All C compiler warnings are disabled are disabled by -w. This change
removes the -w from flex and bison targets. To avoid implicit
declarations header files are declared as targets and included.
Tested with GCC 9.3.0 and clang 9.0.1.
Signed-off-by: Ian Rogers
---
tools/perf/util/Build
This avoids multiple declarations if the flex header is included.
Signed-off-by: Ian Rogers
---
tools/perf/util/pmu.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h
index 85e0c7f2515c..f971d9aa4570 100644
--- a/tools/perf/util/pmu.h
+++ b/tools
Fixes: a26e47162d76 (perf tools: Move ALLOC_LIST into a function)
Signed-off-by: Ian Rogers
---
tools/perf/util/parse-events.y | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y
index d4e076c9c2ab..acef87d9af58
Arrays are pointer types and don't need their address taking.
Fixes: 8255718f4bed (perf pmu: Expand PMU events by prefix match)
Signed-off-by: Ian Rogers
---
tools/perf/util/parse-events.y | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/util/parse-events.y b/tools
e, name) < 0)
> goto error;
>
> if (!strcmp(name, "msr"))
> @@ -2498,6 +2501,9 @@ static int process_pmu_mappings(struct feat_fd *ff,
> void *data __maybe_unused)
> free(name);
> pmu_num--;
>
On Sun, 2020-06-07 at 16:40 +0800, Ian Kent wrote:
> Hi Greg,
>
> On Mon, 2020-05-25 at 13:47 +0800, Ian Kent wrote:
> > @@ -189,9 +189,9 @@ int kernfs_iop_getattr(const struct path *path,
> > struct kstat *stat,
> > struct inode *inode = d_inode(path->dentry);
On Wed, 2020-06-03 at 10:15 +0800, Ian Kent wrote:
> On Tue, 2020-06-02 at 16:55 +0100, David Howells wrote:
> > [[ With regard to the mount/sb notifications and fsinfo(), Karel
> > Zak
> > and
> >Ian Kent have been working on making libmount use them,
> &
Hi Greg,
On Mon, 2020-05-25 at 13:47 +0800, Ian Kent wrote:
> @@ -189,9 +189,9 @@ int kernfs_iop_getattr(const struct path *path,
> struct kstat *stat,
> struct inode *inode = d_inode(path->dentry);
> struct kernfs_node *kn = inode->i_private;
>
>
ommit:
> >
> >
> > commit: ea7c5fc39ab005b501e0c7666c29db36321e4f74 ("[PATCH 1/4]
> > kernfs: switch kernfs to use an rwsem")
> > url:
> > https://github.com/0day-ci/linux/commits/Ian-Kent/kernfs-proposed-locking-and-concurrency-improvement/20200525-134849
> >
>
> Seriou
On 04/06/2020 08:25, Miklos Szeredi wrote:
> On Wed, Jun 3, 2020 at 6:15 PM Colin Ian King
> wrote:
>>
>> On 03/06/2020 17:11, Amir Goldstein wrote:
>>> On Wed, Jun 3, 2020 at 6:46 PM Colin King wrote:
>>>>
>>>> From: Colin Ian King
>&g
On 03/06/2020 17:11, Amir Goldstein wrote:
> On Wed, Jun 3, 2020 at 6:46 PM Colin King wrote:
>>
>> From: Colin Ian King
>>
>> There are two error return paths where the call to path_put is
>> dereferencing the null pointer 'stack'. Fix this by avoiding the
>
On 03/06/2020 15:45, Dan Carpenter wrote:
> On Wed, Jun 03, 2020 at 03:18:46PM +0100, Colin Ian King wrote:
>> On 03/06/2020 15:09, Dan Carpenter wrote:
>>> On Wed, Jun 03, 2020 at 02:51:02PM +0100, Colin King wrote:
>>>> From: Colin Ian King
>>>>
&g
On 03/06/2020 15:09, Dan Carpenter wrote:
> On Wed, Jun 03, 2020 at 02:51:02PM +0100, Colin King wrote:
>> From: Colin Ian King
>>
>> The variable error is being initialized with a value that is
>> never read and the -ENOMEM error return is being returned anyhow
>>
On 03/06/2020 13:15, Colin Ian King wrote:
> On 03/06/2020 13:11, Colin King wrote:
>> From: Colin Ian King
>>
>> There are two spelling mistakes in the Kconfig text. Fix these.
>>
>> Signed-off-by: Colin Ian King
>> ---
>> drivers/usb/host/Kcon
On 03/06/2020 13:11, Colin King wrote:
> From: Colin Ian King
>
> There are two spelling mistakes in the Kconfig text. Fix these.
>
> Signed-off-by: Colin Ian King
> ---
> drivers/usb/host/Kconfig | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
>
On Tue, 2020-06-02 at 16:55 +0100, David Howells wrote:
>
> [[ With regard to the mount/sb notifications and fsinfo(), Karel Zak
> and
>Ian Kent have been working on making libmount use them,
> preparatory to
>working on systemd:
>
> https://github.com/kar
On Wed, Apr 29, 2020 at 12:08 PM Ian Rogers wrote:
>
> Counters may be being used for pinned or other events which inhibit the
> instruction counter in the test from being scheduled - time_enabled > 0
> but time_running == 0. This causes the test to fail with division by 0.
>
On Tue, Jun 2, 2020 at 12:08 PM Jiri Olsa wrote:
>
> On Tue, Jun 02, 2020 at 10:58:32AM -0700, Ian Rogers wrote:
>
> SNIP
>
> > > +static int check_parse_fake(const char *id)
> > > +{
> > > + struct parse_events_error error;
> >
> > ni
- add missing free to patch 1 [Ian]
> - factor pmu-events test functions and reuse it in the new test [Ian]
> - add fake_pmu bool to parse_events interface [Ian]
> - simplify metric tests
> - use proper cover letter subject ;-)
>
> I actually reworked the 2 patches Ian ac
if (ret && same_cpu) {
> pr_warning("Parse event failed metric '%s' id '%s' expr
> '%s'\n",
> pe->metric_name, id, pe->metric_expr);
> @@ -430,6 +431,18 @@ static int check_parse_cpu(const char *id, bool
> same_cpu, s
nsure this condition is met and removing the
> group (with warning) if we detect mixed cpus, like:
>
> $ sudo perf stat -e
> '{power/energy-cores/,cycles},{instructions,power/energy-cores/}'
> WARNING: event cpu maps do not match, disabling group:
> anon group { power/energy-cores/, c
On Mon, Jun 1, 2020 at 6:21 AM Jiri Olsa wrote:
>
> On Mon, Jun 01, 2020 at 12:44:15AM -0700, Ian Rogers wrote:
>
> SNIP
>
> > > + memset(, 0, sizeof(error));
> > > + ret = parse_events_fake(evlist, id, );
> > > + if (ret) {
> &g
nsure this condition is met and removing the
> group (with warning) if we detect mixed cpus, like:
>
> $ sudo perf stat -e
> '{power/energy-cores/,cycles},{instructions,power/energy-cores/}'
> WARNING: event cpu maps do not match, disabling group:
> anon group { power/energy-cores/, c
On Mon, Jun 1, 2020 at 6:09 AM Jiri Olsa wrote:
>
> On Mon, Jun 01, 2020 at 12:55:44AM -0700, Ian Rogers wrote:
> > On Sun, May 24, 2020 at 3:43 PM Jiri Olsa wrote:
> > >
> > > Adding new test that process metrics code and checks
> > > the expected
The following commit has been merged into the timers/core branch of tip:
Commit-ID: ac593e62b0cfcbc53502be8b6c7e40fed8baff8c
Gitweb:
https://git.kernel.org/tip/ac593e62b0cfcbc53502be8b6c7e40fed8baff8c
Author:Colin Ian King
AuthorDate:Tue, 19 May 2020 23:44:28 +01:00
On Sun, May 31, 2020 at 11:17 PM Nick Gasson wrote:
>
> On 05/28/20 17:32 PM, Ian Rogers wrote:
> >
> > So on tip/perf/core with:
> > 1c0cd2dbb993 perf jvmti: Fix jitdump for methods without debug info
> > 3ce17c1e52f4 perf jvmti: remove redundant jitdump line t
On Sun, May 24, 2020 at 3:43 PM Jiri Olsa wrote:
>
> Adding new metri test for frontend metric. It's stolen
s/metri/metric/
> from x86 pmu events.
>
> Signed-off-by: Jiri Olsa
Acked-by: Ian Rogers
Thanks,
Ian
> ---
> tools/perf/test
On Sun, May 24, 2020 at 3:43 PM Jiri Olsa wrote:
>
> Adding new test that process metrics code and checks
> the expected results. Starting with easy ipc metric.
>
> Signed-off-by: Jiri Olsa
Acked-by: Ian Rogers
I wonder if there's a better organization with testing in
pmu-e
("parsing '%s'\n", str);
> +
> + expr__ctx_init();
> + if (expr__find_other(str, NULL, , 0) < 0) {
> + pr_err("expr__find_other failed\n");
> + return -1;
> + }
> +
> +
for testing purposes coming in following
> changes.
>
> Signed-off-by: Jiri Olsa
Acked-by: Ian Rogers
Alternatively fake_pmu could be an argument to parse_events.
Thanks,
Ian
> ---
> tools/perf/util/parse-events.c | 48 +-
> tools/perf/util/parse-
39 @@ PE_PMU_EVENT_PRE '-' PE_PMU_EVENT_SUF sep_dc
> YYABORT;
> $$ = list;
> }
> +|
> +PE_PMU_EVENT_FAKE sep_dc
> +{
> + struct list_head *list;
> + int err;
> +
> + list = alloc_list();
> + if (!list)
> +
On Sun, May 31, 2020 at 7:29 AM Namhyung Kim wrote:
>
> Hi Ian,
>
> On Sat, May 30, 2020 at 7:53 AM Ian Rogers wrote:
> >
> > This is currently working due to extra include paths in the build.
> >
> > Before:
> > $ cd tools/perf/arch/arm64/util
>
y-cores/, cycles }
> anon group { instructions, power/energy-cores/ }
This is really cool! I wonder if there is a better wording for 'event
cpu maps' ? It may be useful to list what the cpu maps are for the
events as a diagnostic aid.
Thanks,
Ian
> Fixes: 6a4bb04caacc8 ("perf tools: E
Tail call optimizations can remove stack frames that are used in
unwinding tests. Add an attribute that can be used to disable the tail
call optimization. Tested on clang and GCC.
Signed-off-by: Ian Rogers
---
tools/include/linux/compiler-gcc.h | 8
tools/include/linux/compiler.h
Avoid a false positive caused by assembly code in arch/x86.
In tests, zero the perf_event to avoid uninitialized memory uses.
Warnings were caught using clang with -fsanitize=memory.
Signed-off-by: Ian Rogers
---
tools/perf/arch/x86/tests/dwarf-unwind.c | 8
tools/perf/tests/dwarf
Fix stack frame count and memory sanitizer issues when running the
dwarf unwinding test with the elfutils/libdw unwinder (libunwind
disabled).
Ian Rogers (3):
tools compiler.h: Add attribute to disable tail calls
perf tests: Don't tail call optimize in unwind test
perf test: Initialize
The tail call optimization can unexpectedly make the stack smaller and
cause the test to fail.
Signed-off-by: Ian Rogers
---
tools/perf/tests/dwarf-unwind.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/tools/perf/tests/dwarf-unwind.c b/tools/perf/tests/dwarf
This is currently working due to extra include paths in the build.
Before:
$ cd tools/perf/arch/arm64/util
$ ls -la ../../util/unwind-libdw.h
ls: cannot access '../../util/unwind-libdw.h': No such file or directory
After:
$ ls -la ../../../util/unwind-libdw.h
-rw-r- 1 irogers irogers 553 Apr
On Fri, May 29, 2020 at 10:36 AM Arnaldo Carvalho de Melo
wrote:
>
> Em Fri, May 29, 2020 at 02:23:10PM -0300, Arnaldo Carvalho de Melo escreveu:
> > Em Fri, May 29, 2020 at 10:03:51AM -0700, Ian Rogers escreveu:
> > > On Tue, May 5, 2020 at 11:29 AM Ian Rogers wrote:
On Fri, May 29, 2020 at 10:23 AM Arnaldo Carvalho de Melo
wrote:
>
> Em Fri, May 29, 2020 at 10:03:51AM -0700, Ian Rogers escreveu:
> > On Tue, May 5, 2020 at 11:29 AM Ian Rogers wrote:
> > >
> > > From: Stephane Eranian
> > >
> > >
On Tue, May 5, 2020 at 11:29 AM Ian Rogers wrote:
>
> From: Stephane Eranian
>
> This patch links perf with the libpfm4 library if it is available
> and LIBPFM4 is passed to the build. The libpfm4 library
> contains hardware event tables for all processors supporte
06-09.
These should be Cc'd to Greg KH and to de...@driverdev.osuosl.org.
Cheers,
Ian
--
-=( Ian Abbott || Web: www.mev.co.uk )=-
-=( MEV Ltd. is a company registered in England & Wales. )=-
-=( Registered number: 02862268. Registered address:)=-
-=( 15 West Park Road, Bramhall, STOC
On 29/05/2020 01:35, Al Viro wrote:
From: Al Viro
Signed-off-by: Al Viro
---
drivers/staging/comedi/comedi_fops.c | 48 ++--
1 file changed, 24 insertions(+), 24 deletions(-)
Signed-off-by: Ian Abbott
--
-=( Ian Abbott || Web: www.mev.co.uk )=-
-=( MEV
On 29/05/2020 01:35, Al Viro wrote:
From: Al Viro
Signed-off-by: Al Viro
---
drivers/staging/comedi/comedi_fops.c | 138 ---
1 file changed, 48 insertions(+), 90 deletions(-)
Signed-off-by: Ian Abbott
--
-=( Ian Abbott || Web: www.mev.co.uk )=-
-=( MEV
1 file changed, 30 insertions(+), 38 deletions(-)
Signed-off-by: Ian Abbott
--
-=( Ian Abbott || Web: www.mev.co.uk )=-
-=( MEV Ltd. is a company registered in England & Wales. )=-
-=( Registered number: 02862268. Registered address:)=-
-=( 15 West
/staging/comedi/comedi_fops.c | 451 +-
4 files changed, 448 insertions(+), 487 deletions(-)
delete mode 100644 drivers/staging/comedi/comedi_compat32.c
delete mode 100644 drivers/staging/comedi/comedi_compat32.h
Signed-off-by: Ian Abbott
--
-=( Ian Abbott || Web
On 29/05/2020 01:35, Al Viro wrote:
From: Al Viro
Signed-off-by: Al Viro
---
drivers/staging/comedi/comedi_fops.c | 26 --
1 file changed, 8 insertions(+), 18 deletions(-)
Signed-off-by: Ian Abbott
--
-=( Ian Abbott || Web: www.mev.co.uk )=-
-=( MEV Ltd
++--
drivers/staging/comedi/comedi_internal.h | 2 +-
drivers/staging/comedi/range.c | 17 ++---
3 files changed, 27 insertions(+), 35 deletions(-)
Signed-off-by: Ian Abbott
--
-=( Ian Abbott || Web: www.mev.co.uk )=-
-=( MEV Ltd. is a company
On 29/05/2020 01:35, Al Viro wrote:
From: Al Viro
Signed-off-by: Al Viro
---
drivers/staging/comedi/comedi_fops.c | 45 ++--
1 file changed, 22 insertions(+), 23 deletions(-)
Signed-off-by: Ian Abbott
--
-=( Ian Abbott || Web: www.mev.co.uk )=-
-=( MEV
On 29/05/2020 01:35, Al Viro wrote:
From: Al Viro
Signed-off-by: Al Viro
---
drivers/staging/comedi/comedi_fops.c | 20
1 file changed, 12 insertions(+), 8 deletions(-)
Signed-off-by: Ian Abbott
--
-=( Ian Abbott || Web: www.mev.co.uk )=-
-=( MEV Ltd. is a company
= -EFAULT;
Missing an 'else' here:
+ rc = do_insn_ioctl(dev, , file);
break;
+ }
case COMEDI_POLL:
--
-=( Ian Abbott || Web: www.mev.co.uk )=-
-=( MEV Ltd. is a company registered in England & Wales. )=-
-=( Registered number: 02862268. Registered add
On 28/05/2020 16:55, Nikolay Aleksandrov wrote:
> On 28/05/2020 18:53, Colin Ian King wrote:
>> On 28/05/2020 15:55, Nikolay Aleksandrov wrote:
>>> On 28/05/2020 17:53, Nikolay Aleksandrov wrote:
>>>> On 28/05/2020 17:51, Colin King wrote:
>>>>> Fro
On 28/05/2020 15:55, Nikolay Aleksandrov wrote:
> On 28/05/2020 17:53, Nikolay Aleksandrov wrote:
>> On 28/05/2020 17:51, Colin King wrote:
>>> From: Colin Ian King
>>>
>>> The allocation failure check for nhg->spare is currently checking
>>> the p
On Wed, May 27, 2020 at 1:59 PM Ian Rogers wrote:
>
> On Wed, May 27, 2020 at 12:27 PM Steve MacLean
> wrote:
> >
> > >> ** Implemented solution
> > >>
> > >> This patch solves the issue by removing // anon mmap events for
riginal code
came from oprofile's jit support:
https://sourceforge.net/p/oprofile/oprofile/ci/master/tree/opjitconv/debug_line.c#l325
Acked-by: Ian Rogers
> Signed-off-by: Nick Gasson
> ---
> tools/perf/util/genelf_debug.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-
pair. Reported by Ian Rogers. This reduces the .dump size for
> SPECjbb from ~230MB to ~40MB.
>
> Signed-off-by: Nick Gasson
> ---
> Changes in v2:
> - Split the unrelated DWARF debug fix into a separate patch
> - Added a comment about the use of c->method
On Wed, May 27, 2020 at 9:39 PM Nick Gasson wrote:
>
> On 05/28/20 02:08 AM, Ian Rogers wrote:
> >>
> >> I noticed it loses information when the Hotspot code cache is
> >> resized. I've been working around that by setting
> >> -XX:InitialCodeCacheS
;, discard "//anon*" mmap events for any pid
> >> which has sucessfully processed a jitdump file.
> >
> >
> > Thanks Steve this is an important fix! As //anon could be for malloc or
> > other uses, should the stripping behavior be behind a flag?
> >
non mmap events for any
> process which has a valid jit-.dump file.
>
> It tracks on a per process basis to handle the case where some running
> apps support jit-.dump, but some only support perf-.map.
>
> It adds new assumptions:
> * // anon mmap events are only required
On Tue, May 26, 2020 at 10:40 PM Nick Gasson wrote:
>
> On 05/27/20 13:03 PM, Ian Rogers wrote:
> >
> > Great result, thanks! I note there is a lack of symbolization when
> > benchmarking a few Java applications. I'll try to see if there's a
> > sensible resolutio
pair. Reported by Ian Rogers. This reduces the .dump size for
> SPECjbb from ~230MB to ~40MB.
Great result, thanks! I note there is a lack of symbolization when
benchmarking a few Java applications. I'll try to see if there's a
sensible resolution for those.
> Also fix an error in the DWARF lin
Hi,
Static analysis with Coverity has detected a potential issue with the
commit in linux-next:
commit 368982cd7d1bd41cd39049c794990aca3770db44
Author: Pablo Neira Ayuso
Date: Wed May 23 09:17:24 2018 +0200
netfilter: nfnetlink_queue: resolve clash for unconfirmed conntracks
Namely:
On Mon, 2020-05-25 at 08:16 +0200, Greg Kroah-Hartman wrote:
> On Mon, May 25, 2020 at 01:46:59PM +0800, Ian Kent wrote:
> > For very large systems with hundreds of CPUs and TBs of RAM booting
> > can
> > take a very long time.
> >
> > Initial reports sh
linux-next for broader exposure and hopefully have
them merged into mainline.
---
Ian Kent (4):
kernfs: switch kernfs to use an rwsem
kernfs: move revalidate to be near lookup
kernfs: improve kernfs path resolution
kernfs: use revision to identify directory node changes
fs
The kernfs global lock restricts the ability to perform kernfs node
lookup operations in parallel.
Change the kernfs mutex to an rwsem so that, when oppertunity arises,
node searches can be done in parallel.
Signed-off-by: Ian Kent
---
fs/kernfs/dir.c | 119
If a kernfs directory node hasn't changed there's no need to search for
an added (or removed) child dentry.
Add a revision counter to kernfs directory nodes so it can be used
to detect if a directory node has changed.
Signed-off-by: Ian Kent
---
fs/kernfs/dir.c | 17
where possible and use the dentry cache
handling of negative hashed dentries to avoid allocating (and freeing
shortly after) new dentries on every negative lookup.
Signed-off-by: Ian Kent
---
fs/kernfs/dir.c | 87 ++-
1 file changed, 72
eed for a forward declaration.
There's no functional change from this patch.
Signed-off-by: Ian Kent
---
fs/kernfs/dir.c | 86 ---
1 file changed, 43 insertions(+), 43 deletions(-)
diff --git a/fs/kernfs/dir.c b/fs/kernfs/dir.c
index d8213fc65
On Fri, May 22, 2020 at 7:59 AM Ian Rogers wrote:
>
>
>
> On Fri, May 22, 2020, 7:49 AM Arnaldo Carvalho de Melo
> wrote:
>>
>> Em Fri, May 22, 2020 at 12:13:11PM +0200, Jiri Olsa escreveu:
>> > On Thu, May 21, 2020 at 02:22:35PM -0300, Arnaldo Carvalho
Before:
Usage: perf list [] [hw|sw|cache|tracepoint|pmu|sdt|event_glob]
After:
Usage: perf list []
[hw|sw|cache|tracepoint|pmu|sdt|metric|metricgroup|event_glob]
Signed-off-by: Ian Rogers
---
tools/perf/builtin-list.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools
On Thu, May 21, 2020 at 10:14 PM Tibor Billes wrote:
>
> Hi,
>
> On Mon, 18 May 2020, Ian Rogers wrote:
>
> > On Sat, May 16, 2020 at 6:36 AM Billes Tibor wrote:
> > >
> > > Hi,
> > >
> > > I've been hitting a freeze on my laptop sinc
On Thu, May 21, 2020 at 3:25 PM Ian Rogers wrote:
>
> This patch documents a flag added in the following kernel commit:
>
> commit d2cd9ede6e193dd7d88b6d27399e96229a551b19
> Author: Rik van Riel
> Date: Wed Sep 6 16:25:15 2017 -0700
>
> mm,fork: i
c0c4f6c29c494c466f3a2a6273c5b55b76a72927
Author: Rik van Riel
Date: Tue Sep 19 20:32:00 2017 +0200
madvise.2: Document MADV_WIPEONFORK and MADV_KEEPONFORK
Signed-off-by: Ian Rogers
---
man5/proc.5 | 1 +
1 file changed, 1 insertion(+)
diff --git a/man5/proc.5 b/man5/proc.5
index 46d603f28..3e5b91af0
On Thu, May 21, 2020 at 3:54 AM Jiri Olsa wrote:
>
> On Wed, May 20, 2020 at 03:42:02PM -0700, Ian Rogers wrote:
>
> SNIP
>
> > >
> > > hum, I think that's also concern if you are multiplexing 2 groups and one
> > > metric getting events from bo
ode has changed a lot recently, please take
> a look at my perf/core branch and keep Ian Rogers in the loop,
>
> - Arnaldo
Thanks, I'd posted the same change here:
https://lore.kernel.org/lkml/20200508053629.210324-10-irog...@google.com/
but it looks like it was mislaid. Arnaldo, I can re-post
On Wed, May 20, 2020 at 3:09 PM Jiri Olsa wrote:
>
> On Wed, May 20, 2020 at 09:50:13AM -0700, Ian Rogers wrote:
>
> SNIP
>
> > > > + }
> > >
> > > is the groupping still enabled when we merge groups? could part
> > > of t
On Wed, May 20, 2020 at 6:14 AM Jiri Olsa wrote:
>
> On Wed, May 20, 2020 at 12:28:10AM -0700, Ian Rogers wrote:
> > Currently event groups are placed into groups_list at the same time as
> > the events string containing the events is built. Separate these two
> &
If a metric contains the duration_time event then the event is placed
outside of the metric's group of events. Rather than split the group,
make it so the duration_time is immediately after the group.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 18 +-
1 file
Use early return in metricgroup__add_metric and try to make the intent
of the returns more intention revealing.
Suggested-by: Jiri Olsa
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 22 +++---
1 file changed, 15 insertions(+), 7 deletions(-)
diff --git a/tools
event appears once, before it
appeared 5 times.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 91 ---
1 file changed, 62 insertions(+), 29 deletions(-)
diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util/metricgroup.c
index bac63524d12f
redundant - it can reuse the events from the large group.
A later patch will add this sharing.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 16 +++-
1 file changed, 15 insertions(+), 1 deletion(-)
diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util
Remove unnecessary commas from events before they are parsed. This
avoids ',' being echoed by parse-events.l.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util
is the same or larger
than that of the second. Sharing may increase or lower accuracy and so
is now configurable.
Signed-off-by: Ian Rogers
---
tools/perf/Documentation/perf-stat.txt | 19 +++
tools/perf/builtin-stat.c | 11 -
tools/perf/util/metricgroup.c | 68
y commas were passed to
parse-events and were echoed. Fix a bug where the same event could be
matched more than once with --metric-no-group, causing there to be
events missing.
https://lore.kernel.org/lkml/20200508053629.210324-1-irog...@google.com/
Ian Rogers (7):
perf metricgroup: Always plac
in a later patch.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 33 +
1 file changed, 21 insertions(+), 12 deletions(-)
diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util/metricgroup.c
index 5c0603ef4c75..dca433520b92 100644
--- a/tools/perf
On Wed, May 20, 2020 at 6:49 AM Jiri Olsa wrote:
>
> On Wed, May 20, 2020 at 12:28:12AM -0700, Ian Rogers wrote:
>
> SNIP
>
> >
> > @@ -157,7 +183,7 @@ static int metricgroup__setup_events(struct list_head
> > *groups,
> > int i = 0;
> >
On Wed, May 20, 2020 at 6:14 AM Jiri Olsa wrote:
>
> On Wed, May 20, 2020 at 12:28:07AM -0700, Ian Rogers wrote:
> > Metric groups contain metrics. Metrics create groups of events to
> > ideally be scheduled together. Often metrics refer to the same events,
> > for exampl
t was to compare it against
> a hopefully better schedule generated by or-tools, but I never
> got around to coding that up.
>
> -Andi
Agreed, and this relates to your comments about doing the algorithm as
a separate pass outside of find_evsel_group. For that, I don't
disagree but would like to land what we have and then follow up with
improvements.
Thanks,
Ian
On Wed, May 20, 2020 at 12:41 AM Jiri Olsa wrote:
>
> Ian reported that we allow to parse following:
>
> $ perf stat -e ,cycles true
>
> which is wrong and we should fail, like we do with this fix:
>
> $ perf stat -e ,cycles true
> e
in a later patch.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 38 +++
1 file changed, 25 insertions(+), 13 deletions(-)
diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util/metricgroup.c
index 7a43ee0a2e40..afd960d03a77 100644
--- a/tools
Remove unnecessary commas from events before they are parsed. This
avoids ',' being echoed by parse-events.l.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 14 +++---
1 file changed, 11 insertions(+), 3 deletions(-)
diff --git a/tools/perf/util/metricgroup.c b/tools
is the same or larger
than that of the second. Sharing may increase or lower accuracy and so
is now configurable.
Signed-off-by: Ian Rogers
---
tools/perf/Documentation/perf-stat.txt | 19 +++
tools/perf/builtin-stat.c | 11 -
tools/perf/util/metricgroup.c | 68
If a metric contains the duration_time event then the event is placed
outside of the metric's group of events. Rather than split the group,
make it so the duration_time is immediately after the group.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 18 +-
1 file
event appears once, before it
appeared 5 times.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 91 ---
1 file changed, 62 insertions(+), 29 deletions(-)
diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util/metricgroup.c
index 52e4c3e4748a
nel/git/acme/linux.git branch tmp.perf/core
Compared to RFC v3: fix a bug where unnecessary commas were passed to
parse-events and were echoed. Fix a bug where the same event could be
matched more than once with --metric-no-group, causing there to be
events missing.
https://lore.kernel.org/lkml/
redundant - it can reuse the events from the large group.
A later patch will add this sharing.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 16 +++-
1 file changed, 15 insertions(+), 1 deletion(-)
diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util
Use a bitmap rather than an array of bools.
Signed-off-by: Ian Rogers
---
tools/perf/util/metricgroup.c | 18 --
1 file changed, 8 insertions(+), 10 deletions(-)
diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util/metricgroup.c
index 6772d256dfdf..a16f60da06ab 100644
On Tue, May 19, 2020 at 6:54 PM Arnaldo Carvalho de Melo
wrote:
>
> Em Wed, Apr 22, 2020 at 10:36:15AM -0700, Ian Rogers escreveu:
> > During the processing of /proc during event synthesis new processes may
> > start. Add a dummy event if /proc is to be processed
On Tue, May 19, 2020 at 12:06 PM Arnaldo Carvalho de Melo
wrote:
>
> Em Fri, May 15, 2020 at 03:17:31PM -0700, Ian Rogers escreveu:
> > Break pmu-events test into 2 and add a test to verify that all pmu
> > metric expressions simply parse. Try to parse all metric ids/eve
ersion I was running.
>
> I tried reverting fd7d55172d1e2e501e6da0a5c1de25f06612dc2e which is the
> first bad commit
> according to my bisect, but there were conflicts, and I'm not a kernel
> developer, I couldn't resolve
> them.
The bisect shows bad and good alongside each other a
601 - 700 of 7811 matches
Mail list logo