Re: [PATCH 02/52] tools/perf/build: Add feature check core code

2013-10-10 Thread Namhyung Kim 
On Thu, 10 Oct 2013 09:21:21 +0200, Ingo Molnar wrote:
> * Namhyung Kim  wrote:
>
>> Hi Ingo,
>> 
>> On Tue,  8 Oct 2013 12:10:32 +0200, Ingo Molnar wrote:
>> > +feature_check = $(eval $(feature_check_code)); $(info CHK: 
>> > config/feature-checks/test-$(1))
>> > +define feature_check_code
>> > +  feature-$(2) := $(shell make -C config/feature-checks test-$1 
>> > >/dev/null 2>/dev/null && echo 1 || echo 0)
>> > +endef
>> > +
>> > +#
>> > +# Build the feature check binaries in parallel, ignore errors, ignore 
>> > return value and suppress output:
>> > +#
>> > +$(info Testing features:)
>> > +$(shell make -i -j -C config/feature-checks >/dev/null 2>&1)
>> > +$(info done)
>> > +
>> > +FEATURE_TESTS = hello
>> > +
>> > +$(foreach test,$(FEATURE_TESTS),$(call feature_check,$(test),$(test)))
>> 
>> So as far as I can see, all the feature checking goes twice - once in 
>> $(shell make ...) and another in $(foreach test,...).  They look like 
>> doing exactly same thing.  And the final Makefile in this series still 
>> has it twice.  Is it intentional or did I miss something?
>
> Yes - and this is a performance trick.
>
> The first run is to build them all in parallel:
>
>  $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) LDFLAGS=$(LDFLAGS) -i -j -C 
> config/feature-checks $(CORE_FEATURE_TESTS) >/dev/null 2>&1)
>
> Note how we build in parallel (-j), ignore errors (-i) and suppress all 
> output.
>
> This is very fast even if all testcases are rebuilt, but we cannot 
> possibly recover which feature test failed.
>
> So we run it once more, this time serially, and recover the test results. 
> The vast majority of testcases will pass, and the repeat build will do, in 
> essence:
>
>  comet:~/tip/tools/perf/config/feature-checks> perf stat --null --repeat 10 
> make test-libelf >/dev/null
>
>  Performance counter stats for 'make test-libelf' (10 runs):
>
>0.014682647 seconds time elapsed   
>( +-  0.14% )
>
> so it's just 14 msecs to recover each testcase result.

Okay, got it.  Thanks for the explanation! :)

>
> In theory we could optimize this further, by generating an actual test 
> success/failure file via the initial parallel make attempt, and recovering 
> those flags via the serial loop. That should be even faster.

Well, I think it's already fast enough. ;)

Thanks,
Namhyung
--
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: [PATCH 02/52] tools/perf/build: Add feature check core code

2013-10-10 Thread Ingo Molnar 

* Namhyung Kim  wrote:

> Hi Ingo,
> 
> On Tue,  8 Oct 2013 12:10:32 +0200, Ingo Molnar wrote:
> > +feature_check = $(eval $(feature_check_code)); $(info CHK: 
> > config/feature-checks/test-$(1))
> > +define feature_check_code
> > +  feature-$(2) := $(shell make -C config/feature-checks test-$1 >/dev/null 
> > 2>/dev/null && echo 1 || echo 0)
> > +endef
> > +
> > +#
> > +# Build the feature check binaries in parallel, ignore errors, ignore 
> > return value and suppress output:
> > +#
> > +$(info Testing features:)
> > +$(shell make -i -j -C config/feature-checks >/dev/null 2>&1)
> > +$(info done)
> > +
> > +FEATURE_TESTS = hello
> > +
> > +$(foreach test,$(FEATURE_TESTS),$(call feature_check,$(test),$(test)))
> 
> So as far as I can see, all the feature checking goes twice - once in 
> $(shell make ...) and another in $(foreach test,...).  They look like 
> doing exactly same thing.  And the final Makefile in this series still 
> has it twice.  Is it intentional or did I miss something?

Yes - and this is a performance trick.

The first run is to build them all in parallel:

 $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) LDFLAGS=$(LDFLAGS) -i -j -C 
config/feature-checks $(CORE_FEATURE_TESTS) >/dev/null 2>&1)

Note how we build in parallel (-j), ignore errors (-i) and suppress all 
output.

This is very fast even if all testcases are rebuilt, but we cannot 
possibly recover which feature test failed.

So we run it once more, this time serially, and recover the test results. 
The vast majority of testcases will pass, and the repeat build will do, in 
essence:

 comet:~/tip/tools/perf/config/feature-checks> perf stat --null --repeat 10 
make test-libelf >/dev/null

 Performance counter stats for 'make test-libelf' (10 runs):

   0.014682647 seconds time elapsed 
 ( +-  0.14% )

so it's just 14 msecs to recover each testcase result.

In theory we could optimize this further, by generating an actual test 
success/failure file via the initial parallel make attempt, and recovering 
those flags via the serial loop. That should be even faster.

Thanks,

Ingo
--
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: [PATCH 02/52] tools/perf/build: Add feature check core code

2013-10-10 Thread Ingo Molnar 

* Namhyung Kim namhy...@kernel.org wrote:

 Hi Ingo,
 
 On Tue,  8 Oct 2013 12:10:32 +0200, Ingo Molnar wrote:
  +feature_check = $(eval $(feature_check_code)); $(info CHK: 
  config/feature-checks/test-$(1))
  +define feature_check_code
  +  feature-$(2) := $(shell make -C config/feature-checks test-$1 /dev/null 
  2/dev/null  echo 1 || echo 0)
  +endef
  +
  +#
  +# Build the feature check binaries in parallel, ignore errors, ignore 
  return value and suppress output:
  +#
  +$(info Testing features:)
  +$(shell make -i -j -C config/feature-checks /dev/null 21)
  +$(info done)
  +
  +FEATURE_TESTS = hello
  +
  +$(foreach test,$(FEATURE_TESTS),$(call feature_check,$(test),$(test)))
 
 So as far as I can see, all the feature checking goes twice - once in 
 $(shell make ...) and another in $(foreach test,...).  They look like 
 doing exactly same thing.  And the final Makefile in this series still 
 has it twice.  Is it intentional or did I miss something?

Yes - and this is a performance trick.

The first run is to build them all in parallel:

 $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) LDFLAGS=$(LDFLAGS) -i -j -C 
config/feature-checks $(CORE_FEATURE_TESTS) /dev/null 21)

Note how we build in parallel (-j), ignore errors (-i) and suppress all 
output.

This is very fast even if all testcases are rebuilt, but we cannot 
possibly recover which feature test failed.

So we run it once more, this time serially, and recover the test results. 
The vast majority of testcases will pass, and the repeat build will do, in 
essence:

 comet:~/tip/tools/perf/config/feature-checks perf stat --null --repeat 10 
make test-libelf /dev/null

 Performance counter stats for 'make test-libelf' (10 runs):

   0.014682647 seconds time elapsed 
 ( +-  0.14% )

so it's just 14 msecs to recover each testcase result.

In theory we could optimize this further, by generating an actual test 
success/failure file via the initial parallel make attempt, and recovering 
those flags via the serial loop. That should be even faster.

Thanks,

Ingo
--
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: [PATCH 02/52] tools/perf/build: Add feature check core code

2013-10-10 Thread Namhyung Kim 
On Thu, 10 Oct 2013 09:21:21 +0200, Ingo Molnar wrote:
 * Namhyung Kim namhy...@kernel.org wrote:

 Hi Ingo,
 
 On Tue,  8 Oct 2013 12:10:32 +0200, Ingo Molnar wrote:
  +feature_check = $(eval $(feature_check_code)); $(info CHK: 
  config/feature-checks/test-$(1))
  +define feature_check_code
  +  feature-$(2) := $(shell make -C config/feature-checks test-$1 
  /dev/null 2/dev/null  echo 1 || echo 0)
  +endef
  +
  +#
  +# Build the feature check binaries in parallel, ignore errors, ignore 
  return value and suppress output:
  +#
  +$(info Testing features:)
  +$(shell make -i -j -C config/feature-checks /dev/null 21)
  +$(info done)
  +
  +FEATURE_TESTS = hello
  +
  +$(foreach test,$(FEATURE_TESTS),$(call feature_check,$(test),$(test)))
 
 So as far as I can see, all the feature checking goes twice - once in 
 $(shell make ...) and another in $(foreach test,...).  They look like 
 doing exactly same thing.  And the final Makefile in this series still 
 has it twice.  Is it intentional or did I miss something?

 Yes - and this is a performance trick.

 The first run is to build them all in parallel:

  $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) LDFLAGS=$(LDFLAGS) -i -j -C 
 config/feature-checks $(CORE_FEATURE_TESTS) /dev/null 21)

 Note how we build in parallel (-j), ignore errors (-i) and suppress all 
 output.

 This is very fast even if all testcases are rebuilt, but we cannot 
 possibly recover which feature test failed.

 So we run it once more, this time serially, and recover the test results. 
 The vast majority of testcases will pass, and the repeat build will do, in 
 essence:

  comet:~/tip/tools/perf/config/feature-checks perf stat --null --repeat 10 
 make test-libelf /dev/null

  Performance counter stats for 'make test-libelf' (10 runs):

0.014682647 seconds time elapsed   
( +-  0.14% )

 so it's just 14 msecs to recover each testcase result.

Okay, got it.  Thanks for the explanation! :)


 In theory we could optimize this further, by generating an actual test 
 success/failure file via the initial parallel make attempt, and recovering 
 those flags via the serial loop. That should be even faster.

Well, I think it's already fast enough. ;)

Thanks,
Namhyung
--
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: [PATCH 02/52] tools/perf/build: Add feature check core code

2013-10-09 Thread Namhyung Kim 
Hi Ingo,

On Tue,  8 Oct 2013 12:10:32 +0200, Ingo Molnar wrote:
> +feature_check = $(eval $(feature_check_code)); $(info CHK: 
> config/feature-checks/test-$(1))
> +define feature_check_code
> +  feature-$(2) := $(shell make -C config/feature-checks test-$1 >/dev/null 
> 2>/dev/null && echo 1 || echo 0)
> +endef
> +
> +#
> +# Build the feature check binaries in parallel, ignore errors, ignore return 
> value and suppress output:
> +#
> +$(info Testing features:)
> +$(shell make -i -j -C config/feature-checks >/dev/null 2>&1)
> +$(info done)
> +
> +FEATURE_TESTS = hello
> +
> +$(foreach test,$(FEATURE_TESTS),$(call feature_check,$(test),$(test)))

So as far as I can see, all the feature checking goes twice - once in
$(shell make ...) and another in $(foreach test,...).  They look like
doing exactly same thing.  And the final Makefile in this series still
has it twice.  Is it intentional or did I miss something?

Thanks,
Namhyung
--
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: [PATCH 02/52] tools/perf/build: Add feature check core code

2013-10-09 Thread Namhyung Kim 
Hi Ingo,

On Tue,  8 Oct 2013 12:10:32 +0200, Ingo Molnar wrote:
 +feature_check = $(eval $(feature_check_code)); $(info CHK: 
 config/feature-checks/test-$(1))
 +define feature_check_code
 +  feature-$(2) := $(shell make -C config/feature-checks test-$1 /dev/null 
 2/dev/null  echo 1 || echo 0)
 +endef
 +
 +#
 +# Build the feature check binaries in parallel, ignore errors, ignore return 
 value and suppress output:
 +#
 +$(info Testing features:)
 +$(shell make -i -j -C config/feature-checks /dev/null 21)
 +$(info done)
 +
 +FEATURE_TESTS = hello
 +
 +$(foreach test,$(FEATURE_TESTS),$(call feature_check,$(test),$(test)))

So as far as I can see, all the feature checking goes twice - once in
$(shell make ...) and another in $(foreach test,...).  They look like
doing exactly same thing.  And the final Makefile in this series still
has it twice.  Is it intentional or did I miss something?

Thanks,
Namhyung
--
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/

[PATCH 02/52] tools/perf/build: Add feature check core code

2013-10-08 Thread Ingo Molnar 
Start the split-out of the feature check code by adding a list of features to be
tested, and rules to process that list by building its matching feature-check
file in config/feature-checks/test-.c.

Add 'hello' as the initial feature.

This structure will allow us to build split-out feature checks in parallel and
thus speed up feature detection dramatically.

No change in functionality: no feature check is used by the build rules yet.

Cc: Arnaldo Carvalho de Melo 
Cc: Peter Zijlstra 
Cc: Namhyung Kim 
Cc: David Ahern 
Cc: Jiri Olsa 
Link: http://lkml.kernel.org/n/tip-pixkihgscfaohffigq5yt...@git.kernel.org
Signed-off-by: Ingo Molnar 
---
 tools/perf/config/Makefile| 16 
 tools/perf/config/feature-checks/Makefile | 16 
 tools/perf/config/feature-checks/test-hello.c |  6 ++
 3 files changed, 38 insertions(+)

diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
index 34be743..daefe2d 100644
--- a/tools/perf/config/Makefile
+++ b/tools/perf/config/Makefile
@@ -89,6 +89,22 @@ CFLAGS += -std=gnu99
 
 EXTLIBS = -lelf -lpthread -lrt -lm -ldl
 
+feature_check = $(eval $(feature_check_code)); $(info CHK: 
config/feature-checks/test-$(1))
+define feature_check_code
+  feature-$(2) := $(shell make -C config/feature-checks test-$1 >/dev/null 
2>/dev/null && echo 1 || echo 0)
+endef
+
+#
+# Build the feature check binaries in parallel, ignore errors, ignore return 
value and suppress output:
+#
+$(info Testing features:)
+$(shell make -i -j -C config/feature-checks >/dev/null 2>&1)
+$(info done)
+
+FEATURE_TESTS = hello
+
+$(foreach test,$(FEATURE_TESTS),$(call feature_check,$(test),$(test)))
+
 ifeq ($(call try-cc,$(SOURCE_HELLO),$(CFLAGS) -Werror 
-fstack-protector-all,-fstack-protector-all),y)
   CFLAGS += -fstack-protector-all
 endif
diff --git a/tools/perf/config/feature-checks/Makefile 
b/tools/perf/config/feature-checks/Makefile
new file mode 100644
index 000..b3f6372
--- /dev/null
+++ b/tools/perf/config/feature-checks/Makefile
@@ -0,0 +1,16 @@
+
+FILES=test-hello
+
+all: $(FILES)
+
+BUILD = $(CC) -o $(OUTPUT)$@ $@.c
+
+###
+
+test-hello: test-hello.c
+   $(BUILD)
+
+###
+
+clean:
+   rm -f $(FILES)
diff --git a/tools/perf/config/feature-checks/test-hello.c 
b/tools/perf/config/feature-checks/test-hello.c
new file mode 100644
index 000..c9f398d
--- /dev/null
+++ b/tools/perf/config/feature-checks/test-hello.c
@@ -0,0 +1,6 @@
+#include 
+
+int main(void)
+{
+   return puts("hi");
+}
-- 
1.8.3.1

--
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/

[PATCH 02/52] tools/perf/build: Add feature check core code

2013-10-08 Thread Ingo Molnar 
Start the split-out of the feature check code by adding a list of features to be
tested, and rules to process that list by building its matching feature-check
file in config/feature-checks/test-feature.c.

Add 'hello' as the initial feature.

This structure will allow us to build split-out feature checks in parallel and
thus speed up feature detection dramatically.

No change in functionality: no feature check is used by the build rules yet.

Cc: Arnaldo Carvalho de Melo a...@redhat.com
Cc: Peter Zijlstra a.p.zijls...@chello.nl
Cc: Namhyung Kim namhy...@kernel.org
Cc: David Ahern dsah...@gmail.com
Cc: Jiri Olsa jo...@redhat.com
Link: http://lkml.kernel.org/n/tip-pixkihgscfaohffigq5yt...@git.kernel.org
Signed-off-by: Ingo Molnar mi...@kernel.org
---
 tools/perf/config/Makefile| 16 
 tools/perf/config/feature-checks/Makefile | 16 
 tools/perf/config/feature-checks/test-hello.c |  6 ++
 3 files changed, 38 insertions(+)

diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
index 34be743..daefe2d 100644
--- a/tools/perf/config/Makefile
+++ b/tools/perf/config/Makefile
@@ -89,6 +89,22 @@ CFLAGS += -std=gnu99
 
 EXTLIBS = -lelf -lpthread -lrt -lm -ldl
 
+feature_check = $(eval $(feature_check_code)); $(info CHK: 
config/feature-checks/test-$(1))
+define feature_check_code
+  feature-$(2) := $(shell make -C config/feature-checks test-$1 /dev/null 
2/dev/null  echo 1 || echo 0)
+endef
+
+#
+# Build the feature check binaries in parallel, ignore errors, ignore return 
value and suppress output:
+#
+$(info Testing features:)
+$(shell make -i -j -C config/feature-checks /dev/null 21)
+$(info done)
+
+FEATURE_TESTS = hello
+
+$(foreach test,$(FEATURE_TESTS),$(call feature_check,$(test),$(test)))
+
 ifeq ($(call try-cc,$(SOURCE_HELLO),$(CFLAGS) -Werror 
-fstack-protector-all,-fstack-protector-all),y)
   CFLAGS += -fstack-protector-all
 endif
diff --git a/tools/perf/config/feature-checks/Makefile 
b/tools/perf/config/feature-checks/Makefile
new file mode 100644
index 000..b3f6372
--- /dev/null
+++ b/tools/perf/config/feature-checks/Makefile
@@ -0,0 +1,16 @@
+
+FILES=test-hello
+
+all: $(FILES)
+
+BUILD = $(CC) -o $(OUTPUT)$@ $@.c
+
+###
+
+test-hello: test-hello.c
+   $(BUILD)
+
+###
+
+clean:
+   rm -f $(FILES)
diff --git a/tools/perf/config/feature-checks/test-hello.c 
b/tools/perf/config/feature-checks/test-hello.c
new file mode 100644
index 000..c9f398d
--- /dev/null
+++ b/tools/perf/config/feature-checks/test-hello.c
@@ -0,0 +1,6 @@
+#include stdio.h
+
+int main(void)
+{
+   return puts(hi);
+}
-- 
1.8.3.1

--
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/