CoreSight November 2021

coresight@lists.linaro.org

8 participants
31 discussions

[FYI][PATCH 1/1] tools build: Fix removal of feature-sync-compare-and-swap feature detection

by Arnaldo Carvalho de Melo

The patch removing the feature-sync-compare-and-swap feature detection didn't remove the call to main_test_sync_compare_and_swap(), making the 'test-all' case fail an all the feature tests to be performed individually: $ cat /tmp/build/perf/feature/test-all.make.output In file included from test-all.c:18: test-libpython-version.c:5:10: error: #error 5 | #error | ^~~~~ test-all.c: In function ‘main’: test-all.c:203:9: error: implicit declaration of function ‘main_test_sync_compare_and_swap’ [-Werror=implicit-function-declaration] 203 | main_test_sync_compare_and_swap(argc, argv); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors $ Fix it, now to figure out what is that test-libpython-version.c problem... Fixes: 60fa754b2a5a4e0c ("tools: Remove feature-sync-compare-and-swap feature detection") Cc: Jiri Olsa <jolsa(a)redhat.com> Cc: Leo Yan <leo.yan(a)linaro.org> Cc: Mark Rutland <mark.rutland(a)arm.com> Cc: Mathieu Poirier <mathieu.poirier(a)linaro.org> Cc: Mike Leach <mike.leach(a)linaro.org> Cc: Namhyung Kim <namhyung(a)kernel.org> Signed-off-by: Arnaldo Carvalho de Melo <acme(a)redhat.com> --- tools/build/feature/test-all.c | 1 - 1 file changed, 1 deletion(-) diff --git a/tools/build/feature/test-all.c b/tools/build/feature/test-all.c index 9204395272912c13..0b243ce842be3383 100644 --- a/tools/build/feature/test-all.c +++ b/tools/build/feature/test-all.c @@ -200,7 +200,6 @@ int main(int argc, char *argv[]) main_test_timerfd(); main_test_stackprotector_all(); main_test_libdw_dwarf_unwind(); - main_test_sync_compare_and_swap(argc, argv); main_test_zlib(); main_test_pthread_attr_setaffinity_np(); main_test_pthread_barrier(); -- 2.31.1

2 years, 5 months

[PATCH v2 0/5] coresight: syscfg: dynamic load

by Mike Leach

1) API updated to allow dynamic load and unload of configurations and features. Dependency management between loaded sets is added. 2) New configuration and feature sets can be added using a loadable module. An example in /samples/coresight is provided to demonstrate this. 3) configfs can be used to activate a configuration which will then be used when controlling tracing using sysfs. Applies and tested on coresight/next - which is 5.15-rc1 Changes since v1: a) Original set split to divide related changes into smaller sets. Removed RFC flag. b) Revised config activation for sysfs to simplify common function. (patch 4). c) Minor changes requested by Mathieu added. Mike Leach (5): coresight: syscfg: Update API to allow dynamic load and unload coresight: syscfg: Update load API for config loadable modules coresight: syscfg: Example CoreSight configuration loadable module coresight: configfs: Allow configfs to activate configuration Documentation: coresight: Update coresight configuration docs .../trace/coresight/coresight-config.rst | 19 ++ MAINTAINERS | 1 + .../coresight/coresight-cfg-preload.c | 9 +- .../hwtracing/coresight/coresight-config.h | 9 +- .../coresight/coresight-etm4x-core.c | 11 +- .../coresight/coresight-syscfg-configfs.c | 87 +++++ .../coresight/coresight-syscfg-configfs.h | 4 + .../hwtracing/coresight/coresight-syscfg.c | 315 ++++++++++++++++-- .../hwtracing/coresight/coresight-syscfg.h | 39 ++- samples/Kconfig | 9 + samples/Makefile | 1 + samples/coresight/Makefile | 4 + samples/coresight/coresight-cfg-sample.c | 73 ++++ 13 files changed, 546 insertions(+), 35 deletions(-) create mode 100644 samples/coresight/Makefile create mode 100644 samples/coresight/coresight-cfg-sample.c -- 2.17.1

2 years, 5 months

[PATCH v1 0/4] coresight: etm: Correct (virtual) contextID tracing for namespace

by Leo Yan

If a profiling program runs in a non-root PID namespace, if CoreSight driver enables contextID tracing, it can lead to mismatching issue between the context ID in hardware trace data and the allocated PID in the non-root namespace. CoreSight driver has tried to address this issue for the contextID related interfaces under sysfs, but it misses other parts: it doesn't prevent user to set VMID (virtual contextID) for kernel runs in EL2 with VHE, and furthermore, it misses to handle the perf mode when the profiling tool (e.g. perf) doesn't run in root PID namespace. For this reason, this patch series is to correct contextID tracing for non-root namespace. Patch 01 is to use spinlock to protect reading virtual context ID comparator. Patch 02 corrects the virtual contextID tracing for non-root PID namespace. Patch 03/04 are used to fix the contextID tracing for perf mode. I only verified this patch series on Juno board in the root PID namespace and confirmed the patches don't introduce any regression for root PID namespace. Leo Yan (4): coresight: etm4x: Add lock for reading virtual context ID comparator coresight: etm4x: Don't use virtual contextID for non-root PID namespace coresight: etm4x: Don't trace contextID for non-root namespace in perf mode coresight: etm3x: Don't trace contextID for non-root namespace in perf mode .../coresight/coresight-etm3x-core.c | 4 +++ .../coresight/coresight-etm4x-core.c | 10 +++++-- .../coresight/coresight-etm4x-sysfs.c | 30 +++++++++++++++++++ 3 files changed, 42 insertions(+), 2 deletions(-) -- 2.25.1

2 years, 5 months

Re: [PATCH 05/10] Coresight: Add interface for TPDM BC subunit

by Mathieu Poirier

On Fri, Nov 12, 2021 at 05:10:38PM +0800, Jinlong wrote: > On Fri, Nov 12, 2021 at 04:42:31PM +0800, Jinlong wrote: > > [....] > > > > > > &tpdm_attr_grp, > > > > > + &tpdm_bc_attr_grp, > > > > > > > > It is quite tedious to review all these options at the same time as the core > > > > drivers. I suggest to concentrate on the base functionality for now. When that > > > > is merged we can add configuration options such as these. > > > > > > > > I am out of time for this patchset and as such will not review the remaining > > > > patches - those will have to wait for another iteration. > > > > > > > > Thanks, > > > > Mathieu > > > > > > > > > Thank your for all the comments. > > > We will update the patches. > > > > > > Thanks > > > Jinlong Mao > > > > > > > Correct the typo. > > Hi Mathieu, > > We will reduce the code changes. > There will be code changes below for next version: > > 1. coresight: add support to enable more coresight paths > 2. Basic function for TPDM(only enable/disable functions) > 3. TPDA driver > > Is that ok for you ? Yes, that sounds much better. Regards, Mathieu > > > > > > > > > > > NULL, > > > > > }; > > > > > > > > > > -- > > > > > 2.17.1 > > > > >

2 years, 5 months

Re: [PATCH] perf cs-etm: use swap() to make code cleaner

by Leo Yan

Hi Yihao, On Thu, Nov 11, 2021 at 06:33:44PM -0800, Yihao Han wrote: > Use the macro 'swap()' defined in 'include/linux/minmax.h' to avoid > opencoding it. The header 'include/linux/minmax.h' is used for kernel but it's not provided for user space program? > Signed-off-by: Yihao Han <hanyihao(a)vivo.com> > --- > tools/perf/util/cs-etm.c | 6 +----- > 1 file changed, 1 insertion(+), 5 deletions(-) > > diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c > index f323adb1af85..aaa3ec725002 100644 > --- a/tools/perf/util/cs-etm.c > +++ b/tools/perf/util/cs-etm.c > @@ -408,17 +408,13 @@ struct cs_etm_packet_queue > static void cs_etm__packet_swap(struct cs_etm_auxtrace *etm, > struct cs_etm_traceid_queue *tidq) > { > - struct cs_etm_packet *tmp; > - > if (etm->sample_branches || etm->synth_opts.last_branch || > etm->sample_instructions) { > /* > * Swap PACKET with PREV_PACKET: PACKET becomes PREV_PACKET for > * the next incoming packet. > */ > - tmp = tidq->packet; > - tidq->packet = tidq->prev_packet; > - tidq->prev_packet = tmp; > + swap(tidq->packet, tidq->prev_packet); NAK - This patch will introduce compilation error: util/cs-etm.c: In function ‘cs_etm__packet_swap’: util/cs-etm.c:417:3: error: implicit declaration of function ‘swap’; did you mean ‘swab’? [-Werror=implicit-function-declaration] swap(tidq->packet, tidq->prev_packet); ^~~~ swab Thanks, Leo > } > } > > -- > 2.17.1 >

2 years, 5 months

Re: [PATCH v2] coresight: Use devm_bitmap_zalloc when applicable

by Mathieu Poirier

Hi Christophe, On Wed, 3 Nov 2021 at 12:39, Christophe JAILLET <christophe.jaillet(a)wanadoo.fr> wrote: > > 'drvdata->chs.guaranteed' is a bitmap. So use 'devm_bitmap_kzalloc()' to > simplify code, improve the semantic and avoid some open-coded arithmetic > in allocator arguments. > > Signed-off-by: Christophe JAILLET <christophe.jaillet(a)wanadoo.fr> > --- > v1 --> v2: remove the 'guaranteed' variable to be even less verbose > --- > drivers/hwtracing/coresight/coresight-stm.c | 10 +++------- > 1 file changed, 3 insertions(+), 7 deletions(-) > > diff --git a/drivers/hwtracing/coresight/coresight-stm.c b/drivers/hwtracing/coresight/coresight-stm.c > index 58062a5a8238..bb14a3a8a921 100644 > --- a/drivers/hwtracing/coresight/coresight-stm.c > +++ b/drivers/hwtracing/coresight/coresight-stm.c > @@ -856,13 +856,11 @@ static int stm_probe(struct amba_device *adev, const struct amba_id *id) > { > int ret; > void __iomem *base; > - unsigned long *guaranteed; > struct device *dev = &adev->dev; > struct coresight_platform_data *pdata = NULL; > struct stm_drvdata *drvdata; > struct resource *res = &adev->res; > struct resource ch_res; > - size_t bitmap_size; > struct coresight_desc desc = { 0 }; > > desc.name = coresight_alloc_device_name(&stm_devs, dev); > @@ -904,12 +902,10 @@ static int stm_probe(struct amba_device *adev, const struct amba_id *id) > else > drvdata->numsp = stm_num_stimulus_port(drvdata); > > - bitmap_size = BITS_TO_LONGS(drvdata->numsp) * sizeof(long); > - > - guaranteed = devm_kzalloc(dev, bitmap_size, GFP_KERNEL); > - if (!guaranteed) > + drvdata->chs.guaranteed = devm_bitmap_zalloc(dev, drvdata->numsp, > + GFP_KERNEL); > + if (!drvdata->chs.guaranteed) > return -ENOMEM; > - drvdata->chs.guaranteed = guaranteed; I will pick up this patch when -rc1 gets released. Thanks, Mathieu > > spin_lock_init(&drvdata->spinlock); > > -- > 2.30.2 >

2 years, 5 months

[PATCH v2 0/2] arm-spe/cs-etm: Print size using consistent format

by German Gomez

Sent as a splitted re-roll of patchset [1] to keep the patches more semantically grouped. The patches themselves have been tested and reviewed in the previous submission already. Changes since v1: - Sent as a separate patchset. - Added acme to the recipients list as I forgot to include him in [1]. [1] https://lore.kernel.org/all/20210916154635.1525-1-german.gomez@arm.com/ Andrew Kilroy (2): perf cs-etm: Print size using consistent format perf arm-spe: Print size using consistent format tools/perf/util/arm-spe.c | 2 +- tools/perf/util/cs-etm.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) -- 2.25.1

2 years, 5 months

[PATCH 1/5] perf cs-etm: Print size using consistent format

by German Gomez

From: Andrew Kilroy <andrew.kilroy(a)arm.com> Since the size is already printed earlier in hex, print the same data using the same format, in hex. Reviewed-by: James Clark <james.clark(a)arm.com> Signed-off-by: Andrew Kilroy <andrew.kilroy(a)arm.com> Signed-off-by: German Gomez <german.gomez(a)arm.com> --- tools/perf/util/cs-etm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c index f323adb1af85..4f672f7d008c 100644 --- a/tools/perf/util/cs-etm.c +++ b/tools/perf/util/cs-etm.c @@ -537,7 +537,7 @@ static void cs_etm__dump_event(struct cs_etm_queue *etmq, fprintf(stdout, "\n"); color_fprintf(stdout, color, - ". ... CoreSight %s Trace data: size %zu bytes\n", + ". ... CoreSight %s Trace data: size %#zx bytes\n", cs_etm_decoder__get_name(etmq->decoder), buffer->size); do { -- 2.17.1

2 years, 5 months

Re: [PATCH v2] drivers/coresight: ultrasoc: Add System Memory Buffer driver

by liuqi (BA)

On 2021/11/5 9:36, Randy Dunlap wrote: > On 11/4/21 6:20 PM, Qi Liu wrote: >> diff --git a/drivers/hwtracing/coresight/ultrasoc/Kconfig >> b/drivers/hwtracing/coresight/ultrasoc/Kconfig >> new file mode 100644 >> index 000000000000..c18c25bf9df3 >> --- /dev/null >> +++ b/drivers/hwtracing/coresight/ultrasoc/Kconfig >> @@ -0,0 +1,12 @@ >> +# SPDX-License-Identifier: MIT >> +# >> +# ultrasoc configuration >> +# >> +config ULTRASOC_SMB >> + tristate "Ultrasoc system memory buffer drivers" >> + depends on ARM64 && CORESIGHT > > The "depends" line should be indented with a tab instead of spaces... > ah, yes, my editor might be set incorrectly, thanks for the reminder, I'll correct it next time. Qi >> + help >> + This enables support for the Ultrasoc system memory buffer >> + driver that is responsible for receiving the trace data from >> + Coresight ETM devices and storing them to a system buffer >> + respectively. > >

2 years, 6 months

[GIT PULL] Coresight fix for v5.16

by Mathieu Poirier

The following changes since commit 561ced0bb90a4be298b7db5fb54f29731d74a3f6: arm64: errata: Enable TRBE workaround for write to out-of-range address (2021-10-27 11:46:06 -0600) are available in the Git repository at: git@gitolite.kernel.org:pub/scm/linux/kernel/git/coresight/linux.git tags/coresight-fixes-v5.16 for you to fetch changes up to 20b02fe36530a1e48dde73c80b882ae276346ea3: arm64: cpufeature: Export this_cpu_has_cap helper (2021-11-04 10:46:21 -0600) --------------------------------------------------------------- coresight: Fix for v5.16 This fix enables to compile the TRBE driver as a module by exporting function this_cpu_has_cap(). Signed-off-by: Mathieu Poirier <mathieu.poirier(a)linaro.org> ---------------------------------------------------------------- Arnd Bergmann (1): arm64: cpufeature: Export this_cpu_has_cap helper arch/arm64/kernel/cpufeature.c | 1 + 1 file changed, 1 insertion(+)

2 years, 6 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

CoreSight November 2021