CoreSight

coresight@lists.linaro.org

6 participants
2346 discussions

Re: [PATCH v2 1/4] Coresight: Add trace_id function to collect trace ID

by Mike Leach

Hi, This patch has a number of issues:- 1) The new dynamic trace ID patchset to use per sink ID maps makes this set unusable. perf supplies a trace ID map for each sink used. - see https://lists.linaro.org/archives/list/coresight@lists.linaro.org/thread/JE… 2) See etm4_enable_perf() - in the perf context a locked version of the read trace ID cannot be used - therefore any path that calls etm4_read_alloc_trace_id() (or equivalent for other sources) in perf mode may result in lockdep issues. 3) on enable: given a cpu number, the trace ID can be read from the id maps rather than needing a new function in ops 4) on disable: trace id can be read directly from the source driver data - again removing need for a new trace_id function in ops. Regards Mike On Fri, 5 Jul 2024 at 10:01, Jie Gan <quic_jiegan(a)quicinc.com> wrote: > > Add 'trace_id' function pointer in ops. It's responsible for collect the > trace ID of the device. > > Add 'struct cs_sink_data' to store the data used by coresight_enable_path/ > coresight_disable_path. The structure will be transmitted to the helper and > sink device. > > Signed-off-by: Jie Gan <quic_jiegan(a)quicinc.com> > --- > drivers/hwtracing/coresight/coresight-core.c | 53 +++++++++++++++---- > drivers/hwtracing/coresight/coresight-etb10.c | 3 +- > .../hwtracing/coresight/coresight-etm-perf.c | 34 ++++++++++-- > .../coresight/coresight-etm3x-core.c | 14 +++++ > .../coresight/coresight-etm4x-core.c | 13 +++++ > drivers/hwtracing/coresight/coresight-priv.h | 12 ++++- > drivers/hwtracing/coresight/coresight-stm.c | 13 +++++ > drivers/hwtracing/coresight/coresight-sysfs.c | 24 +++++++-- > .../hwtracing/coresight/coresight-tmc-etf.c | 3 +- > .../hwtracing/coresight/coresight-tmc-etr.c | 6 ++- > drivers/hwtracing/coresight/coresight-tpda.c | 13 +++++ > drivers/hwtracing/coresight/coresight-trbe.c | 4 +- > drivers/hwtracing/coresight/ultrasoc-smb.c | 3 +- > include/linux/coresight.h | 4 ++ > 14 files changed, 174 insertions(+), 25 deletions(-) > > diff --git a/drivers/hwtracing/coresight/coresight-core.c b/drivers/hwtracing/coresight/coresight-core.c > index 9fc6f6b863e0..f414e66f4cda 100644 > --- a/drivers/hwtracing/coresight/coresight-core.c > +++ b/drivers/hwtracing/coresight/coresight-core.c > @@ -297,12 +297,12 @@ static int coresight_enable_helper(struct coresight_device *csdev, > return helper_ops(csdev)->enable(csdev, mode, data); > } > > -static void coresight_disable_helper(struct coresight_device *csdev) > +static void coresight_disable_helper(struct coresight_device *csdev, void *data) > { > - helper_ops(csdev)->disable(csdev, NULL); > + helper_ops(csdev)->disable(csdev, data); > } > > -static void coresight_disable_helpers(struct coresight_device *csdev) > +static void coresight_disable_helpers(struct coresight_device *csdev, void *data) > { > int i; > struct coresight_device *helper; > @@ -310,7 +310,7 @@ static void coresight_disable_helpers(struct coresight_device *csdev) > for (i = 0; i < csdev->pdata->nr_outconns; ++i) { > helper = csdev->pdata->out_conns[i]->dest_dev; > if (helper && coresight_is_helper(helper)) > - coresight_disable_helper(helper); > + coresight_disable_helper(helper, data); > } > } > > @@ -327,7 +327,7 @@ static void coresight_disable_helpers(struct coresight_device *csdev) > void coresight_disable_source(struct coresight_device *csdev, void *data) > { > source_ops(csdev)->disable(csdev, data); > - coresight_disable_helpers(csdev); > + coresight_disable_helpers(csdev, NULL); > } > EXPORT_SYMBOL_GPL(coresight_disable_source); > > @@ -337,7 +337,8 @@ EXPORT_SYMBOL_GPL(coresight_disable_source); > * disabled. > */ > static void coresight_disable_path_from(struct list_head *path, > - struct coresight_node *nd) > + struct coresight_node *nd, > + void *sink_data) > { > u32 type; > struct coresight_device *csdev, *parent, *child; > @@ -382,13 +383,13 @@ static void coresight_disable_path_from(struct list_head *path, > } > > /* Disable all helpers adjacent along the path last */ > - coresight_disable_helpers(csdev); > + coresight_disable_helpers(csdev, sink_data); > } > } > > -void coresight_disable_path(struct list_head *path) > +void coresight_disable_path(struct list_head *path, void *sink_data) > { > - coresight_disable_path_from(path, NULL); > + coresight_disable_path_from(path, NULL, sink_data); > } > EXPORT_SYMBOL_GPL(coresight_disable_path); > > @@ -468,10 +469,42 @@ int coresight_enable_path(struct list_head *path, enum cs_mode mode, > out: > return ret; > err: > - coresight_disable_path_from(path, nd); > + coresight_disable_path_from(path, nd, sink_data); > goto out; > } > > +int coresight_read_traceid(struct list_head *path) > +{ > + int trace_id, type; > + struct coresight_device *csdev; > + struct coresight_node *nd; > + > + list_for_each_entry(nd, path, link) { > + csdev = nd->csdev; > + type = csdev->type; > + > + switch(type) { > + case CORESIGHT_DEV_TYPE_SOURCE: > + if (source_ops(csdev)->trace_id != NULL) { > + trace_id = source_ops(csdev)->trace_id(csdev); > + if (trace_id > 0) > + return trace_id; > + } > + break; > + case CORESIGHT_DEV_TYPE_LINK: > + if (link_ops(csdev)->trace_id != NULL) { > + trace_id = link_ops(csdev)->trace_id(csdev); > + if (trace_id > 0) > + return trace_id; > + } > + break; > + default: > + break; > + } > + } > + return -EINVAL; > +} > + > struct coresight_device *coresight_get_sink(struct list_head *path) > { > struct coresight_device *csdev; > diff --git a/drivers/hwtracing/coresight/coresight-etb10.c b/drivers/hwtracing/coresight/coresight-etb10.c > index 7edd3f1d0d46..05e620529c14 100644 > --- a/drivers/hwtracing/coresight/coresight-etb10.c > +++ b/drivers/hwtracing/coresight/coresight-etb10.c > @@ -173,7 +173,8 @@ static int etb_enable_perf(struct coresight_device *csdev, void *data) > pid_t pid; > unsigned long flags; > struct etb_drvdata *drvdata = dev_get_drvdata(csdev->dev.parent); > - struct perf_output_handle *handle = data; > + struct cs_sink_data *sink_data = (struct cs_sink_data *)data; > + struct perf_output_handle *handle = sink_data->handle; > struct cs_buffers *buf = etm_perf_sink_config(handle); > > spin_lock_irqsave(&drvdata->spinlock, flags); > diff --git a/drivers/hwtracing/coresight/coresight-etm-perf.c b/drivers/hwtracing/coresight/coresight-etm-perf.c > index c0c60e6a1703..8b155765b959 100644 > --- a/drivers/hwtracing/coresight/coresight-etm-perf.c > +++ b/drivers/hwtracing/coresight/coresight-etm-perf.c > @@ -452,6 +452,7 @@ static void etm_event_start(struct perf_event *event, int flags) > struct perf_output_handle *handle = &ctxt->handle; > struct coresight_device *sink, *csdev = per_cpu(csdev_src, cpu); > struct list_head *path; > + struct cs_sink_data *sink_data = NULL; > u64 hw_id; > > if (!csdev) > @@ -490,9 +491,18 @@ static void etm_event_start(struct perf_event *event, int flags) > if (WARN_ON_ONCE(!sink)) > goto fail_end_stop; > > + sink_data = kzalloc(sizeof(*sink_data), GFP_KERNEL); > + if (!sink_data) > + goto fail_end_stop; > + > + sink_data->sink = sink; > + sink_data->traceid = coresight_read_traceid(path); > + sink_data->handle = handle; > /* Nothing will happen without a path */ > - if (coresight_enable_path(path, CS_MODE_PERF, handle)) > + if (coresight_enable_path(path, CS_MODE_PERF, sink_data)) { > + kfree(sink_data); > goto fail_end_stop; > + } > > /* Finally enable the tracer */ > if (source_ops(csdev)->enable(csdev, event, CS_MODE_PERF)) > @@ -511,6 +521,7 @@ static void etm_event_start(struct perf_event *event, int flags) > perf_report_aux_output_id(event, hw_id); > } > > + kfree(sink_data); > out: > /* Tell the perf core the event is alive */ > event->hw.state = 0; > @@ -519,7 +530,8 @@ static void etm_event_start(struct perf_event *event, int flags) > return; > > fail_disable_path: > - coresight_disable_path(path); > + coresight_disable_path(path, sink_data); > + kfree(sink_data); > fail_end_stop: > /* > * Check if the handle is still associated with the event, > @@ -544,6 +556,7 @@ static void etm_event_stop(struct perf_event *event, int mode) > struct perf_output_handle *handle = &ctxt->handle; > struct etm_event_data *event_data; > struct list_head *path; > + struct cs_sink_data *sink_data = NULL; > > /* > * If we still have access to the event_data via handle, > @@ -588,6 +601,10 @@ static void etm_event_stop(struct perf_event *event, int mode) > if (!sink) > return; > > + sink_data = kzalloc(sizeof(*sink_data), GFP_KERNEL); > + if (!sink_data) > + return; > + > /* stop tracer */ > coresight_disable_source(csdev, event); > > @@ -601,12 +618,16 @@ static void etm_event_stop(struct perf_event *event, int mode) > * have to do anything here. > */ > if (handle->event && (mode & PERF_EF_UPDATE)) { > - if (WARN_ON_ONCE(handle->event != event)) > + if (WARN_ON_ONCE(handle->event != event)) { > + kfree(sink_data); > return; > + } > > /* update trace information */ > - if (!sink_ops(sink)->update_buffer) > + if (!sink_ops(sink)->update_buffer) { > + kfree(sink_data); > return; > + } > > size = sink_ops(sink)->update_buffer(sink, handle, > event_data->snk_config); > @@ -627,8 +648,11 @@ static void etm_event_stop(struct perf_event *event, int mode) > WARN_ON(size); > } > > + sink_data->sink = sink; > + sink_data->traceid = coresight_read_traceid(path); > /* Disabling the path make its elements available to other sessions */ > - coresight_disable_path(path); > + coresight_disable_path(path, sink_data); > + kfree(sink_data); > } > > static int etm_event_add(struct perf_event *event, int mode) > diff --git a/drivers/hwtracing/coresight/coresight-etm3x-core.c b/drivers/hwtracing/coresight/coresight-etm3x-core.c > index 8b362605d242..27e973749050 100644 > --- a/drivers/hwtracing/coresight/coresight-etm3x-core.c > +++ b/drivers/hwtracing/coresight/coresight-etm3x-core.c > @@ -696,10 +696,24 @@ static void etm_disable(struct coresight_device *csdev, > coresight_set_mode(csdev, CS_MODE_DISABLED); > } > > +static int etm_trace_id(struct coresight_device *csdev) > +{ > + struct etm_drvdata *drvdata; > + > + if (csdev == NULL) > + return -EINVAL; > + > + drvdata = dev_get_drvdata(csdev->dev.parent); > + > + return etm_read_alloc_trace_id(drvdata); > +} > + > + > static const struct coresight_ops_source etm_source_ops = { > .cpu_id = etm_cpu_id, > .enable = etm_enable, > .disable = etm_disable, > + .trace_id = etm_trace_id, > }; > > static const struct coresight_ops etm_cs_ops = { > diff --git a/drivers/hwtracing/coresight/coresight-etm4x-core.c b/drivers/hwtracing/coresight/coresight-etm4x-core.c > index bf01f01964cf..8c3e9bfb9a9c 100644 > --- a/drivers/hwtracing/coresight/coresight-etm4x-core.c > +++ b/drivers/hwtracing/coresight/coresight-etm4x-core.c > @@ -1024,10 +1024,23 @@ static void etm4_disable(struct coresight_device *csdev, > coresight_set_mode(csdev, CS_MODE_DISABLED); > } > > +static int etm4_trace_id(struct coresight_device *csdev) > +{ > + struct etmv4_drvdata *drvdata; > + > + if (csdev == NULL) > + return -EINVAL; > + > + drvdata = dev_get_drvdata(csdev->dev.parent); > + > + return etm4_read_alloc_trace_id(drvdata); > +} > + > static const struct coresight_ops_source etm4_source_ops = { > .cpu_id = etm4_cpu_id, > .enable = etm4_enable, > .disable = etm4_disable, > + .trace_id = etm4_trace_id, > }; > > static const struct coresight_ops etm4_cs_ops = { > diff --git a/drivers/hwtracing/coresight/coresight-priv.h b/drivers/hwtracing/coresight/coresight-priv.h > index 61a46d3bdcc8..e2576531f796 100644 > --- a/drivers/hwtracing/coresight/coresight-priv.h > +++ b/drivers/hwtracing/coresight/coresight-priv.h > @@ -105,6 +105,15 @@ struct cs_buffers { > void **data_pages; > }; > > +/** > + * struct cs_sink_data - data used by coresight_enable_path/coresight_disable_path > + */ > +struct cs_sink_data { > + struct perf_output_handle *handle; > + struct coresight_device *sink; > + u32 traceid; > +}; > + > static inline void coresight_insert_barrier_packet(void *buf) > { > if (buf) > @@ -129,9 +138,10 @@ static inline void CS_UNLOCK(void __iomem *addr) > } while (0); > } > > -void coresight_disable_path(struct list_head *path); > +void coresight_disable_path(struct list_head *path, void *sink_data); > int coresight_enable_path(struct list_head *path, enum cs_mode mode, > void *sink_data); > +int coresight_read_traceid(struct list_head *path); > struct coresight_device *coresight_get_sink(struct list_head *path); > struct coresight_device *coresight_get_sink_by_id(u32 id); > struct coresight_device * > diff --git a/drivers/hwtracing/coresight/coresight-stm.c b/drivers/hwtracing/coresight/coresight-stm.c > index 117dbb484543..3817743fc0c6 100644 > --- a/drivers/hwtracing/coresight/coresight-stm.c > +++ b/drivers/hwtracing/coresight/coresight-stm.c > @@ -280,9 +280,22 @@ static void stm_disable(struct coresight_device *csdev, > } > } > > +static int stm_trace_id(struct coresight_device *csdev) > +{ > + struct stm_drvdata *drvdata; > + > + if (csdev == NULL) > + return -EINVAL; > + > + drvdata = dev_get_drvdata(csdev->dev.parent); > + > + return drvdata->traceid; > +} > + > static const struct coresight_ops_source stm_source_ops = { > .enable = stm_enable, > .disable = stm_disable, > + .trace_id = stm_trace_id, > }; > > static const struct coresight_ops stm_cs_ops = { > diff --git a/drivers/hwtracing/coresight/coresight-sysfs.c b/drivers/hwtracing/coresight/coresight-sysfs.c > index 1e67cc7758d7..a95afc890587 100644 > --- a/drivers/hwtracing/coresight/coresight-sysfs.c > +++ b/drivers/hwtracing/coresight/coresight-sysfs.c > @@ -167,6 +167,7 @@ int coresight_enable_sysfs(struct coresight_device *csdev) > int cpu, ret = 0; > struct coresight_device *sink; > struct list_head *path; > + struct cs_sink_data *sink_data; > enum coresight_dev_subtype_source subtype; > u32 hash; > > @@ -208,7 +209,14 @@ int coresight_enable_sysfs(struct coresight_device *csdev) > goto out; > } > > - ret = coresight_enable_path(path, CS_MODE_SYSFS, NULL); > + sink_data = kzalloc(sizeof(*sink_data), GFP_KERNEL); > + if (!sink_data) { > + ret = -ENOMEM; > + goto out; > + } > + sink_data->traceid = coresight_read_traceid(path); > + sink_data->sink = sink; > + ret = coresight_enable_path(path, CS_MODE_SYSFS, sink_data); > if (ret) > goto err_path; > > @@ -245,15 +253,17 @@ int coresight_enable_sysfs(struct coresight_device *csdev) > break; > } > > + kfree(sink_data); > out: > mutex_unlock(&coresight_mutex); > return ret; > > err_source: > - coresight_disable_path(path); > + coresight_disable_path(path, sink_data); > > err_path: > coresight_release_path(path); > + kfree(sink_data); > goto out; > } > EXPORT_SYMBOL_GPL(coresight_enable_sysfs); > @@ -262,6 +272,7 @@ void coresight_disable_sysfs(struct coresight_device *csdev) > { > int cpu, ret; > struct list_head *path = NULL; > + struct cs_sink_data *sink_data = NULL; > u32 hash; > > mutex_lock(&coresight_mutex); > @@ -273,6 +284,10 @@ void coresight_disable_sysfs(struct coresight_device *csdev) > if (!coresight_disable_source_sysfs(csdev, NULL)) > goto out; > > + sink_data = kzalloc(sizeof(*sink_data), GFP_KERNEL); > + if (!sink_data) > + goto out; > + > switch (csdev->subtype.source_subtype) { > case CORESIGHT_DEV_SUBTYPE_SOURCE_PROC: > cpu = source_ops(csdev)->cpu_id(csdev); > @@ -296,8 +311,11 @@ void coresight_disable_sysfs(struct coresight_device *csdev) > break; > } > > - coresight_disable_path(path); > + sink_data->sink = coresight_find_activated_sysfs_sink(csdev); > + sink_data->traceid = coresight_read_traceid(path); > + coresight_disable_path(path, sink_data); > coresight_release_path(path); > + kfree(sink_data); > > out: > mutex_unlock(&coresight_mutex); > diff --git a/drivers/hwtracing/coresight/coresight-tmc-etf.c b/drivers/hwtracing/coresight/coresight-tmc-etf.c > index d4f641cd9de6..7dc536eba3e2 100644 > --- a/drivers/hwtracing/coresight/coresight-tmc-etf.c > +++ b/drivers/hwtracing/coresight/coresight-tmc-etf.c > @@ -250,7 +250,8 @@ static int tmc_enable_etf_sink_perf(struct coresight_device *csdev, void *data) > pid_t pid; > unsigned long flags; > struct tmc_drvdata *drvdata = dev_get_drvdata(csdev->dev.parent); > - struct perf_output_handle *handle = data; > + struct cs_sink_data *sink_data = (struct cs_sink_data *)data; > + struct perf_output_handle *handle = sink_data->handle; > struct cs_buffers *buf = etm_perf_sink_config(handle); > > spin_lock_irqsave(&drvdata->spinlock, flags); > diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c > index e75428fa1592..0c24520645e2 100644 > --- a/drivers/hwtracing/coresight/coresight-tmc-etr.c > +++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c > @@ -1253,7 +1253,8 @@ static int tmc_enable_etr_sink_sysfs(struct coresight_device *csdev) > struct etr_buf *tmc_etr_get_buffer(struct coresight_device *csdev, > enum cs_mode mode, void *data) > { > - struct perf_output_handle *handle = data; > + struct cs_sink_data *sink_data = (struct cs_sink_data *)data; > + struct perf_output_handle *handle = sink_data->handle; > struct etr_perf_buffer *etr_perf; > > switch (mode) { > @@ -1647,7 +1648,8 @@ static int tmc_enable_etr_sink_perf(struct coresight_device *csdev, void *data) > pid_t pid; > unsigned long flags; > struct tmc_drvdata *drvdata = dev_get_drvdata(csdev->dev.parent); > - struct perf_output_handle *handle = data; > + struct cs_sink_data *sink_data = (struct cs_sink_data *)data; > + struct perf_output_handle *handle = sink_data->handle; > struct etr_perf_buffer *etr_perf = etm_perf_sink_config(handle); > > spin_lock_irqsave(&drvdata->spinlock, flags); > diff --git a/drivers/hwtracing/coresight/coresight-tpda.c b/drivers/hwtracing/coresight/coresight-tpda.c > index bfca103f9f84..20f0ab73159c 100644 > --- a/drivers/hwtracing/coresight/coresight-tpda.c > +++ b/drivers/hwtracing/coresight/coresight-tpda.c > @@ -232,9 +232,22 @@ static void tpda_disable(struct coresight_device *csdev, > dev_dbg(drvdata->dev, "TPDA inport %d disabled\n", in->dest_port); > } > > +static int tpda_trace_id(struct coresight_device *csdev) > +{ > + struct tpda_drvdata *drvdata; > + > + if (csdev == NULL) > + return -EINVAL; > + > + drvdata = dev_get_drvdata(csdev->dev.parent); > + > + return drvdata->atid; > +} > + > static const struct coresight_ops_link tpda_link_ops = { > .enable = tpda_enable, > .disable = tpda_disable, > + .trace_id = tpda_trace_id, > }; > > static const struct coresight_ops tpda_cs_ops = { > diff --git a/drivers/hwtracing/coresight/coresight-trbe.c b/drivers/hwtracing/coresight/coresight-trbe.c > index 96a32b213669..7f4560b067a8 100644 > --- a/drivers/hwtracing/coresight/coresight-trbe.c > +++ b/drivers/hwtracing/coresight/coresight-trbe.c > @@ -21,6 +21,7 @@ > > #include "coresight-self-hosted-trace.h" > #include "coresight-trbe.h" > +#include "coresight-priv.h" > > #define PERF_IDX2OFF(idx, buf) ((idx) % ((buf)->nr_pages << PAGE_SHIFT)) > > @@ -1012,7 +1013,8 @@ static int arm_trbe_enable(struct coresight_device *csdev, enum cs_mode mode, > { > struct trbe_drvdata *drvdata = dev_get_drvdata(csdev->dev.parent); > struct trbe_cpudata *cpudata = dev_get_drvdata(&csdev->dev); > - struct perf_output_handle *handle = data; > + struct cs_sink_data *sink_data = (struct cs_sink_data *)data; > + struct perf_output_handle *handle = sink_data->handle; > struct trbe_buf *buf = etm_perf_sink_config(handle); > > WARN_ON(cpudata->cpu != smp_processor_id()); > diff --git a/drivers/hwtracing/coresight/ultrasoc-smb.c b/drivers/hwtracing/coresight/ultrasoc-smb.c > index f9ebf20c91e6..92d8a9fb844e 100644 > --- a/drivers/hwtracing/coresight/ultrasoc-smb.c > +++ b/drivers/hwtracing/coresight/ultrasoc-smb.c > @@ -217,7 +217,8 @@ static void smb_enable_sysfs(struct coresight_device *csdev) > static int smb_enable_perf(struct coresight_device *csdev, void *data) > { > struct smb_drv_data *drvdata = dev_get_drvdata(csdev->dev.parent); > - struct perf_output_handle *handle = data; > + struct cs_sink_data *sink_data = (struct cs_sink_data *)data; > + struct perf_output_handle *handle = sink_data->handle; > struct cs_buffers *buf = etm_perf_sink_config(handle); > pid_t pid; > > diff --git a/include/linux/coresight.h b/include/linux/coresight.h > index f09ace92176e..fb1c225076a5 100644 > --- a/include/linux/coresight.h > +++ b/include/linux/coresight.h > @@ -344,6 +344,7 @@ struct coresight_ops_sink { > * Operations available for links. > * @enable: enables flow between iport and oport. > * @disable: disables flow between iport and oport. > + * @trace_id: Collect the traceid. > */ > struct coresight_ops_link { > int (*enable)(struct coresight_device *csdev, > @@ -352,6 +353,7 @@ struct coresight_ops_link { > void (*disable)(struct coresight_device *csdev, > struct coresight_connection *in, > struct coresight_connection *out); > + int (*trace_id)(struct coresight_device *csdev); > }; > > /** > @@ -361,6 +363,7 @@ struct coresight_ops_link { > * is associated to. > * @enable: enables tracing for a source. > * @disable: disables tracing for a source. > + * @trace_id: collect the traceid. > */ > struct coresight_ops_source { > int (*cpu_id)(struct coresight_device *csdev); > @@ -368,6 +371,7 @@ struct coresight_ops_source { > enum cs_mode mode); > void (*disable)(struct coresight_device *csdev, > struct perf_event *event); > + int (*trace_id)(struct coresight_device *csdev); > }; > > /** > -- > 2.34.1 > -- Mike Leach Principal Engineer, ARM Ltd. Manchester Design Centre. UK

9 months, 3 weeks

[PATCH v3 0/3] coresight: Add preferred trace id support

by Mao Jinlong

Some HW has static trace id which cannot be changed via software programming. For this case, configure the trace id in device tree with "arm,trace-id = <xxx>", and call coresight_trace_id_get_system_id with the trace id value in device probe function. The id will be reserved for the HW all the time if the device is probed. Changes since V2: 1. Change "trace-id" to "arm,trace-id". 2. Add trace id flag for getting preferred id or ODD id. Changes since V1: 1. Add argument to coresight_trace_id_get_system_id for preferred id instead of adding new function coresight_trace_id_reserve_system_id. 2. Add constraint to trace-id in dt-binding file. Mao Jinlong (3): dt-bindings: arm: Add arm,trace-id for coresight dummy source coresight: Add support to get preferred id for system trace sources coresight: dummy: Add reserve atid support for dummy source .../sysfs-bus-coresight-devices-dummy-source | 15 +++++ .../arm/arm,coresight-dummy-source.yaml | 6 ++ drivers/hwtracing/coresight/coresight-dummy.c | 59 +++++++++++++++++-- .../hwtracing/coresight/coresight-platform.c | 25 ++++++++ drivers/hwtracing/coresight/coresight-stm.c | 2 +- drivers/hwtracing/coresight/coresight-tpda.c | 2 +- .../hwtracing/coresight/coresight-trace-id.c | 35 +++++++---- .../hwtracing/coresight/coresight-trace-id.h | 11 +++- include/linux/coresight.h | 1 + 9 files changed, 137 insertions(+), 19 deletions(-) create mode 100644 Documentation/ABI/testing/sysfs-bus-coresight-devices-dummy-source -- 2.41.0

9 months, 3 weeks

Re: [PATCH] perf scripts python arm-cs-trace-disasm.py: Skip disasm if address continuity is broken

by James Clark

On 19/07/2024 10:26 am, Ganapatrao Kulkarni wrote: > To generate the instruction tracing, script uses 2 contiguous packets > address range. If there a continuity brake due to discontiguous branch > address, it is required to reset the tracing and start tracing with the > new set of contiguous packets. > > Adding change to identify the break and complete the remaining tracing > of current packets and restart tracing from new set of packets, if > continuity is established. > Hi Ganapatrao, Can you add a before and after example of what's changed to the commit message? It wasn't immediately obvious to me if this is adding missing output, or it was correcting the tail end of the output that was previously wrong. > Signed-off-by: Ganapatrao Kulkarni <gankulkarni(a)os.amperecomputing.com> > --- > tools/perf/scripts/python/arm-cs-trace-disasm.py | 10 ++++++++++ > 1 file changed, 10 insertions(+) > > diff --git a/tools/perf/scripts/python/arm-cs-trace-disasm.py b/tools/perf/scripts/python/arm-cs-trace-disasm.py > index d973c2baed1c..ad10cee2c35e 100755 > --- a/tools/perf/scripts/python/arm-cs-trace-disasm.py > +++ b/tools/perf/scripts/python/arm-cs-trace-disasm.py > @@ -198,6 +198,10 @@ def process_event(param_dict): > cpu_data[str(cpu) + 'addr'] = addr > return > > + if (cpu_data.get(str(cpu) + 'ip') == None): > + cpu_data[str(cpu) + 'ip'] = ip > + Do you need to write into the global cpu_data here? Doesn't it get overwritten after you load it back into 'prev_ip' prev_ip = cpu_data[str(cpu) + 'ip'] ... then ... # Record for previous sample packet cpu_data[str(cpu) + 'addr'] = addr cpu_data[str(cpu) + 'ip'] = stop_addr Would a local variable not accomplish the same thing? > + prev_ip = cpu_data[str(cpu) + 'ip'] > > if (options.verbose == True): > print("Event type: %s" % name) > @@ -243,12 +247,18 @@ def process_event(param_dict): > > # Record for previous sample packet > cpu_data[str(cpu) + 'addr'] = addr > + cpu_data[str(cpu) + 'ip'] = stop_addr > > # Handle CS_ETM_TRACE_ON packet if start_addr=0 and stop_addr=4 > if (start_addr == 0 and stop_addr == 4): > print("CPU%d: CS_ETM_TRACE_ON packet is inserted" % cpu) > return > > + if (stop_addr < start_addr): > + # Continuity of the Packets broken, set start_addr to previous > + # packet ip to complete the remaining tracing of the address range. > + start_addr = prev_ip > + > if (start_addr < int(dso_start) or start_addr > int(dso_end)): > print("Start address 0x%x is out of range [ 0x%x .. 0x%x ] for dso %s" % (start_addr, int(dso_start), int(dso_end), dso)) > return

9 months, 4 weeks

[PATCH v5 00/17] coresight: Use per-sink trace ID maps for Perf sessions

by James Clark

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). Per-thread mode works but only until there are any overlapping IDs, at which point Perf will error out. Both per-thread mode and sysfs mode are left to future changes, but both can be added on top of this initial implementation and only sysfs mode requires further driver changes. The HW_ID version field hasn't been bumped in order to not break Perf which already has an error condition for other values of that field. Instead a new minor version has been added which signifies that there are new fields but the old fields are backwards compatible. Changes since v4: * Fix compilation failure when TRACE_ID_DEBUG is set * Expand comment about not freeing individual trace IDs in free_event_data() Changes since v3: * Fix issue where trace IDs were overwritten by possibly invalid ones by Perf in unformatted mode. Now the HW_IDs are also used for unformatted mode unless the kernel didn't emit any. * Add a commit to check the OpenCSD version. * Add a commit to not save invalid IDs in the Perf header. * Replace cs_etm_queue's formatted and formatted_set members with a single enum which is easier to use. * Drop CORESIGHT_TRACE_ID_UNUSED_FLAG as it's no longer needed. * Add a commit to print the queue number in the raw dump. * Don't assert on the number of unformatted decoders if decoders == 0. Changes since v2: * Rebase on coresight-next 6.10-rc2 (b9b25c8496). * Fix double free of csdev if device registration fails. * Fix leak of coresight_trace_id_perf_start() if trace ID allocation fails. * Don't resend HW_ID for sink changes in per-thread mode. The existing CPU field on AUX records can be used to track this instead. * Tidy function doc for coresight_trace_id_release_all() * Drop first two commits now that they are in coresight-next * Add a commit to make the trace ID spinlock local to the map Changes since V1: * Rename coresight_device.perf_id_map to perf_sink_id_map. * Instead of outputting a HW_ID for each reachable ETM, output the sink ID and continue to output only the HW_ID once for each mapping. * Keep the first two Perf patches so that it applies cleanly on coresight-next, although they have been applied on perf-tools-next * Add new *_map() functions to the trace ID public API instead of modifying existing ones. * Collapse "coresight: Pass trace ID map into source enable" into "coresight: Use per-sink trace ID maps for Perf sessions" because the first commit relied on the default map being accessible which is no longer necessary due to the previous bullet point. James Clark (17): perf: cs-etm: Create decoders after both AUX and HW_ID search passes perf: cs-etm: Allocate queues for all CPUs perf: cs-etm: Move traceid_list to each queue perf: cs-etm: Create decoders based on the trace ID mappings perf: cs-etm: Only save valid trace IDs into files perf: cs-etm: Support version 0.1 of HW_ID packets perf: cs-etm: Print queue number in raw trace dump perf: cs-etm: Add runtime version check for OpenCSD coresight: Remove unused ETM Perf stubs coresight: Clarify comments around the PID of the sink owner coresight: Move struct coresight_trace_id_map to common header coresight: Expose map arguments in trace ID API coresight: Make CPU id map a property of a trace ID map coresight: Use per-sink trace ID maps for Perf sessions coresight: Remove pending trace ID release mechanism coresight: Emit sink ID in the HW_ID packets coresight: Make trace ID map spinlock local to the map drivers/hwtracing/coresight/coresight-core.c | 37 +- drivers/hwtracing/coresight/coresight-dummy.c | 3 +- .../hwtracing/coresight/coresight-etm-perf.c | 43 +- .../hwtracing/coresight/coresight-etm-perf.h | 18 - .../coresight/coresight-etm3x-core.c | 9 +- .../coresight/coresight-etm4x-core.c | 9 +- drivers/hwtracing/coresight/coresight-priv.h | 1 + drivers/hwtracing/coresight/coresight-stm.c | 3 +- drivers/hwtracing/coresight/coresight-sysfs.c | 3 +- .../hwtracing/coresight/coresight-tmc-etr.c | 5 +- drivers/hwtracing/coresight/coresight-tmc.h | 5 +- drivers/hwtracing/coresight/coresight-tpdm.c | 3 +- .../hwtracing/coresight/coresight-trace-id.c | 138 ++-- .../hwtracing/coresight/coresight-trace-id.h | 70 +- include/linux/coresight-pmu.h | 17 +- include/linux/coresight.h | 21 +- tools/build/feature/test-libopencsd.c | 4 +- tools/include/linux/coresight-pmu.h | 17 +- tools/perf/Makefile.config | 2 +- tools/perf/arch/arm/util/cs-etm.c | 11 +- .../perf/util/cs-etm-decoder/cs-etm-decoder.c | 49 +- .../perf/util/cs-etm-decoder/cs-etm-decoder.h | 3 +- .../util/cs-etm-decoder/cs-etm-min-version.h | 13 + tools/perf/util/cs-etm.c | 625 +++++++++++------- tools/perf/util/cs-etm.h | 12 +- 25 files changed, 645 insertions(+), 476 deletions(-) create mode 100644 tools/perf/util/cs-etm-decoder/cs-etm-min-version.h -- 2.34.1

9 months, 4 weeks

Re: [PATCH] kselftest/arm64: Add coresight test

by James Clark

+cc linux-kselftest(a)vger.kernel.org On 19/07/2024 11:21 am, Jeremy Szu wrote: > > > James Clark 於 7/17/24 10:48 PM 寫道: >> >> >> On 16/07/2024 3:06 am, Jeremy Szu wrote: >>> Hi James, >>> >>> Thank you for your reply. >>> >>> James Clark 於 7/12/24 5:01 PM 寫道: >>>> >>>> >>>> On 11/07/2024 7:03 pm, Catalin Marinas wrote: >>>>> On Wed, Jul 10, 2024 at 02:27:32PM +0800, Jeremy Szu wrote: >>>>>> Add a script to test the coresight functionalities by performing the >>>>>> perf test 'coresight'. >>>>>> >>>>>> The script checks the prerequisites and launches a bundle of >>>>>> Coresight-related tests with a 180-second timeout. >>>>>> >>>> >>>> Hi Jeremy, >>>> >>>> On the whole I'm not sure running the Perf tests under kself tests >>>> is a good fit. We're already running all the Perf tests in various >>>> CIs, so this is going to duplicate effort. Especially with setup and >>>> parsing of the results output. >>>> >>>> There is also no clean line between what's a kernel issue and whats >>>> a Perf issue when these fail. >>>> >>>> And thirdly why only run the Coresight tests? Most of the Perf tests >>>> test some part of the kernel, so if we were going to do this I think >>>> it would make sense to make some kind of proper harness and run them >>>> all. I have some recollection that someone said it might be >>>> something we could do, but I can't remember the discussion. >>> >>> The idea I'm trying to pursue is to use arm64 kselftest to run as >>> many test cases as possible for ARM SoCs across different designs and >>> distros. I believe it could provide an alert if there is an issue, >>> whether it originates from userspace or kernel, similar to how perf >>> is used in other categories. >>> >>> I'm not sure if all perf tests could be counted in soc >>> (selftests/arm64) category such as some tests may target to storage, >>> memory or devices. I >> >> Could we not put the Perf tests in .../selftests/perf.sh, then it >> doesn't really matter which subsystem they're targeting and we can run >> all the Perf tests? >> > > The .../sefltests/ seems for the kselftest framework only, not sure if > having a new .../selftests/perf will make more sense? > Yeah that sounds better, but it probably requires changing the title of the patch to "[RFC] kselftest: Run Perf tests under kselftest" or something like that. >>> could replace 'arm64/coresight' with 'arm64/perf' if it makes more >>> sense. I believe it could help users verify functionality more >>> conveniently. >>> >>>> >>>> Ignoring the main issue above I've left some comments about this >>>> patch inline below: >>>> >>>>>> Signed-off-by: Jeremy Szu <jszu(a)nvidia.com> >>>>> >>>>> I have not idea how to test coresight, so adding Suzuki as well. >>>>> >>>>>> --- >>>>>> tools/testing/selftests/arm64/Makefile | 2 +- >>>>>> .../selftests/arm64/coresight/Makefile | 5 +++ >>>>>> .../selftests/arm64/coresight/coresight.sh | 40 >>>>>> +++++++++++++++++++ >>>>>> .../selftests/arm64/coresight/settings | 1 + >>>>>> 4 files changed, 47 insertions(+), 1 deletion(-) >>>>>> create mode 100644 tools/testing/selftests/arm64/coresight/Makefile >>>>>> create mode 100755 >>>>>> tools/testing/selftests/arm64/coresight/coresight.sh >>>>>> create mode 100644 tools/testing/selftests/arm64/coresight/settings >>>>>> >>>>>> diff --git a/tools/testing/selftests/arm64/Makefile >>>>>> b/tools/testing/selftests/arm64/Makefile >>>>>> index 28b93cab8c0dd..2b788d7bab22d 100644 >>>>>> --- a/tools/testing/selftests/arm64/Makefile >>>>>> +++ b/tools/testing/selftests/arm64/Makefile >>>>>> @@ -4,7 +4,7 @@ >>>>>> ARCH ?= $(shell uname -m 2>/dev/null || echo not) >>>>>> ifneq (,$(filter $(ARCH),aarch64 arm64)) >>>>>> -ARM64_SUBTARGETS ?= tags signal pauth fp mte bti abi >>>>>> +ARM64_SUBTARGETS ?= tags signal pauth fp mte bti abi coresight >>>>>> else >>>>>> ARM64_SUBTARGETS := >>>>>> endif >>>>>> diff --git a/tools/testing/selftests/arm64/coresight/Makefile >>>>>> b/tools/testing/selftests/arm64/coresight/Makefile >>>>>> new file mode 100644 >>>>>> index 0000000000000..1cc8c1f2a997e >>>>>> --- /dev/null >>>>>> +++ b/tools/testing/selftests/arm64/coresight/Makefile >>>>>> @@ -0,0 +1,5 @@ >>>>>> +# SPDX-License-Identifier: GPL-2.0 >>>>>> + >>>>>> +TEST_PROGS := coresight.sh >>>>>> + >>>>>> +include ../../lib.mk >>>>>> diff --git a/tools/testing/selftests/arm64/coresight/coresight.sh >>>>>> b/tools/testing/selftests/arm64/coresight/coresight.sh >>>>>> new file mode 100755 >>>>>> index 0000000000000..e550957cf593b >>>>>> --- /dev/null >>>>>> +++ b/tools/testing/selftests/arm64/coresight/coresight.sh >>>>>> @@ -0,0 +1,40 @@ >>>>>> +#!/bin/bash >>>>>> +# SPDX-License-Identifier: GPL-2.0 >>>>>> + >>>>>> +skip() { >>>>>> + echo "SKIP: $1" >>>>>> + exit 4 >>>>>> +} >>>>>> + >>>>>> +fail() { >>>>>> + echo "FAIL: $1" >>>>>> + exit 255 >>>>>> +} >>>>>> + >>>>>> +is_coresight_supported() { >>>>>> + if [ -d "/sys/bus/coresight/devices" ]; then >>>>>> + return 0 >>>>>> + fi >>>>>> + return 255 >>>>>> +} >>>> >>>> The Perf coresight tests already have a skip mechanism built in so >>>> can we rely on that instead of duplicating it here? There are also >>>> other scenarios for skipping like Perf not linked with OpenCSD which >>>> aren't covered here. >>>> >>> >>> Will it return the different error code to indicate if it's failed to >>> be executed or the coresight is not supported? >>> >> >> I think the exit code is only used for more serious errors. For things >> like missing tests, SKIP and FAIL it still exits with 0 but you have >> to grep for the strings. Actually for a missing test it prints nothing >> and exits 0. >> >>>>> + >>>>>> +if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then >>>>>> + [ "$(id -u)" -ne 0 ] && \ >>>>>> + skip "this test must be run as root." >>>>>> + which perf >/dev/null 2>&1 || \ >>>>>> + skip "perf is not installed." >>>>>> + perf test list 2>&1 | grep -qi 'coresight' || \ >>>>>> + skip "perf doesn't support testing coresight." >>>> >>>> Can this be an error instead? The coresight tests were added in 5.10 >>>> and I don't think new kselftest needs to support such old kernels. >>>> This skip risks turning an error with installing the tests into a >>>> silent failure. >>>> >>>> Also as far as I know a lot of distros will refuse to open Perf >>>> unless it matches the kernel version if it was installed from the >>>> package manager, so we don't need to worry about old versions. >>>> >>> >>> The idea to skip it here is because I thought either a distro >>> (custom) kernel doesn't enable the kconfig or the distro built the >>> perf with CORESIGHT=0. >>> >>> I could make it as an error and put it after "is_coresight_support" >>> if it makes more sense. >>> >> >> But the Coresight test already checks these things, which is my point. >> You can grep for "SKIP" which it will print for any case where the >> coresight test can't be run due to some missing config. >> > > Oh, I guess you mean to rely on something like the `perf list cs_etm | > grep -q cs_etm || exit 2`. Yes, that makes more sense. > If we're leaning towards running all the Perf tests then none of this is required anymore. This is another example of why it's better to run them all. We can't realistically hard code a kself test with loads of logic for each Perf subtest when Perf will happily run all the tests on it's own without any extra work. >>>>>> + is_coresight_supported || \ >>>>>> + skip "coresight is not supported." >>>>>> + >>>>>> + cmd_output=$(perf test -vv 'coresight' 2>&1) >>>>>> + perf_ret=$? >>>>>> + >>>>>> + if [ $perf_ret -ne 0 ]; then >>>>>> + fail "perf command returns non-zero." >>>>>> + elif [[ $cmd_output == *"FAILED!"* ]]; then >>>>>> + echo $cmd_output >>>> >>>> It's probably helpful to print cmd_output in both failure cases. >>>> >>> >>> ok, will do. >>> >>>>>> + fail "perf test 'arm coresight' test failed!" > > I think I should remove the `is_coresight_supported` there and checks > the output as a "else" to see if the test is PASS or SKIP. Does it make > sense to you? > Same as above applies, the only thing that's really required is parsing for "FAILED". >>>>>> + fi >>>>>> +fi >>>>>> diff --git a/tools/testing/selftests/arm64/coresight/settings >>>>>> b/tools/testing/selftests/arm64/coresight/settings >>>>>> new file mode 100644 >>>>>> index 0000000000000..a953c96aa16e1 >>>>>> --- /dev/null >>>>>> +++ b/tools/testing/selftests/arm64/coresight/settings >>>>>> @@ -0,0 +1 @@ >>>>>> +timeout=180 >>>> >>>> I timed 331 seconds on n1sdp, and probably even longer on Juno. >>>> >>>> It doesn't need to run for this long and it's an issue with the >>>> tests, but currently that's how long it takes so the timeout needs >>>> to be longer. >>>> >>> >>> ok, will extend it to 600. >>> >>>>>> -- >>>>>> 2.34.1 >>>>>

9 months, 4 weeks

Re: [PATCH] kselftest/arm64: Add coresight test

by James Clark

On 16/07/2024 3:06 am, Jeremy Szu wrote: > Hi James, > > Thank you for your reply. > > James Clark 於 7/12/24 5:01 PM 寫道: >> >> >> On 11/07/2024 7:03 pm, Catalin Marinas wrote: >>> On Wed, Jul 10, 2024 at 02:27:32PM +0800, Jeremy Szu wrote: >>>> Add a script to test the coresight functionalities by performing the >>>> perf test 'coresight'. >>>> >>>> The script checks the prerequisites and launches a bundle of >>>> Coresight-related tests with a 180-second timeout. >>>> >> >> Hi Jeremy, >> >> On the whole I'm not sure running the Perf tests under kself tests is >> a good fit. We're already running all the Perf tests in various CIs, >> so this is going to duplicate effort. Especially with setup and >> parsing of the results output. >> >> There is also no clean line between what's a kernel issue and whats a >> Perf issue when these fail. >> >> And thirdly why only run the Coresight tests? Most of the Perf tests >> test some part of the kernel, so if we were going to do this I think >> it would make sense to make some kind of proper harness and run them >> all. I have some recollection that someone said it might be something >> we could do, but I can't remember the discussion. > > The idea I'm trying to pursue is to use arm64 kselftest to run as many > test cases as possible for ARM SoCs across different designs and > distros. I believe it could provide an alert if there is an issue, > whether it originates from userspace or kernel, similar to how perf is > used in other categories. > > I'm not sure if all perf tests could be counted in soc (selftests/arm64) > category such as some tests may target to storage, memory or devices. I Could we not put the Perf tests in .../selftests/perf.sh, then it doesn't really matter which subsystem they're targeting and we can run all the Perf tests? > could replace 'arm64/coresight' with 'arm64/perf' if it makes more > sense. I believe it could help users verify functionality more > conveniently. > >> >> Ignoring the main issue above I've left some comments about this patch >> inline below: >> >>>> Signed-off-by: Jeremy Szu <jszu(a)nvidia.com> >>> >>> I have not idea how to test coresight, so adding Suzuki as well. >>> >>>> --- >>>> tools/testing/selftests/arm64/Makefile | 2 +- >>>> .../selftests/arm64/coresight/Makefile | 5 +++ >>>> .../selftests/arm64/coresight/coresight.sh | 40 >>>> +++++++++++++++++++ >>>> .../selftests/arm64/coresight/settings | 1 + >>>> 4 files changed, 47 insertions(+), 1 deletion(-) >>>> create mode 100644 tools/testing/selftests/arm64/coresight/Makefile >>>> create mode 100755 >>>> tools/testing/selftests/arm64/coresight/coresight.sh >>>> create mode 100644 tools/testing/selftests/arm64/coresight/settings >>>> >>>> diff --git a/tools/testing/selftests/arm64/Makefile >>>> b/tools/testing/selftests/arm64/Makefile >>>> index 28b93cab8c0dd..2b788d7bab22d 100644 >>>> --- a/tools/testing/selftests/arm64/Makefile >>>> +++ b/tools/testing/selftests/arm64/Makefile >>>> @@ -4,7 +4,7 @@ >>>> ARCH ?= $(shell uname -m 2>/dev/null || echo not) >>>> ifneq (,$(filter $(ARCH),aarch64 arm64)) >>>> -ARM64_SUBTARGETS ?= tags signal pauth fp mte bti abi >>>> +ARM64_SUBTARGETS ?= tags signal pauth fp mte bti abi coresight >>>> else >>>> ARM64_SUBTARGETS := >>>> endif >>>> diff --git a/tools/testing/selftests/arm64/coresight/Makefile >>>> b/tools/testing/selftests/arm64/coresight/Makefile >>>> new file mode 100644 >>>> index 0000000000000..1cc8c1f2a997e >>>> --- /dev/null >>>> +++ b/tools/testing/selftests/arm64/coresight/Makefile >>>> @@ -0,0 +1,5 @@ >>>> +# SPDX-License-Identifier: GPL-2.0 >>>> + >>>> +TEST_PROGS := coresight.sh >>>> + >>>> +include ../../lib.mk >>>> diff --git a/tools/testing/selftests/arm64/coresight/coresight.sh >>>> b/tools/testing/selftests/arm64/coresight/coresight.sh >>>> new file mode 100755 >>>> index 0000000000000..e550957cf593b >>>> --- /dev/null >>>> +++ b/tools/testing/selftests/arm64/coresight/coresight.sh >>>> @@ -0,0 +1,40 @@ >>>> +#!/bin/bash >>>> +# SPDX-License-Identifier: GPL-2.0 >>>> + >>>> +skip() { >>>> + echo "SKIP: $1" >>>> + exit 4 >>>> +} >>>> + >>>> +fail() { >>>> + echo "FAIL: $1" >>>> + exit 255 >>>> +} >>>> + >>>> +is_coresight_supported() { >>>> + if [ -d "/sys/bus/coresight/devices" ]; then >>>> + return 0 >>>> + fi >>>> + return 255 >>>> +} >> >> The Perf coresight tests already have a skip mechanism built in so can >> we rely on that instead of duplicating it here? There are also other >> scenarios for skipping like Perf not linked with OpenCSD which aren't >> covered here. >> > > Will it return the different error code to indicate if it's failed to be > executed or the coresight is not supported? > I think the exit code is only used for more serious errors. For things like missing tests, SKIP and FAIL it still exits with 0 but you have to grep for the strings. Actually for a missing test it prints nothing and exits 0. >>>> + >>>> +if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then >>>> + [ "$(id -u)" -ne 0 ] && \ >>>> + skip "this test must be run as root." >>>> + which perf >/dev/null 2>&1 || \ >>>> + skip "perf is not installed." >>>> + perf test list 2>&1 | grep -qi 'coresight' || \ >>>> + skip "perf doesn't support testing coresight." >> >> Can this be an error instead? The coresight tests were added in 5.10 >> and I don't think new kselftest needs to support such old kernels. >> This skip risks turning an error with installing the tests into a >> silent failure. >> >> Also as far as I know a lot of distros will refuse to open Perf unless >> it matches the kernel version if it was installed from the package >> manager, so we don't need to worry about old versions. >> > > The idea to skip it here is because I thought either a distro (custom) > kernel doesn't enable the kconfig or the distro built the perf with > CORESIGHT=0. > > I could make it as an error and put it after "is_coresight_support" if > it makes more sense. > But the Coresight test already checks these things, which is my point. You can grep for "SKIP" which it will print for any case where the coresight test can't be run due to some missing config. >>>> + is_coresight_supported || \ >>>> + skip "coresight is not supported." >>>> + >>>> + cmd_output=$(perf test -vv 'coresight' 2>&1) >>>> + perf_ret=$? >>>> + >>>> + if [ $perf_ret -ne 0 ]; then >>>> + fail "perf command returns non-zero." >>>> + elif [[ $cmd_output == *"FAILED!"* ]]; then >>>> + echo $cmd_output >> >> It's probably helpful to print cmd_output in both failure cases. >> > > ok, will do. > >>>> + fail "perf test 'arm coresight' test failed!" >>>> + fi >>>> +fi >>>> diff --git a/tools/testing/selftests/arm64/coresight/settings >>>> b/tools/testing/selftests/arm64/coresight/settings >>>> new file mode 100644 >>>> index 0000000000000..a953c96aa16e1 >>>> --- /dev/null >>>> +++ b/tools/testing/selftests/arm64/coresight/settings >>>> @@ -0,0 +1 @@ >>>> +timeout=180 >> >> I timed 331 seconds on n1sdp, and probably even longer on Juno. >> >> It doesn't need to run for this long and it's an issue with the tests, >> but currently that's how long it takes so the timeout needs to be longer. >> > > ok, will extend it to 600. > >>>> -- >>>> 2.34.1 >>>

10 months

Re: [PATCH] kselftest/arm64: Add coresight test

by James Clark

On 11/07/2024 7:03 pm, Catalin Marinas wrote: > On Wed, Jul 10, 2024 at 02:27:32PM +0800, Jeremy Szu wrote: >> Add a script to test the coresight functionalities by performing the >> perf test 'coresight'. >> >> The script checks the prerequisites and launches a bundle of >> Coresight-related tests with a 180-second timeout. >> Hi Jeremy, On the whole I'm not sure running the Perf tests under kself tests is a good fit. We're already running all the Perf tests in various CIs, so this is going to duplicate effort. Especially with setup and parsing of the results output. There is also no clean line between what's a kernel issue and whats a Perf issue when these fail. And thirdly why only run the Coresight tests? Most of the Perf tests test some part of the kernel, so if we were going to do this I think it would make sense to make some kind of proper harness and run them all. I have some recollection that someone said it might be something we could do, but I can't remember the discussion. Ignoring the main issue above I've left some comments about this patch inline below: >> Signed-off-by: Jeremy Szu <jszu(a)nvidia.com> > > I have not idea how to test coresight, so adding Suzuki as well. > >> --- >> tools/testing/selftests/arm64/Makefile | 2 +- >> .../selftests/arm64/coresight/Makefile | 5 +++ >> .../selftests/arm64/coresight/coresight.sh | 40 +++++++++++++++++++ >> .../selftests/arm64/coresight/settings | 1 + >> 4 files changed, 47 insertions(+), 1 deletion(-) >> create mode 100644 tools/testing/selftests/arm64/coresight/Makefile >> create mode 100755 tools/testing/selftests/arm64/coresight/coresight.sh >> create mode 100644 tools/testing/selftests/arm64/coresight/settings >> >> diff --git a/tools/testing/selftests/arm64/Makefile b/tools/testing/selftests/arm64/Makefile >> index 28b93cab8c0dd..2b788d7bab22d 100644 >> --- a/tools/testing/selftests/arm64/Makefile >> +++ b/tools/testing/selftests/arm64/Makefile >> @@ -4,7 +4,7 @@ >> ARCH ?= $(shell uname -m 2>/dev/null || echo not) >> >> ifneq (,$(filter $(ARCH),aarch64 arm64)) >> -ARM64_SUBTARGETS ?= tags signal pauth fp mte bti abi >> +ARM64_SUBTARGETS ?= tags signal pauth fp mte bti abi coresight >> else >> ARM64_SUBTARGETS := >> endif >> diff --git a/tools/testing/selftests/arm64/coresight/Makefile b/tools/testing/selftests/arm64/coresight/Makefile >> new file mode 100644 >> index 0000000000000..1cc8c1f2a997e >> --- /dev/null >> +++ b/tools/testing/selftests/arm64/coresight/Makefile >> @@ -0,0 +1,5 @@ >> +# SPDX-License-Identifier: GPL-2.0 >> + >> +TEST_PROGS := coresight.sh >> + >> +include ../../lib.mk >> diff --git a/tools/testing/selftests/arm64/coresight/coresight.sh b/tools/testing/selftests/arm64/coresight/coresight.sh >> new file mode 100755 >> index 0000000000000..e550957cf593b >> --- /dev/null >> +++ b/tools/testing/selftests/arm64/coresight/coresight.sh >> @@ -0,0 +1,40 @@ >> +#!/bin/bash >> +# SPDX-License-Identifier: GPL-2.0 >> + >> +skip() { >> + echo "SKIP: $1" >> + exit 4 >> +} >> + >> +fail() { >> + echo "FAIL: $1" >> + exit 255 >> +} >> + >> +is_coresight_supported() { >> + if [ -d "/sys/bus/coresight/devices" ]; then >> + return 0 >> + fi >> + return 255 >> +} The Perf coresight tests already have a skip mechanism built in so can we rely on that instead of duplicating it here? There are also other scenarios for skipping like Perf not linked with OpenCSD which aren't covered here. >> + >> +if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then >> + [ "$(id -u)" -ne 0 ] && \ >> + skip "this test must be run as root." >> + which perf >/dev/null 2>&1 || \ >> + skip "perf is not installed." >> + perf test list 2>&1 | grep -qi 'coresight' || \ >> + skip "perf doesn't support testing coresight." Can this be an error instead? The coresight tests were added in 5.10 and I don't think new kselftest needs to support such old kernels. This skip risks turning an error with installing the tests into a silent failure. Also as far as I know a lot of distros will refuse to open Perf unless it matches the kernel version if it was installed from the package manager, so we don't need to worry about old versions. >> + is_coresight_supported || \ >> + skip "coresight is not supported." >> + >> + cmd_output=$(perf test -vv 'coresight' 2>&1) >> + perf_ret=$? >> + >> + if [ $perf_ret -ne 0 ]; then >> + fail "perf command returns non-zero." >> + elif [[ $cmd_output == *"FAILED!"* ]]; then >> + echo $cmd_output It's probably helpful to print cmd_output in both failure cases. >> + fail "perf test 'arm coresight' test failed!" >> + fi >> +fi >> diff --git a/tools/testing/selftests/arm64/coresight/settings b/tools/testing/selftests/arm64/coresight/settings >> new file mode 100644 >> index 0000000000000..a953c96aa16e1 >> --- /dev/null >> +++ b/tools/testing/selftests/arm64/coresight/settings >> @@ -0,0 +1 @@ >> +timeout=180 I timed 331 seconds on n1sdp, and probably even longer on Juno. It doesn't need to run for this long and it's an issue with the tests, but currently that's how long it takes so the timeout needs to be longer. >> -- >> 2.34.1 >

10 months

Re: [PATCH v2 2/4] dt-bindings: arm: Add binding document for Coresight Control Unit device.

by Suzuki K Poulose

On 11/07/2024 09:36, JieGan wrote: > On Tue, Jul 09, 2024 at 02:00:25PM +0800, JieGan wrote: >> On Mon, Jul 08, 2024 at 01:50:11PM +0100, Suzuki K Poulose wrote: >>> On 08/07/2024 11:25, JieGan wrote: >>>> On Mon, Jul 08, 2024 at 06:10:28PM +0800, JieGan wrote: >>>>> On Mon, Jul 08, 2024 at 10:41:55AM +0100, Suzuki K Poulose wrote: >>>>>> On 05/07/2024 10:00, Jie Gan wrote: >>>>>>> Add binding document for Coresight Control Unit device. >>>>>> >>>>>> nit: This is again too generic ? corsight-tmc-control-unit ? After all >>>>>> thats what it is and not a *generic* coresight control unit ? >>>>>> >>>>> coresight-tmc-control-unit is much better. We will check it. >>>>>>> >>>>>>> Signed-off-by: Jie Gan <quic_jiegan(a)quicinc.com> >>>>>>> --- >>>>>>> .../bindings/arm/qcom,coresight-ccu.yaml | 87 +++++++++++++++++++ >>>>>>> 1 file changed, 87 insertions(+) >>>>>>> create mode 100644 Documentation/devicetree/bindings/arm/qcom,coresight-ccu.yaml >>>>>>> >>>>>>> diff --git a/Documentation/devicetree/bindings/arm/qcom,coresight-ccu.yaml b/Documentation/devicetree/bindings/arm/qcom,coresight-ccu.yaml >>>>>>> new file mode 100644 >>>>>>> index 000000000000..9bb8ced393a7 >>>>>>> --- /dev/null >>>>>>> +++ b/Documentation/devicetree/bindings/arm/qcom,coresight-ccu.yaml >>>>>>> @@ -0,0 +1,87 @@ >>>>>>> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) >>>>>>> +%YAML 1.2 >>>>>>> +--- >>>>>>> +$id: http://devicetree.org/schemas/arm/qcom,coresight-ccu.yaml# >>>>>>> +$schema: http://devicetree.org/meta-schemas/core.yaml# >>>>>>> + >>>>>>> +title: CoreSight Control Unit >>>>>>> + >>>>>>> +maintainers: >>>>>>> + - Yuanfang Zhang <quic_yuanfang(a)quicinc.com> >>>>>>> + - Mao Jinlong <quic_jinlmao(a)quicinc.com> >>>>>>> + - Jie Gan <quic_jiegan(a)quicinc.com> >>>>>>> + >>>>>>> +description: >>>>>>> + The Coresight Control unit controls various Coresight behaviors. >>>>>>> + Used to enable/disable ETR’s data filter function based on trace ID. >>>>>>> + >>>>>>> +properties: >>>>>>> + compatible: >>>>>>> + const: qcom,coresight-ccu >>>>>>> + >>>>>>> + reg: >>>>>>> + maxItems: 1 >>>>>>> + >>>>>>> + clocks: >>>>>>> + maxItems: 1 >>>>>>> + >>>>>>> + clock-names: >>>>>>> + items: >>>>>>> + - const: apb_pclk >>>>>>> + >>>>>>> + reg-names: >>>>>>> + items: >>>>>>> + - const: ccu-base >>>>>>> + >>>>>>> + in-ports: >>>>>>> + $ref: /schemas/graph.yaml#/properties/ports >>>>>>> + >>>>>>> + unevaluatedProperties: >>>>>>> + patternProperties: >>>>>>> + '^port(@[0-7])?$': >>>>>>> + description: Input connections from CoreSight Trace bus >>>>>>> + $ref: /schemas/graph.yaml#/properties/port >>>>>>> + >>>>>>> + properties: >>>>>>> + qcom,ccu-atid-offset: >>>>>> >>>>>> Why do we need this atid offset ? Couldn't this be mapped to the "port" >>>>>> number ? >>>>>> >>>>>> e.g, input-port 0 on CCU => Offset x >>>>>> input-port 1 on CCU => (Offset x + Size of 1 region) >>>>> If the first ATID offset remains constant, it appears to be feasible. >>>>> We will consider the possibility of this solution. >>>> We just checked the ATID offset varies across different hardware platforms. >>>> It defined as 0xf4 on some platforms, and some others defined as 0xf8. >>> >>> What do you mean ? The offset where you apply the filter changes across >>> different platforms ? or different "tmc-control-unit" implementations ? >>> Is this discoverable from the device ? We could use different >>> compatibles for different "types" of the "devices". Simply adding >>> something in the DT is not the right way. >> >> I got your point here. I believe we should consult our hardware engineers first to check it. >> We need to figure out the design of ATID offset from hardware perspective. Then we can >> propose an alternative approach, such as associating the offset with a compitable value, >> to resolve this issue. >> >>> >>>> >>>> So I think it should be better to define it in device tree node. >>> >>> No. See above. > > > Hi Suzuki > > There is a new solution for CCU device. We would like to separate one CCU device into several helper > devices, each responsible for one feature of the CCU device. > > For the data filter feature, we will define the address of the AITD Register that included by CCU in DT > as base address of the helper node. So the driver code can easily program the register field with the base address. > With this solution, we may need define several helper nodes in DT file to support different features for CCU and each > helper device needs a driver to support its behavior. > > for example, ATID register for ETR0 with base address 0x10000f8: (tmp name used, TDFU for tmc data filter unit) That looks like a hack to me and don't prefer that and there would be multiple mappings for a single device and that could complicate device resource accounting. > > TDFU@10000f8 { > ... > } > > ATID register for ETR1 with base address 0x1000108: > TDFU@1000108 { > ... > } > > The CCU device supports various features and the data filter feature for ETR is one of those features. How to support > those features with one helper_enable function is a serious challenge. That's why we would like to separate it. > Meanwhile, This solution can also resolve the offset issue. > > We are looking forward your opinions with this proposal. We need to know what other functionalities are supported and how that will be used ? In the worst case, you could define register bases for each port in the CCU device node. TDFU@.. { reg = <base-address 4K> <I don't know the standard property name for> = List of { <port number>, <Offset from base>, } } Suzuki

10 months

[PATCH v1 0/3] Add source filtering for multi-port output

by Tao Zhang

In our hardware design, by combining a funnel and a replicator, it implement a hardware device with one-to-one correspondence between output ports and input ports. The programming usage on this device is the same as funnel. The software uses a funnel and a static replicator to implement the driver of this device. Since original funnels only support a single output connection and original replicator only support a single input connection, the code needs to be modified to support this new feature. The following is a typical topology diagram of multi-port output mechanism. |----------| |---------| |----------| |---------| | TPDM 0 | | Source0 | | Source 1 | | TPDM 1 | |----------| |---------| |----------| |---------| | | | | | | | | | --------- | | | | | | | | | | | | | | | \-------------/ ---------------------- | \ Funnel 0 / | | ----------- | ------------------------------ | | | | | | \------------------/ \ Funnel 1 / ----| \--------------/ | | |----> Combine a funnel and a | | static replicator /-----------------\ | / replicator 0 \ ----| /---------------------\ | | | | | |-----------| | |---------| | | |TPDM0 |TPDM1 | \-----------------/ | \ TPDA 0 / | \-------------/ | | | | |Source0/1 | \-------------------------------/ \ Funnel 2 / \---------------------------/ Changes in V1: 1. Add a static replicator connect to a funnel to implement the correspondence between the output ports and the input ports on funnels. -- Suzuki K Poulose 2. Add filter_src_dev and filter_src_dev phandle to "coresight_connection" struct, and populate them if there is one. -- Suzuki K Poulose 3. To look at the phandle and then fixup/remove the filter_src device in fixup/remove connections. -- Suzuki K Poulose 4. When TPDA reads DSB/CMB element size, it is implemented by looking up filter src device in the connections. -- Suzuki K Poulose Tao Zhang (3): dt-bindings: arm: qcom,coresight-static-replicator: Add property for source filtering coresight: Add source filtering for multi-port output coresight-tpda: Optimize the function of reading element size .../arm/arm,coresight-static-replicator.yaml | 18 +++- drivers/hwtracing/coresight/coresight-core.c | 89 ++++++++++++++++--- .../hwtracing/coresight/coresight-platform.c | 13 +++ drivers/hwtracing/coresight/coresight-tpda.c | 6 +- include/linux/coresight.h | 5 ++ 5 files changed, 115 insertions(+), 16 deletions(-) -- 2.17.1

10 months

[PATCH v9 0/7] Coresight for Kernel panic and watchdog reset

by Linu Cherian

This patch series is rebased on coresight-next-v6.10. Changelog from v8: * Added missing exit path on error in __tmc_probe. * Few whitespace fixes, checkpatch fixes. * With perf sessions honouring stop_on_flush sysfs attribute, removed redundant variable stop_on_flush_en. Changelog from v7: * Fixed breakage on perf test -vvvv "arm coresight". No issues seen with and without "resrv" buffer mode * Moved the crashdev registration into a seperate function. * Removed redundant variable in tmc_etr_setup_crashdata_buf * Avoided a redundant memcpy in tmc_panic_sync_etf. * Tested kernel panic with trace session started uisng perf. Please see the title "Perf based testing" below for details. For this, stop_on_flush sysfs attribute is taken into consideration while starting perf sessions as well. Changelog from v6: * Added special device files for reading crashdata, so that read_prevboot mode flag is removed. * Added new sysfs TMC device attribute, stop_on_flush. Stop on flush trigger event is disabled by default. User need to explicitly enable this from sysfs for panic stop to work. * Address parameter for panicstop ETM configuration is chosen as kernel "panic" address by default. * Added missing tmc_wait_for_tmcready during panic handling * Few other misc code rearrangements. Changelog from v5: * Fixed issues reported by CONFIG_DEBUG_ATOMIC_SLEEP * Fixed a memory leak while reading data from /dev/tmc_etrx in READ_PREVBOOT mode * Tested reading trace data from crashdump kernel Changelog from v4: * Device tree binding - Description is made more explicit on the usage of reserved memory region - Mismatch in memory region names in dts binding and driver fixed - Removed "mem" suffix from the memory region names * Rename "struct tmc_register_snapshot" -> "struct tmc_crash_metadata", since it contains more than register snapshot. Related variables are named accordingly. * Rename struct tmc_drvdata members resrv_buf -> crash_tbuf metadata -> crash_mdata * Size field in metadata refers to RSZ register and hence indicates the size in 32 bit words. ETR metadata follows this convention, the same has been extended to ETF metadata as well. * Added crc32 for more robust metadata and tracedata validation. * Added/modified dev_dbg messages during metadata validation * Fixed a typo in patch 5 commit description Changelog from v3: * Converted the Coresight ETM driver change to a named configuration. RFC tag has been removed with this change. * Fixed yaml issues reported by "make dt_binding_check" * Added names for reserved memory regions 0 and 1 * Added prevalidation checks for metadata processing * Fixed a regression introduced in RFC v3 - TMC Status register was getting saved wrongly * Reverted memremap attribute changes from _WB to _WC to match with the dma map attributes * Introduced reserved buffer mode specific .sync op. This fixes a possible crash when reserved buffer mode was used in normal trace capture, due to unwanted dma maintenance operations. v8 is posted here: https://lore.kernel.org/lkml/20240531042745.494222-4-lcherian@marvell.com/T/ Using Coresight for Kernel panic and Watchdog reset =================================================== This patch series is about extending Linux coresight driver support to address kernel panic and watchdog reset scenarios. This would help coresight users to debug kernel panic and watchdog reset using coresight trace data. Coresight trace capture: Kernel panic ------------------------------------- From the coresight driver point of view, addressing the kernel panic situation has four main requirements. a. Support for allocation of trace buffer pages from reserved memory area. Platform can advertise this using a new device tree property added to relevant coresight nodes. b. Support for stopping coresight blocks at the time of panic c. Saving required metadata in the specified format d. Support for reading trace data captured at the time of panic Allocation of trace buffer pages from reserved RAM ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ A new optional device tree property "memory-region" is added to the ETR/ETF device nodes, that would give the base address and size of trace buffer. Static allocation of trace buffers would ensure that both IOMMU enabled and disabled cases are handled. Also, platforms that support persistent RAM will allow users to read trace data in the subsequent boot without booting the crashdump kernel. Note: For ETR sink devices, this reserved region will be used for both trace capture and trace data retrieval. For ETF sink devices, internal SRAM would be used for trace capture, and they would be synced to reserved region for retrieval. Note: Patches 1 & 2 adds support for this. Disabling coresight blocks at the time of panic ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In order to avoid the situation of losing relevant trace data after a kernel panic, it would be desirable to stop the coresight blocks at the time of panic. This can be achieved by configuring the comparator, CTI and sink devices as below, Comparator(triggers on kernel panic) --->External out --->CTI -- | ETR/ETF stop <------External In <-------------- Note: * Patch 6 provides the necessary ETR configuration. * Patch 7 provides the necessary ETM configuration. Saving metadata at the time of kernel panic ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Coresight metadata involves all additional data that are required for a successful trace decode in addition to the trace data. This involves ETR/ETF, ETE register snapshot etc. A new optional device property "memory-region" is added to the ETR/ETF/ETE device nodes for this. Note: Patches 3 & 4 adds support for this. Reading trace data captured at the time of panic ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Trace data captured at the time of panic, can be read from rebooted kernel or from crashdump kernel using the below mentioned interface. Note: Patch 5 adds support for this. Steps for reading trace data captured in previous boot ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1. cd /sys/bus/coresight/devices/tmc_etrXX/ 2. Dump trace buffer crashdata to a file, #dd if=/dev/crash_tmc_etrXX of=~/cstrace.bin General flow of trace capture and decode incase of kernel panic ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1. Enable source and sink on all the cores using the sysfs interface. ETR sink will have trace buffers allocated from reserved memory, by selecting "resrv" buffer mode from sysfs. 2. Run relevant tests. 3. On a kernel panic, all coresight blocks are disabled, necessary metadata is synced by kernel panic handler. System would eventually reboot or boot a crashdump kernel. 4. For platforms that supports crashdump kernel, raw trace data can be dumped using the coresight sysfs interface from the crashdump kernel itself. Persistent RAM is not a requirement in this case. 5. For platforms that supports persistent RAM, trace data can be dumped using the coresight sysfs interface in the subsequent Linux boot. Crashdump kernel is not a requirement in this case. Persistent RAM ensures that trace data is intact across reboot. Coresight trace capture: Watchdog reset --------------------------------------- The main difference between addressing the watchdog reset and kernel panic case are below, a. Saving coresight metadata need to be taken care by the SCP(system control processor) firmware in the specified format, instead of kernel. b. Reserved memory region given by firmware for trace buffer and metadata has to be in persistent RAM. Note: This is a requirement for watchdog reset case but optional in kernel panic case. Watchdog reset can be supported only on platforms that meet the above two requirements. Testing Kernel panic on Linux 6.8 --------------------------------- 1. Enable the preloaded ETM configuration #echo 1 > /sys/kernel/config/cs-syscfg/configurations/panicstop/enable 2. Configure CTI using sysfs interface #./cti_setup.sh #cat cti_setup.sh cd /sys/bus/coresight/devices/ ap_cti_config () { #ETM trig out[0] trigger to Channel 0 echo 0 4 > channels/trigin_attach } etf_cti_config () { #ETF Flush in trigger from Channel 0 echo 0 1 > channels/trigout_attach echo 1 > channels/trig_filter_enable } etr_cti_config () { #ETR Flush in from Channel 0 echo 0 1 > channels/trigout_attach echo 1 > channels/trig_filter_enable } ctidevs=`find . -name "cti*"` for i in $ctidevs do cd $i connection=`find . -name "ete*"` if [ ! -z "$connection" ] then echo "AP CTI config for $i" ap_cti_config fi connection=`find . -name "tmc_etf*"` if [ ! -z "$connection" ] then echo "ETF CTI config for $i" etf_cti_config fi connection=`find . -name "tmc_etr*"` if [ ! -z "$connection" ] then echo "ETR CTI config for $i" etr_cti_config fi cd .. done Note: CTI connections are SOC specific and hence the above script is added just for reference. 3. Choose reserved buffer mode for ETR buffer #echo "resrv" > /sys/bus/coresight/devices/tmc_etr0/buf_mode_preferred 4. Enable stop on flush trigger configuration #echo 1 > /sys/bus/coresight/devices/tmc_etr0/stop_on_flush 4. Start Coresight tracing on cores 1 and 2 using sysfs interface 5. Run some application on core 1 #taskset -c 1 dd if=/dev/urandom of=/dev/null & 6. Invoke kernel panic on core 2 #echo 1 > /proc/sys/kernel/panic #taskset -c 2 echo c > /proc/sysrq-trigger 7. From rebooted kernel or crashdump kernel, read crashdata Note: For crashdump kernel option, please make sure "crash_kexec_post_notifiers" is added to the kernel bootargs. #dd if=/dev/crash_tmc_etr0 of=/trace/cstrace.bin 8. Run opencsd decoder tools/scripts to generate the instruction trace. Sample Core 1 instruction trace dump: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ A etm4_enable_hw: ffff800008ae1dd4 CONTEXT EL2 etm4_enable_hw: ffff800008ae1dd4 I etm4_enable_hw: ffff800008ae1dd4: d503201f nop I etm4_enable_hw: ffff800008ae1dd8: d503201f nop I etm4_enable_hw: ffff800008ae1ddc: d503201f nop I etm4_enable_hw: ffff800008ae1de0: d503201f nop I etm4_enable_hw: ffff800008ae1de4: d503201f nop I etm4_enable_hw: ffff800008ae1de8: d503233f paciasp I etm4_enable_hw: ffff800008ae1dec: a9be7bfd stp x29, x30, [sp, #-32]! I etm4_enable_hw: ffff800008ae1df0: 910003fd mov x29, sp I etm4_enable_hw: ffff800008ae1df4: a90153f3 stp x19, x20, [sp, #16] I etm4_enable_hw: ffff800008ae1df8: 2a0003f4 mov w20, w0 I etm4_enable_hw: ffff800008ae1dfc: 900085b3 adrp x19, ffff800009b95000 <reserved_mem+0xc48> I etm4_enable_hw: ffff800008ae1e00: 910f4273 add x19, x19, #0x3d0 I etm4_enable_hw: ffff800008ae1e04: f8747a60 ldr x0, [x19, x20, lsl #3] E etm4_enable_hw: ffff800008ae1e08: b4000140 cbz x0, ffff800008ae1e30 <etm4_starting_cpu+0x50> I 149.039572921 etm4_enable_hw: ffff800008ae1e30: a94153f3 ldp x19, x20, [sp, #16] I 149.039572921 etm4_enable_hw: ffff800008ae1e34: 52800000 mov w0, #0x0 // #0 I 149.039572921 etm4_enable_hw: ffff800008ae1e38: a8c27bfd ldp x29, x30, [sp], #32 ..snip 149.052324811 chacha_block_generic: ffff800008642d80: 9100a3e0 add x0, I 149.052324811 chacha_block_generic: ffff800008642d84: b86178a2 ldr w2, [x5, x1, lsl #2] I 149.052324811 chacha_block_generic: ffff800008642d88: 8b010803 add x3, x0, x1, lsl #2 I 149.052324811 chacha_block_generic: ffff800008642d8c: b85fc063 ldur w3, [x3, #-4] I 149.052324811 chacha_block_generic: ffff800008642d90: 0b030042 add w2, w2, w3 I 149.052324811 chacha_block_generic: ffff800008642d94: b8217882 str w2, [x4, x1, lsl #2] I 149.052324811 chacha_block_generic: ffff800008642d98: 91000421 add x1, x1, #0x1 I 149.052324811 chacha_block_generic: ffff800008642d9c: f100443f cmp x1, #0x11 Sample Core 2 instruction trace dump(kernel panic triggered core): ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ A etm4_enable_hw: ffff800008ae1dd4 CONTEXT EL2 etm4_enable_hw: ffff800008ae1dd4 I etm4_enable_hw: ffff800008ae1dd4: d503201f nop I etm4_enable_hw: ffff800008ae1dd8: d503201f nop I etm4_enable_hw: ffff800008ae1ddc: d503201f nop I etm4_enable_hw: ffff800008ae1de0: d503201f nop I etm4_enable_hw: ffff800008ae1de4: d503201f nop I etm4_enable_hw: ffff800008ae1de8: d503233f paciasp I etm4_enable_hw: ffff800008ae1dec: a9be7bfd stp x29, x30, [sp, #-32]! I etm4_enable_hw: ffff800008ae1df0: 910003fd mov x29, sp I etm4_enable_hw: ffff800008ae1df4: a90153f3 stp x19, x20, [sp, #16] I etm4_enable_hw: ffff800008ae1df8: 2a0003f4 mov w20, w0 I etm4_enable_hw: ffff800008ae1dfc: 900085b3 adrp x19, ffff800009b95000 <reserved_mem+0xc48> I etm4_enable_hw: ffff800008ae1e00: 910f4273 add x19, x19, #0x3d0 I etm4_enable_hw: ffff800008ae1e04: f8747a60 ldr x0, [x19, x20, lsl #3] E etm4_enable_hw: ffff800008ae1e08: b4000140 cbz x0, ffff800008ae1e30 <etm4_starting_cpu+0x50> I 149.046243445 etm4_enable_hw: ffff800008ae1e30: a94153f3 ldp x19, x20, [sp, #16] I 149.046243445 etm4_enable_hw: ffff800008ae1e34: 52800000 mov w0, #0x0 // #0 I 149.046243445 etm4_enable_hw: ffff800008ae1e38: a8c27bfd ldp x29, x30, [sp], #32 I 149.046243445 etm4_enable_hw: ffff800008ae1e3c: d50323bf autiasp E 149.046243445 etm4_enable_hw: ffff800008ae1e40: d65f03c0 ret A ete_sysreg_write: ffff800008adfa18 ..snip I 149.05422547 panic: ffff800008096300: a90363f7 stp x23, x24, [sp, #48] I 149.05422547 panic: ffff800008096304: 6b00003f cmp w1, w0 I 149.05422547 panic: ffff800008096308: 3a411804 ccmn w0, #0x1, #0x4, ne // ne = any N 149.05422547 panic: ffff80000809630c: 540001e0 b.eq ffff800008096348 <panic+0xe0> // b.none I 149.05422547 panic: ffff800008096310: f90023f9 str x25, [sp, #64] E 149.05422547 panic: ffff800008096314: 97fe44ef bl ffff8000080276d0 <panic_smp_self_stop> A panic: ffff80000809634c I 149.05422547 panic: ffff80000809634c: 910102d5 add x21, x22, #0x40 I 149.05422547 panic: ffff800008096350: 52800020 mov w0, #0x1 // #1 E 149.05422547 panic: ffff800008096354: 94166b8b bl ffff800008631180 <bust_spinlocks> N 149.054225518 bust_spinlocks: ffff800008631180: 340000c0 cbz w0, ffff800008631198 <bust_spinlocks+0x18> I 149.054225518 bust_spinlocks: ffff800008631184: f000a321 adrp x1, ffff800009a98000 <pbufs.0+0xbb8> I 149.054225518 bust_spinlocks: ffff800008631188: b9405c20 ldr w0, [x1, #92] I 149.054225518 bust_spinlocks: ffff80000863118c: 11000400 add w0, w0, #0x1 I 149.054225518 bust_spinlocks: ffff800008631190: b9005c20 str w0, [x1, #92] E 149.054225518 bust_spinlocks: ffff800008631194: d65f03c0 ret A panic: ffff800008096358 Perf based testing ------------------ Kernel panic during perf trace sessions has been tested with this series. Starting perf session ~~~~~~~~~~~~~~~~~~~~~ ETF: ./tools/perf/perf record -e cs_etm/panicstop,@tmc_etf1/ -C 1 ./tools/perf/perf record -e cs_etm/panicstop,@tmc_etf2/ -C 2 ETR: ./tools/perf/perf record -e cs_etm/panicstop,@tmc_etr0/ -C 1,2 Reading trace data after panic ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Same sysfs based method explained above can be used to retrieve and decode the trace data after the reboot on kernel panic. Future Improvements ------------------- * Explore changing CTI sysfs script to system configuration manager profile Linu Cherian (7): dt-bindings: arm: coresight-tmc: Add "memory-region" property coresight: tmc-etr: Add support to use reserved trace memory coresight: core: Add provision for panic callbacks coresight: tmc: Enable panic sync handling coresight: tmc: Add support for reading crash data coresight: tmc: Stop trace capture on FlIn coresight: config: Add preloaded configuration .../bindings/arm/arm,coresight-tmc.yaml | 26 ++ drivers/hwtracing/coresight/Makefile | 2 +- .../coresight/coresight-cfg-preload.c | 2 + .../coresight/coresight-cfg-preload.h | 2 + .../hwtracing/coresight/coresight-cfg-pstop.c | 83 +++++ drivers/hwtracing/coresight/coresight-core.c | 37 +++ .../coresight/coresight-etm4x-core.c | 1 + .../hwtracing/coresight/coresight-tmc-core.c | 253 +++++++++++++- .../hwtracing/coresight/coresight-tmc-etf.c | 156 ++++++++- .../hwtracing/coresight/coresight-tmc-etr.c | 312 +++++++++++++++++- drivers/hwtracing/coresight/coresight-tmc.h | 81 +++++ include/linux/coresight.h | 25 ++ 12 files changed, 962 insertions(+), 18 deletions(-) create mode 100644 drivers/hwtracing/coresight/coresight-cfg-pstop.c -- 2.34.1

10 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

CoreSight