Hi Krzysztof,

On 2026-11-04 at 11:11:03 +0000, Krzysztof Karas wrote:
> Move away from generic pr prints to GT aware gt_* versions,
> which tag each message with device and the tile it came from.
> 
> Signed-off-by: Krzysztof Karas <[email protected]>
> Acked-by: Jani Nikula <[email protected]>
> ---
>  .../drm/i915/gt/selftest_engine_heartbeat.c   |  7 +++---
>  drivers/gpu/drm/i915/gt/selftest_llc.c        | 25 ++++++++++++-------
>  2 files changed, 20 insertions(+), 12 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c 
> b/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c
> index b893a8fde895..1164dd4aea11 100644
> --- a/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c
> +++ b/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c
> @@ -7,6 +7,7 @@
>  
>  #include "i915_drv.h"
>  
> +#include "intel_gt_print.h"
>  #include "intel_gt_requests.h"
>  #include "i915_selftest.h"
>  #include "selftest_engine_heartbeat.h"
> @@ -202,7 +203,7 @@ static int __live_heartbeat_off(struct intel_engine_cs 
> *engine)
>       engine->serial++;
>       flush_delayed_work(&engine->heartbeat.work);
>       if (!delayed_work_pending(&engine->heartbeat.work)) {
> -             pr_err("%s: heartbeat not running\n",
> +             gt_err(engine->gt, "%s: heartbeat not running\n",
>                      engine->name);
>               err = -EINVAL;
>               goto err_pm;
> @@ -215,14 +216,14 @@ static int __live_heartbeat_off(struct intel_engine_cs 
> *engine)
>       engine->serial++;
>       flush_delayed_work(&engine->heartbeat.work);
>       if (delayed_work_pending(&engine->heartbeat.work)) {
> -             pr_err("%s: heartbeat still running\n",
> +             gt_err(engine->gt, "%s: heartbeat still running\n",
>                      engine->name);
>               err = -EINVAL;
>               goto err_beat;
>       }
>  
>       if (READ_ONCE(engine->heartbeat.systole)) {
> -             pr_err("%s: heartbeat still allocated\n",
> +             gt_err(engine->gt, "%s: heartbeat still allocated\n",
>                      engine->name);
>               err = -EINVAL;
>               goto err_beat;
> diff --git a/drivers/gpu/drm/i915/gt/selftest_llc.c 
> b/drivers/gpu/drm/i915/gt/selftest_llc.c
> index 779fadcec7c4..71ea688be154 100644
> --- a/drivers/gpu/drm/i915/gt/selftest_llc.c
> +++ b/drivers/gpu/drm/i915/gt/selftest_llc.c
> @@ -3,6 +3,7 @@
>   * Copyright © 2019 Intel Corporation
>   */
>  
> +#include "intel_gt_print.h"
>  #include "selftest_llc.h"
>  #include "intel_rps.h"
>  
> @@ -22,8 +23,8 @@ static int gen6_verify_ring_freq(struct intel_llc *llc)
>       for (gpu_freq = consts.min_gpu_freq;
>            gpu_freq <= consts.max_gpu_freq;
>            gpu_freq++) {
> +             u32 scaler = GRAPHICS_VER(i915) >= 9 ? GEN9_FREQ_SCALER : 1;

Giga nit: gpu_freq is "unsigned int", not explicltly u32, and when the
ternary expression was inline I'm pretty sure the (implicit) cast is into
unsigned int, and not to u32. Personally I'd follow the type of gpu_freq
but I am aware that this is complete pedantry since currently unsigned int
is (basically) guaranteed to be 32bit anyway (and GEN9_FREQ_SCALER is
currently equal to the huge value of 3 so it probably doesn't matter much
anyway)

Disregarding my rambling:

Reviewed-by: Krzysztof Niemiec <[email protected]>

Thanks
Krzysztof

Reply via email to