Re: [Intel-gfx] [PATCH v5 08/11] drm/i915: Add dev.i915.perf_event_paranoid sysctl option

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 14 September 2016 at 15:19, Robert Bragg <robert@xxxxxxxxxxxxx> wrote:
> Consistent with the kernel.perf_event_paranoid sysctl option that can
> allow non-root users to access system wide cpu metrics, this can
> optionally allow non-root users to access system wide OA counter metrics
> from Gen graphics hardware.
>
> Signed-off-by: Robert Bragg <robert@xxxxxxxxxxxxx>
> ---
>  drivers/gpu/drm/i915/i915_drv.h  |  1 +
>  drivers/gpu/drm/i915/i915_perf.c | 45 +++++++++++++++++++++++++++++++++++++++-
>  2 files changed, 45 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
> index f5ddf70..eaba7a9 100644
> --- a/drivers/gpu/drm/i915/i915_drv.h
> +++ b/drivers/gpu/drm/i915/i915_drv.h
> @@ -2142,6 +2142,7 @@ struct drm_i915_private {
>                 bool initialized;
>
>                 struct kobject *metrics_kobj;
> +               struct ctl_table_header *sysctl_header;
>
>                 struct mutex lock;
>                 struct list_head streams;
> diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
> index e890c38..38b13fa 100644
> --- a/drivers/gpu/drm/i915/i915_perf.c
> +++ b/drivers/gpu/drm/i915/i915_perf.c
> @@ -62,6 +62,8 @@
>  #define POLL_FREQUENCY 200
>  #define POLL_PERIOD (NSEC_PER_SEC / POLL_FREQUENCY)
>
> +static u32 i915_perf_stream_paranoid = true;
> +
>  /* The maximum exponent the hardware accepts is 63 (essentially it selects one
>   * of the 64bit timestamp bits to trigger reports from) but there's currently
>   * no known use case for sampling as infrequently as once per 47 thousand years.
> @@ -1170,7 +1172,13 @@ int i915_perf_open_ioctl_locked(struct drm_device *dev,
>                 }
>         }
>
> -       if (!specific_ctx && !capable(CAP_SYS_ADMIN)) {
> +       /* Similar to perf's kernel.perf_paranoid_cpu sysctl option
> +        * we check a dev.i915.perf_stream_paranoid sysctl option
> +        * to determine if it's ok to access system wide OA counters
> +        * without CAP_SYS_ADMIN privileges.
> +        */
> +       if (!specific_ctx &&
> +           i915_perf_stream_paranoid && !capable(CAP_SYS_ADMIN)) {
>                 DRM_ERROR("Insufficient privileges to open system-wide i915 perf stream\n");
>                 ret = -EACCES;
>                 goto err_ctx;
> @@ -1417,6 +1425,37 @@ void i915_perf_unregister(struct drm_i915_private *dev_priv)
>         dev_priv->perf.metrics_kobj = NULL;
>  }
>
> +static struct ctl_table oa_table[] = {
> +       {
> +        .procname = "perf_stream_paranoid",
> +        .data = &i915_perf_stream_paranoid,
> +        .maxlen = sizeof(i915_perf_stream_paranoid),
> +        .mode = 0644,
> +        .proc_handler = proc_dointvec,
If this really is just a bool, why not:

  .proc_handler = proc_dointvec_minmax,
  .extra1 = &zero,
  .extra2 = &one

Reviewed-by: Matthew Auld <matthew.auld@xxxxxxxxx>
_______________________________________________
dri-devel mailing list
dri-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/dri-devel




[Index of Archives]     [Linux DRI Users]     [Linux Intel Graphics]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [XFree86]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [XFree86]
  Powered by Linux