[PATCH] x86: Skip latched NMIs on early boot in kdump

sting and debugging were based off a 3.10 kernel (RHEL-7) but has included Seiji's tracepoint cleanups to arch/x86/kernel/head_64.S|head64.c. Not much has changed upstream here. Also 3.14-rc4 still has the same hang. Signed-off-by: Don Zickus --- arch/x86/kernel/head_64.S |2 ++

Re: [PATCH] x86: Skip latched NMIs on early boot in kdump

On Fri, Mar 07, 2014 at 01:15:35PM -0800, H. Peter Anvin wrote: > On 03/07/2014 11:39 AM, Don Zickus wrote: > > A customer generated an external NMI using their iLO to test kdump worked. > > Unfortunately, the machine hung. Disabling the nmi_watchdog made things > > work.

Re: [PATCHSET 00/17] perf report: Add -F option for specifying output fields (v4)

On Thu, Apr 24, 2014 at 05:00:15PM -0400, Don Zickus wrote: > On Thu, Apr 24, 2014 at 10:41:39PM +0900, Namhyung Kim wrote: > > Hi Don, > > > > 2014-04-23 (수), 08:58 -0400, Don Zickus: > > > On Wed, Apr 23, 2014 at 03:15:35PM +0900, Namhyung Kim wrote: > > &g

Re: linux-next: build failure after merge of the akpm-current tree

On Tue, Apr 29, 2014 at 05:13:21PM +1000, Stephen Rothwell wrote: > Hi Andrew, > > After merging the akpm tree, today's linux-next build (sparc64 defconfig) > failed like this: > > arch/sparc/kernel/process_64.c: In function 'arch_trigger_all_cpu_backtrace': > arch/sparc/kernel/process_64.c:267:3

Re: [PATCH 2/6] Revert "perf: Disable PERF_RECORD_MMAP2 support"

On Fri, May 16, 2014 at 01:25:14PM +0200, Jiri Olsa wrote: > On Tue, May 13, 2014 at 12:48:13PM -0400, Don Zickus wrote: > > SNIP > > > - /* > > -* Anon maps don't have the execname. > > -*/ > > - if (n <

Re: [PATCH 6/6] perf: Add dcacheline sort

On Fri, May 16, 2014 at 01:47:57PM +0200, Jiri Olsa wrote: > On Tue, May 13, 2014 at 12:48:17PM -0400, Don Zickus wrote: > > In perf's 'mem-mode', one can get access to a whole bunch of details > > specific to a > > particular sample instruction. A bunch

Re: [PATCH 1/6] events, perf: Pass protection and flags bits through mmap2 interface

On Fri, May 16, 2014 at 02:22:19PM +0200, Peter Zijlstra wrote: > On Tue, May 13, 2014 at 12:48:12PM -0400, Don Zickus wrote: > > From: Peter Zijlstra > > > > The mmap2 interface was missing the protection and flags bits needed to > > accurately determine if a mma

Re: [PATCH] x86, perf: Update Haswell PEBS event table

On Thu, Apr 17, 2014 at 05:03:44PM -0700, Andi Kleen wrote: > From: Andi Kleen > > - We were allowing some sub events in c4 that are not in the event list. > Tighten the check slightly. > - We were missing some valid subevents in d1-d3. Allow all subevents. > > Signed-off-by: Andi Kleen > --- >

Re: [PATCH 6/6] perf: Add dcacheline sort

On Fri, May 16, 2014 at 04:09:59PM +0200, Stephane Eranian wrote: > > diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c > > index 635cd8f..0e91ba9 100644 > > --- a/tools/perf/util/sort.c > > +++ b/tools/perf/util/sort.c > > @@ -1,3 +1,4 @@ > > +#include > > #include "sort.h" > > #incl

Re: [PATCH 6/6] perf: Add dcacheline sort

On Fri, May 16, 2014 at 04:05:51PM +0200, Jiri Olsa wrote: > On Fri, May 16, 2014 at 09:30:58AM -0400, Don Zickus wrote: > > On Fri, May 16, 2014 at 01:47:57PM +0200, Jiri Olsa wrote: > > > On Tue, May 13, 2014 at 12:48:17PM -0400, Don Zickus wrote: > > > > In perf&#x

Re: [PATCH 1/6] events, perf: Pass protection and flags bits through mmap2 interface

On Fri, May 16, 2014 at 05:45:38PM +0200, Peter Zijlstra wrote: > On Fri, May 16, 2014 at 09:33:00AM -0400, Don Zickus wrote: > > On Fri, May 16, 2014 at 02:22:19PM +0200, Peter Zijlstra wrote: > > > On Tue, May 13, 2014 at 12:48:12PM -0400, Don Zickus wrote: > >

Re: [PATCH 6/6] perf: Add dcacheline sort

On Fri, May 16, 2014 at 06:02:43PM +0200, Stephane Eranian wrote: > On Fri, May 16, 2014 at 5:59 PM, Peter Zijlstra wrote: > > On Fri, May 16, 2014 at 04:09:59PM +0200, Stephane Eranian wrote: > >> > +#define CACHE_LINESIZE 64 > >> I had something similar to your patch here in my original se

Re: [PATCH 6/6] perf: Add dcacheline sort

On Mon, May 19, 2014 at 01:25:45PM +0200, Jiri Olsa wrote: > On Fri, May 16, 2014 at 12:24:41PM -0400, Don Zickus wrote: > > On Fri, May 16, 2014 at 06:02:43PM +0200, Stephane Eranian wrote: > > > On Fri, May 16, 2014 at 5:59 PM, Peter Zijlstra > > > wrote: > >

Re: [PATCH 6/6] perf: Add dcacheline sort

On Mon, May 19, 2014 at 03:34:14PM +0200, Jiri Olsa wrote: > On Fri, May 16, 2014 at 10:30:02AM -0400, Don Zickus wrote: > > On Fri, May 16, 2014 at 04:05:51PM +0200, Jiri Olsa wrote: > > > On Fri, May 16, 2014 at 09:30:58AM -0400, Don Zickus wrote: > > > > On Fri, Ma

[PATCH 3/7] perf: Update mmap2 interface with protection and flag bits

The kernel piece passes more info now. Update the perf tool to reflect that and adjust the synthesized maps to play along. Signed-off-by: Don Zickus --- tools/perf/util/event.c | 23 +-- tools/perf/util/event.h | 2 ++ tools/perf/util/machine.c | 4 +++- tools/perf

[PATCH 4/7] perf report: Add mem-mode documentation to report command

Add mem-mode sorting types and mem-mode itself to perf-report documentation. Signed-off-by: Don Zickus --- tools/perf/Documentation/perf-report.txt | 22 ++ 1 file changed, 22 insertions(+) diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation

[PATCH 6/7] perf: Add support to dynamically get cacheline size

Different arches may have different cacheline sizes. Look it up and set a global variable for reference. Signed-off-by: Don Zickus --- tools/perf/util/cpumap.c | 31 +++ tools/perf/util/cpumap.h | 12 tools/perf/util/sort.c | 3 +++ 3 files changed

[PATCH 1/7] events, perf: Pass protection and flags bits through mmap2 interface

From: Peter Zijlstra The mmap2 interface was missing the protection and flags bits needed to accurately determine if a mmap memory area was shared or private and if it was readable or not. Signed-off-by: Peter Zijlstra [tweaked patch to compile and wrote changelog] Signed-off-by: Don Zickus

[PATCH 2/7] Revert "perf: Disable PERF_RECORD_MMAP2 support"

ded updating to use mmap2 interface Signed-off-by: Don Zickus --- v2: added a better changelog fix a unwind test (thanks Jiri) --- kernel/events/core.c| 4 tools/perf/tests/dwarf-unwind.c | 2 +- tools/perf/util/event.c | 36 +++- to

[PATCH 0/7 V3] x86, nmi: Various fixes and cleanups

-devel) fix header column length V2: adding irq_work items to handled possible lost NMIs (new patch 1) modified output of /proc/interrupts based on feedback (patch 6) Don Zickus (5): x86, nmi: Add new nmi type 'external' x86, nmi: Add boot line option 'panic_on_unrecovered

[PATCH 7/7] perf: Add dcacheline sort

x27; to symlen size in hists__calc_col_len to prevent the next column from prematurely tabbing over and mis-aligning. Not sure what the problem is. Signed-off-by: Don Zickus --- v3: fix header column length V2: update using cpu__cacheline_size() --- tools/perf/Documentation/perf-report.txt | 3 +-

[PATCH 5/7] perf: Add cpumode to struct hist_entry

The next patch needs to sort on cpumode, so add it to hist_entry to be tracked. Signed-off-by: Don Zickus --- tools/perf/util/hist.c | 7 --- tools/perf/util/sort.h | 1 + 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c index

Re: [PATCHSET 00/17] perf report: Add -F option for specifying output fields (v4)

On Tue, Apr 29, 2014 at 10:13:35AM +0900, Namhyung Kim wrote: > > Our orignal concept for the c2c tool was to sort hist entries into > > cachelines, filter in only the HITMs and stores and re-sort based on > > cachelines with the most weight. > > > > So using today's perf with a new search called '

Re: [PATCHSET 00/17] perf report: Add -F option for specifying output fields (v4)

On Wed, Apr 30, 2014 at 08:38:10AM +0900, Namhyung Kim wrote: > Hi Don, > > On Tue, 29 Apr 2014 13:27:35 -0400, Don Zickus wrote: > > On Tue, Apr 29, 2014 at 10:13:35AM +0900, Namhyung Kim wrote: > >> >/* > >> > * Addresses with n

Re: [PATCH] watchdog: print all locks on a softlock

ks to see if a task hasn't scheduled in 2 minutes or so. That could be the result of long lock (but that output already dumps the lockdep stuff). Cheers, Don > > Signed-off-by: Eric Paris > Cc: Frederic Weisbecker > Cc: Andrew Morton > Cc: Don Zickus > Cc: Michal Ho

Re: [PATCH] watchdog: print all locks on a softlock

On Thu, May 01, 2014 at 10:09:01PM +0200, Frederic Weisbecker wrote: > On Thu, May 01, 2014 at 03:17:20PM -0400, Don Zickus wrote: > > On Thu, May 01, 2014 at 02:55:35PM -0400, Eric Paris wrote: > > > If the CPU hits a softlockup this patch will also have it print the > >

[PATCH 3/6] perf: Update mmap2 interface with protection and flag bits

The kernel piece passes more info now. Update the perf tool to reflect that and adjust the synthesized maps to play along. Signed-off-by: Don Zickus --- tools/perf/util/event.c | 23 +-- tools/perf/util/event.h | 2 ++ tools/perf/util/machine.c | 4 +++- tools/perf

[PATCH 5/6] perf: Add cpumode to struct hist_entry

The next patch needs to sort on cpumode, so add it to hist_entry to be tracked. Signed-off-by: Don Zickus --- tools/perf/util/hist.c | 7 --- tools/perf/util/sort.h | 1 + 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c index

[PATCH 2/6] Revert "perf: Disable PERF_RECORD_MMAP2 support"

This reverts commit 3090ffb5a2515990182f3f55b0688a7817325488. Conflicts: tools/perf/util/event.c --- kernel/events/core.c| 4 tools/perf/util/event.c | 36 +++- tools/perf/util/evsel.c | 1 + 3 files changed, 20 insertions(+), 21 deletions(-) di

[PATCH 4/6] perf report: Add mem-mode documentation to report command

Add mem-mode sorting types and mem-mode itself to perf-report documentation. Signed-off-by: Don Zickus --- tools/perf/Documentation/perf-report.txt | 22 ++ 1 file changed, 22 insertions(+) diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation

[PATCH 1/6] events, perf: Pass protection and flags bits through mmap2 interface

From: Peter Zijlstra The mmap2 interface was missing the protection and flags bits needed to accurately determine if a mmap memory area was shared or private and if it was readable or not. [tweaked patch to compile and wrote changelog] Signed-off-by: Don Zickus -- Peter you mentioned writing

[PATCH 0/6 V2] perf: Enable mmap2 and add dcacheline sorting

These patches enable mmap2 support in the kernel and perf to allow support for data cacheline sorting. V2: change cacheline to dcacheline (Namhyung Kim) change dcacheline alogorithm a bit re-order mmap2 patches (Namhyung Kim) provide documentation (Andi Kleen) Don Zickus (6

[PATCH 6/6] perf: Add dcacheline sort

e generation numbers o if userspace anon, then sort on pid o sort on cachelines based on data addresses The 'dcacheline' sort option in 'perf report' only works in 'mem-mode'. Signed-off-by: Don Zickus --- tools/perf/Documentation/perf-report.txt | 3 +- tools/perf/

Re: [PATCH 0/6 V2] perf: Enable mmap2 and add dcacheline sorting

On Tue, May 13, 2014 at 12:48:11PM -0400, Don Zickus wrote: > These patches enable mmap2 support in the kernel and perf to > allow support for data cacheline sorting. > > V2: change cacheline to dcacheline (Namhyung Kim) > change dcacheline alogorithm a bit > re-o

Haswell mem-store question

Hi Andi, Joe was playing with our c2c tool today and noticed we were losing store events from perf's mem-stores event. Upon investigation we stumbled into some differences in data that Haswell reports vs. Ivy/Sandy Bridge. This leaves our tool needing two different paths depending on the archite

[PATCH 5/6] x86, nmi: Move default external NMI handler to its own routine

unregister the default handler on its registration (and vice versa). This allows the driver to take control of that io port (which it ultimately wanted to do originally), but in a cleaner way. Tested by HP to make sure I didn't break anything. :-) Cc: thomas.mingare...@hp.com Signed-off-by: Don Z

[PATCH 4/6] x86, nmi: Remove 'reason' value from unknown nmi output

available. Signed-off-by: Don Zickus --- arch/x86/kernel/nmi.c |8 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c index 0467f42..a5835ad 100644 --- a/arch/x86/kernel/nmi.c +++ b/arch/x86/kernel/nmi.c @@ -388,7 +388,7 @

[PATCH 2/6] x86, nmi: Add new nmi type 'external'

-by: Don Zickus --- arch/x86/include/asm/nmi.h |1 + arch/x86/kernel/nmi.c | 41 ++--- drivers/acpi/apei/ghes.c |4 ++-- drivers/watchdog/hpwdt.c | 10 ++ 4 files changed, 31 insertions(+), 25 deletions(-) diff --git a/arch/x86

[PATCH 6/6 V2] x86, nmi: Add better NMI stats to /proc/interrupts and show handlers

number of NMIs 'handled' whereas the general NMI count represents how many actual NMIs were processed. IOW, two NMIs came in at once during one call. I am open to better suggestions. Signed-off-by: Don Zickus --- V2: modified output based on feedback (Ingo Molnar, Rob Elliott) ---

[PATCH 1/6] x86, nmi: Implement delayed irq_work mechanism to handle lost NMIs

queue prevent up to 30-40 events from being queued because of the small delay utilized during one iteration of an irq_work item. Signed-off-by: Don Zickus --- arch/x86/kernel/nmi.c | 109 + 1 files changed, 109 insertions(+), 0 deletions(-) diff

[PATCH 3/6] x86, nmi: Add boot line option 'panic_on_unrecovered_nmi' and 'panic_on_io_nmi'

These options are accessable through /proc/sys/kernel but not on the command line. The main use is for on board controllers (iLO, DRAC, BMC) to be able to issue an external NMI to bring down a hung box. This just makes configuring a box a little easier. Signed-off-by: Don Zickus

[PATCH 0/6 V2] x86, nmi: Various fixes and cleanups

boxes to test performance results. V2: adding irq_work items to handled possible lost NMIs (new patch 1) modified output of /proc/interrupts based on feedback (patch 6) Don Zickus (5): x86, nmi: Add new nmi type 'external' x86, nmi: Add boot line option 'panic_on_unrec

Re: [PATCH v2]: fix Haswell precise store data source encoding

documentation? Just wanted to make sure we are not missing one more case. Thanks for the quick patch Stephane! Tested-and-Reviewed-by: Don Zickus > > Signed-off-by: Stephane Eranian > > diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c > b/arch/x86/kern

Re: [PATCH v2]: fix Haswell precise store data source encoding

On Thu, May 15, 2014 at 10:08:51PM +0200, Stephane Eranian wrote: > On Thu, May 15, 2014 at 9:56 PM, Don Zickus wrote: > > On Thu, May 15, 2014 at 05:56:44PM +0200, Stephane Eranian wrote: > >> > >> This patch fixes a bug in precise_store_data_hsw() whereby > &

Re: [PATCH 0/6 V2] x86, nmi: Various fixes and cleanups

On Thu, May 15, 2014 at 03:25:43PM -0400, Don Zickus wrote: > I started this patch by fixing a performance problem with the GHES > NMI handler and then things evolved to more patches as I was poking > around in the code. > > The main focus was moving the GHES NMI driver to its

Re: [PATCH 1/5] x86, nmi: Add new nmi type 'external'

On Wed, May 07, 2014 at 06:27:46PM +0200, Ingo Molnar wrote: > > [...] But I guess in theory, if a PMI NMI comes in and before the > > cpu can accept it and GHES NMI comes in, then it would suffice to > > say it may get dropped. That would be not be good. Though the race > > would be very sma

Re: [PATCH 1/5] x86, nmi: Add new nmi type 'external'

On Thu, May 08, 2014 at 07:35:01PM +0200, Ingo Molnar wrote: > > * Don Zickus wrote: > > > > > Again, I don't have a solution to juggle between PMI performance > > > > and reliable delivery. We could do away with the spinlocks and > > > >

Re: [PATCHSET 00/17] perf report: Add -F option for specifying output fields (v4)

On Fri, May 09, 2014 at 03:11:28PM +0900, Namhyung Kim wrote: > >> > I am willing to believe I missed scenario when sorting, I just can't > >> > think > >> > of it (so I wouldn't know how to fix it). That's why I was looking for > >> > an > >> > example to make it more obvious to me. Sorry for

Re: [PATCH 1/5] x86, nmi: Add new nmi type 'external'

On Fri, May 09, 2014 at 09:10:50AM +0200, Ingo Molnar wrote: > > * Don Zickus wrote: > > > On Thu, May 08, 2014 at 07:35:01PM +0200, Ingo Molnar wrote: > > > > > > * Don Zickus wrote: > > > > > > > > > Again, I don't have a sol

Re: [PATCH 05/15 V3] perf, c2c: Rework setup code to prepare for features

On Tue, Apr 08, 2014 at 04:41:29PM +0900, Namhyung Kim wrote: > On Sat, 29 Mar 2014 18:10:18 +0100, Jiri Olsa wrote: > > On Mon, Mar 24, 2014 at 03:36:56PM -0400, Don Zickus wrote: > > > > SNIP > > > >> > >> static int perf

Re: [PATCH 07/15 V3] perf, c2c: Add in sort on physid

On Tue, Apr 08, 2014 at 04:56:25PM +0900, Namhyung Kim wrote: > On Mon, 24 Mar 2014 15:36:58 -0400, Don Zickus wrote: > > Now that the infrastructure is set, add in the support to use > > hist_entry to sort on physid. > > > > V2: use new mmap2 sort > &

Re: [PATCH 03/15 V3] perf c2c: Shared data analyser

On Tue, Apr 08, 2014 at 03:59:15PM +0900, Namhyung Kim wrote: > Hi Don, Oh by the way, thank you for your review. I will clean up a bunch of stuff based on your suggestions. Cheers, Don > > On Mon, 24 Mar 2014 15:36:54 -0400, Don Zickus wrote: > > From: Arnaldo Carvalho de Mel

Re: [PATCH 03/15 V3] perf c2c: Shared data analyser

On Tue, Apr 08, 2014 at 03:59:15PM +0900, Namhyung Kim wrote: > Hi Don, > > On Mon, 24 Mar 2014 15:36:54 -0400, Don Zickus wrote: > > From: Arnaldo Carvalho de Melo > > > > This is the start of a new perf tool that will collect information about > > memory acces

Re: [PATCH 05/15 V3] perf, c2c: Rework setup code to prepare for features

On Wed, Apr 09, 2014 at 10:12:32AM +0900, Namhyung Kim wrote: > On Tue, 8 Apr 2014 10:11:07 -0400, Don Zickus wrote: > > On Tue, Apr 08, 2014 at 04:41:29PM +0900, Namhyung Kim wrote: > >> On Sat, 29 Mar 2014 18:10:18 +0100, Jiri Olsa wrote: > >> > On Mon, Mar 24,

Re: [PATCH 07/15 V3] perf, c2c: Add in sort on physid

On Wed, Apr 09, 2014 at 10:30:56AM +0900, Namhyung Kim wrote: > On Tue, 8 Apr 2014 10:17:58 -0400, Don Zickus wrote: > > On Tue, Apr 08, 2014 at 04:56:25PM +0900, Namhyung Kim wrote: > >> On Mon, 24 Mar 2014 15:36:58 -0400, Don Zickus wrote: > >> > +static int perf_

Re: [PATCH 4/6] perf, sort: Add physid sorting based on mmap2 data

On Mon, Mar 24, 2014 at 03:34:34PM -0400, Don Zickus wrote: > In order for the c2c tool to work correctly, it needs to properly > sort all the records on uniquely identifiable data addresses. These > unique addresses are converted from virtual addresses provided by the > hardware i

Re: [PATCH 6/6] perf, sort: Allow unique sorting instead of combining hist_entries

On Wed, Apr 09, 2014 at 02:31:00PM +0900, Namhyung Kim wrote: > On Mon, 24 Mar 2014 15:34:36 -0400, Don Zickus wrote: > > The cache contention tools needs to keep all the perf records unique in > > order > > to properly parse all the data. Currently add_hist_entry()

Re: [PATCH 1/4 v2] perf: Allow ability to map cpus to nodes easily

On Thu, Apr 10, 2014 at 11:31:46AM +0200, Jiri Olsa wrote: > On Mon, Apr 07, 2014 at 02:55:21PM -0400, Don Zickus wrote: > > SNIP > > > +static int init_cpunode_map(void) > > +{ > > + int i; > > + > > + set_max_cpu_num(); > > + set_ma

Re: [PATCH 1/4 v2] perf: Allow ability to map cpus to nodes easily

On Thu, Apr 10, 2014 at 03:24:58PM +0200, Jiri Olsa wrote: > On Thu, Apr 10, 2014 at 09:15:15AM -0400, Don Zickus wrote: > > On Thu, Apr 10, 2014 at 11:31:46AM +0200, Jiri Olsa wrote: > > > On Mon, Apr 07, 2014 at 02:55:21PM -0400, Don Zickus wrote: > > > > >

[RFC 5/5] perf: Enable multiple hist_entry_group output

604 2.87% 7 L1 hit[.] 0x8630 1.44% 7 L1 hit[.] 0x000085f8 Signed-off-by: Don Zickus --- tools/perf/builtin-report.c | 2 + tools/perf/ui/gtk/hists.c | 10

[RFC 4/5] perf: Switch to using hist_entry_group

group list and entry list and still fulfill the spirit of the function (most of it was my lack of understanding how the function works). Signed-off-by: Don Zickus --- tools/perf/builtin-annotate.c | 11 +- tools/perf/builtin-diff.c | 77 ++ tools/perf/builtin-top.c | 12 +-

[RFC 1/5] perf: Wrap __hists__add_entry to prep for group entry change

This patch is mainly mechanical and just wraps __hists__add_entry with hists__add_entry. Later on, we can modify hists__add_entry to include group entry changes without disturbing the builtin-* files. Signed-off-by: Don Zickus --- tools/perf/builtin-annotate.c | 2 +- tools/perf/builtin

[RFC 3/5] perf: Add in stub hist_entry_group code

iant conversion. Signed-off-by: Don Zickus --- tools/perf/util/hist.c | 135 + tools/perf/util/hist.h | 2 + tools/perf/util/sort.c | 2 + tools/perf/util/sort.h | 24 + 4 files changed, 163 insertions(+) diff --git a/tools/perf/util/hist

[RFC 0/5] perf: Create hist_entry groups

lso, this patchset breaks most tools (mainly because I don't understand all the interactions), hence the RFC. I mostly tested with 'perf report --stdio' and 'perf mem report --stdio'. Please let me know if this is an interesting idea to go forward with or not. Don Zickus (5

[RFC 2/5] perf: Use macros to walk hist entries

cases stayed unchanged. Should be a mostly mechanical change and nothing should change from a technical perspective. Signed-off-by: Don Zickus --- tools/perf/builtin-diff.c | 48 ++ tools/perf/builtin-top.c | 6 + tools/perf/te

Re: [RFC 5/5] perf: Enable multiple hist_entry_group output

On Fri, Apr 11, 2014 at 07:30:00PM +0200, Jiri Olsa wrote: > On Thu, Apr 10, 2014 at 04:11:01PM -0400, Don Zickus wrote: > > Enable multiple hist_entry_group groups in the output based on a sort > > method. > > > > Currently only 'perf report' is hooked

Re: [RFC 5/5] perf: Enable multiple hist_entry_group output

On Fri, Apr 11, 2014 at 07:30:00PM +0200, Jiri Olsa wrote: > On Thu, Apr 10, 2014 at 04:11:01PM -0400, Don Zickus wrote: > > Enable multiple hist_entry_group groups in the output based on a sort > > method. > > > > Currently only 'perf report' is hooked

[PATCH 1/4] perf: Allow ability to map cpus to nodes easily

instead of max). It also calculates the max number of nodes. V3: simplify function names Signed-off-by: Don Zickus --- tools/perf/util/cpumap.c | 139 +++ tools/perf/util/cpumap.h | 35 2 files changed, 174 insertions(+) diff --git a/tools

[PATCH 2/4] perf, kmem: Utilize the new generic cpunode_map

Use the previous patch implementation of cpunode_map for builtin-kmem.c Should not be any functional difference. Cc: Li Zefan Signed-off-by: Don Zickus --- tools/perf/builtin-kmem.c | 78 ++- 1 file changed, 3 insertions(+), 75 deletions(-) diff

[PATCH 3/4] perf, callchain: Add generic report parse callchain callback function

This takes the parse_callchain_opt function and copies it into the callchain.c file. Now the c2c tool can use it too without duplicating. Signed-off-by: Don Zickus --- tools/perf/util/callchain.c | 83 + tools/perf/util/callchain.h | 1 + 2 files

[PATCH 0/4] perf: Make some functions generic

This patch just converts some private functions into global ones that can be used by other tools like the c2c tool I am trying to merge. Don Zickus (4): perf: Allow ability to map cpus to nodes easily perf, kmem: Utilize the new generic cpunode_map perf, callchain: Add generic report parse

[PATCH 4/4] perf, report: Use new generic report parse callchain callback

Use the new routine. Signed-off-by: Don Zickus --- tools/perf/builtin-report.c | 77 + 1 file changed, 1 insertion(+), 76 deletions(-) diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index c8f2113..c87412b 100644 --- a/tools

[PATCH 5/6] perf: Update sort to handle MAP_SHARED bits

tool in this example the memory is private and will not cause contention with other processes accessing the same shared library area. Signed-off-by: Don Zickus --- tools/perf/util/sort.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/perf/util/sort.c b/tools/perf

[PATCH 6/6] perf, sort: Allow unique sorting instead of combining hist_entries

the data source). Create a flag to force the records to stay unique. Signed-off-by: Don Zickus --- tools/perf/util/hist.c | 3 ++- tools/perf/util/sort.c | 1 + tools/perf/util/sort.h | 1 + 3 files changed, 4 insertions(+), 1 deletion(-) diff --git a/tools/perf/util/hist.c b/tools/perf/util

[PATCH 3/6] Revert "perf: Disable PERF_RECORD_MMAP2 support"

This reverts commit 3090ffb5a2515990182f3f55b0688a7817325488. Conflicts: tools/perf/util/event.c --- kernel/events/core.c| 4 tools/perf/util/event.c | 36 +++- tools/perf/util/evsel.c | 1 + 3 files changed, 20 insertions(+), 21 deletions(-) di

[PATCH 04/15 V3] perf c2c: Dump raw records, decode data_src bits

available at this point. Cc: David Ahern Cc: Don Zickus Cc: Frederic Weisbecker Cc: Joe Mario Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Richard Fowles Cc: Stephane Eranian Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/builti

[PATCH 03/15 V3] perf c2c: Shared data analyser

ta file with methods to handle them, right now just printing the resolved perf_sample data structure after each event name. [dcz: refreshed to latest upstream changes] Cc: David Ahern Cc: Don Zickus Cc: Frederic Weisbecker Cc: Joe Mario Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zi

[PATCH 15/15 V3] perf, c2c: Add shared cachline summary table

-by: Don Zickus --- tools/perf/builtin-c2c.c | 136 +++ 1 file changed, 136 insertions(+) diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c index e30ce2f..e33f548 100644 --- a/tools/perf/builtin-c2c.c +++ b/tools/perf/builtin-c2c.c

[PATCH 11/15 V3] perf, c2c: Add callchain support

rect. The amount of memory corruption (from the callchain dynamic array) seems to have dwindled done to nothing. :-) V2: update to latest api V3: remove call_graph variable, unused Signed-off-by: Don Zickus --- tools/perf/builtin-c2c.c | 67 +++- 1 file c

[PATCH 14/15 V3] perf, c2c: Add symbol count table

jbd2] /lib/modules/3.10.0c2c_all+/kernel/fs/jbd2/jbd2.ko 3 65392 [ext4] /lib/modules/3.10.0c2c_all+/kernel/fs/ext4/ext4.ko V2: refresh to latest upstream changes and hist_entry Original-by: Dick Fowles Signed-off-by: Don Zickus --- tools/perf/builtin-c

[PATCH 06/15 V3] perf, c2c: Add in new options to configure latency and stores

Modified the code to allow latency settings to be tweaked on the command line and also the ability to dynamically profile stores (or disable using stores). This allows the tool to be used on older Intel platforms like Westmere. Signed-off-by: Don Zickus --- tools/perf/builtin-c2c.c | 73

[PATCH 12/15 V3] perf, c2c: Output summary stats

: Dick Fowles Signed-off-by: Don Zickus --- tools/perf/builtin-c2c.c | 47 ++- 1 file changed, 46 insertions(+), 1 deletion(-) diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c index 363deec..37bf0bd 100644 --- a/tools/perf/builtin-c2c.c

[PATCH 08/15 V3] perf, c2c: Add stats to track data source bits and cpu to node maps

Fowles for determining which bits are important and how to properly track them. Ported to perf by me. V2: refresh with hist_entry Original-by: Dick Fowles Signed-off-by: Don Zickus --- tools/perf/builtin-c2c.c | 187 +++ 1 file changed, 187 insertions

[PATCH 13/15 V3] perf, c2c: Dump rbtree for debugging

18179 7ff9d7fbaf98 812ad509 68100242 [LOAD,LFB,HIT,SNP NONE]1 V2: refresh with hist_entry Signed-off-by: Don Zickus --- tools/perf/builtin-c2c.c | 51 1 file changed, 51 insertions(+) diff --git a/tools/perf

[PATCH 10/15 V3] perf, c2c: Display cacheline HITM analysis to stdout

0.0% 0.0%89.9% 0.0% 0x28 18156 18156 0x815c4be9 n/a n/a n/a 50 Original-by: Dick Fowles Signed-off-by: Don Zickus --- tools/perf/builtin-c2c.c | 519 +++ 1 file

[PATCH 09/15 V3] perf, c2c: Sort based on hottest cache line

-by: Don Zickus --- tools/perf/builtin-c2c.c | 201 +++ 1 file changed, 201 insertions(+) diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c index 55c5ce9..8674626 100644 --- a/tools/perf/builtin-c2c.c +++ b/tools/perf/builtin-c2c.c

[PATCH 00/15 V3] perf, c2c: Add new tool to analyze cacheline contention on NUMA systems

uggestion dropped latency analyze for now until this patchset is accepted little fixes and tweaks Signed-off-by: Don Zickus Arnaldo Carvalho de Melo (2): perf c2c: Shared data analyser perf c2c: Dump raw records, decode data_src bits Don Zickus (19): Revert "perf: Disable PE

[PATCH 05/15 V3] perf, c2c: Rework setup code to prepare for features

gunk o add some generic options There isn't much meat in this patch just a bunch of code movement and cleanups. V2: refresh to latest upstream changes Signed-off-by: Don Zickus --- tools/perf/builtin-c2c.c | 125 +++ 1 file changed, 93 inser

[PATCH 02/15 V3] perf, callchain: Add generic callchain print handler for stdio

. Tweaks and recommendations for a better approach are welcomed. :-) Signed-off-by: Don Zickus --- tools/perf/ui/stdio/hist.c | 37 + tools/perf/util/hist.h | 4 2 files changed, 41 insertions(+) diff --git a/tools/perf/ui/stdio/hist.c b/tools/perf

[PATCH 07/15 V3] perf, c2c: Add in sort on physid

Now that the infrastructure is set, add in the support to use hist_entry to sort on physid. V2: use new mmap2 sort Signed-off-by: Don Zickus --- tools/perf/builtin-c2c.c | 52 +++- 1 file changed, 51 insertions(+), 1 deletion(-) diff --git a/tools

[PATCH 01/15 V3] perf: Fix stddev calculation

. Signed-off-by: Don Zickus --- tools/perf/util/stat.c | 13 + tools/perf/util/stat.h | 1 + 2 files changed, 14 insertions(+) diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index 6506b3d..0cb4dbc 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -33,6 +33,7

[PATCH 0/6] perf, events: Enable mmap2 support

;s to cleanup the output a bit Don Zickus (6): events, perf: Pass protection and flags bits through mmap2 interface perf: Update mmap2 interface with protection and flag bits Revert "perf: Disable PERF_RECORD_MMAP2 support" perf, sort: Add physid sorting based on mmap2 data perf:

[PATCH 4/6] perf, sort: Add physid sorting based on mmap2 data

s makes it look far less ugly create a new 'physid mode' to group all the sorting rules together (mimics the mem-mode) Signed-off-by: Don Zickus --- tools/perf/builtin-report.c | 20 ++- tools/perf/util/hist.c | 27 +++- tools/perf/util/hist.h | 8 +

[PATCH 1/6] events, perf: Pass protection and flags bits through mmap2 interface

From: Peter Zijlstra The mmap2 interface was missing the protection and flags bits needed to accurately determine if a mmap memory area was shared or private and if it was readable or not. [tweaked patch to compile and wrote changelog - Don Signed-off-by: Don Zickus -- Peter you mentioned

[PATCH 2/6] perf: Update mmap2 interface with protection and flag bits

The kernel piece passes more info now. Update the perf tool to reflect that and adjust the synthesized maps to play along. Signed-off-by: Don Zickus --- tools/perf/util/event.c | 23 +-- tools/perf/util/event.h | 2 ++ tools/perf/util/machine.c | 4 +++- tools/perf

Re: [PATCH 4/6] perf, sort: Add physid sorting based on mmap2 data

On Mon, Mar 24, 2014 at 12:54:31PM -0700, Andi Kleen wrote: > Don Zickus writes: > > > In order for the c2c tool to work correctly, it needs to properly > > sort all the records on uniquely identifiable data addresses. These > > unique addresses are converted from virtu

Re: [PATCH 4/6] perf, sort: Add physid sorting based on mmap2 data

On Mon, Mar 24, 2014 at 09:20:45PM +0100, Andi Kleen wrote: > On Mon, Mar 24, 2014 at 04:17:57PM -0400, Don Zickus wrote: > > On Mon, Mar 24, 2014 at 12:54:31PM -0700, Andi Kleen wrote: > > > Don Zickus writes: > > > > > > > In order for the c2c tool

[PATCH 01/15 V3] perf: Fix stddev calculation

. Signed-off-by: Don Zickus --- tools/perf/util/stat.c | 13 + tools/perf/util/stat.h | 1 + 2 files changed, 14 insertions(+) diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index 6506b3d..0cb4dbc 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -33,6 +33,7

Re: [PATCH 01/15 V3] perf: Fix stddev calculation

On Mon, Mar 24, 2014 at 04:54:38PM -0400, Don Zickus wrote: > The stddev calculation written matched standard error. As a result when > using this result to find the relative stddev between runs, it was not > accurate. > This isn't the patch I that had my updates... So

[PATCH 4/6 V2] perf, sort: Add physid sorting based on mmap2 data

s makes it look far less ugly create a new 'physid mode' to group all the sorting rules together (mimics the mem-mode) Signed-off-by: Don Zickus --- tools/perf/Documentation/perf-report.txt | 23 +++ tools/perf/builtin-report.c | 20 ++- tools/perf/util/hist.

<    1   2   3   4   5   6   7   >