Commit 43ca8efa authored by pding's avatar pding Committed by Alex Deucher
Browse files

drm/amdgpu: busywait KIQ register accessing (v4) 



Register accessing is performed when IRQ is disabled. Never sleep in
this function.

Known issue: dead sleep in many use cases of index/data registers.

v2:
 - wrap polling fence functions.
 - don't trigger IRQ for polling in case of wrongly fence signal.

v3:
 - handle wrap round gracefully.
 - add comments for polling function

v4:
 - don't return negative timeout confused with error code

Signed-off-by: default avatarpding <Pixel.Ding@amd.com>
Reviewed-by: default avatarChristian König <christian.koenig@amd.com>
Signed-off-by: default avatarAlex Deucher <alexander.deucher@amd.com>
parent e71de076

drivers/gpu/drm/amd/amdgpu/amdgpu.h

+1 −1
Original line number Diff line number Diff line
@@ -879,7 +879,7 @@ struct amdgpu_mec { 
struct amdgpu_kiq {

	u64			eop_gpu_addr;

	struct amdgpu_bo	*eop_obj;

	struct mutex		ring_mutex;

	spinlock_t              ring_lock;

	struct amdgpu_ring	ring;

	struct amdgpu_irq_src	irq;

};

drivers/gpu/drm/amd/amdgpu/amdgpu_device.c

+2 −6
Original line number Diff line number Diff line
@@ -109,10 +109,8 @@ uint32_t amdgpu_mm_rreg(struct amdgpu_device *adev, uint32_t reg, 
{

	uint32_t ret;



	if (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev)) {

		BUG_ON(in_interrupt());

	if (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev))

		return amdgpu_virt_kiq_rreg(adev, reg);

	}



	if ((reg * 4) < adev->rmmio_size && !(acc_flags & AMDGPU_REGS_IDX))

		ret = readl(((void __iomem *)adev->rmmio) + (reg * 4));
@@ -137,10 +135,8 @@ void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, 
		adev->last_mm_index = v;

	}



	if (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev)) {

		BUG_ON(in_interrupt());

	if (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev))

		return amdgpu_virt_kiq_wreg(adev, reg, v);

	}



	if ((reg * 4) < adev->rmmio_size && !(acc_flags & AMDGPU_REGS_IDX))

		writel(v, ((void __iomem *)adev->rmmio) + (reg * 4));

drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c

+50 −0
Original line number Diff line number Diff line
@@ -168,6 +168,32 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f) 
	return 0;

}



/**

 * amdgpu_fence_emit_polling - emit a fence on the requeste ring

 *

 * @ring: ring the fence is associated with

 * @s: resulting sequence number

 *

 * Emits a fence command on the requested ring (all asics).

 * Used For polling fence.

 * Returns 0 on success, -ENOMEM on failure.

 */

int amdgpu_fence_emit_polling(struct amdgpu_ring *ring, uint32_t *s)

{

	uint32_t seq;



	if (!s)

		return -EINVAL;



	seq = ++ring->fence_drv.sync_seq;

	amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr,

			       seq, AMDGPU_FENCE_FLAG_INT);



	*s = seq;



	return 0;

}



/**

 * amdgpu_fence_schedule_fallback - schedule fallback check

 *
@@ -281,6 +307,30 @@ int amdgpu_fence_wait_empty(struct amdgpu_ring *ring) 
	return r;

}



/**

 * amdgpu_fence_wait_polling - busy wait for givn sequence number

 *

 * @ring: ring index the fence is associated with

 * @wait_seq: sequence number to wait

 * @timeout: the timeout for waiting in usecs

 *

 * Wait for all fences on the requested ring to signal (all asics).

 * Returns left time if no timeout, 0 or minus if timeout.

 */

signed long amdgpu_fence_wait_polling(struct amdgpu_ring *ring,

				      uint32_t wait_seq,

				      signed long timeout)

{

	uint32_t seq;



	do {

		seq = amdgpu_fence_read(ring);

		udelay(5);

		timeout -= 5;

	} while ((int32_t)(wait_seq - seq) > 0 && timeout > 0);



	return timeout > 0 ? timeout : 0;

}

/**

 * amdgpu_fence_count_emitted - get the count of emitted fences

 *

drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c

+1 −1
Original line number Diff line number Diff line
@@ -201,7 +201,7 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev, 
	struct amdgpu_kiq *kiq = &adev->gfx.kiq;

	int r = 0;



	mutex_init(&kiq->ring_mutex);

	spin_lock_init(&kiq->ring_lock);



	r = amdgpu_wb_get(adev, &adev->virt.reg_val_offs);

	if (r)

drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h

+4 −0
Original line number Diff line number Diff line
@@ -90,8 +90,12 @@ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring, 
void amdgpu_fence_driver_suspend(struct amdgpu_device *adev);

void amdgpu_fence_driver_resume(struct amdgpu_device *adev);

int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **fence);

int amdgpu_fence_emit_polling(struct amdgpu_ring *ring, uint32_t *s);

void amdgpu_fence_process(struct amdgpu_ring *ring);

int amdgpu_fence_wait_empty(struct amdgpu_ring *ring);

signed long amdgpu_fence_wait_polling(struct amdgpu_ring *ring,

				      uint32_t wait_seq,

				      signed long timeout);

unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring);



/*

CRA Git | Maintained and supported by SUSTech CRA and CCSE