<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux/tools/perf/util, branch v2.6.37</title>
<subtitle>Linux kernel source tree</subtitle>
<id>https://git.amat.us/linux/atom/tools/perf/util?h=v2.6.37</id>
<link rel='self' href='https://git.amat.us/linux/atom/tools/perf/util?h=v2.6.37'/>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/'/>
<updated>2011-01-03T15:13:11Z</updated>
<entry>
<title>perf: Fix callchain hit bad cast on ascii display</title>
<updated>2011-01-03T15:13:11Z</updated>
<author>
<name>Frederic Weisbecker</name>
<email>fweisbec@gmail.com</email>
</author>
<published>2011-01-03T15:13:11Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=d425de5436a620de506f9e4119bf8daa1cb93718'/>
<id>urn:sha1:d425de5436a620de506f9e4119bf8daa1cb93718</id>
<content type='text'>
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 &lt;npiggin@kernel.dk&gt;
Signed-off-by: Frederic Weisbecker &lt;fweisbec@gmail.com&gt;
Cc: Ingo Molnar &lt;mingo@elte.hu&gt;
Cc: Peter Zijlstra &lt;a.p.zijlstra@chello.nl&gt;
Cc: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: Paul Mackerras &lt;paulus@samba.org&gt;
</content>
</entry>
<entry>
<title>perf probe: Fix to support libdwfl older than 0.148</title>
<updated>2010-12-21T21:24:57Z</updated>
<author>
<name>Masami Hiramatsu</name>
<email>masami.hiramatsu.pt@hitachi.com</email>
</author>
<published>2010-12-17T13:12:18Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=3b4694de3596edac0159524b15a4a06945875421'/>
<id>urn:sha1:3b4694de3596edac0159524b15a4a06945875421</id>
<content type='text'>
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 &lt;mingo@elte.hu&gt;
Cc: Paul Mackerras &lt;paulus@samba.org&gt;
Cc: Peter Zijlstra &lt;a.p.zijlstra@chello.nl&gt;
Cc: Srikar Dronamraju &lt;srikar@linux.vnet.ibm.com&gt;
Cc: Steven Rostedt &lt;rostedt@goodmis.org&gt;
LKML-Reference: &lt;20101217131218.24123.62424.stgit@ltc236.sdl.hitachi.co.jp&gt;
Signed-off-by: Masami Hiramatsu &lt;masami.hiramatsu.pt@hitachi.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
</entry>
<entry>
<title>perf tools: Fix lazy wildcard matching</title>
<updated>2010-12-21T21:15:42Z</updated>
<author>
<name>Masami Hiramatsu</name>
<email>masami.hiramatsu.pt@hitachi.com</email>
</author>
<published>2010-12-17T13:12:00Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=ea187cfbb9a3de73e7bd7b7125ae345d92d4384b'/>
<id>urn:sha1:ea187cfbb9a3de73e7bd7b7125ae345d92d4384b</id>
<content type='text'>
Fix lazy wildcard matching to ignore space after wild card.

Cc: 2nddept-manager@sdl.hitachi.co.jp
Cc: Frederic Weisbecker &lt;fweisbec@gmail.com&gt;
Cc: Hitoshi Mitake &lt;mitake@dcl.info.waseda.ac.jp&gt;
Cc: Ingo Molnar &lt;mingo@elte.hu&gt;
Cc: Paul Mackerras &lt;paulus@samba.org&gt;
Cc: Peter Zijlstra &lt;a.p.zijlstra@chello.nl&gt;
Cc: Srikar Dronamraju &lt;srikar@linux.vnet.ibm.com&gt;
Cc: Steven Rostedt &lt;rostedt@goodmis.org&gt;
LKML-Reference: &lt;20101217131200.24123.8202.stgit@ltc236.sdl.hitachi.co.jp&gt;
Signed-off-by: Masami Hiramatsu &lt;masami.hiramatsu.pt@hitachi.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
</entry>
<entry>
<title>perf buildid-cache: Fix symbolic link handling</title>
<updated>2010-12-16T11:41:45Z</updated>
<author>
<name>Franck Bui-Huu</name>
<email>fbuihuu@gmail.com</email>
</author>
<published>2010-12-10T21:06:26Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=68a7a771ad0e2959983729bf88cbc74a7014438f'/>
<id>urn:sha1:68a7a771ad0e2959983729bf88cbc74a7014438f</id>
<content type='text'>
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: &lt;m38vzxxrql.fsf@gmail.com&gt;
Signed-off-by: Franck Bui-Huu &lt;fbuihuu@gmail.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
</entry>
<entry>
<title>perf symbols: Stop using vmlinux files with no symbols</title>
<updated>2010-12-16T11:41:45Z</updated>
<author>
<name>Franck Bui-Huu</name>
<email>fbuihuu@gmail.com</email>
</author>
<published>2010-12-10T13:07:14Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=c3a34e06db25a8c74e196517732d65cdb56028ec'/>
<id>urn:sha1:c3a34e06db25a8c74e196517732d65cdb56028ec</id>
<content type='text'>
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 &lt;masami.hiramatsu.pt@hitachi.com&gt;
Cc: 2nddept-manager@sdl.hitachi.co.jp
Cc: Francis Moreau &lt;francis.moro@gmail.com&gt;
Cc: Franck Bui-Huu &lt;vagabon.xyz@gmail.com&gt;
Cc: Masami Hiramatsu &lt;masami.hiramatsu.pt@hitachi.com&gt;
LPU-Reference: &lt;m3d3p9ydx9.fsf_-_@gmail.com&gt;
Signed-off-by: Franck Bui-Huu &lt;fbuihuu@gmail.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
</entry>
<entry>
<title>perf probe: Fix use of kernel image path given by 'k' option</title>
<updated>2010-12-16T11:41:45Z</updated>
<author>
<name>Franck Bui-Huu</name>
<email>fbuihuu@gmail.com</email>
</author>
<published>2010-12-10T13:06:03Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=fd930ff91e7fda18f7790984a98489a85edb7c71'/>
<id>urn:sha1:fd930ff91e7fda18f7790984a98489a85edb7c71</id>
<content type='text'>
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 &lt;francis.moro@gmail.com&gt;
Acked-by: Masami Hiramatsu &lt;masami.hiramatsu.pt@hitachi.com&gt;
Cc: 2nddept-manager@sdl.hitachi.co.jp
Cc: Francis Moreau &lt;francis.moro@gmail.com&gt;
Cc: Franck Bui-Huu &lt;vagabon.xyz@gmail.com&gt;
Cc: Masami Hiramatsu &lt;masami.hiramatsu.pt@hitachi.com&gt;
LPU-Reference: &lt;m3hbelydz8.fsf_-_@gmail.com&gt;
Signed-off-by: Franck Bui-Huu &lt;fbuihuu@gmail.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
</entry>
<entry>
<title>perf header: Don't assume there's no attr info if no sample ids is provided</title>
<updated>2010-11-30T16:48:07Z</updated>
<author>
<name>Franck Bui-Huu</name>
<email>fbuihuu@gmail.com</email>
</author>
<published>2010-11-30T11:50:14Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=60e677373be9c0bf7c9a22937601d5a40e51c042'/>
<id>urn:sha1:60e677373be9c0bf7c9a22937601d5a40e51c042</id>
<content type='text'>
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: &lt;m3mxor6nex.fsf@gmail.com&gt;
Signed-off-by: Franck Bui-Huu &lt;fbuihuu@gmail.com&gt;
</content>
</entry>
<entry>
<title>perf symbols: Figure out start address of kernel map from kallsyms</title>
<updated>2010-11-30T16:47:58Z</updated>
<author>
<name>Ming Lei</name>
<email>tom.leiming@gmail.com</email>
</author>
<published>2010-11-25T11:27:25Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=d214afbd81405d4da2c5745fe867e6b313fd4178'/>
<id>urn:sha1:d214afbd81405d4da2c5745fe867e6b313fd4178</id>
<content type='text'>
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 &lt;imunsie@au1.ibm.com&gt;
Cc: Ingo Molnar &lt;mingo@elte.hu&gt;
Cc: Paul Mackerras &lt;paulus@samba.org&gt;
Cc: Peter Zijlstra &lt;a.p.zijlstra@chello.nl&gt;
Cc: Thomas Gleixner &lt;tglx@linutronix.de&gt;
Cc: Tom Zanussi &lt;tzanussi@gmail.com&gt;
LKML-Reference: &lt;20101125192725.62d31b42@tom-lei&gt;
Signed-off-by: Ming Lei &lt;tom.leiming@gmail.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
</entry>
<entry>
<title>perf symbols: Fix kallsyms kernel/module map splitting</title>
<updated>2010-11-30T16:47:51Z</updated>
<author>
<name>Arnaldo Carvalho de Melo</name>
<email>acme@redhat.com</email>
</author>
<published>2010-11-29T14:44:15Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=8a9533123f43f2cdb3eb601c17ff2ad336882eff'/>
<id>urn:sha1:8a9533123f43f2cdb3eb601c17ff2ad336882eff</id>
<content type='text'>
On ARM, module addresss space is ahead of kernel space, so the module
symbols are handled before kernel symbol in dso__split_kallsyms, then
was causing one map to be created for each kernel symbol.

Reported-by: Ming Lei &lt;tom.leiming@gmail.com&gt;
Tested-by: Ming Lei &lt;tom.leiming@gmail.com&gt;
Cc: Frederic Weisbecker &lt;fweisbec@gmail.com&gt;
Cc: Ingo Molnar &lt;mingo@elte.hu&gt;
Cc: Mike Galbraith &lt;efault@gmx.de&gt;
Cc: Ming Lei &lt;tom.leiming@gmail.com&gt;
Cc: Paul Mackerras &lt;paulus@samba.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Stephane Eranian &lt;eranian@google.com&gt;
Cc: Tom Zanussi &lt;tzanussi@gmail.com&gt;
LKML-Reference: &lt;20101124144540.GB15875@ghostprotocols.net&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
</entry>
<entry>
<title>perf symbols: Remove incorrect open-coded container_of()</title>
<updated>2010-11-23T18:30:32Z</updated>
<author>
<name>Rabin Vincent</name>
<email>rabin@rab.in</email>
</author>
<published>2010-11-23T16:38:18Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=02a9d03772aa1ff33a26180a2da0bfb191240eda'/>
<id>urn:sha1:02a9d03772aa1ff33a26180a2da0bfb191240eda</id>
<content type='text'>
At least on ARM, padding is inserted between rb_node and sym in struct
symbol_name_rb_node, causing "((void *)sym) - sizeof(struct rb_node)" to
point inside rb_node rather than to the symbol_name_rb_node.  Fix this
by converting the code to use container_of().

Cc: Ian Munsie &lt;imunsie@au1.ibm.com&gt;
Cc: Ingo Molnar &lt;mingo@elte.hu&gt;
Cc: Ming Lei &lt;tom.leiming@gmail.com&gt;
Cc: Paul Mackerras &lt;paulus@samba.org&gt;
Cc: Peter Zijlstra &lt;a.p.zijlstra@chello.nl&gt;
Cc: Thomas Gleixner &lt;tglx@linutronix.de&gt;
Cc: Tom Zanussi &lt;tzanussi@gmail.com&gt;
LKML-Reference: &lt;20101123163106.GA25677@debian&gt;
Signed-off-by: Rabin Vincent &lt;rabin@rab.in&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
</entry>
</feed>
