From: Zane Leung <liangzhen@linux.spacemit.com>
To: gregkh@linuxfoundation.org, palmer@dabbelt.com, pjw@kernel.org,
alexander.shishkin@linux.intel.com, suzuki.poulose@arm.com,
mchitale@gmail.com, anup@brainfault.org, irogers@google.com
Cc: mike.leach@arm.com, james.clark@linaro.org, leo.yan@arm.com,
peterz@infradead.org, mingo@redhat.com, namhyung@kernel.org,
mark.rutland@arm.com, jolsa@kernel.org, adrian.hunter@intel.com,
alex@ghiti.fr, atish.patra@linux.dev,
andrew.jones@oss.qualcomm.com, sunilvl@oss.qualcomm.com,
liangzhen@linux.spacemit.com, zhuangqiubin@linux.spacemit.com,
coresight@lists.linaro.org, linux-riscv@lists.infradead.org,
linux-kernel@vger.kernel.org
Subject: [PATCH 08/24] gtrace: add perf symlink support for gtrace components
Date: Fri, 23 Oct 2026 11:14:49 +0800 [thread overview]
Message-ID: <20261023031505.2020925-9-liangzhen@linux.spacemit.com> (raw)
In-Reply-To: <20261023031505.2020925-1-liangzhen@linux.spacemit.com>
Implement gtrace_perf_symlink() to let RISC-V Trace encoder
link themselves to the gtrace PMU directory as "cpuX" based
on their bound CPU.
---
drivers/hwtracing/gtrace/gtrace-perf.c | 29 +++++++++++++++++--
.../hwtracing/gtrace/rvtrace-encoder-core.c | 10 +++++++
include/linux/gtrace.h | 2 ++
3 files changed, 39 insertions(+), 2 deletions(-)
diff --git a/drivers/hwtracing/gtrace/gtrace-perf.c b/drivers/hwtracing/gtrace/gtrace-perf.c
index 2fcfbae9acf6..604b673a29d5 100644
--- a/drivers/hwtracing/gtrace/gtrace-perf.c
+++ b/drivers/hwtracing/gtrace/gtrace-perf.c
@@ -23,6 +23,7 @@
#define GTRACE_PMU_NAME "gtrace"
static struct pmu gtrace_pmu;
static struct workqueue_struct *gtrace_wq;
+static bool gtrace_perf_up;
/**
* struct gtrace_event_data - generic hardware trace perf event data
@@ -351,6 +352,27 @@ static const struct attribute_group *gtrace_pmu_attr_groups[] = {
NULL,
};
+int gtrace_perf_symlink(struct gtrace_component *comp, bool link)
+{
+ char entry[sizeof("cpu9999999")];
+ int ret = 0, cpu = comp->pdata->bound_cpu;
+ struct device *pmu_dev = gtrace_pmu.dev;
+ struct device *comp_dev = &comp->dev;
+
+ sprintf(entry, "cpu%d", cpu);
+
+ if (!gtrace_perf_up)
+ return -EPROBE_DEFER;
+
+ if (link)
+ ret = sysfs_create_link(&pmu_dev->kobj, &comp_dev->kobj, entry);
+ else
+ sysfs_remove_link(&pmu_dev->kobj, entry);
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(gtrace_perf_symlink);
+
int __init gtrace_perf_init(void)
{
int ret;
@@ -373,10 +395,13 @@ int __init gtrace_perf_init(void)
return -ENOMEM;
ret = perf_pmu_register(>race_pmu, GTRACE_PMU_NAME, -1);
- if (ret)
+ if (ret) {
destroy_workqueue(gtrace_wq);
+ return ret;
+ }
- return ret;
+ gtrace_perf_up = true;
+ return 0;
}
void __exit gtrace_perf_exit(void)
diff --git a/drivers/hwtracing/gtrace/rvtrace-encoder-core.c b/drivers/hwtracing/gtrace/rvtrace-encoder-core.c
index 817535ff9558..17a3dde1314c 100644
--- a/drivers/hwtracing/gtrace/rvtrace-encoder-core.c
+++ b/drivers/hwtracing/gtrace/rvtrace-encoder-core.c
@@ -129,6 +129,10 @@ static int rvtrace_encoder_probe(struct gtrace_component *comp)
}
dev_set_drvdata(&comp->dev, priv);
+ ret = gtrace_perf_symlink(comp, true);
+ if (ret)
+ return ret;
+
if (boot_enable) {
ret = gtrace_enable_sysfs(comp);
if (ret) {
@@ -140,6 +144,11 @@ static int rvtrace_encoder_probe(struct gtrace_component *comp)
return 0;
}
+static void rvtrace_encoder_remove(struct gtrace_component *comp)
+{
+ gtrace_perf_symlink(comp, false);
+}
+
static void rvtrace_encoder_parse_event_config(struct rvtrace_encoder_priv *priv,
struct perf_event *event)
{
@@ -240,6 +249,7 @@ static const struct gtrace_component_id rvtrace_encoder_ids[] = {
static struct gtrace_driver rvtrace_encoder_driver = {
.id_table = rvtrace_encoder_ids,
.probe = rvtrace_encoder_probe,
+ .remove = rvtrace_encoder_remove,
.start = rvtrace_encoder_start,
.stop = rvtrace_encoder_stop,
.driver = {
diff --git a/include/linux/gtrace.h b/include/linux/gtrace.h
index 3d769b405030..4bbe5ac835ec 100644
--- a/include/linux/gtrace.h
+++ b/include/linux/gtrace.h
@@ -437,4 +437,6 @@ int gtrace_disable_sysfs(struct gtrace_component *comp);
int gtrace_sysfs_add(struct gtrace_component *comp);
void gtrace_sysfs_remove(struct gtrace_component *comp);
+int gtrace_perf_symlink(struct gtrace_component *comp, bool link);
+
#endif /* __LINUX_GTRACE_H__ */
--
2.34.1
next prev parent reply other threads:[~2026-10-23 3:16 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-23 3:14 [RFC PATCH 00/24] Add Linux RISC-V trace support via the gtrace framework Zane Leung
2026-10-23 3:14 ` [PATCH 01/24] gtrace: Move connection release into component unregister Zane Leung
2026-10-23 3:14 ` [PATCH 02/24] gtrace: Initial implementation of gtrace sysfs framework Zane Leung
2026-10-23 3:14 ` [PATCH 03/24] gtrace: Drop trace format field and helper function Zane Leung
2026-10-23 3:14 ` [PATCH 04/24] gtrace: Add RISC-V trace encoder sysfs support Zane Leung
2026-10-23 3:14 ` [PATCH 05/24] gtrace: Introduce some PMU format attributes Zane Leung
2026-10-23 3:14 ` [PATCH 06/24] gtrace: Replace owner PID with perf_event pointer in gtrace_path Zane Leung
2026-10-23 3:14 ` [PATCH 07/24] gtrace: Parse RISC-V Trace encoder config from perf event Zane Leung
2026-10-23 3:14 ` Zane Leung [this message]
2026-10-23 3:14 ` [PATCH 09/24] gtrace: Add port-level start/stop driver callbacks Zane Leung
2026-10-23 3:14 ` [PATCH 10/24] gtrace: Add RISC-V trace funnel driver Zane Leung
2026-10-23 3:14 ` [PATCH 11/24] gtrace: Add RISC-V trace timestamp support Zane Leung
2026-10-23 3:14 ` [PATCH 12/24] gtrace: Add RISC-V trace ATB bridge driver Zane Leung
2026-10-23 3:14 ` [PATCH 13/24] gtrace: Add CoreSight platform core APIs Zane Leung
2026-10-23 3:14 ` [PATCH 14/24] gtrace: Add CoreSight funnel driver Zane Leung
2026-10-23 3:14 ` [PATCH 15/24] gtrace: Only copy AUX data from the path sink Zane Leung
2026-10-23 3:14 ` [PATCH 16/24] perf: Flag whether gtrace AUX data carries CoreSight framing Zane Leung
2026-10-23 3:14 ` [PATCH 17/24] gtrace: Unify default sink selection across CPUs in perf setup Zane Leung
2026-10-23 3:14 ` [PATCH 18/24] gtrace: Add sink-specific buffer alloc/free callbacks Zane Leung
2026-10-23 3:15 ` [PATCH 19/24] gtrace: Add setup_sysfs_buf callback to setup sysfs sink buffer Zane Leung
2026-10-23 3:15 ` [PATCH 20/24] gtrace: Add CoreSight TMC driver support Zane Leung
2026-10-23 3:15 ` [PATCH 21/24] perf tools: riscv: making gtrace PMU listable Zane Leung
2026-10-23 3:15 ` [PATCH 22/24] perf tools: Add Nexus RISC-V Trace decoder Zane Leung
2026-10-23 3:15 ` [PATCH 23/24] perf tools: gtrace: Record and parse trace encoder metadata Zane Leung
2026-10-23 3:15 ` [PATCH 24/24] perf tools: gtrace: Add AUX trace decoder and branch sample synthesis Zane Leung
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=20261023031505.2020925-9-liangzhen@linux.spacemit.com \
--to=liangzhen@linux.spacemit.com \
--cc=adrian.hunter@intel.com \
--cc=alex@ghiti.fr \
--cc=alexander.shishkin@linux.intel.com \
--cc=andrew.jones@oss.qualcomm.com \
--cc=anup@brainfault.org \
--cc=atish.patra@linux.dev \
--cc=coresight@lists.linaro.org \
--cc=gregkh@linuxfoundation.org \
--cc=irogers@google.com \
--cc=james.clark@linaro.org \
--cc=jolsa@kernel.org \
--cc=leo.yan@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-riscv@lists.infradead.org \
--cc=mark.rutland@arm.com \
--cc=mchitale@gmail.com \
--cc=mike.leach@arm.com \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=palmer@dabbelt.com \
--cc=peterz@infradead.org \
--cc=pjw@kernel.org \
--cc=sunilvl@oss.qualcomm.com \
--cc=suzuki.poulose@arm.com \
--cc=zhuangqiubin@linux.spacemit.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®