending callers by disabling the counter before using it -- powerpc appears to do this nicely. Signed-off-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by:
Paul Mackerras <paulus@samba.org> Orig-LKML-Reference: <20090330171023.241410660@chello.nl> Signed-off-by:
Ingo Molnar <mingo@elte.hu>
-
Peter Zijlstra authored
While going over the wakeup code I noticed delayed wakeups only work for hardware counters but basically all software counters rely on them. This patch unifies and generalizes the delayed wakeup to fix this issue. Since we're dealing with NMI context bits here, use a cmpxchg() based single link list implementation to track counters that have pending wakeups. [ This should really be generic code for delayed wakeups, but since we cannot use cmpxchg()/xchg() in generic code, I've let it live in the perf_counter code. -- Eric Dumazet could use it to aggregate the network wakeups. ] Furthermore, the x86 method of using TIF flags was flawed in that its quite possible to end up setting the bit on the idle task, loosing the wakeup. The powerpc method uses per-cpu storage and does appear to be sufficient. Signed-off-by:
-
Paul Mackerras authored
Impact: new functionality Currently, if there are more counters enabled than can fit on the CPU, the kernel will multiplex the counters on to the hardware using round-robin scheduling. That isn't too bad for sampling counters, but for counting counters it means that the value read from a counter represents some unknown fraction of the true count of events that occurred while the counter was enabled. This remedies the situation by keeping track of how long each counter is enabled for, and how long it is actually on the cpu and counting events. These times are recorded in nanoseconds using the task clock for per-task counters and the cpu clock for per-cpu counters. These values can be supplied to userspace on a read from the counter. Userspace requests that they be supplied after the counter value by setting the PERF_FORMAT_TOTAL_TIME_ENABLED and/or PERF_FORMAT_TOTAL_TIME_RUNNING bits in the hw_event.read_format field when creating the counter. (There is no way to change the read format after the counter is created, though it would be possible to add some way to do that.) Using this information it is possible for userspace to scale the count it reads from the counter to get an estimate of the true count: true_count_estimate = count * total_time_enabled / total_time_running This also lets userspace detect the situation where the counter never got to go on the cpu: total_time_running == 0. This functionality has been requested by the PAPI developers, and will be generally needed for interpreting the count values from counting counters correctly. In the implementation, this keeps 5 time values (in nanoseconds) for each counter: total_time_enabled and total_time_running are used when the counter is in state OFF or ERROR and for reporting back to userspace. When the counter is in state INACTIVE or ACTIVE, it is the tstamp_enabled, tstamp_running and tstamp_stopped values that are relevant, and total_time_enabled and total_time_running are determined from them. (tstamp_stopped is only used in INACTIVE state.) The reason for doing it like this is that it means that only counters being enabled or disabled at sched-in and sched-out time need to be updated. There are no new loops that iterate over all counters to update total_time_enabled or total_time_running. This also keeps separate child_total_time_running and child_total_time_enabled fields that get added in when reporting the totals to userspace. They are separate fields so that they can be atomic. We don't want to use atomics for total_time_running, total_time_enabled etc., because then we would have to use atomic sequences to update them, which are slower than regular arithmetic and memory accesses. It is possible to measure total_time_running by adding a task_clock counter to each group of counters, and total_time_enabled can be measured approximately with a top-level task_clock counter (though inaccuracies will creep in if you need to disable and enable groups since it is not possible in general to disable/enable the top-level task_clock counter simultaneously with another group). However, that adds extra overhead - I measured around 15% increase in the context switch latency reported by lat_ctx (from lmbench) when a task_clock counter was added to each of 2 groups, and around 25% increase when a task_clock counter was added to each of 4 groups. (In both cases a top-level task-clock counter was also added.) In contrast, the code added in this commit gives better information with no overhead that I could measure (in fact in some cases I measured lower times with this code, but the differences were all less than one standard deviation). [ v2: address review comments by Andrew Morton. ] Signed-off-by:
-
Peter Zijlstra authored
A brainfart stopped single page mmap()s working. The rest of the code should be perfectly fine with not having any data pages. Reported-by:
-
Peter Zijlstra authored
Teach kerneltop about the new output ABI. XXX: anybody fancy integrating the PID/TID data into the output? Bump the mmap_data pages a little because we bloated the output and have to be more careful about overruns with structured data. Signed-off-by:
-
Peter Zijlstra authored
provide a knob to set the number of mmap data pages. Signed-off-by:
Mike Galbraith <efault@gmx.de> Signed-off-by:
-
Peter Zijlstra authored
Allow cpu wide counters to profile userspace by providing what process the sample belongs to. This raises the first issue with the output type, lots of these options: group, tid, callchain, etc.. are non-exclusive and could be combined, suggesting a bitfield. However, things like the mmap() data stream doesn't fit in that. How to split the type field... Signed-off-by:
-
Peter Zijlstra authored
Ensure we never write more than we said we would. Signed-off-by:
-
Peter Zijlstra authored
Provide a {type,size} header for each output entry. This should provide extensible output, and the ability to mix multiple streams. Signed-off-by: -
Peter Zijlstra authored
Provide a begin, copy, end interface to the output buffer. begin() reserves the space, copy() copies the data over, considering page boundaries, end() finalizes the event and does the wakeup. Signed-off-by:
-
Peter Zijlstra authored
Impact: fix kerneltop 100% CPU usage Only return a poll event when there's actually been one, poll_wait() doesn't actually wait for the waitq you pass it, it only enqueues you on it. Only once all FDs have been iterated and none of thm returned a poll-event will it schedule(). Also make it return POLL_HUP when there's not mmap() area to read from. Further, fix a silly bug in the write code. Reported-by:
-
Paul Mackerras authored
Impact: documentation fix This updates the perfcounter documentation to reflect recent changes. Signed-off-by: -
Peter Zijlstra authored
Paul Mackerras wrote: > I noticed the poll stuff is bogus - we have a 2D array of struct > pollfds (MAX_NR_CPUS x MAX_COUNTERS), we fill in a sub-array (with the > rest being uninitialized, since the array is on the stack) and then > pass the first nr_cpus elements to poll. Not what we really meant, I > suspect. :) Not even if we only have one counter, since it's the > counter dimension that varies fastest. This should fix the most obvious poll fubar.. not enough to fix the full problem though.. Reported-by:
-
Paul Mackerras authored
The glib dependency in kerneltop.c is only for a little bit of list manipulation, and I find it inconvenient. This adds a 'next' field to struct source_line, which lets us link them together into a list. The code to do the linking ourselves turns out to be no longer or more difficult than using glib. This also fixes a few other problems: - We need to #include <limits.h> to get PATH_MAX on powerpc. - We need to #include <linux/types.h> rather than have our own definitions of __u64 and __s64; on powerpc the installed headers define them to be unsigned long and long respectively, and if we have our own, different definition here that causes a compile error. - This takes out the x86 setting of errno from -ret in sys_perf_counter_open. My experiments on x86 indicate that the glibc syscall() does this for us already. - We had two CPU migration counters in the default set, which seems unnecessary; I changed one of them to a context switch counter. - In perfstat mode we were printing CPU cycles and instructions as milliseconds, and the cpu clock and task clock counters as events. This fixes that. - In perfstat mode we were still printing a blank line after the first counter, which was a holdover from when a task clock counter was automatically included as the first counter. This removes the blank line. - On a test machine here, parse_symbols() and parse_vmlinux() were taking long enough (almost 0.5 seconds) for the mmap buffer to overflow before we got to the first mmap_read() call, so this moves them before we open all the counters. - The error message if sys_perf_counter_open fails needs to use errno, not -fd[i][counter]. Signed-off-by:
-
Ingo Molnar authored
Commit b7368fdd7d decreased the CPU cycles interval 100-fold, but this is causig kerneltop failures on my Nehalem box: aldebaran:/home/mingo/linux/linux/Documentation/perf_counter> ./kerneltop KernelTop refresh period: 2 seconds ERROR: failed to keep up with mmap data 10,000 cycles is way too short. What we should do instead on mostly-idle systems is some sort of read/poll timeout, so that we display something every 2 seconds for sure. Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: -
Ingo Molnar authored
Fix: kerneltop.c: In function ‘record_ip’: kerneltop.c:1005: warning: format ‘%016llx’ expects type ‘long long unsigned int’, but argument 2 has type ‘uint64_t’ Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Paul Mackerras <paulus@samba.org> Acked-by:
-
Ingo Molnar authored
Remove now unified perfstat.c and perf_counter.h, and link to the in-kernel perf_counter.h. Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Paul Mackerras <paulus@samba.org> Acked-by:
-
Peter Zijlstra authored
update kerneltop to use the mmap() output to gather overflow information Signed-off-by:
-
Peter Zijlstra authored
update the kerneltop userspace to work with the latest syscall ABI Signed-off-by:
-
Peter Zijlstra authored
Impact: Rework the perfcounter output ABI use sys_read() only for instant data and provide mmap() output for all async overflow data. The first mmap() determines the size of the output buffer. The mmap() size must be a PAGE_SIZE multiple of 1+pages, where pages must be a power of 2 or 0. Further mmap()s of the same fd must have the same size. Once all maps are gone, you can again mmap() with a new size. In case of 0 extra pages there is no data output and the first page only contains meta data. When there are data pages, a poll() event will be generated for each full page of data. Furthermore, the output is circular. This means that although 1 page is a valid configuration, its useless, since we'll start overwriting it the instant we report a full page. Future work will focus on the output format (currently maintained) where we'll likey want each entry denoted by a header which includes a type and length. Further future work will allow to splice() the fd, also containing the async overflow data -- splice() would be mutually exclusive with mmap() of the data. Signed-off-by:
-
H. Peter Anvin authored
Impact: build fix mutex_lock() is was defined inline in kernel/mutex.c, but wasn't declared so not in <linux/mutex.h>. This didn't cause a problem until checkin 3a2d367d9aabac486ac4444c6c7ec7a1dab16267 added the atomic_dec_and_mutex_lock() inline in between declaration and definion. This broke building with CONFIG_ALLOW_WARNINGS=n, e.g. make allnoconfig. Either from the source code nor the allnoconfig binary output I cannot find any internal references to mutex_lock() in kernel/mutex.c, so presumably this "inline" is now-useless legacy. Cc: Eric Paris <eparis@redhat.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Orig-LKML-Reference: <tip-3a2d367d9aabac486ac4444c6c7ec7a1dab16267@git.kernel.org> Signed-off-by:H. Peter Anvin <hpa@linux.intel.com>
-
Eric Paris authored
Much like the atomic_dec_and_lock() function in which we take an hold a spin_lock if we drop the atomic to 0 this function takes and holds the mutex if we dec the atomic to 0. Signed-off-by:
Eric Paris <eparis@redhat.com> Signed-off-by:
-
Paul Mackerras authored
Impact: new feature giving performance improvement This adds the ability for userspace to do an mmap on a hardware counter fd and get access to a read-only page that contains the information needed to translate a hardware counter value to the full 64-bit counter value that would be returned by a read on the fd. This is useful on architectures that allow user programs to read the hardware counters, such as PowerPC. The mmap will only succeed if the counter is a hardware counter monitoring the current process. On my quad 2.5GHz PowerPC 970MP machine, userspace can read a counter and translate it to the full 64-bit value in about 30ns using the mmapped page, compared to about 830ns for the read syscall on the counter, so this does give a significant performance improvement. Signed-off-by:
-
Peter Zijlstra authored
Tracepoint events like lock_acquire and software counters like pagefaults can recurse into the perf counter code again, avoid that. Signed-off-by:
-
Peter Zijlstra authored
Since the bitfields turned into a bit of a mess, remove them and rely on good old masks. Signed-off-by:
-
Wu Fengguang authored
Signed-off-by:
Wu Fengguang <fengguang.wu@intel.com> Acked-by:
-
Wu Fengguang authored
In my system, it takes kerneltop dozens of minutes to show up usable numbers. Make the default count 100 times smaller fixed this long startup latency. I'm not sure if it's the right solution though. Signed-off-by:
-
Wu Fengguang authored
Signed-off-by:
-
Wu Fengguang authored
Signed-off-by:
-
Wu Fengguang authored
- perfstat.c can be safely removed now - perfstat: -s => -a for system wide accounting - kerneltop: add -S/--stat for perfstat mode - minor adjustments to kerneltop --help, perfstat --help Signed-off-by:
-
Wu Fengguang authored
- can handle sw counters now - the outputs will look slightly different Acked-by:
-
Wu Fengguang authored
- kerneltop: --event_id => --event - kerneltop: can accept SW event types now - perfstat: it used to implicitly add event -2(task-clock), the new code no longer does this. Shall we? Signed-off-by:
-
Wu Fengguang authored
Signed-off-by:
-
Wu Fengguang authored
kerneltop's MAX_COUNTERS is increased from 8 to 64(the value used by perfstat). Signed-off-by:
-
Ingo Molnar authored
Initial version of kerneltop.c and perfstat.c. Acked-by:
-
Ingo Molnar authored
We'll have more files in that directory, prepare for that. Acked-by:
-
Paul Mackerras authored
Impact: build fix for powerpc Commit db3a944aca35ae61 ("perf_counter: revamp syscall input ABI") expanded the hw_event.type field into a union of structs containing bitfields. In particular it introduced a type field and a raw_type field, with the intention that the 1-bit raw_type field should overlay the most-significant bit of the 8-bit type field, and in fact perf_counter_alloc() now assumes that (or at least, assumes that raw_type doesn't overlay any of the bits that are 1 in the values of PERF_TYPE_{HARDWARE,SOFTWARE,TRACEPOINT}). Unfortunately this is not true on big-endian systems such as PowerPC, where bitfields are laid out from left to right, i.e. from most significant bit to least significant. This means that setting hw_event.type = PERF_TYPE_SOFTWARE will set hw_event.raw_type to 1. This fixes it by making the layout depend on whether or not __BIG_ENDIAN_BITFIELD is defined. It's a bit ugly, but that's what we get for using bitfields in a user/kernel ABI. Also, that commit didn't fix up some places in arch/powerpc/kernel/ perf_counter.c where hw_event.raw and hw_event.event_id were used. This fixes them too. Signed-off-by: -
Paul Mackerras authored
Impact: cleanup This updates the powerpc perf_counter_interrupt following on from the "perf_counter: unify irq output code" patch. Since we now use the generic perf_counter_output code, which sets the perf_counter_pending flag directly, we no longer need the need_wakeup variable. This removes need_wakeup and makes perf_counter_interrupt use get_perf_counter_pending() instead. Signed-off-by:
-
Peter Zijlstra authored
Impact: cleanup Having 3 slightly different copies of the same code around does nobody any good. First step in revamping the output format. Signed-off-by:
-
Peter Zijlstra authored
Impact: modify ABI The hardware/software classification in hw_event->type became a little strained due to the addition of tracepoint tracing. Instead split up the field and provide a type field to explicitly specify the counter type, while using the event_id field to specify which event to use. Raw counters still work as before, only the raw config now goes into raw_event. Signed-off-by:
-
