drm/radeon: Only flush HDP cache for indirect buffers from userspace

It isn't necessary for command streams generated by the kernel (at least
not while we aren't storing ring or indirect buffers in VRAM).

Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index e7d99e1547ccbe1930be44b2b70d75842028fa16..d61bc1841d5e5acecc739efabf2c195d73d26511 100644 (file)
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -3801,7 +3801,7 @@ int cik_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
        radeon_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG, 1));
        radeon_ring_write(ring, ((scratch - PACKET3_SET_UCONFIG_REG_START) >> 2));
        radeon_ring_write(ring, 0xDEADBEEF);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        for (i = 0; i < rdev->usec_timeout; i++) {
                tmp = RREG32(scratch);
@@ -4004,7 +4004,7 @@ int cik_copy_cpdma(struct radeon_device *rdev,
                return r;
        }
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        radeon_semaphore_free(rdev, &sem, *fence);
 
        return r;
@@ -4103,7 +4103,7 @@ int cik_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
        ib.ptr[1] = ((scratch - PACKET3_SET_UCONFIG_REG_START) >> 2);
        ib.ptr[2] = 0xDEADBEEF;
        ib.length_dw = 3;
-       r = radeon_ib_schedule(rdev, &ib, NULL);
+       r = radeon_ib_schedule(rdev, &ib, NULL, false);
        if (r) {
                radeon_scratch_free(rdev, scratch);
                radeon_ib_free(rdev, &ib);
@@ -4324,7 +4324,7 @@ static int cik_cp_gfx_start(struct radeon_device *rdev)
        radeon_ring_write(ring, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
        radeon_ring_write(ring, 0x00000010); /* VGT_OUT_DEALLOC_CNTL */
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        return 0;
 }

diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index bcf480510ac228af4bdf6e177eb0ccfb5951b24c..192278bc993ce581e42b9eb2e47a207533a45cd5 100644 (file)
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -596,7 +596,7 @@ int cik_copy_dma(struct radeon_device *rdev,
                return r;
        }
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        radeon_semaphore_free(rdev, &sem, *fence);
 
        return r;
@@ -638,7 +638,7 @@ int cik_sdma_ring_test(struct radeon_device *rdev,
        radeon_ring_write(ring, upper_32_bits(rdev->vram_scratch.gpu_addr));
        radeon_ring_write(ring, 1); /* number of DWs to follow */
        radeon_ring_write(ring, 0xDEADBEEF);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        for (i = 0; i < rdev->usec_timeout; i++) {
                tmp = readl(ptr);
@@ -695,7 +695,7 @@ int cik_sdma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
        ib.ptr[4] = 0xDEADBEEF;
        ib.length_dw = 5;
 
-       r = radeon_ib_schedule(rdev, &ib, NULL);
+       r = radeon_ib_schedule(rdev, &ib, NULL, false);
        if (r) {
                radeon_ib_free(rdev, &ib);
                DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);

diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 4fedd14e670aeb33e07fadcf5ff2564c2cc2681c..dbca60c7d097664e381d9bd5c6539d140aa72c16 100644 (file)
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -2869,7 +2869,7 @@ static int evergreen_cp_start(struct radeon_device *rdev)
        radeon_ring_write(ring, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
        radeon_ring_write(ring, 0);
        radeon_ring_write(ring, 0);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        cp_me = 0xff;
        WREG32(CP_ME_CNTL, cp_me);
@@ -2912,7 +2912,7 @@ static int evergreen_cp_start(struct radeon_device *rdev)
        radeon_ring_write(ring, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
        radeon_ring_write(ring, 0x00000010); /*  */
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        return 0;
 }

diff --git a/drivers/gpu/drm/radeon/evergreen_dma.c b/drivers/gpu/drm/radeon/evergreen_dma.c
index 478caefe0fef918011fadd78e1419ac92183ae54..afaba388c36dec2fcd6cd5b986a9ab4d70406385 100644 (file)
--- a/drivers/gpu/drm/radeon/evergreen_dma.c
+++ b/drivers/gpu/drm/radeon/evergreen_dma.c
@@ -155,7 +155,7 @@ int evergreen_copy_dma(struct radeon_device *rdev,
                return r;
        }
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        radeon_semaphore_free(rdev, &sem, *fence);
 
        return r;

diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index 327b85f7fd0d45eb434a8f8232a206ac736ab0be..ba89375f197f9146ddd62cd64368075bb7fc9113 100644 (file)
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -1505,7 +1505,7 @@ static int cayman_cp_start(struct radeon_device *rdev)
        radeon_ring_write(ring, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
        radeon_ring_write(ring, 0);
        radeon_ring_write(ring, 0);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        cayman_cp_enable(rdev, true);
 
@@ -1547,7 +1547,7 @@ static int cayman_cp_start(struct radeon_device *rdev)
        radeon_ring_write(ring, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
        radeon_ring_write(ring, 0x00000010); /*  */
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        /* XXX init other rings */
 

diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 04b5940b89234893f3f3042e5a00f30f97e899cf..4c5ec44ff328b3c63203e4c7c92f5283126e65e9 100644 (file)
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -925,7 +925,7 @@ int r100_copy_blit(struct radeon_device *rdev,
        if (fence) {
                r = radeon_fence_emit(rdev, fence, RADEON_RING_TYPE_GFX_INDEX);
        }
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        return r;
 }
 
@@ -958,7 +958,7 @@ void r100_ring_start(struct radeon_device *rdev, struct radeon_ring *ring)
                          RADEON_ISYNC_ANY3D_IDLE2D |
                          RADEON_ISYNC_WAIT_IDLEGUI |
                          RADEON_ISYNC_CPSCRATCH_IDLEGUI);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 }
 
 
@@ -3638,7 +3638,7 @@ int r100_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
        }
        radeon_ring_write(ring, PACKET0(scratch, 0));
        radeon_ring_write(ring, 0xDEADBEEF);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        for (i = 0; i < rdev->usec_timeout; i++) {
                tmp = RREG32(scratch);
                if (tmp == 0xDEADBEEF) {
@@ -3700,7 +3700,7 @@ int r100_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
        ib.ptr[6] = PACKET2(0);
        ib.ptr[7] = PACKET2(0);
        ib.length_dw = 8;
-       r = radeon_ib_schedule(rdev, &ib, NULL);
+       r = radeon_ib_schedule(rdev, &ib, NULL, false);
        if (r) {
                DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
                goto free_ib;

diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c
index 58f0473aa73fba3cd1e62a786d6776c7a20ccd3b..67780374a65203fc57736510685f917b14d4ce89 100644 (file)
--- a/drivers/gpu/drm/radeon/r200.c
+++ b/drivers/gpu/drm/radeon/r200.c
@@ -121,7 +121,7 @@ int r200_copy_dma(struct radeon_device *rdev,
        if (fence) {
                r = radeon_fence_emit(rdev, fence, RADEON_RING_TYPE_GFX_INDEX);
        }
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        return r;
 }
 

diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 75b30338c226f01ef263edcac893e16dec496643..1bc4704034ce9d90f9563f598bb6135b44878f06 100644 (file)
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -295,7 +295,7 @@ void r300_ring_start(struct radeon_device *rdev, struct radeon_ring *ring)
        radeon_ring_write(ring,
                          R300_GEOMETRY_ROUND_NEAREST |
                          R300_COLOR_ROUND_NEAREST);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 }
 
 static void r300_errata(struct radeon_device *rdev)

diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c
index 802b19220a21358712dc304a894db892ec3537fb..2828605aef3fcdeb76b529e7b478bdc866d1a27f 100644 (file)
--- a/drivers/gpu/drm/radeon/r420.c
+++ b/drivers/gpu/drm/radeon/r420.c
@@ -219,7 +219,7 @@ static void r420_cp_errata_init(struct radeon_device *rdev)
        radeon_ring_write(ring, PACKET0(R300_CP_RESYNC_ADDR, 1));
        radeon_ring_write(ring, rdev->config.r300.resync_scratch);
        radeon_ring_write(ring, 0xDEADBEEF);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 }
 
 static void r420_cp_errata_fini(struct radeon_device *rdev)
@@ -232,7 +232,7 @@ static void r420_cp_errata_fini(struct radeon_device *rdev)
        radeon_ring_lock(rdev, ring, 8);
        radeon_ring_write(ring, PACKET0(R300_RB3D_DSTCACHE_CTLSTAT, 0));
        radeon_ring_write(ring, R300_RB3D_DC_FINISH);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        radeon_scratch_free(rdev, rdev->config.r300.resync_scratch);
 }
 

diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index c70a504d96af5a8ac6f8a9f8e40dd2039917e467..1e4770138fd420f999c1393e3e01633714456cb6 100644 (file)
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -2547,7 +2547,7 @@ int r600_cp_start(struct radeon_device *rdev)
        radeon_ring_write(ring, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
        radeon_ring_write(ring, 0);
        radeon_ring_write(ring, 0);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        cp_me = 0xff;
        WREG32(R_0086D8_CP_ME_CNTL, cp_me);
@@ -2683,7 +2683,7 @@ int r600_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
        radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
        radeon_ring_write(ring, ((scratch - PACKET3_SET_CONFIG_REG_OFFSET) >> 2));
        radeon_ring_write(ring, 0xDEADBEEF);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        for (i = 0; i < rdev->usec_timeout; i++) {
                tmp = RREG32(scratch);
                if (tmp == 0xDEADBEEF)
@@ -2845,7 +2845,7 @@ int r600_copy_cpdma(struct radeon_device *rdev,
                return r;
        }
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        radeon_semaphore_free(rdev, &sem, *fence);
 
        return r;
@@ -3165,7 +3165,7 @@ int r600_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
        ib.ptr[1] = ((scratch - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
        ib.ptr[2] = 0xDEADBEEF;
        ib.length_dw = 3;
-       r = radeon_ib_schedule(rdev, &ib, NULL);
+       r = radeon_ib_schedule(rdev, &ib, NULL, false);
        if (r) {
                DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
                goto free_ib;

diff --git a/drivers/gpu/drm/radeon/r600_dma.c b/drivers/gpu/drm/radeon/r600_dma.c
index 4969cef44a1911b706e933fb8252397ddd893785..51fd98553eaf6c4a34e73c5baeecf95b94220e29 100644 (file)
--- a/drivers/gpu/drm/radeon/r600_dma.c
+++ b/drivers/gpu/drm/radeon/r600_dma.c
@@ -261,7 +261,7 @@ int r600_dma_ring_test(struct radeon_device *rdev,
        radeon_ring_write(ring, rdev->vram_scratch.gpu_addr & 0xfffffffc);
        radeon_ring_write(ring, upper_32_bits(rdev->vram_scratch.gpu_addr) & 0xff);
        radeon_ring_write(ring, 0xDEADBEEF);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        for (i = 0; i < rdev->usec_timeout; i++) {
                tmp = readl(ptr);
@@ -368,7 +368,7 @@ int r600_dma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
        ib.ptr[3] = 0xDEADBEEF;
        ib.length_dw = 4;
 
-       r = radeon_ib_schedule(rdev, &ib, NULL);
+       r = radeon_ib_schedule(rdev, &ib, NULL, false);
        if (r) {
                radeon_ib_free(rdev, &ib);
                DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
@@ -493,7 +493,7 @@ int r600_copy_dma(struct radeon_device *rdev,
                return r;
        }
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        radeon_semaphore_free(rdev, &sem, *fence);
 
        return r;

diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index e715e0c1f5d8c774b0f2a024cb0e256bbfe86c01..b281886f6f5194609caab9bddb632eadea191004 100644 (file)
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -968,7 +968,7 @@ int radeon_ib_get(struct radeon_device *rdev, int ring,
                  unsigned size);
 void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib);
 int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
-                      struct radeon_ib *const_ib);
+                      struct radeon_ib *const_ib, bool hdp_flush);
 int radeon_ib_pool_init(struct radeon_device *rdev);
 void radeon_ib_pool_fini(struct radeon_device *rdev);
 int radeon_ib_ring_tests(struct radeon_device *rdev);
@@ -978,8 +978,10 @@ bool radeon_ring_supports_scratch_reg(struct radeon_device *rdev,
 void radeon_ring_free_size(struct radeon_device *rdev, struct radeon_ring *cp);
 int radeon_ring_alloc(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ndw);
 int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ndw);
-void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *cp);
-void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *cp);
+void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *cp,
+                       bool hdp_flush);
+void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *cp,
+                              bool hdp_flush);
 void radeon_ring_undo(struct radeon_ring *ring);
 void radeon_ring_unlock_undo(struct radeon_device *rdev, struct radeon_ring *cp);
 int radeon_ring_test(struct radeon_device *rdev, struct radeon_ring *cp);

diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index cb12df784d83ba9f21ce9a513c4cb625d485f8e4..83f382e8e40e35cb380d4fc6fd835a7ddac9c01a 100644 (file)
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -451,7 +451,7 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev,
                radeon_vce_note_usage(rdev);
 
        radeon_cs_sync_rings(parser);
-       r = radeon_ib_schedule(rdev, &parser->ib, NULL);
+       r = radeon_ib_schedule(rdev, &parser->ib, NULL, true);
        if (r) {
                DRM_ERROR("Failed to schedule IB !\n");
        }
@@ -542,9 +542,9 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev,
 
        if ((rdev->family >= CHIP_TAHITI) &&
            (parser->chunk_const_ib_idx != -1)) {
-               r = radeon_ib_schedule(rdev, &parser->ib, &parser->const_ib);
+               r = radeon_ib_schedule(rdev, &parser->ib, &parser->const_ib, true);
        } else {
-               r = radeon_ib_schedule(rdev, &parser->ib, NULL);
+               r = radeon_ib_schedule(rdev, &parser->ib, NULL, true);
        }
 
 out:

diff --git a/drivers/gpu/drm/radeon/radeon_ib.c b/drivers/gpu/drm/radeon/radeon_ib.c
index 65b0c213488dcf084e08c2ddf95c92c28f367c83..5bf2c0a05827f57b459ca1be85e268af0bbf32ce 100644 (file)
--- a/drivers/gpu/drm/radeon/radeon_ib.c
+++ b/drivers/gpu/drm/radeon/radeon_ib.c
@@ -107,6 +107,7 @@ void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib)
  * @rdev: radeon_device pointer
  * @ib: IB object to schedule
  * @const_ib: Const IB to schedule (SI only)
+ * @hdp_flush: Whether or not to perform an HDP cache flush
  *
  * Schedule an IB on the associated ring (all asics).
  * Returns 0 on success, error on failure.
@@ -122,7 +123,7 @@ void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib)
  * to SI there was just a DE IB.
  */
 int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
-                      struct radeon_ib *const_ib)
+                      struct radeon_ib *const_ib, bool hdp_flush)
 {
        struct radeon_ring *ring = &rdev->ring[ib->ring];
        int r = 0;
@@ -176,7 +177,7 @@ int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
        if (ib->vm)
                radeon_vm_fence(rdev, ib->vm, ib->fence);
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, hdp_flush);
        return 0;
 }
 

diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index 5b4e0cf231a04d0a104868ed590f1324afe70be1..0678998dbcb877695e77ef323110be636a8d1f7a 100644 (file)
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -177,16 +177,18 @@ int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *ring, unsig
  *
  * @rdev: radeon_device pointer
  * @ring: radeon_ring structure holding ring information
+ * @hdp_flush: Whether or not to perform an HDP cache flush
  *
  * Update the wptr (write pointer) to tell the GPU to
  * execute new commands on the ring buffer (all asics).
  */
-void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring)
+void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring,
+                       bool hdp_flush)
 {
        /* If we are emitting the HDP flush via the ring buffer, we need to
         * do it before padding.
         */
-       if (rdev->asic->ring[ring->idx]->hdp_flush)
+       if (hdp_flush && rdev->asic->ring[ring->idx]->hdp_flush)
                rdev->asic->ring[ring->idx]->hdp_flush(rdev, ring);
        /* We pad to match fetch size */
        while (ring->wptr & ring->align_mask) {
@@ -196,7 +198,7 @@ void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring)
        /* If we are emitting the HDP flush via MMIO, we need to do it after
         * all CPU writes to VRAM finished.
         */
-       if (rdev->asic->mmio_hdp_flush)
+       if (hdp_flush && rdev->asic->mmio_hdp_flush)
                rdev->asic->mmio_hdp_flush(rdev);
        radeon_ring_set_wptr(rdev, ring);
 }
@@ -207,12 +209,14 @@ void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring)
  *
  * @rdev: radeon_device pointer
  * @ring: radeon_ring structure holding ring information
+ * @hdp_flush: Whether or not to perform an HDP cache flush
  *
  * Call radeon_ring_commit() then unlock the ring (all asics).
  */
-void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *ring)
+void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *ring,
+                              bool hdp_flush)
 {
-       radeon_ring_commit(rdev, ring);
+       radeon_ring_commit(rdev, ring, hdp_flush);
        mutex_unlock(&rdev->ring_lock);
 }
 
@@ -372,7 +376,7 @@ int radeon_ring_restore(struct radeon_device *rdev, struct radeon_ring *ring,
                radeon_ring_write(ring, data[i]);
        }
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        kfree(data);
        return 0;
 }

diff --git a/drivers/gpu/drm/radeon/radeon_semaphore.c b/drivers/gpu/drm/radeon/radeon_semaphore.c
index dbd6bcde92de412a87379d7da23a4cea74b40649..56d9fd66d8aed379c417d1ba4a47de7aa65ba472 100644 (file)
--- a/drivers/gpu/drm/radeon/radeon_semaphore.c
+++ b/drivers/gpu/drm/radeon/radeon_semaphore.c
@@ -179,7 +179,7 @@ int radeon_semaphore_sync_rings(struct radeon_device *rdev,
                        continue;
                }
 
-               radeon_ring_commit(rdev, &rdev->ring[i]);
+               radeon_ring_commit(rdev, &rdev->ring[i], false);
                radeon_fence_note_sync(fence, ring);
 
                semaphore->gpu_addr += 8;

diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c
index 5adf4207453d7eb041f6679722283a36ae5a0931..17bc3dced9f1db921a11f80b1167f520261ed0b1 100644 (file)
--- a/drivers/gpu/drm/radeon/radeon_test.c
+++ b/drivers/gpu/drm/radeon/radeon_test.c
@@ -288,7 +288,7 @@ static int radeon_test_create_and_emit_fence(struct radeon_device *rdev,
                        return r;
                }
                radeon_fence_emit(rdev, fence, ring->idx);
-               radeon_ring_unlock_commit(rdev, ring);
+               radeon_ring_unlock_commit(rdev, ring, false);
        }
        return 0;
 }
@@ -313,7 +313,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
                goto out_cleanup;
        }
        radeon_semaphore_emit_wait(rdev, ringA->idx, semaphore);
-       radeon_ring_unlock_commit(rdev, ringA);
+       radeon_ring_unlock_commit(rdev, ringA, false);
 
        r = radeon_test_create_and_emit_fence(rdev, ringA, &fence1);
        if (r)
@@ -325,7 +325,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
                goto out_cleanup;
        }
        radeon_semaphore_emit_wait(rdev, ringA->idx, semaphore);
-       radeon_ring_unlock_commit(rdev, ringA);
+       radeon_ring_unlock_commit(rdev, ringA, false);
 
        r = radeon_test_create_and_emit_fence(rdev, ringA, &fence2);
        if (r)
@@ -344,7 +344,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
                goto out_cleanup;
        }
        radeon_semaphore_emit_signal(rdev, ringB->idx, semaphore);
-       radeon_ring_unlock_commit(rdev, ringB);
+       radeon_ring_unlock_commit(rdev, ringB, false);
 
        r = radeon_fence_wait(fence1, false);
        if (r) {
@@ -365,7 +365,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
                goto out_cleanup;
        }
        radeon_semaphore_emit_signal(rdev, ringB->idx, semaphore);
-       radeon_ring_unlock_commit(rdev, ringB);
+       radeon_ring_unlock_commit(rdev, ringB, false);
 
        r = radeon_fence_wait(fence2, false);
        if (r) {
@@ -408,7 +408,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev,
                goto out_cleanup;
        }
        radeon_semaphore_emit_wait(rdev, ringA->idx, semaphore);
-       radeon_ring_unlock_commit(rdev, ringA);
+       radeon_ring_unlock_commit(rdev, ringA, false);
 
        r = radeon_test_create_and_emit_fence(rdev, ringA, &fenceA);
        if (r)
@@ -420,7 +420,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev,
                goto out_cleanup;
        }
        radeon_semaphore_emit_wait(rdev, ringB->idx, semaphore);
-       radeon_ring_unlock_commit(rdev, ringB);
+       radeon_ring_unlock_commit(rdev, ringB, false);
        r = radeon_test_create_and_emit_fence(rdev, ringB, &fenceB);
        if (r)
                goto out_cleanup;
@@ -442,7 +442,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev,
                goto out_cleanup;
        }
        radeon_semaphore_emit_signal(rdev, ringC->idx, semaphore);
-       radeon_ring_unlock_commit(rdev, ringC);
+       radeon_ring_unlock_commit(rdev, ringC, false);
 
        for (i = 0; i < 30; ++i) {
                mdelay(100);
@@ -468,7 +468,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev,
                goto out_cleanup;
        }
        radeon_semaphore_emit_signal(rdev, ringC->idx, semaphore);
-       radeon_ring_unlock_commit(rdev, ringC);
+       radeon_ring_unlock_commit(rdev, ringC, false);
 
        mdelay(1000);
 

diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c
index 6bf55ec85b62f7b3ee817356c6167181938a3dc1..341848a143761dd8236edc7862f79482d5d2d19d 100644 (file)
--- a/drivers/gpu/drm/radeon/radeon_uvd.c
+++ b/drivers/gpu/drm/radeon/radeon_uvd.c
@@ -646,7 +646,7 @@ static int radeon_uvd_send_msg(struct radeon_device *rdev,
                ib.ptr[i] = PACKET2(0);
        ib.length_dw = 16;
 
-       r = radeon_ib_schedule(rdev, &ib, NULL);
+       r = radeon_ib_schedule(rdev, &ib, NULL, false);
        if (r)
                goto err;
        ttm_eu_fence_buffer_objects(&ticket, &head, ib.fence);

diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c
index f9b70a43aa524f4ccbcff640e28e0d21573463b6..c7190aadbd8981bee97c0c4ed4d923cdf7880b52 100644 (file)
--- a/drivers/gpu/drm/radeon/radeon_vce.c
+++ b/drivers/gpu/drm/radeon/radeon_vce.c
@@ -368,7 +368,7 @@ int radeon_vce_get_create_msg(struct radeon_device *rdev, int ring,
        for (i = ib.length_dw; i < ib_size_dw; ++i)
                ib.ptr[i] = 0x0;
 
-       r = radeon_ib_schedule(rdev, &ib, NULL);
+       r = radeon_ib_schedule(rdev, &ib, NULL, false);
        if (r) {
                DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
        }
@@ -425,7 +425,7 @@ int radeon_vce_get_destroy_msg(struct radeon_device *rdev, int ring,
        for (i = ib.length_dw; i < ib_size_dw; ++i)
                ib.ptr[i] = 0x0;
 
-       r = radeon_ib_schedule(rdev, &ib, NULL);
+       r = radeon_ib_schedule(rdev, &ib, NULL, false);
        if (r) {
                DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
        }
@@ -715,7 +715,7 @@ int radeon_vce_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
                return r;
        }
        radeon_ring_write(ring, VCE_CMD_END);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        for (i = 0; i < rdev->usec_timeout; i++) {
                if (vce_v1_0_get_rptr(rdev, ring) != rptr)

diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index 058f200853696d4b6db63c6d419977b8cc5465f4..832ef320c26de5fd92cb9ad858b9c3aaeb061d70 100644 (file)
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -422,7 +422,7 @@ static int radeon_vm_clear_bo(struct radeon_device *rdev,
        radeon_asic_vm_pad_ib(rdev, &ib);
        WARN_ON(ib.length_dw > 64);
 
-       r = radeon_ib_schedule(rdev, &ib, NULL);
+       r = radeon_ib_schedule(rdev, &ib, NULL, false);
        if (r)
                 goto error;
 
@@ -699,7 +699,7 @@ int radeon_vm_update_page_directory(struct radeon_device *rdev,
                radeon_semaphore_sync_to(ib.semaphore, pd->tbo.sync_obj);
                radeon_semaphore_sync_to(ib.semaphore, vm->last_id_use);
                WARN_ON(ib.length_dw > ndw);
-               r = radeon_ib_schedule(rdev, &ib, NULL);
+               r = radeon_ib_schedule(rdev, &ib, NULL, false);
                if (r) {
                        radeon_ib_free(rdev, &ib);
                        return r;
@@ -963,7 +963,7 @@ int radeon_vm_bo_update(struct radeon_device *rdev,
        WARN_ON(ib.length_dw > ndw);
 
        radeon_semaphore_sync_to(ib.semaphore, vm->fence);
-       r = radeon_ib_schedule(rdev, &ib, NULL);
+       r = radeon_ib_schedule(rdev, &ib, NULL, false);
        if (r) {
                radeon_ib_free(rdev, &ib);
                return r;

diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 3e21e869015fece1124e7093bfd028e99be2e909..8a477bf1fdb31529173234f8f30a0b4e3fb9c608 100644 (file)
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -124,7 +124,7 @@ void rv515_ring_start(struct radeon_device *rdev, struct radeon_ring *ring)
        radeon_ring_write(ring, GEOMETRY_ROUND_NEAREST | COLOR_ROUND_NEAREST);
        radeon_ring_write(ring, PACKET0(0x20C8, 0));
        radeon_ring_write(ring, 0);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 }
 
 int rv515_mc_wait_for_idle(struct radeon_device *rdev)

diff --git a/drivers/gpu/drm/radeon/rv770_dma.c b/drivers/gpu/drm/radeon/rv770_dma.c
index bbf2e076ee457816924a736c57d2192cabba88ac..74426ac2bb5c2f39a26fb07eb76538afcc7061a9 100644 (file)
--- a/drivers/gpu/drm/radeon/rv770_dma.c
+++ b/drivers/gpu/drm/radeon/rv770_dma.c
@@ -90,7 +90,7 @@ int rv770_copy_dma(struct radeon_device *rdev,
                return r;
        }
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        radeon_semaphore_free(rdev, &sem, *fence);
 
        return r;

diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index dbd9d8101f056de631050ff0c266be14536e7b47..7e58423c0bf65b2a85ee349271750901ee42133a 100644 (file)
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -3541,7 +3541,7 @@ static int si_cp_start(struct radeon_device *rdev)
        radeon_ring_write(ring, PACKET3_BASE_INDEX(CE_PARTITION_BASE));
        radeon_ring_write(ring, 0xc000);
        radeon_ring_write(ring, 0xe000);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        si_cp_enable(rdev, true);
 
@@ -3570,7 +3570,7 @@ static int si_cp_start(struct radeon_device *rdev)
        radeon_ring_write(ring, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
        radeon_ring_write(ring, 0x00000010); /* VGT_OUT_DEALLOC_CNTL */
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
        for (i = RADEON_RING_TYPE_GFX_INDEX; i <= CAYMAN_RING_TYPE_CP2_INDEX; ++i) {
                ring = &rdev->ring[i];
@@ -3580,7 +3580,7 @@ static int si_cp_start(struct radeon_device *rdev)
                radeon_ring_write(ring, PACKET3_COMPUTE(PACKET3_CLEAR_STATE, 0));
                radeon_ring_write(ring, 0);
 
-               radeon_ring_unlock_commit(rdev, ring);
+               radeon_ring_unlock_commit(rdev, ring, false);
        }
 
        return 0;

diff --git a/drivers/gpu/drm/radeon/si_dma.c b/drivers/gpu/drm/radeon/si_dma.c
index 7165051294504470c24e054b384802f05dc24de7..7c22baaf94dbf9ac598f745bfb35b2f5ba5138f2 100644 (file)
--- a/drivers/gpu/drm/radeon/si_dma.c
+++ b/drivers/gpu/drm/radeon/si_dma.c
@@ -275,7 +275,7 @@ int si_copy_dma(struct radeon_device *rdev,
                return r;
        }
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        radeon_semaphore_free(rdev, &sem, *fence);
 
        return r;

diff --git a/drivers/gpu/drm/radeon/uvd_v1_0.c b/drivers/gpu/drm/radeon/uvd_v1_0.c
index be42c8125203b22bd62d9fc4be8510672ba577b3..cda391347286c3545bdd952234b3b5fb7186a5d9 100644 (file)
--- a/drivers/gpu/drm/radeon/uvd_v1_0.c
+++ b/drivers/gpu/drm/radeon/uvd_v1_0.c
@@ -124,7 +124,7 @@ int uvd_v1_0_init(struct radeon_device *rdev)
        radeon_ring_write(ring, PACKET0(UVD_SEMA_CNTL, 0));
        radeon_ring_write(ring, 3);
 
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
 
 done:
        /* lower clocks again */
@@ -331,7 +331,7 @@ int uvd_v1_0_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
        }
        radeon_ring_write(ring, PACKET0(UVD_CONTEXT_ID, 0));
        radeon_ring_write(ring, 0xDEADBEEF);
-       radeon_ring_unlock_commit(rdev, ring);
+       radeon_ring_unlock_commit(rdev, ring, false);
        for (i = 0; i < rdev->usec_timeout; i++) {
                tmp = RREG32(UVD_CONTEXT_ID);
                if (tmp == 0xDEADBEEF)