CoreSight August 2026

coresight@lists.linaro.org

6 participants
23 discussions

[PATCH 0/6] perf: cs-etm: HW_ID improvements

by James Clark

The driver might not always be able to write HW_IDs, but we don't need them for unformatted mode anyway, so fix that in commit ("perf cs-etm: Synthesize missing HW_ID mappings for raw trace"). At the same time, give the driver another chance to send them in commit ("coresight: perf: Retry failed HW_ID writes"). The other commits are semi-related improvements and fixes. ("perf: cs-etm: Respect --no-itrace option") makes debugging broken Coresight perf.data files easier. Applies on top of "[PATCH v2 00/14] perf cs-etm: Per-thread mode fixes and snapshot wrap support" Signed-off-by: James Clark <james.clark(a)linaro.org> --- James Clark (5): perf: cs-etm: Don't add global v0 HW_IDs to unformatted queues perf cs-etm: Free partially created queues perf: cs-etm: Respect --no-itrace option perf/core: Return errors from perf_report_aux_output_id() coresight: perf: Retry failed HW_ID writes Leo Yan (1): perf cs-etm: Synthesize missing HW_ID mappings for raw trace drivers/hwtracing/coresight/coresight-etm-perf.c | 52 +++++---- include/linux/perf_event.h | 2 +- kernel/events/core.c | 6 +- tools/perf/util/auxtrace.c | 2 +- tools/perf/util/auxtrace.h | 1 + tools/perf/util/cs-etm.c | 143 ++++++++++++++++------- 6 files changed, 139 insertions(+), 67 deletions(-) --- base-commit: 78148c85297024ffe7a709acb7cc4fc907271176 change-id: 20260706-james-cs-hw_id-output-failure-1d06d042ef96 prerequisite-change-id: 20260605-james-cs-unformatted-per-thread-fix-50e723aa7f0e:v2 prerequisite-patch-id: 1aa32269a3a7dc76840dd8a24cb5a8715507e898 prerequisite-patch-id: ef471f468351462f67efa58a09a3461306ac5a0a prerequisite-patch-id: c78946cf4ec1c7722570865403a3562625bdaa33 prerequisite-patch-id: 0b953eee0252db3c7ba3ef1e3397048a94482ca2 prerequisite-patch-id: 27da72c2f01bcc68205bbcc14d5019369c02cb83 prerequisite-patch-id: 365c6d5f71c754e4c690b5ebed3453565bb5b809 prerequisite-patch-id: 64e9419a41082a2db3daf255baf0fb40efbae5dd prerequisite-patch-id: 1dc2e6ef8e76b369736b4c88560302861b2e4faf prerequisite-patch-id: b7de38ec4d90f5b45d56390a4e5919a1d6439951 prerequisite-patch-id: 3548d5b161cfb8c11a166ca2b3228ca4821f1ca0 prerequisite-patch-id: 97e66600218a2b32e5b21d0abcb6321590f98de9 prerequisite-patch-id: 292282b20bd8ad7d096ee7ffbc55dcfb75455f53 prerequisite-patch-id: c80259b47850ba65c985c85a29c8cf4948463d59 Best regards, -- James Clark <james.clark(a)linaro.org>

4 days, 8 hours

Re: [PATCH 2/2] perf: arm_spe: Prefer large AUX mappings

by Leo Yan

On Mon, Aug 10, 2026 at 04:10:48PM +0100, Will Deacon wrote: > On Mon, Aug 10, 2026 at 03:44:42PM +0100, Leo Yan wrote: > > Commit 18049c8cff9c ("perf/aux: Allocate non-contiguous AUX pages by > > default") made the AUX allocator use order-0 pages by default unless a > > PMU explicitly asks for contiguous allocations. > > But that commit specifically calls out SPE as benefitting from > non-contiguous pages: > > "For instance, ARM SPE and TRBE operate with virtual pages, and > Coresight ETR allocates a separate buffer. For these PMUs, > allocating contiguous AUX pages unnecessarily exacerbates memory > fragmentation. This fragmentation can prevent their use on > long-running devices." > > so why doesn't passing PERF_PMU_CAP_AUX_PREFER_LARGE reintroduce the > problems that 18049c8cff9c was trying to solve? The question is how "allocating contiguous AUX pages unnecessarily exacerbates memory fragmentation." The relevant information I could find is [1]: "On Android, we collect ETM data periodically on internal user devices for AutoFDO optimization (for both userspace libraries and the kernel). Allocating a large chunk of contiguous AUX pages (4M for each CPU) periodically is almost unbearable. The kernel may need to kill many processes to fulfill the request. It affects user experience even after using PMU." We might have missed chance to clarify how the fragmentation issue occurs in the first place. Let's say, a phone with 8 CPUs, allocating 4MB per CPU requires 32MB in total, which is a relatively small portion of 4GiB or 8GiB of RAM commonly found in phones. Moreover, once contiguous pages are freed, the buddy allocator can coalesce them again into buddy list. It is not obvious to me that PREFER_LARGE directly causes fragmentation. One case where AUX allocation could exacerbate fragmentation is when the system is already fragmented. If a high-order allocation fails and the allocator falls back to smaller-order blocks, those allocations may consume free blocks scattered across different buddy regions and make subsequent high-order allocations more difficult. If this is the main concern, I'd suggest using a smaller AUX buffer (e.g. 1MB or even 512KB) for TRBE/SPE to reduce memory pressure. Snapshot mode '-S' could also be considered, as it allows the buffer to be allocated once and reused for subsequent recordings by signals. OTOH, using only order-0 pages can significantly increase TTW overhead on the trace path and lead to overflows, we observe this causes huge trace discontinuity. In the end, we need to trace-off the fragmentation concern against the trace discontinuity. Thanks, Leo [1] https://lore.kernel.org/lkml/CALJ9ZPNLgEBxOmDim-vztUknEETwdL-Z2gJ8K9s44TiPg…

1 week

[PATCH v2 00/14] perf cs-etm: Per-thread mode fixes and snapshot wrap support

by James Clark

Fix a few issues with per-thread mode: - With TRBE, thread migrations hit a queue without a decoder assert - On nVHE, everything looks like guest trace instead of host - Tracing multiple threads didn't really work After adding a tests for those, it turns out there were some snapshot bugs not picked up by the other snapshot tests. That showed that the different snapshot searches for Intel BTS and Arm SPE were inconsistent and I think all tracers can benefit from using the same pointer fixup logic, so that gets refactored into the common auxtrace.c. Then SPE can just share the Intel BTS search as it was obviously copied from it originally. IntelPT keeps the more advanced duplicate data search, but in the future that could probably be made the common one and all tracers would benefit from using it (except maybe BTS if it always has a very small buffer, but I doubt the overhead of the duplicate search would be an issue). For now this is more of a refactor rather than behavioral change so don't do that yet. Signed-off-by: James Clark <james.clark(a)linaro.org> --- Changes in v2: - Remove stale "may be reset to -1 in per-thread mode" comment - Set CPU when queueing auxtrace fragments otherwise auxtrace_queues__add_event() overwrites the queue's CPU - Link to v1: https://patch.msgid.link/20260811-james-cs-unformatted-per-thread-fix-v1-0-… --- James Clark (14): perf cs-etm: Fix nVHE per-thread decoding perf cs-etm: Warn for invalid timestamp option perf cs-etm: Turn on context packet timestamps in per-thread mode perf cs-etm: Use per-CPU queues for per-thread mode perf cs-etm: Increase default timestamp generation period perf auxtrace: Turn Intel BTS snapshot search into a generic one perf arm-spe: Use generic snapshot search perf auxtrace: intel-pt: Use new snapshot_has_wrapped callback perf cs-etm: Queue partial AUX records perf cs-etm: Don't print missing buffers in snapshot mode perf auxtrace: cs-etm: Capture wrapped snapshots perf test: Allow infinite named_thread loops perf test: Add test for per-thread mode perf cs-etm: Test multiple per-thread threads Documentation/userspace-api/perf_ring_buffer.rst | 6 +- tools/perf/Documentation/perf-test.txt | 2 +- tools/perf/arch/arm/util/cs-etm.c | 30 ++++- tools/perf/arch/arm64/util/arm-spe.c | 147 +-------------------- tools/perf/arch/x86/util/intel-bts.c | 115 +--------------- tools/perf/arch/x86/util/intel-pt.c | 58 +++----- .../perf/tests/shell/coresight/per-thread-multi.sh | 78 +++++++++++ tools/perf/tests/shell/coresight/per-thread.sh | 48 +++++++ .../perf/tests/shell/coresight/raw_dump_stress.sh | 5 - tools/perf/tests/workloads/named_threads.c | 7 +- tools/perf/util/auxtrace.c | 136 +++++++++++++++++-- tools/perf/util/auxtrace.h | 20 ++- tools/perf/util/cs-etm.c | 101 ++++++++++---- 13 files changed, 384 insertions(+), 369 deletions(-) --- base-commit: 9453bc6a69ef43755f1c28d5688cacdd69fa16bd change-id: 20260605-james-cs-unformatted-per-thread-fix-50e723aa7f0e Best regards, -- James Clark <james.clark(a)linaro.org>

1 week, 5 days

[PATCH v9 00/13] fix several inconsistencies with sysfs configuration in etmX

by Yeoreum Yun

The current ETMx configuration via sysfs can lead to the following inconsistencies: - If a configuration is modified via sysfs while a perf session is active, the running configuration may differ between before a sched-out and after a subsequent sched-in. - If a perf session and sysfs session tries to enable concurrently, configuration from configfs could be corrupted (etm4). - There is chance to corrupt drvdata->config if perf session tries to enabled among handling cscfg_csdev_disable_active_config() in etm4_disable_sysfs() (etm4). To resolve these inconsistencies, the configuration should be separated into: - active_config, which is applied configuration for the current session - config, which stores the settings configured via sysfs. and apply configuration from configfs after taking a mode. Also, This patch set includes some small fixes: - missing trace id release in etm4x. - underflow issue for nrseqstate. - wrong check in etm4x_sspcicrn_present(). - missing call of cscfg_csdev_disable_active_config() This patch based on coresight tree's next Patch History ============= from v8 to v9: - add feat_csdev_lock guard interface. - set feat_csdev->drv_spinlock as NULL for etmv4_drvdata. - https://lore.kernel.org/all/20260629090007.1718746-1-yeoreum.yun@arm.com/ from v7 to v8: - accept @Leo Yan' suggestion to handle error. - small minor fixes following @Suzuki' suggestion. - https://lore.kernel.org/all/20260519154812.254884-1-yeoreum.yun@arm.com/ from v6 to v7: - rebase on coresight/next - add ETM_MAX_SEQ_TRANSITIONS define - remove redundant patch relavent cpu-hotplug as coresight-pm patch merged. - https://lore.kernel.org/all/20260422132203.977549-1-yeoreum.yun@arm.com/ from v5 to v6: - fix missing of calling cscfg_csdev_disable_active_config() - add rb & fixes tags. - add ss_status field in etm4x_drvdata to expose STATUS and PENDING bits. - https://lore.kernel.org/all/20260415165528.3369607-1-yeoreum.yun@arm.com/ from v4 to v5: - add rb-tag. - fix underflow issue for nrseqstate. - fix wrong check in etm4_sspcicrn_present(). - remove redundant fields on etmv4_save_state. - rename caps->ss_status to ss_cmp. - fix wrong location of etm4_release_trace_id. - https://lore.kernel.org/all/20260413142003.3549310-1-yeoreum.yun@arm.com/ from v3 to v4: - change etm_drvdata->spinlock type to raw_spin_lock_t - remove redundant call etmX_enable_hw() with starting_cpu() callsback. - fix missing trace id release. - add missing docs. - https://lore.kernel.org/all/20260412175506.412301-1-yeoreum.yun@arm.com/ from v2 to v3: - fix build error for etm3x. - fix checkpatch warning. - https://lore.kernel.org/all/20260410074310.2693385-1-yeoreum.yun@arm.com/ from v1 to v2 - rebased to v7.0-rc7. - introduce etmX_caps structure to save etmX's capabilities. - remove ss_status from etmv4_config. - modify active_config after taking a mode (perf/sysfs). - https://lore.kernel.org/all/20260317181705.2456271-1-yeoreum.yun@arm.com/ Yeoreum Yun (13): coresight: etm4x: fix wrong check of etm4x_sspcicrn_present() coresight: etm4x: fix underflow for usage of (nrseqstate - 1) coresight: etm4x: fix leaked trace id coresight: etm4x: fix inconsistencies with sysfs configuration coresight: etm4x: missing cscfg_csdev_disable_active_config() in perf enable coresight: etm3x: fix inconsistencies with sysfs configuration coresight: etm3x: change drvdata->spinlock type to raw_spin_lock_t coresight: etm3x: remove redundant cpu online check on etm_enable_sysfs() coresight: etm4x: introduce struct etm4_caps coresight: etm4x: exclude ss_status from drvdata->config coresight: etm4x: remove s_ex_level from config coresight: etm4x: remove redundant fields in etmv4_save_state coresight: etm3x: introduce struct etm_caps .../hwtracing/coresight/coresight-config.c | 18 +- .../hwtracing/coresight/coresight-config.h | 26 + drivers/hwtracing/coresight/coresight-etm.h | 48 +- .../coresight/coresight-etm3x-core.c | 96 ++-- .../coresight/coresight-etm3x-sysfs.c | 159 +++--- .../hwtracing/coresight/coresight-etm4x-cfg.c | 16 +- .../coresight/coresight-etm4x-core.c | 454 ++++++++++-------- .../coresight/coresight-etm4x-sysfs.c | 204 ++++---- drivers/hwtracing/coresight/coresight-etm4x.h | 204 ++++---- 9 files changed, 682 insertions(+), 543 deletions(-) base-commit: 98495b5a4d77dd22e106f462b76e1093a55b29a7 -- LEVI:{C3F47F37-75D8-414A-A8BA-3980EC8A46D7}

1 week, 5 days

[PATCH v3 0/8] coresight: Prevent per-thread events from sharing a sink

by James Clark

Concurrent per-thread events results in a WARN on N1SDP which leads to the realization that per-thread events shouldn't have been sharing sinks in the first place. This slips through because different per-thread events will have the same PID if owned by the same process, and we only check the PID and nothing else. That results in unexpected WARNs because it looks like we assumed it couldn't happen (although exclusive PMU rules allow it). But even if it was supported it would result in trace from the wrong thread in another event's per-thread buffer, so we should disallow it. Fix it everywhere the same PID checking logic was copy pasted. Then the PIDs can be dropped from a few structs as they are now unused. Signed-off-by: James Clark <james.clark(a)linaro.org> --- Changes in v3: - Storing and accessing event owners at runtime causes problems due to various scenarios of: events (and sibling events) exiting, children inheriting event FDs, PID reuse, CPU affine events that also have a target process set but different inherit settings. Fix it by creating a session ID in etm_setup_aux() and holding the references in it for the duration of the whole session. (Leo) - Make the ETR buffer allocator consistent with sink sharing rules by not doing numeric PID comparisons there either. - Fix up some Sashiko reports that it sees after interacting with cscfg and taking extra references to tasks and PIDs. - Link to v2: https://lore.kernel.org/r/20260709-james-cs-multiple-per-threads-v2-0-10ac7… Changes in v2: - Fix inherited events by following event->parent - Link to v1: https://lore.kernel.org/r/20260709-james-cs-multiple-per-threads-v1-0-d384e… --- James Clark (8): coresight: tmc-etr: Don't stop Perf cleanup for active sysfs reads coresight: configfs: Don't assume active until cscfg_mgr is set coresight: etm-perf: Flush workqueue before unloading module coresight: tmc-etr: Prevent per-thread events from sharing a sink coresight: tmc-etr: Use session ID for buffer ownership coresight: tmc-etf: Prevent per-thread events from sharing a sink coresight: etb10: Prevent per-thread events from sharing a sink coresight: ultrasoc-smb: Prevent per-thread events from sharing a sink drivers/hwtracing/coresight/coresight-core.c | 28 +-- drivers/hwtracing/coresight/coresight-etb10.c | 33 ++-- drivers/hwtracing/coresight/coresight-etm-perf.c | 79 ++++++++- drivers/hwtracing/coresight/coresight-etm-perf.h | 15 ++ drivers/hwtracing/coresight/coresight-priv.h | 2 - drivers/hwtracing/coresight/coresight-syscfg.c | 6 +- drivers/hwtracing/coresight/coresight-tmc-core.c | 6 +- drivers/hwtracing/coresight/coresight-tmc-etf.c | 44 ++--- drivers/hwtracing/coresight/coresight-tmc-etr.c | 207 +++++++++++++---------- drivers/hwtracing/coresight/coresight-tmc.h | 30 ++-- drivers/hwtracing/coresight/coresight-trbe.c | 3 +- drivers/hwtracing/coresight/ultrasoc-smb.c | 25 +-- drivers/hwtracing/coresight/ultrasoc-smb.h | 6 +- include/linux/coresight.h | 5 +- 14 files changed, 294 insertions(+), 195 deletions(-) --- base-commit: 98495b5a4d77dd22e106f462b76e1093a55b29a7 change-id: 20260708-james-cs-multiple-per-threads-ed1d25ed1734 Best regards, -- James Clark <james.clark(a)linaro.org>

1 week, 6 days

Re: [PATCH v2 11/14] perf auxtrace: cs-etm: Capture wrapped snapshots

by James Clark

On 24/08/2026 13:21, Adrian Hunter wrote: > On 21/08/2026 12:49, James Clark wrote: >> snapshot_has_wrapped() isn't implemented for Coresight, so when a wrap >> occurs we only save from offset 0 to current head, modulo buffer size, >> rather than capturing the whole buffer. Coresight drivers currently use >> a monotonic head pointer even in snapshot mode, which gives a lot more >> information for taking snapshots than a wrapped pointer. It means we >> don't need to search for data at the end of the buffer to guess if a >> wrap has occurred, and we can go back to saving partial buffers after a >> wrap by checking if the next size to take is less than the buffer >> size. >> >> The drivers will always be able to generate a monotonic head because >> TRBE has a bytes written counter, so even if we stop generating an AUX >> record on every overflow, we can check if the counter indicated an >> overwrite and offset head by a whole buffer size if it wrapped. This >> forces userspace to read the whole buffer rather than between last and >> current head. ETR and other sinks use software double buffering, so can >> continue as they are. >> >> Add a much simpler 'monotonic_snapshot_head' path for tracers to use, >> and use it for Coresight. >> >> Signed-off-by: James Clark <james.clark(a)linaro.org> >> --- >> tools/perf/arch/arm/util/cs-etm.c | 1 + >> .../perf/tests/shell/coresight/raw_dump_stress.sh | 5 ----- >> tools/perf/util/auxtrace.c | 22 ++++++++++++++++------ >> tools/perf/util/auxtrace.h | 2 ++ >> 4 files changed, 19 insertions(+), 11 deletions(-) >> >> diff --git a/tools/perf/arch/arm/util/cs-etm.c b/tools/perf/arch/arm/util/cs-etm.c >> index d36b7e41399b..dbf76dd3349e 100644 >> --- a/tools/perf/arch/arm/util/cs-etm.c >> +++ b/tools/perf/arch/arm/util/cs-etm.c >> @@ -918,6 +918,7 @@ struct auxtrace_record *cs_etm_record_init(int *err) >> ptr->itr.info_fill = cs_etm_info_fill; >> ptr->itr.snapshot_start = cs_etm_snapshot_start; >> ptr->itr.snapshot_finish = cs_etm_snapshot_finish; >> + ptr->itr.monotonic_snapshot_head = true; >> ptr->itr.reference = cs_etm_reference; >> ptr->itr.free = cs_etm_recording_free; >> ptr->itr.read_finish = auxtrace_record__read_finish; >> diff --git a/tools/perf/tests/shell/coresight/raw_dump_stress.sh b/tools/perf/tests/shell/coresight/raw_dump_stress.sh >> index bea70d825596..c9459ded4cbe 100755 >> --- a/tools/perf/tests/shell/coresight/raw_dump_stress.sh >> +++ b/tools/perf/tests/shell/coresight/raw_dump_stress.sh >> @@ -48,11 +48,6 @@ read -r size offset last_idx <<< "$(awk ' >> # everything was dumped. Allow 48 bytes difference to cover 3 frames: current >> # frame length, a partial frame and a final empty one, all of which aren't >> # dumped. >> -# >> -# TODO: for a single snapshot, offset should always be zero. However, we >> -# currently output AUX records in snapshot mode when we shouldn't, which >> -# increments the offset. Allow for that until it's fixed so we can test raw >> -# dumping. >> decode_size=$((1 + last_idx - offset)) >> if [ "$decode_size" -gt "$((size - 48))" ] && [ "$decode_size" -le "$((size))" ]; then >> echo "PASS: AUXTRACE buffer length matches dumped packet index" >> diff --git a/tools/perf/util/auxtrace.c b/tools/perf/util/auxtrace.c >> index 8ca872e6d465..e3c770b46e94 100644 >> --- a/tools/perf/util/auxtrace.c >> +++ b/tools/perf/util/auxtrace.c >> @@ -2051,9 +2051,9 @@ static int __auxtrace_mmap__read(struct mmap *map, >> bool snapshot, size_t snapshot_size) >> { >> struct auxtrace_mmap *mm = &map->auxtrace_mmap; >> - u64 head, old = mm->prev, offset, ref; >> + u64 head, old = mm->prev, offset, ref, size; >> unsigned char *data = mm->base; >> - size_t size, head_off, old_off, len1, len2, padding; >> + size_t head_off, old_off, len1, len2, padding; >> union perf_event ev; >> void *data1, *data2; >> int kernel_is_64_bit = perf_env__kernel_is_64_bit(env); >> @@ -2061,10 +2061,20 @@ static int __auxtrace_mmap__read(struct mmap *map, >> >> head = auxtrace_mmap__read_head(mm, kernel_is_64_bit); >> >> - if (snapshot && itr->snapshot_has_wrapped) { >> - err = auxtrace_find_snapshot(itr, mm->idx, mm, data, &head, &old); >> - if (err) >> - return err; >> + if (snapshot) { >> + if (itr->snapshot_has_wrapped) { >> + err = auxtrace_find_snapshot(itr, mm->idx, mm, data, &head, &old); >> + if (err) >> + return err; >> + } else if (itr->monotonic_snapshot_head) { >> + size = head - old; >> + /* Force a full buffer read if a wrap has occurred */ >> + if (size > mm->len) { >> + pr_debug3("%s: wrap detected, adjusting old from 0x%"PRIx64" to 0x%"PRIx64"\n", >> + __func__, old, head - mm->len); >> + old = head - mm->len; >> + } >> + } > > I feel like this is asking to keep the ->find_snapshot() callback. > i.e. basically the same thing, but add a helper that BTS and SPE > can call directly, like auxtrace_record__default_find_snapshot(). > For PT, add/use auxtrace_record__snapshot_fixup() that does the > common old/head adjustment. > > Then ->monotonic_snapshot_head is not needed. > Hmm yeah I suppose that makes sense. I can give it a go. >> } >> >> if (old == head) >> diff --git a/tools/perf/util/auxtrace.h b/tools/perf/util/auxtrace.h >> index a9d0c84184c5..971b817d3396 100644 >> --- a/tools/perf/util/auxtrace.h >> +++ b/tools/perf/util/auxtrace.h >> @@ -385,6 +385,7 @@ struct auxtrace_mmap_params { >> * @evlist: selected events list >> * @snapshot_wrapped_len: number of bits in @snapshot_wrapped >> * @snapshot_wrapped: bitmap indicating if each aux buffer has wrapped >> + * @monotonic_snapshot_head: driver doesn't wrap head pointer on overflow >> */ >> struct auxtrace_record { >> int (*recording_options)(struct auxtrace_record *itr, >> @@ -412,6 +413,7 @@ struct auxtrace_record { >> int snapshot_wrapped_len; >> unsigned long *snapshot_wrapped; >> int snapshot_search_bytes; >> + bool monotonic_snapshot_head; >> }; >> >> /** >> >

2 weeks, 1 day

Re: [PATCH v2 03/14] perf cs-etm: Turn on context packet timestamps in per-thread mode

by James Clark

On 21/08/2026 11:01, sashiko-bot(a)kernel.org wrote: > Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: > - [Medium] Unconditionally enabling timestamps in per-thread mode breaks recording out-of-the-box on hardware without timestamp support (e.g., ETMv3 or ETMv4 without TRCIDR0.TSSIZE). > -- > > --- Patch [3]: [PATCH v2 03/14] perf cs-etm: Turn on context packet timestamps in per-thread mode --- > commit 23dc37429523715cdb26c5647b22b1a9dc0c89ab > Author: James Clark <james.clark(a)linaro.org> > > perf cs-etm: Turn on context packet timestamps in per-thread mode > > This enables context packet timestamps in per-thread mode, replacing the > previous unset default. It helps keep samples in order when supporting > multiple per-thread mode threads. > >> diff --git a/tools/perf/arch/arm/util/cs-etm.c b/tools/perf/arch/arm/util/cs-etm.c >> index acfdb3f53821e..d0486a164a700 100644 >> --- a/tools/perf/arch/arm/util/cs-etm.c >> +++ b/tools/perf/arch/arm/util/cs-etm.c >> @@ -453,13 +453,22 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, > [ ... ] >> if (!perf_cpu_map__is_any_cpu_or_is_empty(cpus)) { >> + /* >> + * Timestamps are required to interleave samples from different >> + * CPUs. >> + */ >> evsel__set_config_if_unset(cs_etm_evsel, "timestamp", 1); >> + /* Context IDs are required to associate trace to a process */ >> evsel__set_config_if_unset(cs_etm_evsel, "contextid", 1); >> + } else { >> + /* >> + * Enable context packet timestamps only (no periodic >> + * timestamps). Even in per-thread mode with a single process we >> + * still need to correlate trace to a specific mmap around >> + * execs, which can be done with just context packet timestamps. >> + */ >> + evsel__set_config_if_unset(cs_etm_evsel, "timestamp", UINT64_MAX); > > [Severity: Medium] > Will unconditionally setting this timestamp break recording out-of-the-box on > hardware without timestamp support, like ETMv3? > > Looking at cs_etm_validate_timestamp(), if the timestamp config is set but > unsupported, it aborts the trace: > > tools/perf/arch/arm/util/cs-etm.c:cs_etm_validate_timestamp() { > if (!val) > return 0; > > if (cs_etm_get_version(cs_etm_pmu, cpu) == CS_ETMV3) { > pr_err("%s: timestamp not supported in ETMv3...\n"); > return -EINVAL; > } > } > > Does this mean users running perf record --per-thread on such hardware will > now get an error by default unless they explicitly pass timestamp=0? > Yes but the behavior is the same in per-CPU mode so this isn't a new issue. It's easily worked around and nobody has ever complained. >> } >> >> /* >

2 weeks, 4 days

Re: [PATCH v2 3/5] perf cs-etm: Add branch history to existing samples

by James Clark

On 17/08/2026 23:22, Amir Ayupov wrote: > Implement --itrace=L for CoreSight ETM: decode timestamped trace up to > each existing PMU sample and attach the branch history that led to it. > The sample keeps its own ip, callchain and event identity, and a sample > that already carries a branch stack is left alone. > > Samples are correlated with the trace by time, so this requires virtual > ETM timestamps that are correlated to perf time; timeless decoding is > rejected. The decode loop, which the previous patch left on its own in > cs_etm__process_timestamped_queues(), grows a timestamp argument and > stops once the decode frontier reaches it, so on return the > thread stack holds the branches that executed before the sample and none > that executed after. Attaching then reduces to the same > thread_stack__br_sample_late() call intel-pt uses. > > No explicit sample-to-queue matching is needed: > thread_stack__br_sample_late() keys on the thread, and the thread stack > is already emptied whenever the decoder reports a discontinuity. The one > case that was not covered is a queue whose trace runs out: flush the > thread stack there too, otherwise samples recorded after the last trace > would pick up stale history. > > Take the branch history when attaching it rather than leaving it in the > thread stack. With AUX pause and resume, a pause sample ends a completed > trace window and that window belongs to the sample. Execution while AUX > is paused is not traced, so retaining the window would let a later sample > reuse branches from before the untraced gap. Consuming it ensures that a > sample with no newly decoded trace gets an empty branch stack instead. > > As with intel-pt, the internal reconstruction ring is kept deeper than > the requested output depth to cover branches decoded between the sampled > ip and the point at which the sample time was recorded, so --itrace=L<n> > can actually return n entries. Kernel-inclusive trace gets the same > conservative 1024-entry headroom that intel-pt uses. > > Assisted-by: Devmate:GPT-5.6 > Signed-off-by: Amir Ayupov <aaupov(a)fb.com> > --- > tools/perf/util/cs-etm.c | 185 +++++++++++++++++++++++++++++++-- > tools/perf/util/thread-stack.c | 17 +++ > tools/perf/util/thread-stack.h | 1 + > 3 files changed, 195 insertions(+), 8 deletions(-) > > diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c > index 4d895f11deb7f..00407a80933e1 100644 > --- a/tools/perf/util/cs-etm.c > +++ b/tools/perf/util/cs-etm.c > @@ -72,6 +72,11 @@ struct cs_etm_auxtrace { > bool use_callchain; > > int num_cpu; > + /* Output depth requested with --itrace=L<n> */ > + unsigned int br_stack_sz; > + /* Internal reconstruction depth, see cs_etm__br_stack_init() */ > + unsigned int br_stack_sz_plus;> + struct branch_stack *br_stack; > u64 latest_kernel_timestamp; > u32 auxtrace_type; > u32 branches_filter; > @@ -91,6 +96,7 @@ struct cs_etm_traceid_queue { > u64 kernel_start; > union perf_event *event_buf; > unsigned int br_stack_sz; > + unsigned int br_stack_sz_plus; > struct branch_stack *last_branch; > struct ip_callchain *callchain; > struct cs_etm_packet *prev_packet; > @@ -141,7 +147,8 @@ struct cs_etm_queue { > }; > > static int cs_etm__update_queues(struct cs_etm_auxtrace *etm); > -static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm); > +static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm, > + u64 timestamp); > static int cs_etm__flush_timestamped_queues(struct cs_etm_auxtrace *etm); > static int cs_etm__process_timeless_queues(struct cs_etm_auxtrace *etm, > pid_t tid); > @@ -165,6 +172,7 @@ static int cs_etm__metadata_set_trace_id(u8 trace_chan_id, u64 *cpu_metadata); > #define TO_QUEUE_NR(cs_queue_nr) (cs_queue_nr >> 16) > #define TO_TRACE_CHAN_ID(cs_queue_nr) (cs_queue_nr & 0x0000ffff) > #define SINK_UNSET ((u32) -1) > +#define MAX_TIMESTAMP (~0ULL) > > static u32 cs_etm__get_v7_protocol_version(u32 etmidr) > { > @@ -674,7 +682,8 @@ static int cs_etm__init_traceid_queue(struct cs_etm_queue *etmq, > if (!tidq->last_branch) > goto out_free; > > - tidq->br_stack_sz = etm->synth_opts.last_branch_sz; > + tidq->br_stack_sz = etm->br_stack_sz; > + tidq->br_stack_sz_plus = etm->br_stack_sz_plus; > } > > if (etm->synth_opts.callchain) { > @@ -794,7 +803,7 @@ static void cs_etm__packet_swap(struct cs_etm_auxtrace *etm, > struct cs_etm_packet *tmp; > > if (etm->synth_opts.branches || etm->synth_opts.last_branch || > - etm->synth_opts.instructions) { > + etm->synth_opts.add_last_branch || etm->synth_opts.instructions) { > /* > * Swap PACKET with PREV_PACKET: PACKET becomes PREV_PACKET for > * the next incoming packet. > @@ -963,7 +972,7 @@ static int cs_etm__flush_events(struct perf_session *session, > if (ret) > return ret; > > - ret = cs_etm__process_timestamped_queues(etm); > + ret = cs_etm__process_timestamped_queues(etm, MAX_TIMESTAMP); > if (ret) > return ret; > > @@ -1060,6 +1069,7 @@ static void cs_etm__free(struct perf_session *session) > zfree(&aux->metadata[i]); > > zfree(&aux->metadata); > + zfree(&aux->br_stack); > zfree(&aux); > } > > @@ -1597,7 +1607,8 @@ static void cs_etm__add_stack_event(struct cs_etm_queue *etmq, > u64 from, to; > int size; > > - if (!etm->synth_opts.branches && !etm->synth_opts.instructions) > + if (!etm->synth_opts.branches && !etm->synth_opts.instructions && > + !etm->synth_opts.add_last_branch) > return; > > if (!cs_etm__packet_has_taken_branch(tidq->prev_packet)) > @@ -1614,7 +1625,7 @@ static void cs_etm__add_stack_event(struct cs_etm_queue *etmq, > tidq->prev_packet->flags, from, to, size, > etmq->buffer->buffer_nr + 1, > etmq->etm->use_callchain, > - tidq->br_stack_sz, 0); > + tidq->br_stack_sz_plus, 0); > } else { > thread_stack__set_trace_nr(tidq->frontend_thread, > tidq->prev_packet->cpu, > @@ -2817,7 +2828,8 @@ static int cs_etm__update_queues(struct cs_etm_auxtrace *etm) > return ret; > } > > -static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm) > +static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm, > + u64 timestamp) > { > int ret = 0; > unsigned int cs_queue_nr, queue_nr; > @@ -2831,6 +2843,9 @@ static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm) > if (!etm->heap.heap_cnt) > break; > > + if (etm->heap.heap_array[0].ordinal >= timestamp) > + break; > + > /* Take the entry at the top of the min heap */ > cs_queue_nr = etm->heap.heap_array[0].queue_nr; > queue_nr = TO_QUEUE_NR(cs_queue_nr); > @@ -2878,8 +2893,25 @@ static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm) > * No more auxtrace_buffers to process in this etmq, simply > * move on to another entry in the auxtrace_heap. > */ > - if (!ret) > + if (!ret) { > + /* > + * The trace for this physical queue is exhausted. Drop > + * branch history for every trace ID it carried so that > + * samples arriving later cannot pick up entries decoded > + * before the gap. > + */ > + if (etm->synth_opts.add_last_branch) { > + struct int_node *inode; > + > + intlist__for_each_entry(inode, etmq->traceid_queues_list) { > + int idx = (int)(intptr_t)inode->priv; > + > + tidq = etmq->traceid_queues[idx]; > + thread_stack__flush(tidq->frontend_thread); > + } > + } > continue; > + } > > ret = cs_etm__decode_data_block(etmq); > if (ret) > @@ -3011,6 +3043,116 @@ static int cs_etm__process_switch_cpu_wide(struct cs_etm_auxtrace *etm, > return 0; > } > > +static bool cs_etm__tracing_kernel(struct cs_etm_auxtrace *etm, > + struct perf_session *session) > +{ > + struct evsel *evsel; > + > + evlist__for_each_entry(session->evlist, evsel) { > + if (evsel->core.attr.type == etm->pmu_type && > + !evsel->core.attr.exclude_kernel) > + return true; > + } > + > + return false; > +} > + > +static int cs_etm__br_stack_init(struct cs_etm_auxtrace *etm, > + struct perf_session *session) > +{ > + struct evsel *evsel; > + > + evlist__for_each_entry(session->evlist, evsel) { > + /* > + * Only timestamped events can be matched against the decoded > + * trace, so do not advertise a branch stack on any other. > + */ > + if (!(evsel->core.attr.sample_type & PERF_SAMPLE_TIME)) > + continue; Do you not also want to check for the coresight virtual timestamp option here for the same reason? Although I do see that checked somewhere else below. > + if (!(evsel->core.attr.sample_type & PERF_SAMPLE_BRANCH_STACK)) > + evsel->synth_sample_type |= PERF_SAMPLE_BRANCH_STACK; > + } > + > + /* > + * Additional branch stack depth to cater for the branches decoded > + * between the sampled ip and the point at which the sample time was > + * recorded. Those are trimmed by thread_stack__br_sample_late(), so How does the trimmer know what branches came after an IP? I could understand trimming between two timestamps, but not between one IP and one timestamp. > + * the extra depth keeps the requested output depth achievable. If > + * kernel space is not traced, only the branch into the kernel needs > + * to be accounted for. > + */ I'm not sure if this description is missing something, but I can't understand why this needs to be done. Or how kernel tracing affects it. > + if (cs_etm__tracing_kernel(etm, session)) > + etm->br_stack_sz_plus += 1024; > + else > + etm->br_stack_sz_plus += 1; > + > + etm->br_stack = zalloc(sizeof(struct branch_stack) + > + etm->br_stack_sz * sizeof(struct branch_entry)); > + if (!etm->br_stack) > + return -ENOMEM; > + > + return 0; > +} > + > +/* > + * Add decoded branch history to an existing sample. The sample keeps its own > + * ip, callchain and event identity; only an absent branch stack is filled in. > + */ > +static int cs_etm__process_sample(struct cs_etm_auxtrace *etm, > + struct perf_session *session, > + struct perf_sample *sample) > +{ > + struct machine *machine = &session->machines.host; > + struct thread *thread; > + int err; > + > + if (!etm->synth_opts.add_last_branch || sample->branch_stack || > + !sample->ip || !sample->time || sample->time == (u64)-1) > + return 0; > + > + /* Adding branch history to existing samples supports the host only */ > + if (sample->cpumode == PERF_RECORD_MISC_GUEST_KERNEL || > + sample->cpumode == PERF_RECORD_MISC_GUEST_USER) > + return 0; > + > + err = cs_etm__update_queues(etm); > + if (err) > + return err; > + > + /* > + * Decode every queue up to this sample's time. Afterwards the thread > + * stack holds the branches that executed before the sample, and > + * nothing that executed after it. > + */ > + err = cs_etm__process_timestamped_queues(etm, sample->time); > + if (err) > + return err; > + > + thread = machine__findnew_thread(machine, sample->pid, sample->tid); > + if (!thread) > + return -ENOMEM; > + > + /* > + * Take the branch history rather than copying it. The trace window > + * belongs to the sample that ends it, so once it has been attached a > + * later sample with nothing newly decoded finds an empty stack rather > + * than being given an earlier window's branches. That is the common > + * case whenever the trace is duty cycled, by AUX pause/resume or by > + * ETM strobing. > + */ > + thread_stack__br_sample_late(thread, sample->cpu, etm->br_stack, > + etm->br_stack_sz, sample->ip, > + machine__kernel_start(machine)); > + thread_stack__br_stack_consume(thread, sample->cpu); > + > + if (etm->br_stack->nr) > + sample->branch_stack = etm->br_stack; How does this work? We have a queue for each CPU, and decoding happens in parallel (kind of), but when a Perf sample arrives we just attach the stack from the global br_stack? Shouldn't we look at the CPU of the sample and use cs_etm__get_queue() to get the right queue and branch stack? Maybe it's not functionally different if br_stack always happens to be set from the queue related to the last Perf sample, but it would be nicer to not have to assume. You might want to check "[PATCH 00/14] perf cs-etm: Per-thread mode fixes and snapshot wrap support" because it changes to a per-CPU queue even for per-thread mode which could help. > + > + thread__put(thread); > + > + return 0; > +} > + > static int cs_etm__process_event(struct perf_session *session, > union perf_event *event, > struct perf_sample *sample, > @@ -3049,6 +3191,9 @@ static int cs_etm__process_event(struct perf_session *session, > case PERF_RECORD_SWITCH_CPU_WIDE: > return cs_etm__process_switch_cpu_wide(etm, event); > > + case PERF_RECORD_SAMPLE: > + return cs_etm__process_sample(etm, session, sample); > + Don't we want to generalise this and process trace up to the timestamp of _any_ event. Can we move the cs_etm__process_timestamped_queues() call into cs_etm__process_event(). Surely we want to always decode up to any event if coresight virtual timestamps are enabled? That way we access the right mmaps too and the decode order doesn't change depending on the branch stack options. > case PERF_RECORD_AUX: > /* > * Record the latest kernel timestamp available in the header > @@ -3752,11 +3897,34 @@ int cs_etm__process_auxtrace_info_full(union perf_event *event, > > etm->use_thread_stack = etm->synth_opts.thread_stack || > etm->synth_opts.last_branch || > + etm->synth_opts.add_last_branch || > etm->synth_opts.callchain; > > etm->use_callchain = etm->synth_opts.thread_stack || > etm->synth_opts.callchain; > > + if (etm->synth_opts.last_branch || etm->synth_opts.add_last_branch) { > + etm->br_stack_sz = etm->synth_opts.last_branch_sz; > + etm->br_stack_sz_plus = etm->br_stack_sz; > + } > + > + if (etm->synth_opts.add_last_branch) { > + /* > + * Existing samples are matched to decoded trace by time, so > + * the trace must carry timestamps that are correlated to perf > + * time and the queues must be decoded in time order. > + */ > + if (etm->timeless_decoding || !etm->has_virtual_ts) { > + pr_err("CS ETM Trace: --itrace=L requires virtual timestamped trace\n"); > + err = -EINVAL; > + goto err_free_queues; > + } > + > + err = cs_etm__br_stack_init(etm, session); > + if (err) > + goto err_free_queues; > + } > + > err = cs_etm__synth_events(etm, session); > if (err) > goto err_free_queues; > @@ -3812,6 +3980,7 @@ int cs_etm__process_auxtrace_info_full(union perf_event *event, > auxtrace_queues__free(&etm->queues); > session->auxtrace = NULL; > err_free_etm: > + zfree(&etm->br_stack); > zfree(&etm); > err_free_metadata: > /* No need to check @metadata[j], free(NULL) is supported */ > diff --git a/tools/perf/util/thread-stack.c b/tools/perf/util/thread-stack.c > index 1360f44421ef8..2713a2ad70b69 100644 > --- a/tools/perf/util/thread-stack.c > +++ b/tools/perf/util/thread-stack.c > @@ -614,6 +614,23 @@ void thread_stack__sample_late(struct thread *thread, int cpu, > } > } > > +/* > + * Branch history belongs to the sample that ends the trace window, so a > + * decoder that attaches it to an existing sample should take it rather than > + * copy it. A later sample with no newly decoded trace then finds an empty > + * branch stack instead of the previous window's branches. > + */ > +void thread_stack__br_stack_consume(struct thread *thread, int cpu) > +{ > + struct thread_stack *ts = thread__stack(thread, cpu); > + > + if (!ts || !ts->br_stack_rb) > + return; > + > + ts->br_stack_pos = 0; > + ts->br_stack_rb->nr = 0; > +} > + > void thread_stack__br_sample(struct thread *thread, int cpu, > struct branch_stack *dst, unsigned int sz) > { > diff --git a/tools/perf/util/thread-stack.h b/tools/perf/util/thread-stack.h > index b3cd09beb62f0..2aec292bd1bcb 100644 > --- a/tools/perf/util/thread-stack.h > +++ b/tools/perf/util/thread-stack.h > @@ -88,6 +88,7 @@ void thread_stack__sample(struct thread *thread, int cpu, struct ip_callchain *c > void thread_stack__sample_late(struct thread *thread, int cpu, > struct ip_callchain *chain, size_t sz, u64 ip, > u64 kernel_start); > +void thread_stack__br_stack_consume(struct thread *thread, int cpu); > void thread_stack__br_sample(struct thread *thread, int cpu, > struct branch_stack *dst, unsigned int sz); > void thread_stack__br_sample_late(struct thread *thread, int cpu,

3 weeks

Re: [PATCH v2 2/5] perf cs-etm: Split up cs_etm__process_timestamped_queues()

by James Clark

On 17/08/2026 23:22, Amir Ayupov wrote: > cs_etm__process_timestamped_queues() currently does three things: it seeds > the auxtrace heap with one entry per queue, it decodes until the heap is > empty, and it then walks every traceID queue to flush whatever is left in > the branch stacks. That is fine while the only caller is > cs_etm__flush_events(), which runs once, but it does not survive the > function being called repeatedly. > > Seeding cannot be repeated because a queue that still holds a heap slot > would be seeded again, adding duplicate entries and growing the heap > without bound. Flushing cannot be repeated either, because ending a block > finalises state that later trace still needs. > > Move both out. Seeding becomes cs_etm__update_queues(), gated on > queues.new_data so it only runs when new AUX data has been queued, with > etmq->on_heap tracking whether a queue currently occupies a heap slot; > this mirrors intel_pt_update_queues() and intel_pt_queue::on_heap. > Flushing becomes cs_etm__flush_timestamped_queues(). What remains is the > decode loop on its own, which a later patch can then drive incrementally. > > No functional change: the sole caller performs the same three steps in the > same order. > > Assisted-by: Devmate:GPT-5.6 > Signed-off-by: Amir Ayupov <aaupov(a)fb.com> LGTM but I'd still like to run the test. > --- > tools/perf/util/cs-etm.c | 71 ++++++++++++++++++++++++++++++++++------ > 1 file changed, 61 insertions(+), 10 deletions(-) > > diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c > index 114b3cd2da495..4d895f11deb7f 100644 > --- a/tools/perf/util/cs-etm.c > +++ b/tools/perf/util/cs-etm.c > @@ -136,9 +136,13 @@ struct cs_etm_queue { > */ > struct intlist *own_traceid_list; > u32 sink_id; > + /* Whether this queue currently occupies a slot in etm->heap */ > + bool on_heap; > }; > > +static int cs_etm__update_queues(struct cs_etm_auxtrace *etm); > static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm); > +static int cs_etm__flush_timestamped_queues(struct cs_etm_auxtrace *etm); > static int cs_etm__process_timeless_queues(struct cs_etm_auxtrace *etm, > pid_t tid); > static int cs_etm__get_data_block(struct cs_etm_queue *etmq); > @@ -939,6 +943,8 @@ static int cs_etm__flush_events(struct perf_session *session, > struct cs_etm_auxtrace *etm = container_of(session->auxtrace, > struct cs_etm_auxtrace, > auxtrace); > + int ret; > + > if (dump_trace) > return 0; > > @@ -953,7 +959,15 @@ static int cs_etm__flush_events(struct perf_session *session, > return cs_etm__process_timeless_queues(etm, -1); > } > > - return cs_etm__process_timestamped_queues(etm); > + ret = cs_etm__update_queues(etm); > + if (ret) > + return ret; > + > + ret = cs_etm__process_timestamped_queues(etm); > + if (ret) > + return ret; > + > + return cs_etm__flush_timestamped_queues(etm); > } > > static void cs_etm__free_traceid_queues(struct cs_etm_queue *etmq) > @@ -1330,6 +1344,8 @@ static int cs_etm__queue_first_cs_timestamp(struct cs_etm_auxtrace *etm, > */ > cs_queue_nr = TO_CS_QUEUE_NR(queue_nr, trace_chan_id); > ret = auxtrace_heap__add(&etm->heap, cs_queue_nr, cs_timestamp); > + if (!ret) > + etmq->on_heap = true; > out: > return ret; > } > @@ -2767,23 +2783,30 @@ static int cs_etm__process_timeless_queues(struct cs_etm_auxtrace *etm, > return 0; > } > > -static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm) > +/* > + * Seed the heap with one entry from each queue that is not already > + * represented in it, so that decoding proceeds in time order across all > + * queues. Only queues that have newly queued data need to be considered. > + */ > +static int cs_etm__update_queues(struct cs_etm_auxtrace *etm) > { > int ret = 0; > - unsigned int cs_queue_nr, queue_nr, i; > - u8 trace_chan_id; > - u64 cs_timestamp; > - struct auxtrace_queue *queue; > + unsigned int i; > struct cs_etm_queue *etmq; > - struct cs_etm_traceid_queue *tidq; > + > + if (!etm->queues.new_data) > + return 0; > + > + etm->queues.new_data = false; > > /* > * Pre-populate the heap with one entry from each queue so that we can > - * start processing in time order across all queues. > + * start processing in time order across all queues. Skip queues that > + * already occupy a heap slot, otherwise they would be added twice. > */ > for (i = 0; i < etm->queues.nr_queues; i++) { > etmq = etm->queues.queue_array[i].priv; > - if (!etmq) > + if (!etmq || etmq->on_heap) > continue; > > ret = cs_etm__queue_first_cs_timestamp(etm, etmq, i); > @@ -2791,6 +2814,19 @@ static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm) > return ret; > } > > + return ret; > +} > + > +static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm) > +{ > + int ret = 0; > + unsigned int cs_queue_nr, queue_nr; > + u8 trace_chan_id; > + u64 cs_timestamp; > + struct auxtrace_queue *queue; > + struct cs_etm_queue *etmq; > + struct cs_etm_traceid_queue *tidq; > + > while (1) { > if (!etm->heap.heap_cnt) > break; > @@ -2807,6 +2843,7 @@ static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm) > * to process it. > */ > auxtrace_heap__pop(&etm->heap); > + etmq->on_heap = false; > > tidq = cs_etm__etmq_get_traceid_queue(etmq, trace_chan_id); > if (!tidq) { > @@ -2874,7 +2911,21 @@ static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm) > */ > cs_queue_nr = TO_CS_QUEUE_NR(queue_nr, trace_chan_id); > ret = auxtrace_heap__add(&etm->heap, cs_queue_nr, cs_timestamp); > + if (ret) > + goto out; > + etmq->on_heap = true; > } > +out: > + return ret; > +} > + > +/* Flush any branch stack entries left over once all trace is decoded */ > +static int cs_etm__flush_timestamped_queues(struct cs_etm_auxtrace *etm) > +{ > + int ret = 0; > + unsigned int i; > + struct cs_etm_queue *etmq; > + struct cs_etm_traceid_queue *tidq; > > for (i = 0; i < etm->queues.nr_queues; i++) { > struct int_node *inode; > @@ -2893,7 +2944,7 @@ static int cs_etm__process_timestamped_queues(struct cs_etm_auxtrace *etm) > return ret; > } > } > -out: > + > return ret; > } >

3 weeks

Re: [PATCH v2 4/5] perf test cs-etm: Test branch history on existing samples

by James Clark

On 17/08/2026 23:22, Amir Ayupov wrote: > Add a CoreSight shell test for --itrace=L. Record timestamped ETM trace > with explicit -T sample timestamps and AUX pause/resume events, then > check that the pause samples carry both a multi-frame callchain and a > non-empty branch stack for each of the workload's two processes. > > Decode the same recording with L4 and L64 and reject any branch stack > deeper than the requested depth. > > The test skips when cs_etm is absent, when not run as root, or when the > recording turns out to lack virtual timestamps. It exercises the > timestamp-gated path and the requested-depth bound; it does not attempt > to verify that the attached history is correlated to the sample. > > Assisted-by: Devmate:GPT-5.6 > Signed-off-by: Amir Ayupov <aaupov(a)fb.com> > --- > .../tests/shell/coresight/add_last_branch.sh | 203 ++++++++++++++++++ > 1 file changed, 203 insertions(+) > create mode 100755 tools/perf/tests/shell/coresight/add_last_branch.sh > > diff --git a/tools/perf/tests/shell/coresight/add_last_branch.sh b/tools/perf/tests/shell/coresight/add_last_branch.sh > new file mode 100755 > index 0000000000000..6f09e720abe80 > --- /dev/null > +++ b/tools/perf/tests/shell/coresight/add_last_branch.sh > @@ -0,0 +1,203 @@ > +#!/bin/bash -e > +# SPDX-License-Identifier: GPL-2.0 > +# CoreSight branch history on existing samples (exclusive) > + > +perf list pmu | grep -q 'cs_etm//' || exit 2 > + > +if [ "$(id -u)" != 0 ]; then > + echo "[Skip] No root permission" > + exit 2 > +fi > + > +tmpdir=$(mktemp -d /tmp/perf-cs-add-last-branch.XXXXX) > + > +cleanup() > +{ > + rm -rf "$tmpdir" > + trap - EXIT TERM INT > +} > + > +# shellcheck disable=SC2317 # Called through trap. > +trap_cleanup() > +{ > + cleanup > + exit 1 > +} > +trap trap_cleanup EXIT TERM INT > + > +record_data() > +{ > + local cf="$tmpdir/ctl" > + local af="$tmpdir/ack" > + > + mkfifo "$cf" "$af" > + > + # Pin to one CPU so proc1 and proc2 alternate in one per-CPU trace > + # buffer. Start disabled and use the control FIFO to record only the > + # workload, not perf test setup and teardown. This is more of a what comment than a why. I got that we were doing that, but the reason I left the comment on V1 was because I couldn't see why context switching is related to branch history. Doesn't the test still test the same thing if you record a single process without -C? It's a bit hard to see what behavior the test is exercising. I see it also adds call-graph=fp, but it doesn't look for symbol names. How is testing for any non zero callchain related to coresight unless we also check the coresight branch stack matches it exactly via symbol names? Seems like it's just testing some other part of Perf here. I would expect some references to the named symbols "context_switch_loop_proc2" from the workload, but I don't see them. It also seems to be very interested in timestamps as per the commit message. But how do you know it's not attaching the branch stack from proc1 to proc2 because it gets the timestamps wrong? If it's easier to add a new workload with a deterministic branch history written in asm after a deterministic call chain we can do that. Then the test can test if a fragment of the branch history appears after an end fragment of the callchain. > + if perf record -T -o "$tmpdir/data" -C 0 -D -1 \ > + --control fifo:"$cf","$af" \ > + -e cs_etm/aux-action=start-paused/u \ > + -e cycles/aux-action=resume,period=550019/u \ > + -e cycles/aux-action=pause,period=100003,call-graph=fp/u -- \ > + taskset --cpu-list 0 perf test --record-ctl fifo:"$cf","$af" \ > + -w context_switch_loop 10000 \ > + >/dev/null 2>"$tmpdir/stderr"; then > + return 0 > + fi > + > + echo "Failed to record ETM trace with AUX pause/resume" >&2 > + cat "$tmpdir/stderr" >&2 > + return 1 > +} > + > +decode() > +{ > + local size=$1 > + local output=$2 > + > + if perf script -i "$tmpdir/data" --itrace="L$size" \ > + -F comm,pid,tid,event,ip,brstack >"$output" \ > + 2>"$tmpdir/stderr"; then > + return 0 > + fi > + > + if grep -q "itrace=L requires virtual timestamped trace" \ > + "$tmpdir/stderr"; then > + echo "[Skip] Virtual CoreSight timestamps are not available" > + cleanup > + exit 2 > + fi > + > + cat "$tmpdir/stderr" >&2 > + return 1 > +} > + > +check_process_samples() > +{ > + local output=$1 > + local comm > + > + # Expect each process to have a pause-event sample followed by at least > + # one branch entry in 0xFROM/0xTO/... form. Can we have an exact copy paste of the output instead of the text description? I still can't tell if what I'm seeing is expected based on this. For example I get this, which you could paste verbatim into the test: proc2 armv8_pmuv3_0/cycles,aux-action=pause,period=100003,call-graph=fp/u: ffff800080021440 ffff8000813a6a04 (This is output from V1, I couldn't run V2 because of the invalid group desc issue) > + for comm in proc1 proc2; do > + awk -v comm="$comm" ' > + $1 == comm && /cycles\/aux-action=pause/ { > + in_sample = 1 > + next > + } > + !NF { > + in_sample = 0 > + next > + } > + in_sample && /0x[[:xdigit:]]+\/0x[[:xdigit:]]+\// { > + found = 1 > + } > + END { exit !found } > + ' "$output" || { > + echo "No pause-event branch stack found for $comm" >&2 > + grep -A 4 "^$comm .*cycles/aux-action=pause" "$output" \ > + | head -n 20 >&2 || true > + return 1 > + } > + done > +} > + > +check_callchains() > +{ > + local output="$tmpdir/script-callchain" > + local comm > + > + if ! perf script -i "$tmpdir/data" -F comm,event,ip >"$output" \ > + 2>"$tmpdir/stderr"; then > + echo "Failed to dump pause-event callchains" >&2 > + cat "$tmpdir/stderr" >&2 > + return 1 > + fi > + > + # Expect a pause-event header for each process followed by at least two > + # indented instruction-pointer frames. Ditto > + for comm in proc1 proc2; do > + awk -v comm="$comm" ' > + $1 == comm && /cycles\/aux-action=pause/ { > + in_sample = 1 > + frames = 0 > + next > + } > + !NF { > + if (in_sample && frames >= 2) > + found = 1 > + in_sample = 0 > + next > + } > + in_sample && /^[[:space:]]+[[:xdigit:]]+([[:space:]]|$)/ { > + frames++ > + } > + END { > + if (in_sample && frames >= 2) > + found = 1 > + exit !found > + } > + ' "$output" || { > + echo "No multi-frame pause-event callchain found for $comm" >&2 > + grep -A 8 "^$comm .*cycles/aux-action=pause" "$output" \ > + | head -n 40 >&2 || true > + return 1 > + } > + done > +} > + > +check_branch_stacks() > +{ > + local output=$1 > + local max_entries=$2 > + > + local ret > + > + if awk -v max="$max_entries" ' > + /0x[[:xdigit:]]+\/0x[[:xdigit:]]+\// { > + entries = 0 > + for (i = 1; i <= NF; i++) > + if ($i ~ /^0x[[:xdigit:]]+\/0x[[:xdigit:]]+\//) > + entries++ > + if (entries) > + found = 1 > + if (entries > max) { > + status = 2 > + exit > + } > + } > + END { > + if (status) > + exit status > + if (!found) > + exit 1 > + } > + ' "$output"; then > + return 0 > + else > + ret=$? > + fi > + > + case $ret in > + 1) echo "No ETM branch stacks found" >&2 ;; > + 2) echo "Branch stack exceeds requested L$max_entries depth" >&2 ;; > + esac > + # Expected decoded pause-event lines contain at most L<n> branch entries. > + grep 'cycles/aux-action=pause' "$output" | head -n 5 >&2 || true > + return 1 > +} > + > +record_data > +check_callchains > + > +decode 4 "$tmpdir/script-L4" > +check_process_samples "$tmpdir/script-L4" > +check_branch_stacks "$tmpdir/script-L4" 4 > + > +decode 64 "$tmpdir/script-L64" > +check_process_samples "$tmpdir/script-L64" > +check_branch_stacks "$tmpdir/script-L64" 64 > + > +cleanup > +exit 0

3 weeks

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

CoreSight August 2026