Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux

Pull drm updates from Dave Airlie:
 "This is the main drm pull request for 4.6 kernel.

  Overall the coolest thing here for me is the nouveau maxwell signed
  firmware support from NVidia, it's taken a long while to extract this
  from them.

  I also wish the ARM vendors just designed one set of display IP, ARM
  display block proliferation is definitely increasing.

  Core:
     - drm_event cleanups
     - Internal API cleanup making mode_fixup optional.
     - Apple GMUX vga switcheroo support.
     - DP AUX testing interface

  Panel:
     - Refactoring of DSI core for use over more transports.

  New driver:
     - ARM hdlcd driver

  i915:
     - FBC/PSR (framebuffer compression, panel self refresh) enabled by default.
     - Ongoing atomic display support work
     - Ongoing runtime PM work
     - Pixel clock limit checks
     - VBT DSI description support
     - GEM fixes
     - GuC firmware scheduler enhancements

  amdkfd:
     - Deferred probing fixes to avoid make file or link ordering.

  amdgpu/radeon:
     - ACP support for i2s audio support.
     - Command Submission/GPU scheduler/GPUVM optimisations
     - Initial GPU reset support for amdgpu

  vmwgfx:
     - Support for DX10 gen mipmaps
     - Pageflipping and other fixes.

  exynos:
     - Exynos5420 SoC support for FIMD
     - Exynos5422 SoC support for MIPI-DSI

  nouveau:
     - GM20x secure boot support - adds acceleration for Maxwell GPUs.
     - GM200 support
     - GM20B clock driver support
     - Power sensors work

  etnaviv:
     - Correctness fixes for GPU cache flushing
     - Better support for i.MX6 systems.

  imx-drm:
     - VBlank IRQ support
     - Fence support
     - OF endpoint support

  msm:
     - HDMI support for 8996 (snapdragon 820)
     - Adreno 430 support
     - Timestamp queries support

  virtio-gpu:
     - Fixes for Android support.

  rockchip:
     - Add support for Innosilicion HDMI

  rcar-du:
     - Support for 4 crtcs
     - R8A7795 support
     - RCar Gen 3 support

  omapdrm:
     - HDMI interlace output support
     - dma-buf import support
     - Refactoring to remove a lot of legacy code.

  tilcdc:
     - Rewrite of pageflipping code
     - dma-buf support
     - pinctrl support

  vc4:
     - HDMI modesetting bug fixes
     - Significant 3D performance improvement.

  fsl-dcu (FreeScale):
     - Lots of fixes

  tegra:
     - Two small fixes

  sti:
     - Atomic support for planes
     - Improved HDMI support"

* 'drm-next' of git://people.freedesktop.org/~airlied/linux: (1063 commits)
  drm/amdgpu: release_pages requires linux/pagemap.h
  drm/sti: restore mode_fixup callback
  drm/amdgpu/gfx7: add MTYPE definition
  drm/amdgpu: removing BO_VAs shouldn't be interruptible
  drm/amd/powerplay: show uvd/vce power gate enablement for tonga.
  drm/amd/powerplay: show uvd/vce power gate info for fiji
  drm/amdgpu: use sched fence if possible
  drm/amdgpu: move ib.fence to job.fence
  drm/amdgpu: give a fence param to ib_free
  drm/amdgpu: include the right version of gmc header files for iceland
  drm/radeon: fix indentation.
  drm/amd/powerplay: add uvd/vce dpm enabling flag to fix the performance issue for CZ
  drm/amdgpu: switch back to 32bit hw fences v2
  drm/amdgpu: remove amdgpu_fence_is_signaled
  drm/amdgpu: drop the extra fence range check v2
  drm/amdgpu: signal fences directly in amdgpu_fence_process
  drm/amdgpu: cleanup amdgpu_fence_wait_empty v2
  drm/amdgpu: keep all fences in an RCU protected array v2
  drm/amdgpu: add number of hardware submissions to amdgpu_fence_driver_init_ring
  drm/amdgpu: RCU protected amd_sched_fence_release
  ...

1 files changed, 57 insertions, 47 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index c55ecf0ea845..d3ac3298fba8 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -212,7 +212,7 @@ static void cik_sdma_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 static void cik_sdma_ring_emit_ib(struct amdgpu_ring *ring,
 			   struct amdgpu_ib *ib)
 {
-	u32 extra_bits = (ib->vm ? ib->vm->ids[ring->idx].id : 0) & 0xf;
+	u32 extra_bits = ib->vm_id & 0xf;
 	u32 next_rptr = ring->wptr + 5;
 
 	while ((next_rptr & 7) != 4)
@@ -261,6 +261,13 @@ static void cik_sdma_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 	amdgpu_ring_write(ring, (0xfff << 16) | 10); /* retry count, poll interval */
 }
 
+static void cik_sdma_ring_emit_hdp_invalidate(struct amdgpu_ring *ring)
+{
+	amdgpu_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SRBM_WRITE, 0, 0xf000));
+	amdgpu_ring_write(ring, mmHDP_DEBUG0);
+	amdgpu_ring_write(ring, 1);
+}
+
 /**
  * cik_sdma_ring_emit_fence - emit a fence on the DMA ring
  *
@@ -295,30 +302,6 @@ static void cik_sdma_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq
 }
 
 /**
- * cik_sdma_ring_emit_semaphore - emit a semaphore on the dma ring
- *
- * @ring: amdgpu_ring structure holding ring information
- * @semaphore: amdgpu semaphore object
- * @emit_wait: wait or signal semaphore
- *
- * Add a DMA semaphore packet to the ring wait on or signal
- * other rings (CIK).
- */
-static bool cik_sdma_ring_emit_semaphore(struct amdgpu_ring *ring,
-					 struct amdgpu_semaphore *semaphore,
-					 bool emit_wait)
-{
-	u64 addr = semaphore->gpu_addr;
-	u32 extra_bits = emit_wait ? 0 : SDMA_SEMAPHORE_EXTRA_S;
-
-	amdgpu_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SEMAPHORE, 0, extra_bits));
-	amdgpu_ring_write(ring, addr & 0xfffffff8);
-	amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
-
-	return true;
-}
-
-/**
  * cik_sdma_gfx_stop - stop the gfx async dma engines
  *
  * @adev: amdgpu_device pointer
@@ -417,6 +400,9 @@ static int cik_sdma_gfx_resume(struct amdgpu_device *adev)
 		cik_srbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
+		WREG32(mmSDMA0_TILING_CONFIG + sdma_offsets[i],
+		       adev->gfx.config.gb_addr_config & 0x70);
+
 		WREG32(mmSDMA0_SEM_INCOMPLETE_TIMER_CNTL + sdma_offsets[i], 0);
 		WREG32(mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL + sdma_offsets[i], 0);
 
@@ -584,7 +570,7 @@ static int cik_sdma_ring_test_ring(struct amdgpu_ring *ring)
 	tmp = 0xCAFEDEAD;
 	adev->wb.wb[index] = cpu_to_le32(tmp);
 
-	r = amdgpu_ring_lock(ring, 5);
+	r = amdgpu_ring_alloc(ring, 5);
 	if (r) {
 		DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
 		amdgpu_wb_free(adev, index);
@@ -595,7 +581,7 @@ static int cik_sdma_ring_test_ring(struct amdgpu_ring *ring)
 	amdgpu_ring_write(ring, upper_32_bits(gpu_addr));
 	amdgpu_ring_write(ring, 1); /* number of DWs to follow */
 	amdgpu_ring_write(ring, 0xDEADBEEF);
-	amdgpu_ring_unlock_commit(ring);
+	amdgpu_ring_commit(ring);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
 		tmp = le32_to_cpu(adev->wb.wb[index]);
@@ -645,7 +631,7 @@ static int cik_sdma_ring_test_ib(struct amdgpu_ring *ring)
 	tmp = 0xCAFEDEAD;
 	adev->wb.wb[index] = cpu_to_le32(tmp);
 	memset(&ib, 0, sizeof(ib));
-	r = amdgpu_ib_get(ring, NULL, 256, &ib);
+	r = amdgpu_ib_get(adev, NULL, 256, &ib);
 	if (r) {
 		DRM_ERROR("amdgpu: failed to get ib (%d).\n", r);
 		goto err0;
@@ -657,9 +643,7 @@ static int cik_sdma_ring_test_ib(struct amdgpu_ring *ring)
 	ib.ptr[3] = 1;
 	ib.ptr[4] = 0xDEADBEEF;
 	ib.length_dw = 5;
-	r = amdgpu_sched_ib_submit_kernel_helper(adev, ring, &ib, 1, NULL,
-						 AMDGPU_FENCE_OWNER_UNDEFINED,
-						 &f);
+	r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f);
 	if (r)
 		goto err1;
 
@@ -685,7 +669,8 @@ static int cik_sdma_ring_test_ib(struct amdgpu_ring *ring)
 
 err1:
 	fence_put(f);
-	amdgpu_ib_free(adev, &ib);
+	amdgpu_ib_free(adev, &ib, NULL);
+	fence_put(f);
 err0:
 	amdgpu_wb_free(adev, index);
 	return r;
@@ -738,7 +723,7 @@ static void cik_sdma_vm_copy_pte(struct amdgpu_ib *ib,
  * Update PTEs by writing them manually using sDMA (CIK).
  */
 static void cik_sdma_vm_write_pte(struct amdgpu_ib *ib,
-				  uint64_t pe,
+				  const dma_addr_t *pages_addr, uint64_t pe,
 				  uint64_t addr, unsigned count,
 				  uint32_t incr, uint32_t flags)
 {
@@ -757,14 +742,7 @@ static void cik_sdma_vm_write_pte(struct amdgpu_ib *ib,
 		ib->ptr[ib->length_dw++] = upper_32_bits(pe);
 		ib->ptr[ib->length_dw++] = ndw;
 		for (; ndw > 0; ndw -= 2, --count, pe += 8) {
-			if (flags & AMDGPU_PTE_SYSTEM) {
-				value = amdgpu_vm_map_gart(ib->ring->adev, addr);
-				value &= 0xFFFFFFFFFFFFF000ULL;
-			} else if (flags & AMDGPU_PTE_VALID) {
-				value = addr;
-			} else {
-				value = 0;
-			}
+			value = amdgpu_vm_map_gart(pages_addr, addr);
 			addr += incr;
 			value |= flags;
 			ib->ptr[ib->length_dw++] = value;
@@ -827,9 +805,9 @@ static void cik_sdma_vm_set_pte_pde(struct amdgpu_ib *ib,
  * @ib: indirect buffer to fill with padding
  *
  */
-static void cik_sdma_vm_pad_ib(struct amdgpu_ib *ib)
+static void cik_sdma_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib)
 {
-	struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ib->ring);
+	struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
 	u32 pad_count;
 	int i;
 
@@ -845,6 +823,30 @@ static void cik_sdma_vm_pad_ib(struct amdgpu_ib *ib)
 }
 
 /**
+ * cik_sdma_ring_emit_pipeline_sync - sync the pipeline
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Make sure all previous operations are completed (CIK).
+ */
+static void cik_sdma_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
+{
+	uint32_t seq = ring->fence_drv.sync_seq;
+	uint64_t addr = ring->fence_drv.gpu_addr;
+
+	/* wait for idle */
+	amdgpu_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_POLL_REG_MEM, 0,
+					    SDMA_POLL_REG_MEM_EXTRA_OP(0) |
+					    SDMA_POLL_REG_MEM_EXTRA_FUNC(3) | /* equal */
+					    SDMA_POLL_REG_MEM_EXTRA_M));
+	amdgpu_ring_write(ring, addr & 0xfffffffc);
+	amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
+	amdgpu_ring_write(ring, seq); /* reference */
+	amdgpu_ring_write(ring, 0xfffffff); /* mask */
+	amdgpu_ring_write(ring, (0xfff << 16) | 4); /* retry count, poll interval */
+}
+
+/**
  * cik_sdma_ring_emit_vm_flush - cik vm flush using sDMA
  *
  * @ring: amdgpu_ring pointer
@@ -1097,6 +1099,8 @@ static void cik_sdma_print_status(void *handle)
 			 i, RREG32(mmSDMA0_GFX_RB_BASE + sdma_offsets[i]));
 		dev_info(adev->dev, "  SDMA%d_GFX_RB_BASE_HI=0x%08X\n",
 			 i, RREG32(mmSDMA0_GFX_RB_BASE_HI + sdma_offsets[i]));
+		dev_info(adev->dev, "  SDMA%d_TILING_CONFIG=0x%08X\n",
+			 i, RREG32(mmSDMA0_TILING_CONFIG + sdma_offsets[i]));
 		mutex_lock(&adev->srbm_mutex);
 		for (j = 0; j < 16; j++) {
 			cik_srbm_select(adev, 0, 0, 0, j);
@@ -1297,12 +1301,14 @@ static const struct amdgpu_ring_funcs cik_sdma_ring_funcs = {
 	.parse_cs = NULL,
 	.emit_ib = cik_sdma_ring_emit_ib,
 	.emit_fence = cik_sdma_ring_emit_fence,
-	.emit_semaphore = cik_sdma_ring_emit_semaphore,
+	.emit_pipeline_sync = cik_sdma_ring_emit_pipeline_sync,
 	.emit_vm_flush = cik_sdma_ring_emit_vm_flush,
 	.emit_hdp_flush = cik_sdma_ring_emit_hdp_flush,
+	.emit_hdp_invalidate = cik_sdma_ring_emit_hdp_invalidate,
 	.test_ring = cik_sdma_ring_test_ring,
 	.test_ib = cik_sdma_ring_test_ib,
 	.insert_nop = cik_sdma_ring_insert_nop,
+	.pad_ib = cik_sdma_ring_pad_ib,
 };
 
 static void cik_sdma_set_ring_funcs(struct amdgpu_device *adev)
@@ -1399,14 +1405,18 @@ static const struct amdgpu_vm_pte_funcs cik_sdma_vm_pte_funcs = {
 	.copy_pte = cik_sdma_vm_copy_pte,
 	.write_pte = cik_sdma_vm_write_pte,
 	.set_pte_pde = cik_sdma_vm_set_pte_pde,
-	.pad_ib = cik_sdma_vm_pad_ib,
 };
 
 static void cik_sdma_set_vm_pte_funcs(struct amdgpu_device *adev)
 {
+	unsigned i;
+
 	if (adev->vm_manager.vm_pte_funcs == NULL) {
 		adev->vm_manager.vm_pte_funcs = &cik_sdma_vm_pte_funcs;
-		adev->vm_manager.vm_pte_funcs_ring = &adev->sdma.instance[0].ring;
-		adev->vm_manager.vm_pte_funcs_ring->is_pte_ring = true;
+		for (i = 0; i < adev->sdma.num_instances; i++)
+			adev->vm_manager.vm_pte_rings[i] =
+				&adev->sdma.instance[i].ring;
+
+		adev->vm_manager.vm_pte_num_rings = adev->sdma.num_instances;
 	}
 }