- 21 Jan, 2015 27 commits
-
-
Namhyung Kim authored
The prior change fixes default output ordering with each column but it breaks -o/--order option. This patch prepends a new hpp fmt struct to sort list but not to output field list so that it can affect ordering without adding a new output column. The new hpp fmt uses its own compare functions which treats dummy entries (which have no baseline) little differently - the delta field can be computed without baseline but others (ratio and wdiff) are not. The new output will look like below: $ perf diff -o 2 perf.data.{old,cur,new} ... # Baseline/0 Delta/1 Delta/2 Shared Object Symbol # .......... ....... ....... ................. .......................................... 22.98% +0.51% +0.52% libc-2.20.so [.] _int_malloc 5.70% +0.28% +0.30% libc-2.20.so [.] free 4.38% -0.21% +0.25% a.out [.] main 1.32% -0.15% +0.05% a.out [.] free@plt +0.01% [kernel.kallsyms] [k] intel_pstate_timer_func +0.01% [kernel.kallsyms] [k] _raw_spin_lock_irqsave +0.01% [kernel.kallsyms] [k] timekeeping_update.constprop.8 +0.01% +0.01% [kernel.kallsyms] [k] apic_timer_interrupt 0.01% -0.00% [kernel.kallsyms] [k] native_read_msr_safe 0.01% -0.01% -0.01% [kernel.kallsyms] [k] native_write_msr_safe 1.31% +0.03% -0.06% a.out [.] malloc@plt 31.50% -0.74% -0.23% libc-2.20.so [.] _int_free 32.75% +0.28% -0.83% libc-2.20.so [.] malloc 0.01% [kernel.kallsyms] [k] scheduler_tick +0.01% [kernel.kallsyms] [k] read_tsc +0.01% [kernel.kallsyms] [k] perf_adjust_freq_unthr_context.part.82 In above example, the output was sorted by 'Delta/2' column first, and then 'Baseline/0' and finally 'Delta/1'. Signed-off-by:Namhyung Kim <namhyung@kernel.org> Acked-by:
Jiri Olsa <jolsa@kernel.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Kan Liang <kan.liang@intel.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1420677949-6719-8-git-send-email-namhyung@kernel.orgSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
Namhyung Kim authored
When perf diff prints output, it sorts the entries using baseline field by default, but entries which don't have baseline are not sorted properly. This patch makes it sorted by values of next column. Before: # Baseline/0 Delta/1 Delta/2 Shared Object Symbol # .......... ....... ....... ................. .......................................... # 32.75% +0.28% -0.83% libc-2.20.so [.] malloc 31.50% -0.74% -0.23% libc-2.20.so [.] _int_free 22.98% +0.51% +0.52% libc-2.20.so [.] _int_malloc 5.70% +0.28% +0.30% libc-2.20.so [.] free 4.38% -0.21% +0.25% a.out [.] main 1.32% -0.15% +0.05% a.out [.] free@plt 1.31% +0.03% -0.06% a.out [.] malloc@plt 0.01% -0.01% -0.01% [kernel.kallsyms] [k] native_write_msr_safe 0.01% [kernel.kallsyms] [k] scheduler_tick 0.01% -0.00% [kernel.kallsyms] [k] native_read_msr_safe +0.01% [kernel.kallsyms] [k] _raw_spin_lock_irqsave +0.01% +0.01% [kernel.kallsyms] [k] apic_timer_interrupt +0.01% [kernel.kallsyms] [k] intel_pstate_timer_func +0.01% [kernel.kallsyms] [k] perf_adjust_freq_unthr_context.part.82 +0.01% [kernel.kallsyms] [k] read_tsc +0.01% [kernel.kallsyms] [k] timekeeping_update.constprop.8 After: # Baseline/0 Delta/1 Delta/2 Shared Object Symbol # .......... ....... ....... ................. .......................................... # 32.75% +0.28% -0.83% libc-2.20.so [.] malloc 31.50% -0.74% -0.23% libc-2.20.so [.] _int_free 22.98% +0.51% +0.52% libc-2.20.so [.] _int_malloc 5.70% +0.28% +0.30% libc-2.20.so [.] free 4.38% -0.21% +0.25% a.out [.] main 1.32% -0.15% +0.05% a.out [.] free@plt 1.31% +0.03% -0.06% a.out [.] malloc@plt 0.01% -0.01% -0.01% [kernel.kallsyms] [k] native_write_msr_safe 0.01% [kernel.kallsyms] [k] scheduler_tick 0.01% -0.00% [kernel.kallsyms] [k] native_read_msr_safe +0.01% +0.01% [kernel.kallsyms] [k] apic_timer_interrupt +0.01% [kernel.kallsyms] [k] read_tsc +0.01% [kernel.kallsyms] [k] perf_adjust_freq_unthr_context.part.82 +0.01% [kernel.kallsyms] [k] intel_pstate_timer_func +0.01% [kernel.kallsyms] [k] _raw_spin_lock_irqsave +0.01% [kernel.kallsyms] [k] timekeeping_update.constprop.8 Signed-off-by: -
Namhyung Kim authored
Currently ->cmp, ->collapse and ->sort callbacks doesn't pass corresponding fmt. But it'll be needed by upcoming changes in perf diff command. Suggested-by:
-
Namhyung Kim authored
The fmt_to_data_file() is to retrieve struct data__file from perf_hpp_fmt which is embedded in diff_hpp_fmt. It'll be used by sort callback functions later. Signed-off-by:
-
Namhyung Kim authored
Current perf diff result is somewhat confusing since it sometimes hide small result and sometimes there's no result. So do not hide small result (less than 0.01%) and print "N/A" if baseline is not recorded (for ratio and wdiff only). Blank means the baseline is available but its pairs are not. Before: # Baseline Delta Shared Object Symbol # ........ ....... ................. ......................... # ... 0.01% -0.01% [kernel.kallsyms] [k] native_write_msr_safe 0.01% [kernel.kallsyms] [k] scheduler_tick 0.01% [kernel.kallsyms] [k] native_read_msr_safe 0.00% [kernel.kallsyms] [k] __rcu_read_unlock [kernel.kallsyms] [k] _raw_spin_lock +0.01% [kernel.kallsyms] [k] apic_timer_interrupt [kernel.kallsyms] [k] read_tsc After: # Baseline Delta Shared Object Symbol # ........ ....... ................. ......................... # ... 0.01% -0.01% [kernel.kallsyms] [k] native_write_msr_safe 0.01% [kernel.kallsyms] [k] scheduler_tick 0.01% [kernel.kallsyms] [k] native_read_msr_safe 0.00% [kernel.kallsyms] [k] __rcu_read_unlock +0.01% [kernel.kallsyms] [k] _raw_spin_lock +0.01% [kernel.kallsyms] [k] apic_timer_interrupt +0.01% [kernel.kallsyms] [k] read_tsc Signed-off-by: -
Namhyung Kim authored
The hists__compute_resort() is to sort output fields based on the given field/criteria. This was done without the sort list but as we added the field to the sort list, we can do it with normal hists__output_resort() using the ->sort callback. Signed-off-by:
-
Namhyung Kim authored
The exclusive options are to prohibit use of conflicting options at the same time. But it had a side effect that it also limits a such option can be used at most once. Currently the only user of the flag is perf probe and it allows to use such options more than once, but when one tries to use it, perf will fail like below: $ sudo perf probe -x /lib/libc-2.20.so --add malloc --add free Error: option `add' cannot be used with add ... Signed-off-by:Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1420886028-15135-1-git-send-email-namhyung@kernel.orgSigned-off-by:
-
Cody P Schafer authored
Signed-off-by:
Cody P Schafer <cody@linux.vnet.ibm.com> Signed-off-by:
Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> Acked-by:
-
Cody P Schafer authored
Event parameters are a basic way for partial events to be specified in sysfs with per-event names given to the fields that need to be filled in when using a particular event. It is intended for supporting cases where the single 'cpu' parameter is insufficient. For example, POWER 8 has events for physical sockets/cores/cpus that are accessible from with virtual machines. To keep using the single 'cpu' parameter we'd need to perform a mapping between Linux's cpus and the physical machine's cpus (in this case Linux is running under a hypervisor). This isn't possible because bindings between our cpus and physical cpus may not be fixed, and we probably won't have a "cpu" on each physical cpu. Signed-off-by:
-
Cody P Schafer authored
This causes `perf list pmu` to show parameters for parameterized events like: pmu/event_name,param1=?,param2=?/ [Kernel PMU event] An example: hv_24x7/HPM_TLBIE__PHYS_CORE,core=?/ [Kernel PMU event] Signed-off-by:
-
Cody P Schafer authored
Enable event specification like: pmu/event_name,param1=0x1,param2=0x4/ Assuming that /sys/bus/event_source/devices/pmu/events/event_name Contains something like param2=?,bar=1,param1=? Signed-off-by:
-
Rickard Strandqvist authored
Removes some functions that are not used anywhere: color_parse_mem() color_parse() This was partially found by using a static code analysis program called cppcheck. Signed-off-by:
Rickard Strandqvist <rickard_strandqvist@spectrumdigital.se> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Ramkumar Ramachandra <artagnon@gmail.com> Link: http://lkml.kernel.org/r/1419079865-354-1-git-send-email-rickard_strandqvist@spectrumdigital.se [ Remove now unused parse_{attr,color} routines too ] Signed-off-by:
-
Namhyung Kim authored
The report__inc_stat() function collects the number of hist entries in the session in order to calculate the max size of the progess bar. It'd be better if it does it during the addition of hist entries so that it can be used by other places too. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
The code being used when decaying and deleting entries from a hists instance was the same, provide a function to avoid code dup. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-j6ideab7lkakavfvfguw858z@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
No logic changes, just to be consistent. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-f7n5y0mvk6gew5185h6fg316@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
Janitorial stuff: boredom moment. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-u70i7shys3kths4hzru72bha@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
Just like the other parameters, grouping it on the builtin-mem specific config area: struct perf_mem. Acked-by:
Stephane Eranian <eranian@google.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Don Zickus <dzickus@redhat.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Joe Mario <jmario@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Richard Fowles <rfowles@redhat.com> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-ad8ns5l51ongemfsir3zy09x@git.kernel.orgSigned-off-by:
-
Stephane Eranian authored
This patch modifies perf mem to default to sampling loads and stores simultaneously. It could only do one or the other before yet there was no hardware restriction preventing simultaneous collection. With this patch, one run is sufficient to collect both. It is still possible to sample only loads or stores by using the -t option: $ perf mem -t load rec $ perf mem -t load rep Or $ perf mem -t store rec $ perf mem -t store rep The perf report TUI will show one event at a time. The store output will contain a Weight column which will be empty. In V2, we updated the man pages to reflect the change and also simplify the initialization of the argv vector passed to the cmd_*() functions as per LKML feedback. In V3, we fixed typos in the changelog. Signed-off-by:
-
Namhyung Kim authored
The commit dfef99cd ("perf probe: Use ref_reloc_sym based address instead of the symbol name") converts kprobes to use ref_reloc_sym (i.e. _stext) and offset instead of using symbol's name directly. So on my system, adding do_fork ends up with like below: $ sudo perf probe -v --add do_fork%return probe-definition(0): do_fork%return symbol:do_fork file:(null) line:0 offset:0 return:1 lazy:(null) 0 arguments Looking at the vmlinux_path (7 entries long) Using /lib/modules/3.17.6-1-ARCH/build/vmlinux for symbols Could not open debuginfo. Try to use symbols. Opening /sys/kernel/debug/tracing/kprobe_events write=1 Added new event: Writing event: r:probe/do_fork _stext+456136 Failed to write event: Invalid argument Error: Failed to add events. Reason: Operation not permitted (Code: -1) As you can see, the do_fork was translated to _stext+456136. This was because to support (local) symbols that have same name. But the problem is that kretprobe requires to be inserted at function start point so it simply checks whether it's called with offset 0. And if not, it'll return with -EINVAL. You can see it with dmesg. $ dmesg | tail -1 [125621.764103] Return probe must be used without offset. So we need to use the symbol name instead of ref_reloc_sym in case of return probes. Reported-by:
Jiri Olsa <jolsa@redhat.com> Signed-off-by:
-
Arnaldo Carvalho de Melo authored
Removing boilerplate from two places, where one would have to find the first entry, then iterate using symbol__next_by_name + strcmp to see if the next member had the same name. Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Link: http://lkml.kernel.org/n/tip-eh73z8gthv20yowirmx2yk38@git.kernel.orgSigned-off-by:
-
Namhyung Kim authored
The find_probe_trace_events_from_map() searches matching symbol from a map (so from a backing dso). For uprobes, it'll create a new map (and dso) and loads it using a filter. It's a little bit inefficient in that it'll read out the symbol table everytime but works well anyway. For kprobes however, it'll reuse existing kernel map which might be loaded before. In this case map__load() just returns with no result. It makes kprobes always failed to find symbol even if it exists in the map (dso). To fix it, use map__find_symbol_by_name() instead. It'll load a map with full symbols and sorts them by name. It needs to search sibing nodes since there can be multiple (local) symbols with same name. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
Given a symbol, go to the next entry in a rbtree sorted by symbol name. Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Link: http://lkml.kernel.org/n/tip-aq210drxprnu2so4dye5xa3j@git.kernel.orgSigned-off-by:
-
Namhyung Kim authored
When a dso contains multiple symbols which have same name, current dso__find_symbol_by_name() only finds an one of them and there's no way to get the all symbols without going through the rbtree. So make symbols__find_by_name() return the first entry with the given name and the next patch in this series will provide a way to iterate from there, by the name ordered rb_tree, till a suitable symbol is found. Signed-off-by:
-
Rabin Vincent authored
The lock prefix handling fails to free the strdup()'d name as well as the fields allocated by the instruction parsing. Signed-off-by:
Rabin Vincent <rabin@rab.in> Cc: Paul Mackerras <paulus@samba.org> Link: http://lkml.kernel.org/r/1421607621-15005-2-git-send-email-rabin@rab.inSigned-off-by:
-
Rabin Vincent authored
Don't use the ins's ->sncprintf() if the parsing failed. For example, this fixes the display of "imul %edx". Without this patch: | imul (null),(null) After this patch: | imul %edx Signed-off-by: -
Wang Nan authored
When building perf for arm64 I hit a warning (and be treated as an error) like below: aarch64-oe-linux-gcc -o .../scripts/perl/Perf-Trace-Util/Context.o -c -Wbad-function-cast \ ... scripts/perl/Perf-Trace-Util/Context.c In file included from .../usr/lib64/perl/5.14.3/CORE/perl.h:2464:0, from Context.xs:23: /.../usr/lib64/perl/5.14.3/CORE/handy.h:108:0: error: "bool" redefined [-Werror] # define bool char ^ In file included from /.../usr/src/kernel/tools/include/linux/types.h:4:0, from /.../usr/src/kernel/arch/arm64/include/uapi/asm/sigcontext.h:19, from /.../usr/include/bits/sigcontext.h:27, from /.../usr/include/signal.h:340, from /.../usr/include/sys/param.h:28, from /.../usr/lib64/perl/5.14.3/CORE/perl.h:678, from Context.xs:23: /.../usr/lib/aarch64-oe-linux/gcc/aarch64-oe-linux/4.9.2/include/stdbool.h:33:0: note: this is the location of the previous definition #define bool _Bool Looks like the failure is caused by arm64 uapi/asm/sigcontext.h, which includes linux/types.h while other archs not. Current perl consider this problem: http://perl5.git.perl.org/perl.git/commit/bd31be4baa3ee68abdb92c0db3200efe0fad903b However there are users which use old version of perl. This patch includes stdbool.h before Context.xs and define HAS_BOOL to prevent perl'e headers define its own 'bool'. Code is learn from perl's git tree. Signed-off-by:Wang Nan <wangnan0@huawei.com> Acked-by:
-
Arnaldo Carvalho de Melo authored
[acme@mica ~]$ trace -p 3330 Error: Unable to find debugfs Hint: Was your kernel was compiled with debugfs support? ^^^ ^^^ Hint: Is the debugfs filesystem mounted? Hint: Try 'sudo mount -t debugfs nodev /sys/kernel/debug' Fix it. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-kb9s0xy5z8i51abdu4bgm3rv@git.kernel.orgSigned-off-by:
-
- 17 Jan, 2015 1 commit
-
-
Ingo Molnar authored
Merge tag 'perf-urgent-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent Pull perf/urgent fixes from Arnaldo Carvalho de Melo: - Fix segfault when using both the map symtab viewer and annotation in the TUI (Namhyung Kim). - uClibc build fixes (Alexey Brodkin, Vineet Gupta). - bitops/hweight were moved from tools/perf/ too tools/include, move some leftovers (Arnaldo Carvalho de Melo) - Fix dwarf unwind x86_64 build error (Namhyung Kim) - Fix __machine__findnew_thread() error path (Namhyung Kim) - Propagate error code when write(2) failed in 'perf probe' (Namhyung Kim) - Use dwfl_report_elf() instead of offline in powerpc bits to properly handle non prelinked DSOs (Sukadev Bhattiprolu). - Fix dwarf unwind using libunwind in 'perf test' (Wang Nan) Signed-off-by:
Ingo Molnar <mingo@kernel.org>
-
- 16 Jan, 2015 12 commits
-
-
Sukadev Bhattiprolu authored
dwfl_report_offline() works only when libraries are prelinked. Replace dwfl_report_offline() with dwfl_report_elf() so we correctly extract debug info even from libraries that are not prelinked. Reported-by:
-
Namhyung Kim authored
Currently the symbol structure is allocated with symbol_conf.priv_size to carry sideband information like annotation, map browser on TUI and sort-by-name tree node. So retrieving these information from symbol needs to care about the details of such placement. However the annotation code just assumes that the symbol is placed after the struct annotation. But actually there's other info between them. So accessing those struct will lead to an undefined behavior (usually a crash) after they write their info to the same location. To reproduce the problem, please follow the steps below: 1. run perf report (TUI of course) with -v option 2. open map browser (by pressing right arrow key for any entry) 3. search any function (by pressing '/' key and input whatever..) 4. return to the hist browser (by pressing 'q' or left arrow key) 5. open annotation window for the same entry (by pressing 'a' key) Signed-off-by:
-
Wang Nan authored
Perf tool fails to unwind user stack if the event raises in a shared object. This patch improves tests/dwarf-unwind.c to demonstrate the problem by utilizing commonly used glibc function "bsearch". If perf is not statically linked, the testcase will try to unwind a mixed call trace. By debugging libunwind I found that there is a bug in unwind-libunwind: it always passes 0 as segbase to libunwind, cause libunwind unable to locate debug_frame entry fir first level ip address (I add some more debugging output into libunwind to make things clear): >_Uarm_dwarf_find_debug_frame: start_ip = 10be98, end_ip = 10c2a4 >_Uarm_dwarf_find_debug_frame: found debug_frame table `/lib/libc-2.18.so': segbase=0x0, len=7, gp=0x0, table_data=0x449388 >_Uarm_dwarf_search_unwind_table: call lookup:ip = b6cd3bcc, segbase = 0, rel_ip = b6cd3bcc >lookup: e->start_ip_offset = bcf18 (rel_ip = b6cd3bcc) >lookup: e->start_ip_offset = 6d314 (rel_ip = b6cd3bcc) >lookup: e->start_ip_offset = 33d0c (rel_ip = b6cd3bcc) ... >lookup: e->start_ip_offset = 15d0c (rel_ip = b6cd3bcc) >lookup: e->start_ip_offset = 15c40 (rel_ip = b6cd3bcc) >_Uarm_dwarf_search_unwind_table: IP b6cd3bcc inside range b6c12000-b6d4c000, but no explicit unwind info found >put_rs_cache: unmasking signals/interrupts and releasing lock >_Uarm_dwarf_step: returning -10 >_Uarm_step: dwarf_step()=-10 This patch passes map->start as segbase to dwarf_find_debug_frame(), so di will be initialized correctly. In addition, dso and executable are different when setting segbase. This patch first check whether the elf is executable, and pass segbase only for shared object. Signed-off-by: -
Vineet Gupta authored
This is due to duplicated unistd inclusion (via uClibc headers + kernel headers) Also seen on ARM uClibc based tools ------- ARC build ---------->8------------- CC util/evlist.o In file included from ~/arc/k.org/arch/arc/include/uapi/asm/unistd.h:25:0, from util/../perf-sys.h:10, from util/../perf.h:15, from util/event.h:7, from util/event.c:3: ~/arc/k.org/include/uapi/asm-generic/unistd.h:906:0: warning: "__NR_fcntl64" redefined [enabled by default] #define __NR_fcntl64 __NR3264_fcntl ^ In file included from ~/arc/gnu/INSTALL_1412-arc-2014.12-rc1/arc-snps-linux-uclibc/sysroot/usr/include/sys/syscall.h:24:0, from util/../perf-sys.h:6, ----------------->8------------------- ------- ARM build ---------->8------------- CC FPIC plugin_scsi.o In file included from util/../perf-sys.h:9:0, from util/../perf.h:15, from util/cache.h:7, from perf.c:12: ~/arc/k.org/arch/arm/include/uapi/asm/unistd.h:28:0: warning: "__NR_restart_syscall" redefined [enabled by default] In file included from ~/buildroot/host/usr/arm-buildroot-linux-uclibcgnueabi/sysroot/usr/include/sys/syscall.h:25:0, from util/../perf-sys.h:6, from util/../perf.h:15, from util/cache.h:7, from perf.c:12: ~/buildroot/host/usr/arm-buildroot-linux-uclibcgnueabi/sysroot/usr/include/bits/sysnum.h:17:0: note: this is the location of the previous definition ----------------->8------------------- Signed-off-by:Vineet Gupta <vgupta@synopsys.com> Cc: Alexey Brodkin <Alexey.Brodkin@synopsys.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1421156604-30603-4-git-send-email-vgupta@synopsys.comSigned-off-by:
-
Vineet Gupta authored
----------------->8------------------ CC bench/sched-pipe.o In file included from builtin-annotate.c:13:0: util/cache.h:76:15: warning: redundant redeclaration of 'strlcpy' [-Wredundant-decls] extern size_t strlcpy(char *dest, const char *src, size_t size); ^ In file included from util/util.h:55:0, from builtin.h:4, from builtin-annotate.c:8: ~/vineetg/arc/gnu/INSTALL_1412-arc-2014.12-rc1/arc-snps-linux-uclibc/sysroot/usr/include/string.h:396:15: note: previous declaration of 'strlcpy' was here extern size_t strlcpy(char *__restrict dst, const char *__restrict src, ----------------->8------------------ Signed-off-by: -
Alexey Brodkin authored
ARC Linux uses the no legacy syscalls abi and corresponding uClibc headers statfs defines f_type to be U32 which causes perf build breakage http://git.uclibc.org/uClibc/tree/libc/sysdeps/linux/common-generic/bits/statfs.h ----------->8--------------- CC fs/fs.o fs/fs.c: In function 'fs__valid_mount': fs/fs.c:82:24: error: comparison between signed and unsigned integer expressions [-Werror=sign-compare] else if (st_fs.f_type != magic) ^ cc1: all warnings being treated as errors ----------->8--------------- Signed-off-by:
Alexey Brodkin <abrodkin@synopsys.com> Acked-by:
-
Arnaldo Carvalho de Melo authored
We need to use lib/hweight.c for that, just like we do for lib/rbtree.c, so tools need to link hweight.o. For now do it directly, but we need to have a tools/lib/lk.a or .so that collects these goodies... Reported-by:
Jan Beulich <JBeulich@suse.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-a1e91dx3apzqw5kbdt7ut21s@git.kernel.orgSigned-off-by:
-
Namhyung Kim authored
When thread__init_map_groups() fails, a new thread should be removed from the rbtree since it's gonna be freed. Also update last match cache only if the function succeeded. Reported-by:
David Ahern <dsahern@gmail.com> Signed-off-by:
-
Namhyung Kim authored
When build with 'make ARCH=x86' and dwarf unwind is on, there is a compiling error: CC /home/wn/perf/arch/x86/util/unwind-libdw.o CC /home/wn/perf/arch/x86/tests/regs_load.o arch/x86/tests/regs_load.S: Assembler messages: arch/x86/tests/regs_load.S:65: Error: operand type mismatch for `push' arch/x86/tests/regs_load.S:72: Error: operand type mismatch for `pop' make[1]: *** [/home/wn/perf/arch/x86/tests/regs_load.o] Error 1 make[1]: INTERNAL: Exiting with 25 jobserver tokens available; should be 24! make: *** [all] Error 2 ... Which is caused by incorrectly undefine macro HAVE_ARCH_X86_64_SUPPORT. 'config/Makefile.arch' tests __x86_64__ only when 'ARCH=x86_64'. However, when building x86_64 kernel, ARCH=x86 is valid and commonly used. Build systems, such as yocto, uses x86_64 compiler with 'ARCH=x86' to build x86_64 perf, which causes mismatching. As __LP64__ is defined for x86_64 as well, we can consolidate the __x86_64__ check to the __LP64__ check and get rid of the IS_X86_64 IMHO. (This patch is made by Namhyung Kim when replying my v1 patch: https://lkml.org/lkml/2015/1/7/17 I modified the code to remove dependency on RAW_ARCH: https://lkml.org/lkml/2015/1/7/865 Namhyung Kim didn't provide his SOB in his original email. I add mine only for my modification.) Signed-off-by:
-
Namhyung Kim authored
When it failed to write probe commands to the probe_event file in debugfs, it needs to propagate the error code properly. Current code blindly uses the return value of the write(2) so it always uses -1 (-EPERM) and it might confuse users. Signed-off-by:
-
Kan Liang authored
cycles:p and cycles:pp do not work on SLM since commit: 86a04461 ("perf/x86: Revamp PEBS event selection") UOPS_RETIRED.ALL is not a PEBS capable event, so it should not be used to count cycle number. Actually SLM calls intel_pebs_aliases_core2() which uses INST_RETIRED.ANY_P to count the number of cycles. It's a PEBS capable event. But inv and cmask must be set to count cycles. Considering SLM allows all events as PEBS with no flags, only INST_RETIRED.ANY_P, inv=1, cmask=16 needs to handled specially. Signed-off-by:
Kan Liang <kan.liang@intel.com> Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Link: http://lkml.kernel.org/r/1421084541-31639-1-git-send-email-kan.liang@intel.com Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
-
Stephane Eranian authored
This patch fixes a problem with the initialization of the sysfs_show() routine for the RAPL PMU. The current code was wrongly relying on the EVENT_ATTR_STR() macro which uses the events_sysfs_show() function in the x86 PMU code. That function itself was relying on the x86_pmu data structure. Yet RAPL and the core PMU (x86_pmu) have nothing to do with each other. They should therefore not interact with each other. The x86_pmu structure is initialized at boot time based on the host CPU model. When the host CPU is not supported, the x86_pmu remains uninitialized and some of the callbacks it contains are NULL. The false dependency with x86_pmu could potentially cause crashes in case the x86_pmu is not initialized while the RAPL PMU is. This may, for instance, be the case in virtualized environments. This patch fixes the problem by using a private sysfs_show() routine for exporting the RAPL PMU events. Signed-off-by:
-
