[PATCH] f2fs: fix to return 0 if err == -ENOENT in f2fs_readdir

t err to 0. This will cause getdents64 incorretly return -ENOENT when lastdirent is NULL in getdents64. This will lead to a wrong return value for syscall caller. Signed-off-by: Yunlong Song <yunlong.s...@huawei.com> --- fs/f2fs/dir.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/f2fs/dir.c

[PATCH] f2fs: fix to return 0 if err == -ENOENT in f2fs_readdir

t err to 0. This will cause getdents64 incorretly return -ENOENT when lastdirent is NULL in getdents64. This will lead to a wrong return value for syscall caller. Signed-off-by: Yunlong Song --- fs/f2fs/dir.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c index 0ad7b9

Re: [Questions] perf c2c: What's the current status of perf c2c?

-- > // Perf c2c output from a linpack run. Hi, Joe, Got these details, thanks a lot for your interpretation. -_- -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body o

Re: [Questions] perf c2c: What's the current status of perf c2c?

-- > // Perf c2c output from a linpack run. Hi, Joe, Got these details, thanks a lot for your interpretation. -_- -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body o

[Questions] perf c2c: What's the current status of perf c2c?

or a repository somewhere else? And does it support Haswell? -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please re

[Questions] perf c2c: What's the current status of perf c2c?

or a repository somewhere else? And does it support Haswell? -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please re

Re: [PATCH] perf record: Add snapshot mode support for perf's regular events

in some way to avoid globals like this. Thanks for your careful review and helpful comments, my patch stands for certain kind of design for snapshot mode. And other people may have suggestions and different designs, let's talk and compare different designs first. If the design of this patch is taken

Re: [PATCH] perf record: Add snapshot mode support for perf's regular events

in some way to avoid globals like this. Thanks for your careful review and helpful comments, my patch stands for certain kind of design for snapshot mode. And other people may have suggestions and different designs, let's talk and compare different designs first. If the design of this patch is taken

Re: [PATCH] perf record: Add snapshot mode support for perf's regular events

esigned for specific use of perf sched. In general case of perf record, with snapshot mode, we still want a perf.data as before. Your sched daemon concurrently does the work of storing and sched-parsing action for each signal trigger. To get a general style of perf.data, the sched-parsing semantic action

[PATCH] perf record: Add snapshot mode support for perf's regular events

it receives a SIGUSR2 signal in the snapshot mode. Signed-off-by: Yunlong Song --- tools/perf/builtin-record.c | 181 +--- 1 file changed, 170 insertions(+), 11 deletions(-) diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index

[PATCH] perf record: Add snapshot mode support for perf's regular events

patch, we create and maintain a user space ring buffer to store perf's tracing info, instead of directly writing to perf.data file as before. In snapshot mode, only a SIGUSR2 signal can trigger perf to dump the tracing info currently stored in the user space ring buffer to perf.data file. Yunlong

Re: [PATCH] perf record: Add snapshot mode support for perf's regular events

esigned for specific use of perf sched. In general case of perf record, with snapshot mode, we still want a perf.data as before. Your sched daemon concurrently does the work of storing and sched-parsing action for each signal trigger. To get a general style of perf.data, the sched-parsing semantic action

[PATCH] perf record: Add snapshot mode support for perf's regular events

it receives a SIGUSR2 signal in the snapshot mode. Signed-off-by: Yunlong Song <yunlong.s...@huawei.com> --- tools/perf/builtin-record.c | 181 +--- 1 file changed, 170 insertions(+), 11 deletions(-) diff --git a/tools/perf/builtin-record.c b/tool

[PATCH] perf record: Add snapshot mode support for perf's regular events

patch, we create and maintain a user space ring buffer to store perf's tracing info, instead of directly writing to perf.data file as before. In snapshot mode, only a SIGUSR2 signal can trigger perf to dump the tracing info currently stored in the user space ring buffer to perf.data file. Yunlong

[PATCH] perf test: Add llvm-src-base.c and llvm-src-kbuild.c to .gitignore

add kbuild test") dynamically creates file llvm-src-kbuild.c during the perf building. Add them to .gitignore. Signed-off-by: Yunlong Song --- tools/perf/tests/.gitignore | 2 ++ 1 file changed, 2 insertions(+) create mode 100644 tools/perf/tests/.gitignore diff --git a/tools/perf/test

[PATCH] perf test: Add llvm-src-base.c and llvm-src-kbuild.c to .gitignore

add kbuild test") dynamically creates file llvm-src-kbuild.c during the perf building. Add them to .gitignore. Signed-off-by: Yunlong Song <yunlong.s...@huawei.com> --- tools/perf/tests/.gitignore | 2 ++ 1 file changed, 2 insertions(+) create mode 100644 tools/perf/tests/.gitignor

[tip:perf/urgent] tools lib bpf: Change FEATURE-DUMP to FEATURE-DUMP.libbpf

Commit-ID: 0c77c04aa9c2b3a154f5cb0559feaee3fb7a6935 Gitweb: http://git.kernel.org/tip/0c77c04aa9c2b3a154f5cb0559feaee3fb7a6935 Author: Yunlong Song AuthorDate: Thu, 5 Nov 2015 14:56:34 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Thu, 5 Nov 2015 12:47:51 -0300 tools lib bpf

[tip:perf/urgent] tools lib bpf: Change FEATURE-DUMP to FEATURE-DUMP.libbpf

Commit-ID: 0c77c04aa9c2b3a154f5cb0559feaee3fb7a6935 Gitweb: http://git.kernel.org/tip/0c77c04aa9c2b3a154f5cb0559feaee3fb7a6935 Author: Yunlong Song <yunlong.s...@huawei.com> AuthorDate: Thu, 5 Nov 2015 14:56:34 +0800 Committer: Arnaldo Carvalho de Melo <a...@redhat.com> Com

Re: [PATCH] perf tools: Change FEATURE-DUMP to FEATURE-DUMP.libbpf for lib bpf

> > Anyway, fixing it up, thanks for the patch! > > - Arnaldo Yes, right. Thanks for your detailed explanation. -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kerne

Re: [PATCH] perf tools: Change FEATURE-DUMP to FEATURE-DUMP.libbpf for lib bpf

> > Anyway, fixing it up, thanks for the patch! > > - Arnaldo Yes, right. Thanks for your detailed explanation. -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kerne

[PATCH] perf tools: Change FEATURE-DUMP to FEATURE-DUMP.libbpf for lib bpf

/bpf/ also mistakes FEATURE-DUMP.libbpf for FEATURE-DUMP, which causes that "make clean" will leave (or say create) FEATURE-DUMP.libbpf rather than remove FEATURE-DUMP.libbpf file. Signed-off-by: Yunlong Song --- tools/lib/bpf/.gitignore | 2 +- tools/lib/bpf/Makefile | 2 +- 2 files changed

[PATCH] perf tools: Change FEATURE-DUMP to FEATURE-DUMP.libbpf for lib bpf

/bpf/ also mistakes FEATURE-DUMP.libbpf for FEATURE-DUMP, which causes that "make clean" will leave (or say create) FEATURE-DUMP.libbpf rather than remove FEATURE-DUMP.libbpf file. Signed-off-by: Yunlong Song <yunlong.s...@huawei.com> --- tools/lib/bpf/.gitignore | 2 +- tools/lib/bpf/

[tip:perf/core] perf build: Add fixdep to .gitignore

Commit-ID: e1d040cfcf76c1c1b5d71fc08ab577a0cf72fefd Gitweb: http://git.kernel.org/tip/e1d040cfcf76c1c1b5d71fc08ab577a0cf72fefd Author: Yunlong Song AuthorDate: Thu, 15 Oct 2015 16:51:56 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Tue, 20 Oct 2015 10:43:28 -0300 perf build

[tip:perf/core] perf build: Add fixdep to .gitignore

Commit-ID: e1d040cfcf76c1c1b5d71fc08ab577a0cf72fefd Gitweb: http://git.kernel.org/tip/e1d040cfcf76c1c1b5d71fc08ab577a0cf72fefd Author: Yunlong Song <yunlong.s...@huawei.com> AuthorDate: Thu, 15 Oct 2015 16:51:56 +0800 Committer: Arnaldo Carvalho de Melo <a...@redhat.com> Com

[tip:perf/core] perf help: Change 'usage' to 'Usage' for consistency

Commit-ID: 3a134ae96ca0af06804d343019b85026486e6fe1 Gitweb: http://git.kernel.org/tip/3a134ae96ca0af06804d343019b85026486e6fe1 Author: Yunlong Song AuthorDate: Thu, 15 Oct 2015 15:39:51 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Mon, 19 Oct 2015 16:51:44 -0300 perf help

[tip:perf/core] perf help: Change 'usage' to 'Usage' for consistency

Commit-ID: 3a134ae96ca0af06804d343019b85026486e6fe1 Gitweb: http://git.kernel.org/tip/3a134ae96ca0af06804d343019b85026486e6fe1 Author: Yunlong Song <yunlong.s...@huawei.com> AuthorDate: Thu, 15 Oct 2015 15:39:51 +0800 Committer: Arnaldo Carvalho de Melo <a...@redhat.com> Com

Re: [PATCH] perf build: Add fixdep to .gitignore

On 2015/10/15 16:51, Yunlong Song wrote: > Commit 7c422f5572667fef0db38d2046ecce69dcf0afc8 ("tools build: Build fixdep > helper from perf and basic libs") dynamically creates fixdep during the perf > building. Add it to .gitignore. > > Signed-off-by: Yunlong Song > -

Re: [PATCH v2 1/3] perf help: Add options description to 'perf -h'

On 2015/10/19 23:29, Namhyung Kim wrote: > Hi, > > On Thu, Oct 15, 2015 at 03:39:50PM +0800, Yunlong Song wrote: >> Add options description to 'perf -h' to make it consistent with other >> builtins >> (e.g., 'perf stat -h'). >> >> Example: >> >&g

Re: [PATCH] perf build: Add fixdep to .gitignore

On 2015/10/15 16:51, Yunlong Song wrote: > Commit 7c422f5572667fef0db38d2046ecce69dcf0afc8 ("tools build: Build fixdep > helper from perf and basic libs") dynamically creates fixdep during the perf > building. Add it to .gitignore. > > Signed-off-by: Yunlong Song

Re: [PATCH v2 1/3] perf help: Add options description to 'perf -h'

On 2015/10/19 23:29, Namhyung Kim wrote: > Hi, > > On Thu, Oct 15, 2015 at 03:39:50PM +0800, Yunlong Song wrote: >> Add options description to 'perf -h' to make it consistent with other >> builtins >> (e.g., 'perf stat -h'). >> >> Example: >> >&g

[PATCH] perf build: Add fixdep to .gitignore

Commit 7c422f5572667fef0db38d2046ecce69dcf0afc8 ("tools build: Build fixdep helper from perf and basic libs") dynamically creates fixdep during the perf building. Add it to .gitignore. Signed-off-by: Yunlong Song --- tools/build/.gitignore | 1 + 1 file changed, 1 insertion(+) c

[PATCH v2 2/3] perf help: Change 'usage' to 'Usage' for consistency

Capitalize 'usage' to make it consistent with all the other 'Usage' in the codes, e.g., usage_builtin. Signed-off-by: Yunlong Song --- tools/perf/builtin-help.c | 2 +- tools/perf/util/parse-options.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/perf

[PATCH v2 0/3] perf help: Make perf's help consistent with other builtins

Hi, Make some fixes to perf's help for consistency. Yunlong Song (3): perf help: Add options description to 'perf -h' perf help: Change 'usage' to 'Usage' for consistency perf help: Change the usage's stdout to stderr for consistency tools/perf/builtin-help.c | 13

[PATCH v2 1/3] perf help: Add options description to 'perf -h'

new dynamic tracepoints See 'perf help COMMAND' for more information on a specific command. As shown above, the options description really appears now. Signed-off-by: Yunlong Song --- tools/perf/builtin-help.c | 2 +- tools/perf/builtin.h| 3 +++ tools/perf/perf.c

[PATCH v2 3/3] perf help: Change the usage's stdout to stderr for consistency

The builtins use stderr everywhere to show the usage info when the opts or cmds are incorrectly used, for consistency, change perf's stdout to stderr to show its usage info when it is incorrectly (including no command) used. Signed-off-by: Yunlong Song --- tools/perf/builtin-help.c | 9

Re: [PATCH 1/1] perf:Adding --list-opts to usage string

On 2015/10/14 21:40, Arnaldo Carvalho de Melo wrote: > > Le 14 oct. 2015 10:33 AM, "Yunlong Song" <mailto:yunlong.s...@huawei.com>> a écrit : >> >> On 2015/10/14 11:10, Arnaldo Carvalho de Melo wrote: >> >> > One doesn

Re: [PATCH 1/1] perf:Adding --list-opts to usage string

On 2015/10/14 21:40, Arnaldo Carvalho de Melo wrote: > > Le 14 oct. 2015 10:33 AM, "Yunlong Song" <yunlong.s...@huawei.com > <mailto:yunlong.s...@huawei.com>> a écrit : >> >> On 2015/10/14 11:10, Arnaldo Carvalho de Melo wrote: >> >> >

[PATCH v2 1/3] perf help: Add options description to 'perf -h'

new dynamic tracepoints See 'perf help COMMAND' for more information on a specific command. As shown above, the options description really appears now. Signed-off-by: Yunlong Song <yunlong.s...@huawei.com> --- tools/perf/builtin-help.c | 2 +- tools/perf/builtin.h

[PATCH v2 0/3] perf help: Make perf's help consistent with other builtins

Hi, Make some fixes to perf's help for consistency. Yunlong Song (3): perf help: Add options description to 'perf -h' perf help: Change 'usage' to 'Usage' for consistency perf help: Change the usage's stdout to stderr for consistency tools/perf/builtin-help.c | 13

[PATCH v2 2/3] perf help: Change 'usage' to 'Usage' for consistency

Capitalize 'usage' to make it consistent with all the other 'Usage' in the codes, e.g., usage_builtin. Signed-off-by: Yunlong Song <yunlong.s...@huawei.com> --- tools/perf/builtin-help.c | 2 +- tools/perf/util/parse-options.c | 4 ++-- 2 files changed, 3 insertions(+), 3 del

[PATCH] perf build: Add fixdep to .gitignore

Commit 7c422f5572667fef0db38d2046ecce69dcf0afc8 ("tools build: Build fixdep helper from perf and basic libs") dynamically creates fixdep during the perf building. Add it to .gitignore. Signed-off-by: Yunlong Song <yunlong.s...@huawei.com> --- tools/build/.gitignore | 1 + 1

[PATCH v2 3/3] perf help: Change the usage's stdout to stderr for consistency

The builtins use stderr everywhere to show the usage info when the opts or cmds are incorrectly used, for consistency, change perf's stdout to stderr to show its usage info when it is incorrectly (including no command) used. Signed-off-by: Yunlong Song <yunlong.s...@huawei.com> --- tool

[PATCH] perf help: Add options description to 'perf -h'

new dynamic tracepoints See 'perf help COMMAND' for more information on a specific command. As shown above, the options description really appears now. Signed-off-by: Yunlong Song --- tools/perf/builtin-help.c | 2 +- tools/perf/builtin.h| 3 +++ tools/perf/perf.c

Re: [PATCH 1/1] perf:Adding --list-opts to usage string

ut to stdout, just like 'ls --help', what do you think? > > - Arnaldo > I think the reason that options do not show in 'perf -h' but show in other builtins is not the different use of stdout or stderr. I will send a patch to fix this instead. -- Thanks, Yunlong Song -- To unsubscribe

Re: [PATCH 1/1] perf:Adding --list-opts to usage string

ut to stdout, just like 'ls --help', what do you think? > > - Arnaldo > I think the reason that options do not show in 'perf -h' but show in other builtins is not the different use of stdout or stderr. I will send a patch to fix this instead. -- Thanks, Yunlong Song -- To unsubscribe

[PATCH] perf help: Add options description to 'perf -h'

new dynamic tracepoints See 'perf help COMMAND' for more information on a specific command. As shown above, the options description really appears now. Signed-off-by: Yunlong Song <yunlong.s...@huawei.com> --- tools/perf/builtin-help.c | 2 +- tools/perf/builtin.h

Re: [PATCH 1/1] perf:Adding --list-opts to usage string

e existing [OPTIONS] in the perf_usage_string[]. >>> >>> const char perf_usage_string[] = >>> - "perf [--version] [--help] [OPTIONS] COMMAND [ARGS]"; >>> + "perf [--version] [--help] [--list-opts] [OPTIONS] COMMAND [ARGS]"; -- Thanks

Re: [PATCH 1/1] perf:Adding --list-opts to usage string

e existing [OPTIONS] in the perf_usage_string[]. >>> >>> const char perf_usage_string[] = >>> - "perf [--version] [--help] [OPTIONS] COMMAND [ARGS]"; >>> + "perf [--version] [--help] [--list-opts] [OPTIONS] COMMAND [ARGS]"; -- Thanks

[RFC resend] Perf: Trigger and dump sample info to perf.data from user space ring buffer

can be used both in daemon mode or in non-daemon mode. This idea can be another way to filter the tracing events compared to eBPF from different view. -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@v

[RFC] Perf: Trigger and dump sample info to perf.data from user space ring buffer

. Thanks, -- Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/

[RFC resend] Perf: Trigger and dump sample info to perf.data from user space ring buffer

can be used both in daemon mode or in non-daemon mode. This idea can be another way to filter the tracing events compared to eBPF from different view. -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@v

[RFC] Perf: Trigger and dump sample info to perf.data from user space ring buffer

. Thanks, -- Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/5/6 5:53, Rabin Vincent wrote: > On Sat, Apr 25, 2015 at 09:53:57AM -0600, David Ahern wrote: >> On 4/25/15 8:05 AM, Yunlong Song wrote: >>> But this only shows the system call like strace, but we want the call >>> stack of kernel functions in fact. >>&g

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/4/25 23:53, David Ahern wrote: > On 4/25/15 8:05 AM, Yunlong Song wrote: >> On 2015/4/24 21:58, David Ahern wrote: >>> On 4/24/15 7:31 AM, Yunlong Song wrote: >>>> Now we are profiling the performance of ext4 and f2fs on an eMMC card with >>>

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/4/25 23:53, David Ahern wrote: On 4/25/15 8:05 AM, Yunlong Song wrote: On 2015/4/24 21:58, David Ahern wrote: On 4/24/15 7:31 AM, Yunlong Song wrote: Now we are profiling the performance of ext4 and f2fs on an eMMC card with iozone, we find a case that ext4 is better than f2fs

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/5/6 5:53, Rabin Vincent wrote: On Sat, Apr 25, 2015 at 09:53:57AM -0600, David Ahern wrote: On 4/25/15 8:05 AM, Yunlong Song wrote: But this only shows the system call like strace, but we want the call stack of kernel functions in fact. We haven't added the callchain option yet

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/4/24 21:58, David Ahern wrote: > On 4/24/15 7:31 AM, Yunlong Song wrote: >> Now we are profiling the performance of ext4 and f2fs on an eMMC card with >> iozone, >> we find a case that ext4 is better than f2fs in random write under the test >> of >> "

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/4/24 21:49, Yunlong Song wrote: > On 2015/4/24 21:31, Yunlong Song wrote: >> [Profiling Background] >> Now we are profiling the performance of ext4 and f2fs on an eMMC card with >> iozone, >> we find a case that ext4 is better than f2fs in random write under th

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/4/24 21:49, Yunlong Song wrote: On 2015/4/24 21:31, Yunlong Song wrote: [Profiling Background] Now we are profiling the performance of ext4 and f2fs on an eMMC card with iozone, we find a case that ext4 is better than f2fs in random write under the test of iozone -s 262144 -r 64

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/4/24 21:58, David Ahern wrote: On 4/24/15 7:31 AM, Yunlong Song wrote: Now we are profiling the performance of ext4 and f2fs on an eMMC card with iozone, we find a case that ext4 is better than f2fs in random write under the test of iozone -s 262144 -r 64 -i 0 -i 2. We want

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/4/24 21:49, David Ahern wrote: > On 4/24/15 7:31 AM, Yunlong Song wrote: >> [Profiling Problem] >> >> Although perf can record the events (with call stack) of a specified pid, >> e.g. using >> "perf record -g iozone -s 262144 -r 64 -i 0 -i 2&q

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/4/24 21:31, Yunlong Song wrote: > [Profiling Background] > Now we are profiling the performance of ext4 and f2fs on an eMMC card with > iozone, > we find a case that ext4 is better than f2fs in random write under the test of > "iozone -s 262144 -r 64 -i 0 -i 2". W

[Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

pted and CPU is scheduled to other process?" -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/

[Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

process? -- Thanks, Yunlong Song -- To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/4/24 21:31, Yunlong Song wrote: [Profiling Background] Now we are profiling the performance of ext4 and f2fs on an eMMC card with iozone, we find a case that ext4 is better than f2fs in random write under the test of iozone -s 262144 -r 64 -i 0 -i 2. We want to analyze the I/O delay

Re: [Question] How does perf still record the stack of a specified pid even when that process is interrupted and CPU is scheduled to other process

On 2015/4/24 21:49, David Ahern wrote: On 4/24/15 7:31 AM, Yunlong Song wrote: [Profiling Problem] Although perf can record the events (with call stack) of a specified pid, e.g. using perf record -g iozone -s 262144 -r 64 -i 0 -i 2. But we find iozone is interrupted That command tells

[tip:perf/core] perf sched replay: Use replay_repeat to calculate the runavg of cpu usage instead of the default value 10

Commit-ID: ff5f3bbd40bfb8632f826f1f83223d95363f36af Gitweb: http://git.kernel.org/tip/ff5f3bbd40bfb8632f826f1f83223d95363f36af Author: Yunlong Song AuthorDate: Tue, 31 Mar 2015 21:46:36 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Wed, 8 Apr 2015 09:07:27 -0300 perf sched

[tip:perf/core] perf sched replay: Fix the EMFILE error caused by the limitation of the maximum open files

Commit-ID: 939cda521a24ae4dbf3beec983abd519bce56231 Gitweb: http://git.kernel.org/tip/939cda521a24ae4dbf3beec983abd519bce56231 Author: Yunlong Song AuthorDate: Tue, 31 Mar 2015 21:46:34 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Wed, 8 Apr 2015 09:07:26 -0300 perf sched

[tip:perf/core] perf sched replay: Handle the dead halt of sem_wait when create_tasks() fails for any task

Commit-ID: 1aff59be53ef37aa9943fb5f772f03148f789bb6 Gitweb: http://git.kernel.org/tip/1aff59be53ef37aa9943fb5f772f03148f789bb6 Author: Yunlong Song AuthorDate: Tue, 31 Mar 2015 21:46:33 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Wed, 8 Apr 2015 09:07:25 -0300 perf sched

[tip:perf/core] perf sched replay: Support using -f to override perf.data file ownership

Commit-ID: f0dd330fdf07d295ac468660cf60341796d5d501 Gitweb: http://git.kernel.org/tip/f0dd330fdf07d295ac468660cf60341796d5d501 Author: Yunlong Song AuthorDate: Tue, 31 Mar 2015 21:46:35 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Wed, 8 Apr 2015 09:07:26 -0300 perf sched

[tip:perf/core] perf sched replay: Fix the segmentation fault problem caused by pr_err in threads

Commit-ID: 08097abc11bcee21355dd857852a807b2a30b79f Gitweb: http://git.kernel.org/tip/08097abc11bcee21355dd857852a807b2a30b79f Author: Yunlong Song AuthorDate: Tue, 31 Mar 2015 21:46:32 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Wed, 8 Apr 2015 09:07:24 -0300 perf sched

[tip:perf/core] perf sched replay: Increase the MAX_PID value to fix assertion failure problem

Commit-ID: a35e27d0e5d801ff75481a8f639bb4d59ea1aafa Gitweb: http://git.kernel.org/tip/a35e27d0e5d801ff75481a8f639bb4d59ea1aafa Author: Yunlong Song AuthorDate: Tue, 31 Mar 2015 21:46:29 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Wed, 8 Apr 2015 09:07:21 -0300 perf sched

[tip:perf/core] perf sched replay: Realloc the memory of pid_to_task stepwise to adapt to the different pid_max configurations

Commit-ID: 3a423a5c36d1a28a258beaa7db855568b82d07ab Gitweb: http://git.kernel.org/tip/3a423a5c36d1a28a258beaa7db855568b82d07ab Author: Yunlong Song AuthorDate: Tue, 31 Mar 2015 21:46:31 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Wed, 8 Apr 2015 09:07:23 -0300 perf sched

[tip:perf/core] perf sched replay: Alloc the memory of pid_to_task dynamically to adapt to the unexpected change of pid_max

Commit-ID: cb06ac256a16fc1a5ab063107c2b35b3b9e95102 Gitweb: http://git.kernel.org/tip/cb06ac256a16fc1a5ab063107c2b35b3b9e95102 Author: Yunlong Song AuthorDate: Tue, 31 Mar 2015 21:46:30 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Wed, 8 Apr 2015 09:07:22 -0300 perf sched

[tip:perf/core] perf sched replay: Use struct task_desc instead of struct task_task for correct meaning

Commit-ID: 0755bc4dc77a876aa60d4b3d33b5f6506f21f91b Gitweb: http://git.kernel.org/tip/0755bc4dc77a876aa60d4b3d33b5f6506f21f91b Author: Yunlong Song AuthorDate: Tue, 31 Mar 2015 21:46:28 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Wed, 8 Apr 2015 09:07:19 -0300 perf sched

Re: [PATCH] perf record: Conditionally define CLOCK_MONOTONIC_RAW for older OSes

On 2015/4/8 21:03, Arnaldo Carvalho de Melo wrote: > Em Wed, Apr 08, 2015 at 12:02:28PM +0800, Yunlong Song escreveu: >> Commit 31a9883106cc ("perf record: Add clockid parameter") used >> CLOCK_MONOTONIC_RAW in the struct clockid_map clockids[], but the >> CLOCK_MON

Re: [PATCH] perf record: Conditionally define CLOCK_MONOTONIC_RAW for older OSes

On 2015/4/8 16:55, Peter Zijlstra wrote: > On Wed, Apr 08, 2015 at 12:02:28PM +0800, Yunlong Song wrote: >> Commit 31a9883106cc ("perf record: Add clockid parameter") used >> CLOCK_MONOTONIC_RAW in the struct clockid_map clockids[], but the >> CLOCK_MONOTONIC_RA

Re: [PATCH] perf record: Conditionally define CLOCK_MONOTONIC_RAW for older OSes

On 2015/4/8 16:55, Peter Zijlstra wrote: On Wed, Apr 08, 2015 at 12:02:28PM +0800, Yunlong Song wrote: Commit 31a9883106cc (perf record: Add clockid parameter) used CLOCK_MONOTONIC_RAW in the struct clockid_map clockids[], but the CLOCK_MONOTONIC_RAW macro is not defined in older releases (e.g

Re: [PATCH] perf record: Conditionally define CLOCK_MONOTONIC_RAW for older OSes

On 2015/4/8 21:03, Arnaldo Carvalho de Melo wrote: Em Wed, Apr 08, 2015 at 12:02:28PM +0800, Yunlong Song escreveu: Commit 31a9883106cc (perf record: Add clockid parameter) used CLOCK_MONOTONIC_RAW in the struct clockid_map clockids[], but the CLOCK_MONOTONIC_RAW macro is not defined in older

[tip:perf/core] perf sched replay: Increase the MAX_PID value to fix assertion failure problem

Commit-ID: a35e27d0e5d801ff75481a8f639bb4d59ea1aafa Gitweb: http://git.kernel.org/tip/a35e27d0e5d801ff75481a8f639bb4d59ea1aafa Author: Yunlong Song yunlong.s...@huawei.com AuthorDate: Tue, 31 Mar 2015 21:46:29 +0800 Committer: Arnaldo Carvalho de Melo a...@redhat.com CommitDate: Wed, 8

[tip:perf/core] perf sched replay: Fix the segmentation fault problem caused by pr_err in threads

Commit-ID: 08097abc11bcee21355dd857852a807b2a30b79f Gitweb: http://git.kernel.org/tip/08097abc11bcee21355dd857852a807b2a30b79f Author: Yunlong Song yunlong.s...@huawei.com AuthorDate: Tue, 31 Mar 2015 21:46:32 +0800 Committer: Arnaldo Carvalho de Melo a...@redhat.com CommitDate: Wed, 8

[tip:perf/core] perf sched replay: Alloc the memory of pid_to_task dynamically to adapt to the unexpected change of pid_max

Commit-ID: cb06ac256a16fc1a5ab063107c2b35b3b9e95102 Gitweb: http://git.kernel.org/tip/cb06ac256a16fc1a5ab063107c2b35b3b9e95102 Author: Yunlong Song yunlong.s...@huawei.com AuthorDate: Tue, 31 Mar 2015 21:46:30 +0800 Committer: Arnaldo Carvalho de Melo a...@redhat.com CommitDate: Wed, 8

[tip:perf/core] perf sched replay: Realloc the memory of pid_to_task stepwise to adapt to the different pid_max configurations

Commit-ID: 3a423a5c36d1a28a258beaa7db855568b82d07ab Gitweb: http://git.kernel.org/tip/3a423a5c36d1a28a258beaa7db855568b82d07ab Author: Yunlong Song yunlong.s...@huawei.com AuthorDate: Tue, 31 Mar 2015 21:46:31 +0800 Committer: Arnaldo Carvalho de Melo a...@redhat.com CommitDate: Wed, 8

[tip:perf/core] perf sched replay: Fix the EMFILE error caused by the limitation of the maximum open files

Commit-ID: 939cda521a24ae4dbf3beec983abd519bce56231 Gitweb: http://git.kernel.org/tip/939cda521a24ae4dbf3beec983abd519bce56231 Author: Yunlong Song yunlong.s...@huawei.com AuthorDate: Tue, 31 Mar 2015 21:46:34 +0800 Committer: Arnaldo Carvalho de Melo a...@redhat.com CommitDate: Wed, 8

[tip:perf/core] perf sched replay: Use replay_repeat to calculate the runavg of cpu usage instead of the default value 10

Commit-ID: ff5f3bbd40bfb8632f826f1f83223d95363f36af Gitweb: http://git.kernel.org/tip/ff5f3bbd40bfb8632f826f1f83223d95363f36af Author: Yunlong Song yunlong.s...@huawei.com AuthorDate: Tue, 31 Mar 2015 21:46:36 +0800 Committer: Arnaldo Carvalho de Melo a...@redhat.com CommitDate: Wed, 8

[tip:perf/core] perf sched replay: Handle the dead halt of sem_wait when create_tasks() fails for any task

Commit-ID: 1aff59be53ef37aa9943fb5f772f03148f789bb6 Gitweb: http://git.kernel.org/tip/1aff59be53ef37aa9943fb5f772f03148f789bb6 Author: Yunlong Song yunlong.s...@huawei.com AuthorDate: Tue, 31 Mar 2015 21:46:33 +0800 Committer: Arnaldo Carvalho de Melo a...@redhat.com CommitDate: Wed, 8

[tip:perf/core] perf sched replay: Use struct task_desc instead of struct task_task for correct meaning

Commit-ID: 0755bc4dc77a876aa60d4b3d33b5f6506f21f91b Gitweb: http://git.kernel.org/tip/0755bc4dc77a876aa60d4b3d33b5f6506f21f91b Author: Yunlong Song yunlong.s...@huawei.com AuthorDate: Tue, 31 Mar 2015 21:46:28 +0800 Committer: Arnaldo Carvalho de Melo a...@redhat.com CommitDate: Wed, 8

[tip:perf/core] perf sched replay: Support using -f to override perf.data file ownership

Commit-ID: f0dd330fdf07d295ac468660cf60341796d5d501 Gitweb: http://git.kernel.org/tip/f0dd330fdf07d295ac468660cf60341796d5d501 Author: Yunlong Song yunlong.s...@huawei.com AuthorDate: Tue, 31 Mar 2015 21:46:35 +0800 Committer: Arnaldo Carvalho de Melo a...@redhat.com CommitDate: Wed, 8

[PATCH] perf record: Conditionally define CLOCK_MONOTONIC_RAW for older OSes

d. Signed-off-by: Yunlong Song --- tools/perf/builtin-record.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index cfdff50..5b0962a 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -731,6 +731,9

Re: [PATCH 3/9] perf sched replay: Alloc the memory of pid_to_task dynamically to adapt to the unexpected change of pid_max

On 2015/4/1 15:06, Yunlong Song wrote: > On 2015/3/31 23:56, Arnaldo Carvalho de Melo wrote: >> Em Tue, Mar 31, 2015 at 08:32:37AM -0600, David Ahern escreveu: >>> On 3/31/15 7:46 AM, Yunlong Song wrote: >>>> - BUG_ON(pid >= MAX_PID); >&

Re: [PATCH 3/9] perf sched replay: Alloc the memory of pid_to_task dynamically to adapt to the unexpected change of pid_max

On 2015/4/1 15:06, Yunlong Song wrote: On 2015/3/31 23:56, Arnaldo Carvalho de Melo wrote: Em Tue, Mar 31, 2015 at 08:32:37AM -0600, David Ahern escreveu: On 3/31/15 7:46 AM, Yunlong Song wrote: - BUG_ON(pid = MAX_PID); + if (sched-pid_to_task == NULL) { + if (sysctl__read_int

[PATCH] perf record: Conditionally define CLOCK_MONOTONIC_RAW for older OSes

-by: Yunlong Song yunlong.s...@huawei.com --- tools/perf/builtin-record.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index cfdff50..5b0962a 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -731,6

Re: [PATCH 0/9] perf sched replay: Make some improvements and fixes

On 2015/3/31 21:46, Yunlong Song wrote: > Hi, > Found some functions to improve and bugs to fix in perf sched replay. > > Yunlong Song (9): > perf sched replay: Use struct task_desc instead of struct task_task > for correct meaning > perf sched replay: Incre

Re: [PATCH 0/9] perf sched replay: Make some improvements and fixes

On 2015/3/31 21:46, Yunlong Song wrote: Hi, Found some functions to improve and bugs to fix in perf sched replay. Yunlong Song (9): perf sched replay: Use struct task_desc instead of struct task_task for correct meaning perf sched replay: Increase the MAX_PID value to fix

[tip:perf/core] perf data: Support using -f to override perf.data file ownership for 'convert'

Commit-ID: bd05954bfa17f03a7bd4454178ba09786b35e383 Gitweb: http://git.kernel.org/tip/bd05954bfa17f03a7bd4454178ba09786b35e383 Author: Yunlong Song AuthorDate: Thu, 2 Apr 2015 21:47:19 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Thu, 2 Apr 2015 13:18:52 -0300 perf data

[tip:perf/core] perf script: Support using -f to override perf.data file ownership

Commit-ID: 06af0f2c919d7c8f05efebe0d96a6f22297aafd4 Gitweb: http://git.kernel.org/tip/06af0f2c919d7c8f05efebe0d96a6f22297aafd4 Author: Yunlong Song AuthorDate: Thu, 2 Apr 2015 21:47:16 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Thu, 2 Apr 2015 13:18:50 -0300 perf script

[tip:perf/core] perf timechart: Support using -f to override perf.data file ownership

Commit-ID: 44f7e432e3dc8a13f5661e8b722f53645df083d1 Gitweb: http://git.kernel.org/tip/44f7e432e3dc8a13f5661e8b722f53645df083d1 Author: Yunlong Song AuthorDate: Thu, 2 Apr 2015 21:47:17 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Thu, 2 Apr 2015 13:18:50 -0300 perf timechart

[tip:perf/core] perf trace: Support using -f to override perf.data file ownership

Commit-ID: e366a6d8949f3cfab01906b42c591098d59f3f35 Gitweb: http://git.kernel.org/tip/e366a6d8949f3cfab01906b42c591098d59f3f35 Author: Yunlong Song AuthorDate: Thu, 2 Apr 2015 21:47:18 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Thu, 2 Apr 2015 13:18:51 -0300 perf trace

[tip:perf/core] perf lock: Support using -f to override perf.data file ownership

Commit-ID: c4ac732a0377d1544a8385393a9877b693ff0652 Gitweb: http://git.kernel.org/tip/c4ac732a0377d1544a8385393a9877b693ff0652 Author: Yunlong Song AuthorDate: Thu, 2 Apr 2015 21:47:14 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Thu, 2 Apr 2015 13:18:48 -0300 perf lock

[tip:perf/core] perf mem: Support using -f to override perf.data file ownership

Commit-ID: 62a1a63a77451dee8fd318a5106ca108d6a8ebcb Gitweb: http://git.kernel.org/tip/62a1a63a77451dee8fd318a5106ca108d6a8ebcb Author: Yunlong Song AuthorDate: Thu, 2 Apr 2015 21:47:15 +0800 Committer: Arnaldo Carvalho de Melo CommitDate: Thu, 2 Apr 2015 13:18:49 -0300 perf mem

<    1   2   3   4   5   6   7   8   >