[PATCH i-g-t] gputop: Add memory only utilization type
Tvrtko Ursulin
tvrtko.ursulin at igalia.com
Thu Nov 21 10:03:40 UTC 2024
On 21/11/2024 09:38, Christian Gmeiner wrote:
> From: Christian Gmeiner <cgmeiner at igalia.com>
>
> Add special handling for drivers that only provide memory utilization
> fdinfo.
To check I am following correctly - special in this context means avoid
displaying empty fields, or avoid a crash, or?
> Signed-off-by: Christian Gmeiner <cgmeiner at igalia.com>
> ---
> tools/gputop.c | 20 ++++++++++++++++++++
> 1 file changed, 20 insertions(+)
>
> diff --git a/tools/gputop.c b/tools/gputop.c
> index 43b01f566..4135d84c1 100644
> --- a/tools/gputop.c
> +++ b/tools/gputop.c
> @@ -35,6 +35,7 @@
> enum utilization_type {
> UTILIZATION_TYPE_ENGINE_TIME,
> UTILIZATION_TYPE_TOTAL_CYCLES,
> + UTILIZATION_TYPE_MEMORY_ONLY,
> };
>
> static const char *bars[] = { " ", "▏", "▎", "▍", "▌", "▋", "▊", "▉", "█" };
> @@ -141,6 +142,9 @@ engines_identical(const struct igt_drm_client *c,
> c->engines->max_engine_id != pc->engines->max_engine_id)
> return false;
>
> + if (!c->engines->num_engines && !pc->engines->num_engines)
> + return true;
strcmp() below crashes?
I would move the check you added to be first in this function. I think
that makes most sense from the flow wise - checking it before max_engine_id.
With this fixed what remains broken?
The loop in print_client() should ideally be skipped but even as is I
think exits on the capacity check, no? At the moment it seems like this
could be enough but maybe I am missing something.
Regards,
Tvrtko
> +
> for (i = 0; i <= c->engines->max_engine_id; i++)
> if (c->engines->capacity[i] != pc->engines->capacity[i] ||
> !!c->engines->names[i] != !!pc->engines->names[i] ||
> @@ -193,6 +197,8 @@ print_client(struct igt_drm_client *c, struct igt_drm_client **prevc,
> utilization_type = UTILIZATION_TYPE_TOTAL_CYCLES;
> else if (c->utilization_mask & IGT_DRM_CLIENT_UTILIZATION_ENGINE_TIME)
> utilization_type = UTILIZATION_TYPE_ENGINE_TIME;
> + else if (c->regions->num_regions)
> + utilization_type = UTILIZATION_TYPE_MEMORY_ONLY;
> else
> return 0;
>
> @@ -209,6 +215,13 @@ print_client(struct igt_drm_client *c, struct igt_drm_client **prevc,
> if (!c->total_total_cycles)
> return 0;
> break;
> + case UTILIZATION_TYPE_MEMORY_ONLY:
> + for (sz = 0, i = 0; i <= c->regions->max_region_id; i++)
> + sz += c->memory[i].total;
> +
> + if (sz == 0)
> + return 0;
> + break;
> }
>
> /* Print header when moving to a different DRM card. */
> @@ -234,6 +247,9 @@ print_client(struct igt_drm_client *c, struct igt_drm_client **prevc,
>
> lines++;
>
> + if (utilization_type == UTILIZATION_TYPE_MEMORY_ONLY)
> + goto print_name;
> +
> for (i = 0; c->samples > 1 && i <= c->engines->max_engine_id; i++) {
> double pct;
>
> @@ -249,6 +265,9 @@ print_client(struct igt_drm_client *c, struct igt_drm_client **prevc,
> pct = (double)c->utilization[i].delta_cycles / c->utilization[i].delta_total_cycles * 100 /
> c->engines->capacity[i];
> break;
> + case UTILIZATION_TYPE_MEMORY_ONLY:
> + /* Nothing to do. */
> + break;
> }
>
> /*
> @@ -262,6 +281,7 @@ print_client(struct igt_drm_client *c, struct igt_drm_client **prevc,
> len += *engine_w;
> }
>
> +print_name:
> printf(" %-*s\n", con_w - len - 1, c->print_name);
>
> return lines;
More information about the igt-dev
mailing list