From: Athira Rajeev <atrajeev@linux.ibm.com>
To: acme@kernel.org, jolsa@kernel.org, adrian.hunter@intel.com,
maddy@linux.ibm.com, irogers@google.com, namhyung@kernel.org
Cc: linux-perf-users@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
atrajeev@linux.ibm.com, hbathini@linux.vnet.ibm.com,
tejas05@linux.ibm.com, tshah@linux.ibm.com,
venkat88@linux.ibm.com
Subject: [PATCH 3/9] tools/perf: Add arch_record__collect_final_data to collect additional data before closing the event
Date: Wed, 1 Jul 2026 14:11:09 +0530 [thread overview]
Message-ID: <20260701084115.80383-4-atrajeev@linux.ibm.com> (raw)
In-Reply-To: <20260701084115.80383-1-atrajeev@linux.ibm.com>
While collecting samples using "perf record", function
"__cmd_record" checks if monitoring is done. Once recording
is done, event list will be disabled using "evlist__disable".
After this, event fd won't be read and event will be removed.
Before removing the event, if any additional data needs
to be captured/written to perf.data, currently its not
possible. Introduce arch_record__collect_final_data to
collect additional data before closing the event
Signed-off-by: Athira Rajeev <atrajeev@linux.ibm.com>
---
tools/perf/builtin-record.c | 29 +++++++++++++++++++++++++++++
tools/perf/util/record.h | 4 ++++
2 files changed, 33 insertions(+)
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index ebd3ed0c9b3e..1312f7223455 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -2426,6 +2426,34 @@ static unsigned long record__waking(struct record *rec)
return waking;
}
+/*
+ * Weak symbol - architecture can override to indicate if more
+ * data needs to be collected before finishing output.
+ *
+ * Returns: 1 if more data exists, 0 if collection is complete
+ */
+__weak int arch_perf_record__need_read(struct evlist *evlist __maybe_unused)
+{
+ return 0; /* Default: no arch-specific data to collect */
+}
+
+static void record__final_data(struct record *rec)
+{
+ /*
+ * Collect any remaining architecture-specific data.
+ * The arch code checks if more data exists, and we do the actual
+ * reading here since we have access to record__mmap_read_all().
+ */
+ while (arch_perf_record__need_read(rec->evlist)) {
+ if (record__mmap_read_all(rec, false) < 0)
+ break;
+ /* Re-enable events for next batch */
+ evlist__enable(rec->evlist);
+ }
+
+ return;
+}
+
static int __cmd_record(struct record *rec, int argc, const char **argv)
{
int err;
@@ -2853,6 +2881,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
*/
if (done && !disabled && !target__none(&opts->target)) {
trigger_off(&auxtrace_snapshot_trigger);
+ record__final_data(rec);
evlist__disable(rec->evlist);
disabled = true;
}
diff --git a/tools/perf/util/record.h b/tools/perf/util/record.h
index 93627c9a7338..21f51efd36fc 100644
--- a/tools/perf/util/record.h
+++ b/tools/perf/util/record.h
@@ -8,6 +8,8 @@
#include <linux/stddef.h>
#include <linux/perf_event.h>
#include "util/target.h"
+#include "util/evlist.h"
+#include "util/util.h"
struct option;
@@ -95,4 +97,6 @@ static inline bool record_opts__no_switch_events(const struct record_opts *opts)
return opts->record_switch_events_set && !opts->record_switch_events;
}
+int arch_perf_record__need_read(struct evlist *evlist);
+
#endif // _PERF_RECORD_H
--
2.52.0
next prev parent reply other threads:[~2026-07-01 8:41 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-01 8:41 [PATCH 0/9] tools/perf: Add interface to expose HTM trace data via perf Athira Rajeev
2026-07-01 8:41 ` [PATCH 1/9] tool/perf: Move auxtrace_record__init for powerpc-vpadtl as separate utility Athira Rajeev
2026-07-01 8:56 ` sashiko-bot
2026-07-01 8:41 ` [PATCH 2/9] tools/perf: Add CONFIG_AUXTRACE support for HTM pmu on powerpc Athira Rajeev
2026-07-01 8:55 ` sashiko-bot
2026-07-01 8:41 ` Athira Rajeev [this message]
2026-07-01 8:54 ` [PATCH 3/9] tools/perf: Add arch_record__collect_final_data to collect additional data before closing the event sashiko-bot
2026-07-01 8:41 ` [PATCH 4/9] tools/perf: Add powerpc callback support for arch_record__collect_final_data Athira Rajeev
2026-07-01 8:55 ` sashiko-bot
2026-07-01 8:41 ` [PATCH 5/9] tools/perf: process htm auxtrace events and display in perf report -D Athira Rajeev
2026-07-01 9:05 ` sashiko-bot
2026-07-01 8:41 ` [PATCH 6/9] perf tools powerpc: Add HTM trace data processing and decoding support Athira Rajeev
2026-07-01 9:06 ` sashiko-bot
2026-07-01 8:41 ` [PATCH 7/9] perf tools powerpc: Add physical to logical address mapping for HTM traces Athira Rajeev
2026-07-01 9:07 ` sashiko-bot
2026-07-01 8:41 ` [PATCH 8/9] tools/perf/powerpc: Add event name as htm of PERF_TYPE_SYNTH type to present htm samples Athira Rajeev
2026-07-01 9:12 ` sashiko-bot
2026-07-01 8:41 ` [PATCH 9/9] tools/perf/powerpc: Add logical address in decoded traces Athira Rajeev
2026-07-01 9:13 ` sashiko-bot
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260701084115.80383-4-atrajeev@linux.ibm.com \
--to=atrajeev@linux.ibm.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=hbathini@linux.vnet.ibm.com \
--cc=irogers@google.com \
--cc=jolsa@kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=maddy@linux.ibm.com \
--cc=namhyung@kernel.org \
--cc=tejas05@linux.ibm.com \
--cc=tshah@linux.ibm.com \
--cc=venkat88@linux.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.