- 16 Mar, 2012 10 commits
-
-
Namhyung Kim authored
Now user can enter symbol name interested via ui_browser__input_window, and perf can process it using hists__filter_by_symbol(). Giving empty symbol (by pressing 's' followed by ENTER) will disable the filtering. Cc: Ingo Molnar <mingo@elte.hu> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1331887855-874-3-git-send-email-namhyung.kim@lge.comSigned-off-by:
Namhyung Kim <namhyung.kim@lge.com> Signed-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
Namhyung Kim authored
The ui_browser__input_window() function is to get user's key input. Current implementation can handle maximum 49 characters. Cc: Ingo Molnar <mingo@elte.hu> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1331887855-874-2-git-send-email-namhyung.kim@lge.comSigned-off-by:
-
Namhyung Kim authored
This function will be used for simple (sub-)string matching filter based on user input. Cc: Ingo Molnar <mingo@elte.hu> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1331887855-874-1-git-send-email-namhyung.kim@lge.comSigned-off-by:
-
Namhyung Kim authored
When event group is enabled for forked task (i.e. no target task/cpu was specified) all events were disabled and marked ->enable_on_exec. However they wouldn't be counted at all since only group leader will be enabled on exec actually. In contrast to perf stat, perf record doesn't have a real problem as it enables all the event before proceeding. But it needs to be fixed anyway IMHO. Cc: Ingo Molnar <mingo@elte.hu> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1331887340-32448-2-git-send-email-namhyung.kim@lge.comSigned-off-by:
-
Namhyung Kim authored
When event group is enabled for forked task (i.e. no target task was specified) all events were disabled and marked ->enable_on_exec. However they are not counted at all since only group leader will be enabled on exec actually. So the result looked like below: $ ./perf stat --group -- sleep 1 Performance counter stats for 'sleep 1': 0.554926 task-clock # 0.001 CPUs utilized <not counted> context-switches <not counted> CPU-migrations <not counted> page-faults <not counted> cycles <not supported> stalled-cycles-frontend <not supported> stalled-cycles-backend <not counted> instructions <not counted> branches <not counted> branch-misses 1.001228093 seconds time elapsed Fix it by disabling group leader only. Cc: Ingo Molnar <mingo@elte.hu> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1331887340-32448-1-git-send-email-namhyung.kim@lge.comSigned-off-by: -
Jiri Olsa authored
Added new event rule to the event definition grammar: event_def: event_pmu | ... event_pmu: PE_NAME '/' event_config '/' Using this rule, event could be now specified like: cpu/config=1,config1=2,config2=3/u where pmu name 'cpu' is looked up via following path: ${sysfs_mount}/bus/event_source/devices/${pmu} and config options are bound to the pmu's format definiton: ${sysfs_mount}/bus/event_source/devices/${pmu}/format The hardcoded config options still stays and have precedence over any format field defined with same name. Acked-by:Peter Zijlstra <peterz@infradead.org> Signed-off-by:
Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/n/tip-50d8nr94f8k4wkezutrxvthe@git.kernel.orgSigned-off-by:
-
Jiri Olsa authored
Adding pmu object which provides interface to pmu's sysfs event format definition located at: ${sysfs_mount}/bus/event_source/devices/${pmu}/format Following interface is exported: struct perf_pmu* perf_pmu__find(char *name); - this function returns pmu object, which is then passed as a handle to other interface functions int perf_pmu__config(struct perf_pmu *pmu, struct perf_event_attr *attr, struct list_head *head_terms); - this function configures perf_event_attr struct based on pmu's format definitions and config terms data, containined in head_terms list. Parser generator is used to retrive the pmu's format definition. The generated parser is part of the patch. Added makefile rule 'pmu-parser' to generate the parser code out of the bison/flex sources. Added builtin test 'Test perf pmu format parsing', which could be run like: perf test pmu Acked-by: -
Jiri Olsa authored
Adding a new rule to the event grammar to be able to specify values of additional attributes of symbolic event. The new syntax for event symbolic definition is: event_legacy_symbol: PE_NAME_SYM '/' event_config '/' | PE_NAME_SYM sep_slash_dc event_config: event_config ',' event_term | event_term event_term: PE_NAME '=' PE_NAME | PE_NAME '=' PE_VALUE PE_NAME sep_slash_dc: '/' | ':' | At the moment the config options are hardcoded to be used for legacy symbol events to define several perf_event_attr fields. It is: 'config' to define perf_event_attr::config 'config1' to define perf_event_attr::config1 'config2' to define perf_event_attr::config2 'period' to define perf_event_attr::sample_period Legacy events could be now specified as: cycles/period=100000/ If term is specified without the value assignment, then 1 is assigned by default. Acked-by: -
Jiri Olsa authored
Changing event parsing to use flex/bison parse generator. The event syntax stays as it was. grammar description: events: events ',' event | event event: event_def PE_MODIFIER_EVENT | event_def event_def: event_legacy_symbol sep_dc | event_legacy_cache sep_dc | event_legacy_breakpoint sep_dc | event_legacy_tracepoint sep_dc | event_legacy_numeric sep_dc | event_legacy_raw sep_dc event_legacy_symbol: PE_NAME_SYM event_legacy_cache: PE_NAME_CACHE_TYPE '-' PE_NAME_CACHE_OP_RESULT '-' PE_NAME_CACHE_OP_RESULT | PE_NAME_CACHE_TYPE '-' PE_NAME_CACHE_OP_RESULT | PE_NAME_CACHE_TYPE event_legacy_raw: PE_SEP_RAW PE_VALUE event_legacy_numeric: PE_VALUE ':' PE_VALUE event_legacy_breakpoint: PE_SEP_BP ':' PE_VALUE ':' PE_MODIFIER_BP event_breakpoint_type: PE_MODIFIER_BPTYPE | empty PE_NAME_SYM: cpu-cycles|cycles | stalled-cycles-frontend|idle-cycles-frontend | stalled-cycles-backend|idle-cycles-backend | instructions | cache-references | cache-misses | branch-instructions|branches | branch-misses | bus-cycles | cpu-clock | task-clock | page-faults|faults | minor-faults | major-faults | context-switches|cs | cpu-migrations|migrations | alignment-faults | emulation-faults PE_NAME_CACHE_TYPE: L1-dcache|l1-d|l1d|L1-data | L1-icache|l1-i|l1i|L1-instruction | LLC|L2 | dTLB|d-tlb|Data-TLB | iTLB|i-tlb|Instruction-TLB | branch|branches|bpu|btb|bpc | node PE_NAME_CACHE_OP_RESULT: load|loads|read | store|stores|write | prefetch|prefetches | speculative-read|speculative-load | refs|Reference|ops|access | misses|miss PE_MODIFIER_EVENT: [ukhp]{0,5} PE_MODIFIER_BP: [rwx] PE_SEP_BP: 'mem' PE_SEP_RAW: 'r' sep_dc: ':' | Added flex/bison files for event grammar parsing. The generated parser is part of the patch. Added makefile rule 'event-parser' to generate the parser code out of the bison/flex sources. Acked-by: -
Jiri Olsa authored
Adding sysfs group 'format' attribute for pmu device that contains a syntax description on how to construct raw events. The event configuration is described in following struct pefr_event_attr attributes: config config1 config2 Each sysfs attribute within the format attribute group, describes mapping of name and bitfield definition within one of above attributes. eg: "/sys/...<dev>/format/event" contains "config:0-7" "/sys/...<dev>/format/umask" contains "config:8-15" "/sys/...<dev>/format/usr" contains "config:16" the attribute value syntax is: line: config ':' bits config: 'config' | 'config1' | 'config2" bits: bits ',' bit_term | bit_term bit_term: VALUE '-' VALUE | VALUE Adding format attribute definitions for x86 cpu pmus. Acked-by:
-
- 15 Mar, 2012 1 commit
-
-
Jan Beulich authored
Add rules to generate pre-processed files (just like are available for the normal kernel build), and adjust the rule to create assembly files from C ones to produce its output in the output directory rather than in the source tree. E.g. $ make -C tools/perf/ O=/tmp/perf /tmp/perf/builtin-top.i make: Entering directory `/home/git/linux/tools/perf' CC /tmp/perf/builtin-top.i make: Leaving directory `/home/git/linux/tools/perf' $ wc -l /tmp/perf/builtin-top.i 31379 /tmp/perf/builtin-top.i $ Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/4F588A0802000078000770FF@nat28.tlf.novell.comSigned-off-by:Jan Beulich <jbeulich@suse.com> Signed-off-by:
-
- 12 Mar, 2012 6 commits
-
-
Ingo Molnar authored
Merge reason: The 'perf record -b' hardware branch sampling feature is ready for upstream. Signed-off-by:Ingo Molnar <mingo@elte.hu>
-
Stephane Eranian authored
This patch fixes perf report to not go back two levels when pressing the 'q' key while annotating in branch view mode. When pressing 'q' in annotate mode and if the branch source and target belong to different functions, perf now brings up the annotation popup menu again to offer the option to annotate the other branch source or target. As part of the code restructuring in perf_evsel__hists_browse() we also fix a memory leak on options[] in case of error. Signed-off-by:
Stephane Eranian <eranian@google.com> Cc: peterz@infradead.org Cc: acme@redhat.com Cc: asharma@fb.com Cc: ravitillo@lbl.gov Cc: vweaver1@eecs.utk.edu Cc: khandual@linux.vnet.ibm.com Cc: dsahern@gmail.com Link: http://lkml.kernel.org/r/1331565210-10865-3-git-send-email-eranian@google.comSigned-off-by:
-
Stephane Eranian authored
This patch removes the duplicated annotate selection when browsing in branch view mode. If the sym and dso oof the branch source and target are the same, then only one annotate choice is proposed. Signed-off-by:
-
Peter Zijlstra authored
I got somewhat tired of having to decode hex numbers.. Signed-off-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by:
Thomas Gleixner <tglx@linutronix.de> Cc: Stephane Eranian <eranian@google.com> Cc: Robert Richter <robert.richter@amd.com> Link: http://lkml.kernel.org/n/tip-0vsy1sgywc4uar3mu1szm0rg@git.kernel.orgSigned-off-by:
-
Ingo Molnar authored
Merge reason: We are going to queue up a dependent patch. Signed-off-by: -
Peter Zijlstra authored
Verified using the below proglet.. before: [root@westmere ~]# perf stat -e node-stores -e node-store-misses ./numa 0 remote write Performance counter stats for './numa 0': 2,101,554 node-stores 2,096,931 node-store-misses 5.021546079 seconds time elapsed [root@westmere ~]# perf stat -e node-stores -e node-store-misses ./numa 1 local write Performance counter stats for './numa 1': 501,137 node-stores 199 node-store-misses 5.124451068 seconds time elapsed After: [root@westmere ~]# perf stat -e node-stores -e node-store-misses ./numa 0 remote write Performance counter stats for './numa 0': 2,107,516 node-stores 2,097,187 node-store-misses 5.012755149 seconds time elapsed [root@westmere ~]# perf stat -e node-stores -e node-store-misses ./numa 1 local write Performance counter stats for './numa 1': 2,063,355 node-stores 165 node-store-misses 5.082091494 seconds time elapsed #define _GNU_SOURCE #include <sched.h> #include <stdio.h> #include <errno.h> #include <sys/mman.h> #include <sys/types.h> #include <dirent.h> #include <signal.h> #include <unistd.h> #include <numaif.h> #include <stdlib.h> #define SIZE (32*1024*1024) volatile int done; void sig_done(int sig) { done = 1; } int main(int argc, char **argv) { cpu_set_t *mask, *mask2; size_t size; int i, err, t; int nrcpus = 1024; char *mem; unsigned long nodemask = 0x01; /* node 0 */ DIR *node; struct dirent *de; int read = 0; int local = 0; if (argc < 2) { printf("usage: %s [0-3]\n", argv[0]); printf(" bit0 - local/remote\n"); printf(" bit1 - read/write\n"); exit(0); } switch (atoi(argv[1])) { case 0: printf("remote write\n"); break; case 1: printf("local write\n"); local = 1; break; case 2: printf("remote read\n"); read = 1; break; case 3: printf("local read\n"); local = 1; read = 1; break; } mask = CPU_ALLOC(nrcpus); size = CPU_ALLOC_SIZE(nrcpus); CPU_ZERO_S(size, mask); node = opendir("/sys/devices/system/node/node0/"); if (!node) perror("opendir"); while ((de = readdir(node))) { int cpu; if (sscanf(de->d_name, "cpu%d", &cpu) == 1) CPU_SET_S(cpu, size, mask); } closedir(node); mask2 = CPU_ALLOC(nrcpus); CPU_ZERO_S(size, mask2); for (i = 0; i < size; i++) CPU_SET_S(i, size, mask2); CPU_XOR_S(size, mask2, mask2, mask); // invert if (!local) mask = mask2; err = sched_setaffinity(0, size, mask); if (err) perror("sched_setaffinity"); mem = mmap(0, SIZE, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0); err = mbind(mem, SIZE, MPOL_BIND, &nodemask, 8*sizeof(nodemask), MPOL_MF_MOVE); if (err) perror("mbind"); signal(SIGALRM, sig_done); alarm(5); if (!read) { while (!done) { for (i = 0; i < SIZE; i++) mem[i] = 0x01; } } else { while (!done) { for (i = 0; i < SIZE; i++) t += *(volatile char *)(mem + i); } } return 0; } Signed-off-by:
-
- 10 Mar, 2012 4 commits
-
-
Linus Torvalds authored
-
Al Viro authored
Current code has put_ioctx() called asynchronously from aio_fput_routine(); that's done *after* we have killed the request that used to pin ioctx, so there's nothing to stop io_destroy() waiting in wait_for_all_aios() from progressing. As the result, we can end up with async call of put_ioctx() being the last one and possibly happening during exit_mmap() or elf_core_dump(), neither of which expects stray munmap() being done to them... We do need to prevent _freeing_ ioctx until aio_fput_routine() is done with that, but that's all we care about - neither io_destroy() nor exit_aio() will progress past wait_for_all_aios() until aio_fput_routine() does really_put_req(), so the ioctx teardown won't be done until then and we don't care about the contents of ioctx past that point. Since actual freeing of these suckers is RCU-delayed, we don't need to bump ioctx refcount when request goes into list for async removal. All we need is rcu_read_lock held just over the ->ctx_lock-protected area in aio_fput_routine(). Signed-off-by:
Al Viro <viro@zeniv.linux.org.uk> Reviewed-by:
Jeff Moyer <jmoyer@redhat.com> Acked-by:
Benjamin LaHaise <bcrl@kvack.org> Cc: stable@vger.kernel.org Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Al Viro authored
Have ioctx_alloc() return an extra reference, so that caller would drop it on success and not bother with re-grabbing it on failure exit. The current code is obviously broken - io_destroy() from another thread that managed to guess the address io_setup() would've returned would free ioctx right under us; gets especially interesting if aio_context_t * we pass to io_setup() points to PROT_READ mapping, so put_user() fails and we end up doing io_destroy() on kioctx another thread has just got freed... Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfsLinus Torvalds authored
Pull btrfs updates from Chris Mason: "I have two additional and btrfs fixes in my for-linus branch. One is a casting error that leads to memory corruption on i386 during scrub, and the other fixes a corner case in the backref walking code (also triggered by scrub)." * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: Btrfs: fix casting error in scrub reada code btrfs: fix locking issues in find_parent_nodes()
-
- 09 Mar, 2012 19 commits
-
-
Hugh Dickins authored
Respectfully revert commit e6ca7b89 "memcg: fix mapcount check in move charge code for anonymous page" for the 3.3 release, so that it behaves exactly like releases 2.6.35 through 3.2 in this respect. Horiguchi-san's commit is correct in itself, 1 makes much more sense than 2 in that check; but it does not go far enough - swapcount should be considered too - if we really want such a check at all. We appear to have reached agreement now, and expect that 3.4 will remove the mapcount check, but had better not make 3.3 different. Signed-off-by:
Hugh Dickins <hughd@google.com> Reviewed-by:
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Signed-off-by:
-
Thomas Gleixner authored
Commit f0fbf0ab ("x86: integrate delay functions") converted delay_tsc() into a random delay generator for 64 bit. The reason is that it merged the mostly identical versions of delay_32.c and delay_64.c. Though the subtle difference of the result was: static void delay_tsc(unsigned long loops) { - unsigned bclock, now; + unsigned long bclock, now; Now the function uses rdtscl() which returns the lower 32bit of the TSC. On 32bit that's not problematic as unsigned long is 32bit. On 64 bit this fails when the lower 32bit are close to wrap around when bclock is read, because the following check if ((now - bclock) >= loops) break; evaluated to true on 64bit for e.g. bclock = 0xffffffff and now = 0 because the unsigned long (now - bclock) of these values results in 0xffffffff00000001 which is definitely larger than the loops value. That explains Tvortkos observation: "Because I am seeing udelay(500) (_occasionally_) being short, and that by delaying for some duration between 0us (yep) and 491us." Make those variables explicitely u32 again, so this works for both 32 and 64 bit. Reported-by:
Tvrtko Ursulin <tvrtko.ursulin@onelan.co.uk> Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/soundLinus Torvalds authored
Pull sound fixes from Takashi Iwai: "Nothing exciting here: just a few regression fixes for HD-audio and ASoC, also the support of missing 32bit compat ioctl for HDSPM." * tag 'sound-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: ALSA: hdspm - Provide ioctl_compat ALSA: hda/realtek - Apply the coef-setup only to ALC269VB ALSA: hda - add quirk to detect CD input on Gigabyte EP45-DS3 ASoC: neo1973: fix neo1973 wm8753 initialization
-
David Brown authored
The msm git tree moved to git://git.kernel.org/pub/scm/linux/kernel/git/davidb/linux-msm.gitSigned-off-by:
David Brown <davidb@codeaurora.org> Signed-off-by:
-
git://linux-c6x.org/git/projects/linux-c6x-upstreamingLinus Torvalds authored
Pull C6X fix from Mark Salter: "Fix for C6X KSTK_EIP and KSTK_ESP macros." * tag 'for-linus' of git://linux-c6x.org/git/projects/linux-c6x-upstreaming: C6X: fix KSTK_EIP and KSTK_ESP macros
-
git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommuLinus Torvalds authored
Pull two IOMMU fixes from Joerg Roedel: "The first is an additional fix for the OMAP initialization order issue and the second patch fixes a possible section mismatch which can lead to a kernel crash in the AMD IOMMU driver when suspend/resume is used and the compiler has not inlined the iommu_set_device_table function." * tag 'iommu-fixes-v3.3-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: x86/amd: iommu_set_device_table() must not be __init ARM: OMAP: fix iommu, not mailbox
-
git://people.freedesktop.org/~airlied/linuxLinus Torvalds authored
Pull radeon drm stuff from Dave Airlie: "Just some radeon fixes, one is for an oops where we run out of ioremap space on some big hardware systems in 32-bit mode, stuff doesn't work properly but at least the machine will boot. One regression fix, and two bugs, one hw, one blit code." * 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: drm/radeon/kms: fix hdmi duallink checks drm/radeon/kms: set SX_MISC in the r6xx blit code (v2) drm/radeon: deal with errors from framebuffer init path. drm/radeon: fix a semaphore deadlock on pre cayman asics
-
git://git.kernel.org/pub/scm/linux/kernel/git/davem/netLinus Torvalds authored
Pull networking from David Miller: 1) IPV4 routing metrics can become stale when routes are changed by the administrator, fix from Steffen Klassert. 2) atl1c does "val |= XXX;" where XXX is a bit number not a bit mask, fix by using set_bit. From Dan Carpenter. 3) Memory accounting bug in carl9170 driver results in wedged TX queue. Fix from Nicolas Cavallari. 4) iwlwifi accidently uses "sizeof(ptr)" instead of "sizeof(*ptr)", fix from Johannes Berg. 5) Openvswitch doesn't honor dp_ifindex when doing vport lookups, fix from Ben Pfaff. 6) ehea conversion to 64-bit stats lost multicast and rx_errors accounting, fix from Eric Dumazet. 7) Bridge state transition logging in br_stp_disable_port() is busted, it's emitted at the wrong time and the message is in the wrong tense, fix from Paulius Zaleckas. 8) mlx4 device erroneously invokes the queue resize firmware operation twice, fix from Jack Morgenstein. 9) Fix deadlock in usbnet, need to drop lock when invoking usb_unlink_urb() otherwise we recurse into taking it again. Fix from Sebastian Siewior. 10) hyperv network driver uses the wrong driver name string, fix from Haiyang Zhang. * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: net/hyperv: Use the built-in macro KBUILD_MODNAME for this driver net/usbnet: avoid recursive locking in usbnet_stop() route: Remove redirect_genid inetpeer: Invalidate the inetpeer tree along with the routing cache mlx4_core: fix bug in modify_cq wrapper for resize flow. atl1c: set ATL1C_WORK_EVENT_RESET bit correctly bridge: fix state reporting when port is disabled bridge: br_log_state() s/entering/entered/ ehea: restore multicast and rx_errors fields openvswitch: Fix checksum update for actions on UDP packets. openvswitch: Honor dp_ifindex, when specified, for vport lookup by name. iwlwifi: fix wowlan suspend mwifiex: reset encryption mode flag before association carl9170: fix frame delivery if sta is in powersave mode carl9170: Fix memory accounting when sta is in power-save mode. -
Stephane Eranian authored
This patch updates perf report to support TUI mode when the perf.data file contains samples with branch stacks. For each row in the report, it is possible to annotate either the source or target of each branch. Signed-off-by:
-
Stephane Eranian authored
This patch enhances perf report to auto-detect when the perf.data file contains samples with branch stacks. That way it is not necessary to use the -b option. To force branch view mode to off, simply use --no-branch-stack. Signed-off-by:
-
Stephane Eranian authored
This patch adds a new feature bit, namely, HEADER_BRANCH_STACK. When present, it indicates that sample records may contain branch stack. This could be useful to a viewer to switch to branch mode without having to parse all the samples or without a specific cmdline option. This will be used in a subsequent patch to enhance perf report with branch stacks. Signed-off-by:
-
Stephane Eranian authored
This patch chanegs the logic of the -b, --branch-stack options of perf record. Based on users' request, the patch provides a default filter mode with the -b (or --branch-any) option. With the option, any type of taken branches is sampled. With -j (or --branch-filter), the user can specify any valid combination of branch types and privilege levels if supported by the underlying hardware. The -b (--branch any) is a shortcut for: --branch-filter any. $ perf record -b foo or: $ perf record --branch-filter any foo For more specific filtering: $ perf record --branch-filter ind_call,u foo Signed-off-by:
-
Stephane Eranian authored
This patches provides a way to handle legacy perf.data files. Legacy files are those using the older PERFFILE signature. For those, it is still necessary to detect endianness but without comparing their header->attr_size with the tool's own version as it may be different. Instead, we use a reference table for all known sizes from the legacy era. We try all the combinations for sizes and endianness. If we find a match, we proceed, otherwise we return: "incompatible file format". This is also done for the pipe-mode file format. Signed-off-by:
-
Stephane Eranian authored
This patches cleans up local variable types for msz and ret. They need to be size_t and ssize_t respectively. It also fixes a bug whereby perf would not read attr struct with a different size than what it knows about. Signed-off-by:
-
Stephane Eranian authored
This patch allows perf to process perf.data files generated using an ABI that has a different perf_event_attr struct size, i.e., a different ABI version. The perf_event_attr can be extended, yet perf needs to cope with older perf.data files. Similarly, perf must be able to cope with a perf.data file which is using a newer version of the ABI than what it knows about. This patch adds read_attr(), a routine that reads a perf_event_attr struct from a file incrementally based on its advertised size. If the on-file struct is smaller than what perf knows, then the extra fields are zeroed. If the on-file struct is bigger, then perf only uses what it knows about, the rest is skipped. Signed-off-by:
-
Stephane Eranian authored
This patch adds reference sizes for revision 1 and 2 of the perf_event ABI, i.e., the size of the perf_event_attr struct. With Rev1: config2 was added = +8 bytes With Rev2: branch_sample_type was added = +8 bytes Adds the definition for Rev1, Rev2. This is useful for tools trying to decode the revision numbers based on the size of the struct. Signed-off-by:
-
Roberto Agostino Vitillo authored
This patch adds support for taken branch sampling, i.e, the PERF_SAMPLE_BRANCH_STACK feature to perf report. In other words, to display histograms based on taken branches rather than executed instructions addresses. The new option is called -b and it takes no argument. To generate meaningful output, the perf.data must have been obtained using perf record -b xxx ... where xxx is a branch filter option. The output shows symbols, modules, sorted by 'who branches where' the most often. The percentages reported in the first column refer to the total number of branches captured and not the usual number of samples. Here is a quick example. Here branchy is simple test program which looks as follows: void f2(void) {} void f3(void) {} void f1(unsigned long n) { if (n & 1UL) f2(); else f3(); } int main(void) { unsigned long i; for (i=0; i < N; i++) f1(i); return 0; } Here is the output captured on Nehalem, if we are only interested in user level function calls. $ perf record -b any_call,u -e cycles:u branchy $ perf report -b --sort=symbol 52.34% [.] main [.] f1 24.04% [.] f1 [.] f3 23.60% [.] f1 [.] f2 0.01% [k] _IO_new_file_xsputn [k] _IO_file_overflow 0.01% [k] _IO_vfprintf_internal [k] _IO_new_file_xsputn 0.01% [k] _IO_vfprintf_internal [k] strchrnul 0.01% [k] __printf [k] _IO_vfprintf_internal 0.01% [k] main [k] __printf About half (52%) of the call branches captured are from main() -> f1(). The second half (24%+23%) is split in two equal shares between f1() -> f2(), f1() ->f3(). The output is as expected given the code. It should be noted, that using -b in perf record does not eliminate information in the perf.data file. Consequently, a typical profile can also be obtained by perf report by simply not using its -b option. It is possible to sort on branch related columns: - dso_from, symbol_from - dso_to, symbol_to - mispredict Signed-off-by:Roberto Agostino Vitillo <ravitillo@lbl.gov> Signed-off-by:
-
Roberto Agostino Vitillo authored
This patch adds a new option to enable taken branch stack sampling, i.e., leverage the PERF_SAMPLE_BRANCH_STACK feature of perf_events. There is a new option to active this mode: -b. It is possible to pass a set of filters to select the type of branches to sample. The following filters are available: - any : any type of branches - any_call : any function call or system call - any_ret : any function return or system call return - any_ind : any indirect branch - u: only when the branch target is at the user level - k: only when the branch target is in the kernel - hv: only when the branch target is in the hypervisor Filters can be combined by passing a comma separated list to the option: $ perf record -b any_call,u -e cycles:u branchy Signed-off-by:
-
Roberto Agostino Vitillo authored
This patch adds: - ability to parse samples with PERF_SAMPLE_BRANCH_STACK - sort on branches (dso_from, symbol_from, dso_to, symbol_to, mispredict) - build histograms on branches Signed-off-by:
-
