| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Fri, 17 May 2024 12:07:10 +0200
From: James Clark <james.clark@....com>
To: Suzuki K Poulose <suzuki.poulose@....com>
Cc: Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
Maxime Coquelin <mcoquelin.stm32@...il.com>,
Alexandre Torgue <alexandre.torgue@...s.st.com>,
Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...hat.com>,
Arnaldo Carvalho de Melo <acme@...nel.org>,
Namhyung Kim <namhyung@...nel.org>, Mark Rutland <mark.rutland@....com>,
Jiri Olsa <jolsa@...nel.org>, Ian Rogers <irogers@...gle.com>,
Adrian Hunter <adrian.hunter@...el.com>, John Garry
<john.g.garry@...cle.com>, Will Deacon <will@...nel.org>,
Leo Yan <leo.yan@...ux.dev>, linux-arm-kernel@...ts.infradead.org,
linux-kernel@...r.kernel.org, linux-stm32@...md-mailman.stormreply.com,
linux-perf-users@...r.kernel.org, gankulkarni@...amperecomputing.com,
scclevenger@...amperecomputing.com, coresight@...ts.linaro.org,
mike.leach@...aro.org
Subject: Re: [PATCH 14/17] coresight: Use per-sink trace ID maps for Perf
sessions
On 07/05/2024 12:52, Suzuki K Poulose wrote:
> On 29/04/2024 16:22, James Clark wrote:
>> This will allow sessions with more than CORESIGHT_TRACE_IDS_MAX ETMs
>> as long as there are fewer than that many ETMs connected to each sink.
>>
>> Each sink owns its own trace ID map, and any Perf session connecting to
>> that sink will allocate from it, even if the sink is currently in use by
>> other users. This is similar to the existing behavior where the dynamic
>> trace IDs are constant as long as there is any concurrent Perf session
>> active. It's not completely optimal because slightly more IDs will be
>> used than necessary, but the optimal solution involves tracking the PIDs
>> of each session and allocating ID maps based on the session owner. This
>> is difficult to do with the combination of per-thread and per-cpu modes
>> and some scheduling issues. The complexity of this isn't likely to worth
>> it because even with multiple users they'd just see a difference in the
>> ordering of ID allocations rather than hitting any limits (unless the
>> hardware does have too many ETMs connected to one sink).
>>
>> Signed-off-by: James Clark <james.clark@....com>
>> ---
>> drivers/hwtracing/coresight/coresight-core.c | 10 ++++++++++
>> drivers/hwtracing/coresight/coresight-etm-perf.c | 15 ++++++++-------
>> include/linux/coresight.h | 1 +
>> 3 files changed, 19 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/hwtracing/coresight/coresight-core.c
>> b/drivers/hwtracing/coresight/coresight-core.c
>> index 9fc6f6b863e0..d1adff467670 100644
>> --- a/drivers/hwtracing/coresight/coresight-core.c
>> +++ b/drivers/hwtracing/coresight/coresight-core.c
>> @@ -902,6 +902,7 @@ static void coresight_device_release(struct device
>> *dev)
>> struct coresight_device *csdev = to_coresight_device(dev);
>> fwnode_handle_put(csdev->dev.fwnode);
>> + free_percpu(csdev->perf_id_map.cpu_map);
>> kfree(csdev);
>> }
>> @@ -1159,6 +1160,14 @@ struct coresight_device
>> *coresight_register(struct coresight_desc *desc)
>> csdev->dev.fwnode = fwnode_handle_get(dev_fwnode(desc->dev));
>> dev_set_name(&csdev->dev, "%s", desc->name);
>> + if (csdev->type == CORESIGHT_DEV_TYPE_SINK ||
>> + csdev->type == CORESIGHT_DEV_TYPE_LINKSINK) {
>> + csdev->perf_id_map.cpu_map = alloc_percpu(atomic_t);
>> + if (!csdev->perf_id_map.cpu_map) {
>> + ret = -ENOMEM;
>> + goto err_out;
>> + }
>> + }
>> /*
>> * Make sure the device registration and the connection fixup
>> * are synchronised, so that we don't see uninitialised devices
>> @@ -1216,6 +1225,7 @@ struct coresight_device
>> *coresight_register(struct coresight_desc *desc)
>> err_out:
>> /* Cleanup the connection information */
>> coresight_release_platform_data(NULL, desc->dev, desc->pdata);
>> + kfree(csdev);
>> return ERR_PTR(ret);
>> }
>> EXPORT_SYMBOL_GPL(coresight_register);
>> diff --git a/drivers/hwtracing/coresight/coresight-etm-perf.c
>> b/drivers/hwtracing/coresight/coresight-etm-perf.c
>> index 177cecae38d9..86ca1a9d09a7 100644
>> --- a/drivers/hwtracing/coresight/coresight-etm-perf.c
>> +++ b/drivers/hwtracing/coresight/coresight-etm-perf.c
>> @@ -229,10 +229,13 @@ static void free_event_data(struct work_struct
>> *work)
>> struct list_head **ppath;
>> ppath = etm_event_cpu_path_ptr(event_data, cpu);
>> - if (!(IS_ERR_OR_NULL(*ppath)))
>> + if (!(IS_ERR_OR_NULL(*ppath))) {
>> + struct coresight_device *sink = coresight_get_sink(*ppath);
>> +
>> + coresight_trace_id_put_cpu_id(cpu, &sink->perf_id_map);
>> coresight_release_path(*ppath);
>> + }
>> *ppath = NULL;
>> - coresight_trace_id_put_cpu_id(cpu,
>> coresight_trace_id_map_default());
>> }
>> /* mark perf event as done for trace id allocator */
>> @@ -401,8 +404,7 @@ static void *etm_setup_aux(struct perf_event
>> *event, void **pages,
>> }
>> /* ensure we can allocate a trace ID for this CPU */
>> - trace_id = coresight_trace_id_get_cpu_id(cpu,
>> - coresight_trace_id_map_default());
>> + trace_id = coresight_trace_id_get_cpu_id(cpu,
>> &sink->perf_id_map);
>
> We could either store the perf_id_map or the traceid itself in the
> event_data isn't it ? Rather than passing the idmap to enable_source ?
>
> Suzuki
>
Yes the end result would be the same. By doing it this way I was keeping
in mind the potential change for sysfs mode in the future. This way
there is common path between the two modes.
IMO an argument is easier to understand, rather than having to know
where/how/at what point the ID is initialised before calling
enable_source().
James
Powered by blists - more mailing lists