linux/tools, branch v2.6.37

perf: Fix callchain hit bad cast on ascii display

2011-01-03T15:13:11Z

ipchain__fprintf_graph() casts the number of hits in a branch as an int, which means we lose its highests bits. This results in meaningless number of callchain hits in perf.data that have a high number of hits recorded, typically those that have callchain branches hits appearing more than INT_MAX. This happens easily as those are pondered by the event period. Reported-by: Nick Piggin Signed-off-by: Frederic Weisbecker Cc: Ingo Molnar Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Paul Mackerras

perf probe: Fix to support libdwfl older than 0.148

2010-12-21T21:24:57Z

Since the libdwfl library before 0.148 fails to analyze live kernel debuginfo, 'perf probe --list' compiled with those old libdwfl sometimes crashes. To avoid that bug, perf probe does not use libdwfl's live kernel analysis routine when it is compiled with older libdwfl. Side effect: perf with older libdwfl doesn't support listing probe in modules with source code line. Those could be shown by symbol+offset. Cc: 2nddept-manager@sdl.hitachi.co.jp Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Srikar Dronamraju Cc: Steven Rostedt LKML-Reference: <20101217131218.24123.62424.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Masami Hiramatsu Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix lazy wildcard matching

2010-12-21T21:15:42Z

Fix lazy wildcard matching to ignore space after wild card. Cc: 2nddept-manager@sdl.hitachi.co.jp Cc: Frederic Weisbecker Cc: Hitoshi Mitake Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Srikar Dronamraju Cc: Steven Rostedt LKML-Reference: <20101217131200.24123.8202.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Masami Hiramatsu Signed-off-by: Arnaldo Carvalho de Melo

perf buildid-list: Fix error return for success

2010-12-16T11:43:47Z

It was always returning -1 (255), confusing test scripts. Reported-by: Han Pingtian Cc: Frederic Weisbecker Cc: Han Pingtian Cc: Ingo Molnar Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Tom Zanussi LKML-Reference: Signed-off-by: Arnaldo Carvalho de Melo

perf buildid-cache: Fix symbolic link handling

2010-12-16T11:41:45Z

This was broken since link(2) doesn't dereference symbolic links. Instead 'filename' becomes a symbolic link to the same file that 'name' refers to. This had the bad effect to create dangling symlinks in the case that even can't be removed with perf-buildid-cache(1). LKML-Reference: Signed-off-by: Franck Bui-Huu Signed-off-by: Arnaldo Carvalho de Melo

perf symbols: Stop using vmlinux files with no symbols

2010-12-16T11:41:45Z

Fail if the kernel image contains no symbol, allowing using other images in the vmlinux search path that may have a usable symtab. Acked-by: Masami Hiramatsu Cc: 2nddept-manager@sdl.hitachi.co.jp Cc: Francis Moreau Cc: Franck Bui-Huu Cc: Masami Hiramatsu LPU-Reference: Signed-off-by: Franck Bui-Huu Signed-off-by: Arnaldo Carvalho de Melo

perf probe: Fix use of kernel image path given by 'k' option

2010-12-16T11:41:45Z

Users were not being able to have the explicitely specified vmlinux pathname used, instead a search on the vmlinux path was always being made. Reported-by: Francis Moreau Acked-by: Masami Hiramatsu Cc: 2nddept-manager@sdl.hitachi.co.jp Cc: Francis Moreau Cc: Franck Bui-Huu Cc: Masami Hiramatsu LPU-Reference: Signed-off-by: Franck Bui-Huu Signed-off-by: Arnaldo Carvalho de Melo

perf record: Fix eternal wait for stillborn child

2010-12-06T17:13:38Z

When execvp fails to find the specified command on the path we won't get SIGCHLD, so send a SIGUSR1 and exit right away. Current situation would require a SIGINT performed by the user and would produce meaningless summary. Now: [acme@emilia linux]$ ./foo -bash: ./foo: No such file or directory [acme@emilia linux]$ perf record ./foo ./foo: No such file or directory [acme@emilia linux]$ Acked-by: Thomas Gleixner Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Tom Zanussi Cc: Thomas Gleixner LKML-Reference: Signed-off-by: Arnaldo Carvalho de Melo

perf header: Don't assume there's no attr info if no sample ids is provided

2010-11-30T16:48:07Z

This primarily fixes perf-report, which didn't report the correct type of event if perf-record was called to record one event different from 'cycles': $ perf record -e instructions true [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.007 MB perf.data (~295 samples) ] $ perf report | head -n1 # Events: 7 cycles LPU-Reference: Signed-off-by: Franck Bui-Huu

perf symbols: Figure out start address of kernel map from kallsyms

2010-11-30T16:47:58Z

On ARM, module symbol start address is ahead of kernel symbol start address, so we can't suppose that the start address of kernel map always is zero, otherwise may cause incorrect .start and .end of kernel map (caused by fixup) when there are modules loaded, then map_groups__find may return incorrect map for symbol query. This patch always figures out the start address of kernel map from /proc/kallsyms if the file is available, so fix the issues on ARM for module loaded case. This patch fixes the following issues on ARM when modules are loaded: - vmlinux symbol can't be found by kallsyms maps doing 'perf test' - module symbols are parsed mistakenlly when doing 'perf top'/'perf report' Cc: Ian Munsie Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Tom Zanussi LKML-Reference: <20101125192725.62d31b42@tom-lei> Signed-off-by: Ming Lei Signed-off-by: Arnaldo Carvalho de Melo