On Wed, May 01, 2019 at 11:50:48AM -0600, Mathieu Poirier wrote:
In snapshot mode the buffer used by the sink devices need to be equal to the ring buffer size in order for the user space mechanic to work properly.
Signed-off-by: Mathieu Poirier mathieu.poirier@linaro.org
drivers/hwtracing/coresight/coresight-etb10.c | 23 +++++++++++++++++++ .../hwtracing/coresight/coresight-tmc-etf.c | 20 ++++++++++++++++ .../hwtracing/coresight/coresight-tmc-etr.c | 8 +++++-- 3 files changed, 49 insertions(+), 2 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-etb10.c b/drivers/hwtracing/coresight/coresight-etb10.c index 4ee4c80a4354..0764647b92bc 100644 --- a/drivers/hwtracing/coresight/coresight-etb10.c +++ b/drivers/hwtracing/coresight/coresight-etb10.c @@ -374,7 +374,30 @@ static void *etb_alloc_buffer(struct coresight_device *csdev, int nr_pages, bool overwrite) { int node, cpu = event->cpu;
- u32 capacity; struct cs_buffers *buf;
- struct etb_drvdata *drvdata;
- /*
* In snapsot mode the size of the perf ring buffer needs to be equal* to the size of the device's internal memory if we want to reuse the* generic AUX buffer management mechanic.** For example (assuming 4096 byte page size):
Here is delibrately to write as '4096 byte'? Though I think should be '4096 bytes' but I am not confident which is right ...
** # cat /sys/bus/coresight/devices/20010000.etb/mgmt/rdp* 0x2000* # perf record -e cs_etm/@20010000.etf/ -S -m,8 --per-thread $APP**/- drvdata = dev_get_drvdata(csdev->dev.parent);
- capacity = drvdata->buffer_depth * ETB_FRAME_SIZE_WORDS;
- if (overwrite &&
((nr_pages << PAGE_SHIFT) != capacity)) {dev_err(&csdev->dev, "Ring buffer not equal to device buffer");return NULL;- }
if (cpu == -1) cpu = smp_processor_id(); diff --git a/drivers/hwtracing/coresight/coresight-tmc-etf.c b/drivers/hwtracing/coresight/coresight-tmc-etf.c index 2527b5d3b65e..7694833b13cb 100644 --- a/drivers/hwtracing/coresight/coresight-tmc-etf.c +++ b/drivers/hwtracing/coresight/coresight-tmc-etf.c @@ -380,6 +380,26 @@ static void *tmc_alloc_etf_buffer(struct coresight_device *csdev, { int node, cpu = event->cpu; struct cs_buffers *buf;
- struct tmc_drvdata *drvdata;
- /*
* In snapsot mode the size of the perf ring buffer needs to be equal* to the size of the device's internal memory if we want to reuse the* generic AUX buffer management mechanic.** For example (assuming 4096 byte page size):** # cat /sys/bus/coresight/devices/20010000.etf/buffer_size* 0x10000* # perf record -e cs_etm/@20010000.etf/ -S -m,16 --per-thread $APP**/- drvdata = dev_get_drvdata(csdev->dev.parent);
- if (overwrite &&
((nr_pages << PAGE_SHIFT) != drvdata->size)) {dev_err(&csdev->dev, "Ring buffer not equal to device buffer");return NULL;- }
if (cpu == -1) cpu = smp_processor_id(); diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c index df6e4b0b84e9..b9881d6d41ba 100644 --- a/drivers/hwtracing/coresight/coresight-tmc-etr.c +++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c @@ -1188,9 +1188,13 @@ alloc_etr_buf(struct tmc_drvdata *drvdata, struct perf_event *event, /* * Try to match the perf ring buffer size if it is larger
* than the size requested via sysfs.
* than the size requested via sysfs. In snapsot mode the size* of the perf ring buffer needs to be equal to the allocated* size if we want to reuse the generic AUX buffer management* mechanic.
- if ((nr_pages << PAGE_SHIFT) > drvdata->size) {
- if (snapshot ||
(nr_pages << PAGE_SHIFT) > drvdata->size) {
If tmc_alloc_etr_buf() returns failure then it's possible to run into the below sequence to allocate smaller buffer size for snapshot mode; which is not expected for snapshot mode.
So here if tmc_alloc_etr_buf() fails to allocate buffer for snapshot mode, should directly bail out with error code.
Thanks, Leo Yan