[Intel-xe] [PATCH v4 10/11] drm/xe: Add HW Engine snapshot to xe_devcoredump.

[Francois Dugast]

On Tue, May 16, 2023 at 10:54:15AM -0400, Rodrigo Vivi wrote:
> Let's continue to add our existent simple logs to devcoredump one
> by one. Any format change should come on follow-up work.
> 
> v2: remove unnecessary, and now duplicated, dma_fence annotation. (Matthew)
> v3: avoid for_each with faulty_engine since that can be already freed at
>     the time of the read/free. Instead, iterate in the full array of
>     hw_engines. (Kasan)
> 
> Cc: Francois Dugast <francois.dugast at intel.com>
> Cc: Matthew Brost <matthew.brost at intel.com>
> Signed-off-by: Rodrigo Vivi <rodrigo.vivi at intel.com>
> Reviewed-by: Matthew Brost <matthew.brost at intel.com> #v2

Reviewed-by: Francois Dugast <francois.dugast at intel.com>

> ---
>  drivers/gpu/drm/xe/xe_devcoredump.c       | 41 +++++++++++++++++++++++
>  drivers/gpu/drm/xe/xe_devcoredump_types.h |  4 +++
>  2 files changed, 45 insertions(+)
> 
> diff --git a/drivers/gpu/drm/xe/xe_devcoredump.c b/drivers/gpu/drm/xe/xe_devcoredump.c
> index 2da27ce6e116..6057649f1d3a 100644
> --- a/drivers/gpu/drm/xe/xe_devcoredump.c
> +++ b/drivers/gpu/drm/xe/xe_devcoredump.c
> @@ -9,10 +9,13 @@
>  #include <linux/devcoredump.h>
>  #include <generated/utsrelease.h>
>  
> +#include "xe_device.h"
>  #include "xe_engine.h"
> +#include "xe_force_wake.h"
>  #include "xe_gt.h"
>  #include "xe_guc_ct.h"
>  #include "xe_guc_submit.h"
> +#include "xe_hw_engine.h"
>  
>  /**
>   * DOC: Xe device coredump
> @@ -63,6 +66,7 @@ static ssize_t xe_devcoredump_read(char *buffer, loff_t offset,
>  	struct drm_printer p;
>  	struct drm_print_iterator iter;
>  	struct timespec64 ts;
> +	int i;
>  
>  	/* Our device is gone already... */
>  	if (!data || !coredump_to_xe(coredump))
> @@ -91,12 +95,19 @@ static ssize_t xe_devcoredump_read(char *buffer, loff_t offset,
>  	xe_guc_ct_snapshot_print(coredump->snapshot.ct, &p);
>  	xe_guc_engine_snapshot_print(coredump->snapshot.ge, &p);
>  
> +	drm_printf(&p, "\n**** HW Engines ****\n");
> +	for (i = 0; i < XE_NUM_HW_ENGINES; i++)
> +		if (coredump->snapshot.hwe[i])
> +			xe_hw_engine_snapshot_print(coredump->snapshot.hwe[i],
> +						    &p);
> +
>  	return count - iter.remain;
>  }
>  
>  static void xe_devcoredump_free(void *data)
>  {
>  	struct xe_devcoredump *coredump = data;
> +	int i;
>  
>  	/* Our device is gone. Nothing to do... */
>  	if (!data || !coredump_to_xe(coredump))
> @@ -104,6 +115,9 @@ static void xe_devcoredump_free(void *data)
>  
>  	xe_guc_ct_snapshot_free(coredump->snapshot.ct);
>  	xe_guc_engine_snapshot_free(coredump->snapshot.ge);
> +	for (i = 0; i < XE_NUM_HW_ENGINES; i++)
> +		if (coredump->snapshot.hwe[i])
> +			xe_hw_engine_snapshot_free(coredump->snapshot.hwe[i]);
>  
>  	coredump->captured = false;
>  	drm_info(&coredump_to_xe(coredump)->drm,
> @@ -115,14 +129,41 @@ static void devcoredump_snapshot(struct xe_devcoredump *coredump,
>  {
>  	struct xe_devcoredump_snapshot *ss = &coredump->snapshot;
>  	struct xe_guc *guc = engine_to_guc(e);
> +	struct xe_hw_engine *hwe;
> +	enum xe_hw_engine_id id;
> +	u32 adj_logical_mask = e->logical_mask;
> +	u32 width_mask = (0x1 << e->width) - 1;
> +	int i;
>  	bool cookie;
>  
>  	ss->snapshot_time = ktime_get_real();
>  	ss->boot_time = ktime_get_boottime();
>  
>  	cookie = dma_fence_begin_signalling();
> +	for (i = 0; e->width > 1 && i < XE_HW_ENGINE_MAX_INSTANCE;) {
> +		if (adj_logical_mask & BIT(i)) {
> +			adj_logical_mask |= width_mask << i;
> +			i += e->width;
> +		} else {
> +			++i;
> +		}
> +	}
> +
> +	xe_force_wake_get(gt_to_fw(e->gt), XE_FORCEWAKE_ALL);
> +
>  	coredump->snapshot.ct = xe_guc_ct_snapshot_capture(&guc->ct, true);
>  	coredump->snapshot.ge = xe_guc_engine_snapshot_capture(e);
> +
> +	for_each_hw_engine(hwe, e->gt, id) {
> +		if (hwe->class != e->hwe->class ||
> +		    !(BIT(hwe->logical_instance) & adj_logical_mask)) {
> +			coredump->snapshot.hwe[id] = NULL;
> +			continue;
> +		}
> +		coredump->snapshot.hwe[id] = xe_hw_engine_snapshot_capture(hwe);
> +	}
> +
> +	xe_force_wake_put(gt_to_fw(e->gt), XE_FORCEWAKE_ALL);
>  	dma_fence_end_signalling(cookie);
>  }
>  
> diff --git a/drivers/gpu/drm/xe/xe_devcoredump_types.h b/drivers/gpu/drm/xe/xe_devcoredump_types.h
> index 7c6453224139..350b905d1797 100644
> --- a/drivers/gpu/drm/xe/xe_devcoredump_types.h
> +++ b/drivers/gpu/drm/xe/xe_devcoredump_types.h
> @@ -9,6 +9,8 @@
>  #include <linux/ktime.h>
>  #include <linux/mutex.h>
>  
> +#include "xe_hw_engine_types.h"
> +
>  struct xe_device;
>  
>  /**
> @@ -29,6 +31,8 @@ struct xe_devcoredump_snapshot {
>  	struct xe_guc_ct_snapshot *ct;
>  	/** @ge: Guc Engine snapshot */
>  	struct xe_guc_submit_engine_snapshot *ge;
> +	/** @hwe: HW Engine snapshot array */
> +	struct xe_hw_engine_snapshot *hwe[XE_NUM_HW_ENGINES];
>  };
>  
>  /**
> -- 
> 2.39.2
>