Unverified Commit 783d6cdc authored by Rodrigo Vivi's avatar Rodrigo Vivi
Browse files

drm/xe: Kill xe_device_mem_access_{get*,put} 



Let's simply convert all the current callers towards direct
xe_pm_runtime access and remove this extra layer of indirection.

No functional change is expected with this patch since
xe_mem_access_get was already using the xe_pm_runtime_get_noresume
at this point.

v2: Convert all the current callers instead of a big refactor
at once.

v3: - Rebased
    - Squashed the GSC/HDCP
    - Added a new case: sriov_pf_policy
    - Improved commit message to highlight that
      there's no functional change in this patch.

Reviewed-by: Matthew Auld <matthew.auld@intel.com> #v2
Cc: Suraj Kandpal <suraj.kandpal@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: default avatarRodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: default avatarSuraj Kandpal <suraj.kandpal@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240418143049.43231-1-rodrigo.vivi@intel.com
parent 62422b7b

drivers/gpu/drm/xe/display/xe_fb_pin.c

+3 −2
Original line number Diff line number Diff line
@@ -10,6 +10,7 @@ 
#include "intel_fb_pin.h"

#include "xe_ggtt.h"

#include "xe_gt.h"

#include "xe_pm.h"



#include <drm/ttm/ttm_bo.h>
@@ -193,7 +194,7 @@ static int __xe_pin_fb_vma_ggtt(struct intel_framebuffer *fb, 
	/* TODO: Consider sharing framebuffer mapping?

	 * embed i915_vma inside intel_framebuffer

	 */

	xe_device_mem_access_get(tile_to_xe(ggtt->tile));

	xe_pm_runtime_get_noresume(tile_to_xe(ggtt->tile));

	ret = mutex_lock_interruptible(&ggtt->lock);

	if (ret)

		goto out;
@@ -244,7 +245,7 @@ static int __xe_pin_fb_vma_ggtt(struct intel_framebuffer *fb, 
out_unlock:

	mutex_unlock(&ggtt->lock);

out:

	xe_device_mem_access_put(tile_to_xe(ggtt->tile));

	xe_pm_runtime_put(tile_to_xe(ggtt->tile));

	return ret;

}

drivers/gpu/drm/xe/display/xe_hdcp_gsc.c

+2 −2
Original line number Diff line number Diff line
@@ -215,7 +215,7 @@ ssize_t intel_hdcp_gsc_msg_send(struct xe_device *xe, u8 *msg_in, 
	addr_out_off = PAGE_SIZE;



	host_session_id = xe_gsc_create_host_session_id();

	xe_device_mem_access_get(xe);

	xe_pm_runtime_get_noresume(xe);

	addr_in_wr_off = xe_gsc_emit_header(xe, &hdcp_message->hdcp_bo->vmap,

					    addr_in_wr_off, HECI_MEADDRESS_HDCP,

					    host_session_id, msg_in_len);
@@ -247,6 +247,6 @@ ssize_t intel_hdcp_gsc_msg_send(struct xe_device *xe, u8 *msg_in, 
			   msg_out_len);



out:

	xe_device_mem_access_put(xe);

	xe_pm_runtime_put(xe);

	return ret;

}

drivers/gpu/drm/xe/xe_bo.c

+4 −4
Original line number Diff line number Diff line
@@ -716,7 +716,7 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, 


	xe_assert(xe, migrate);

	trace_xe_bo_move(bo, new_mem->mem_type, old_mem_type, move_lacks_source);

	xe_device_mem_access_get(xe);

	xe_pm_runtime_get_noresume(xe);



	if (xe_bo_is_pinned(bo) && !xe_bo_is_user(bo)) {

		/*
@@ -740,7 +740,7 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, 


				if (XE_WARN_ON(new_mem->start == XE_BO_INVALID_OFFSET)) {

					ret = -EINVAL;

					xe_device_mem_access_put(xe);

					xe_pm_runtime_put(xe);

					goto out;

				}
@@ -758,7 +758,7 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, 
						new_mem, handle_system_ccs);

		if (IS_ERR(fence)) {

			ret = PTR_ERR(fence);

			xe_device_mem_access_put(xe);

			xe_pm_runtime_put(xe);

			goto out;

		}

		if (!move_lacks_source) {
@@ -783,7 +783,7 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, 
		dma_fence_put(fence);

	}



	xe_device_mem_access_put(xe);

	xe_pm_runtime_put(xe);



out:

	return ret;

drivers/gpu/drm/xe/xe_device.c

+0 −36
Original line number Diff line number Diff line
@@ -729,42 +729,6 @@ void xe_device_assert_mem_access(struct xe_device *xe) 
	xe_assert(xe, !xe_pm_runtime_suspended(xe));

}



void xe_device_mem_access_get(struct xe_device *xe)

{

	int ref;



	/*

	 * This looks racy, but should be fine since the pm_callback_task only

	 * transitions from NULL -> current (and back to NULL again), during the

	 * runtime_resume() or runtime_suspend() callbacks, for which there can

	 * only be a single one running for our device. We only need to prevent

	 * recursively calling the runtime_get or runtime_put from those

	 * callbacks, as well as preventing triggering any access_ongoing

	 * asserts.

	 */

	if (xe_pm_read_callback_task(xe) == current)

		return;



	xe_pm_runtime_get_noresume(xe);

	ref = atomic_inc_return(&xe->mem_access.ref);



	xe_assert(xe, ref != S32_MAX);



}



void xe_device_mem_access_put(struct xe_device *xe)

{

	int ref;



	if (xe_pm_read_callback_task(xe) == current)

		return;



	ref = atomic_dec_return(&xe->mem_access.ref);

	xe_pm_runtime_put(xe);



	xe_assert(xe, ref >= 0);

}



void xe_device_snapshot_print(struct xe_device *xe, struct drm_printer *p)

{

	struct xe_gt *gt;

drivers/gpu/drm/xe/xe_device.h

+0 −3
Original line number Diff line number Diff line
@@ -133,9 +133,6 @@ static inline struct xe_force_wake *gt_to_fw(struct xe_gt *gt) 
	return &gt->mmio.fw;

}



void xe_device_mem_access_get(struct xe_device *xe);

void xe_device_mem_access_put(struct xe_device *xe);



void xe_device_assert_mem_access(struct xe_device *xe);



static inline bool xe_device_in_fault_mode(struct xe_device *xe)