Merge branch 'drm-next-4.20' of git://people.freedesktop.org/~agd5f/linux into drm-next

This is a new pull for drm-next on top of last weeks with the following
changes:
- Fixed 64 bit divide
- Fixed vram type on vega20
- Misc vega20 fixes
- Misc DC fixes
- Fix GDS/GWS/OA domain handling

Previous changes from last week:
amdgpu/kfd:
- Picasso (new APU) support
- Raven2 (new APU) support
- Vega20 enablement
- ACP powergating improvements
- Add ABGR/XBGR display support
- VCN JPEG engine support
- Initial xGMI support
- Use load balancing for engine scheduling
- Lots of new documentation
- Rework and clean up i2c and aux handling in DC
- Add DP YCbCr 4:2:0 support in DC
- Add DMCU firmware loading for Raven (used for ABM and PSR)
- New debugfs features in DC
- LVDS support in DC
- Implement wave kill for gfx/compute (light weight reset for shaders)
- Use AGP aperture to avoid gart mappings when possible
- GPUVM performance improvements
- Bulk moves for more efficient GPUVM LRU handling
- Merge amdgpu and amdkfd into one module
- Enable gfxoff and stutter mode on Raven
- Misc cleanups

Scheduler:
- Load balancing support
- Bug fixes

ttm:
- Bulk move functionality
- Bug fixes

radeon:
- Misc cleanups
Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Alex Deucher <alexdeucher@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180920150438.12693-1-alexander.deucher@amd.com

Documentation/gpu/drm-mm.rst

@@ -505,7 +505,7 @@ GPU Scheduler
 Overview
 --------
 
-.. kernel-doc:: drivers/gpu/drm/scheduler/gpu_scheduler.c
+.. kernel-doc:: drivers/gpu/drm/scheduler/sched_main.c
    :doc: Overview
 
 Scheduler Function References
@@ -514,5 +514,5 @@ Scheduler Function References
 .. kernel-doc:: include/drm/gpu_scheduler.h
    :internal:
 
-.. kernel-doc:: drivers/gpu/drm/scheduler/gpu_scheduler.c
+.. kernel-doc:: drivers/gpu/drm/scheduler/sched_main.c
    :export:

drivers/gpu/drm/Kconfig

@@ -285,8 +285,6 @@ source "drivers/gpu/drm/bridge/Kconfig"
 
 source "drivers/gpu/drm/sti/Kconfig"
 
-source "drivers/gpu/drm/amd/amdkfd/Kconfig"
-
 source "drivers/gpu/drm/imx/Kconfig"
 
 source "drivers/gpu/drm/v3d/Kconfig"

drivers/gpu/drm/amd/amdgpu/Kconfig

@@ -42,3 +42,4 @@ config DRM_AMDGPU_GART_DEBUGFS
 
 source "drivers/gpu/drm/amd/acp/Kconfig"
 source "drivers/gpu/drm/amd/display/Kconfig"
+source "drivers/gpu/drm/amd/amdkfd/Kconfig"

drivers/gpu/drm/amd/amdgpu/Makefile

@@ -35,7 +35,8 @@ ccflags-y := -I$(FULL_AMD_PATH)/include/asic_reg \
 	-I$(FULL_AMD_DISPLAY_PATH) \
 	-I$(FULL_AMD_DISPLAY_PATH)/include \
 	-I$(FULL_AMD_DISPLAY_PATH)/dc \
-	-I$(FULL_AMD_DISPLAY_PATH)/amdgpu_dm
+	-I$(FULL_AMD_DISPLAY_PATH)/amdgpu_dm \
+	-I$(FULL_AMD_PATH)/amdkfd
 
 amdgpu-y := amdgpu_drv.o
 
@@ -51,8 +52,8 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \
 	amdgpu_prime.o amdgpu_vm.o amdgpu_ib.o amdgpu_pll.o \
 	amdgpu_ucode.o amdgpu_bo_list.o amdgpu_ctx.o amdgpu_sync.o \
 	amdgpu_gtt_mgr.o amdgpu_vram_mgr.o amdgpu_virt.o amdgpu_atomfirmware.o \
-	amdgpu_queue_mgr.o amdgpu_vf_error.o amdgpu_sched.o amdgpu_debugfs.o \
-	amdgpu_ids.o
+	amdgpu_vf_error.o amdgpu_sched.o amdgpu_debugfs.o amdgpu_ids.o \
+	amdgpu_gmc.o amdgpu_xgmi.o
 
 # add asic specific block
 amdgpu-$(CONFIG_DRM_AMDGPU_CIK)+= cik.o cik_ih.o kv_smc.o kv_dpm.o \
@@ -62,7 +63,7 @@ amdgpu-$(CONFIG_DRM_AMDGPU_SI)+= si.o gmc_v6_0.o gfx_v6_0.o si_ih.o si_dma.o dce
 
 amdgpu-y += \
 	vi.o mxgpu_vi.o nbio_v6_1.o soc15.o emu_soc.o mxgpu_ai.o nbio_v7_0.o vega10_reg_init.o \
-	vega20_reg_init.o
+	vega20_reg_init.o nbio_v7_4.o
 
 # add DF block
 amdgpu-y += \
@@ -73,7 +74,7 @@ amdgpu-y += \
 amdgpu-y += \
 	gmc_v7_0.o \
 	gmc_v8_0.o \
-	gfxhub_v1_0.o mmhub_v1_0.o gmc_v9_0.o
+	gfxhub_v1_0.o mmhub_v1_0.o gmc_v9_0.o gfxhub_v1_1.o
 
 # add IH block
 amdgpu-y += \
@@ -88,7 +89,8 @@ amdgpu-y += \
 amdgpu-y += \
 	amdgpu_psp.o \
 	psp_v3_1.o \
-	psp_v10_0.o
+	psp_v10_0.o \
+	psp_v11_0.o
 
 # add SMC block
 amdgpu-y += \
@@ -108,6 +110,7 @@ amdgpu-y += \
 
 # add async DMA block
 amdgpu-y += \
+	amdgpu_sdma.o \
 	sdma_v2_4.o \
 	sdma_v3_0.o \
 	sdma_v4_0.o
@@ -134,6 +137,9 @@ amdgpu-y += \
 amdgpu-y += amdgpu_amdkfd.o
 
 ifneq ($(CONFIG_HSA_AMD),)
+AMDKFD_PATH := ../amdkfd
+include $(FULL_AMD_PATH)/amdkfd/Makefile
+amdgpu-y += $(AMDKFD_FILES)
 amdgpu-y += \
 	 amdgpu_amdkfd_fence.o \
 	 amdgpu_amdkfd_gpuvm.o \

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

@@ -116,136 +116,47 @@ static int acp_sw_fini(void *handle)
 	return 0;
 }
 
-/* power off a tile/block within ACP */
-static int acp_suspend_tile(void *cgs_dev, int tile)
-{
-	u32 val = 0;
-	u32 count = 0;
-
-	if ((tile  < ACP_TILE_P1) || (tile > ACP_TILE_DSP2)) {
-		pr_err("Invalid ACP tile : %d to suspend\n", tile);
-		return -1;
-	}
-
-	val = cgs_read_register(cgs_dev, mmACP_PGFSM_READ_REG_0 + tile);
-	val &= ACP_TILE_ON_MASK;
-
-	if (val == 0x0) {
-		val = cgs_read_register(cgs_dev, mmACP_PGFSM_RETAIN_REG);
-		val = val | (1 << tile);
-		cgs_write_register(cgs_dev, mmACP_PGFSM_RETAIN_REG, val);
-		cgs_write_register(cgs_dev, mmACP_PGFSM_CONFIG_REG,
-					0x500 + tile);
-
-		count = ACP_TIMEOUT_LOOP;
-		while (true) {
-			val = cgs_read_register(cgs_dev, mmACP_PGFSM_READ_REG_0
-								+ tile);
-			val = val & ACP_TILE_ON_MASK;
-			if (val == ACP_TILE_OFF_MASK)
-				break;
-			if (--count == 0) {
-				pr_err("Timeout reading ACP PGFSM status\n");
-				return -ETIMEDOUT;
-			}
-			udelay(100);
-		}
-
-		val = cgs_read_register(cgs_dev, mmACP_PGFSM_RETAIN_REG);
-
-		val |= ACP_TILE_OFF_RETAIN_REG_MASK;
-		cgs_write_register(cgs_dev, mmACP_PGFSM_RETAIN_REG, val);
-	}
-	return 0;
-}
-
-/* power on a tile/block within ACP */
-static int acp_resume_tile(void *cgs_dev, int tile)
-{
-	u32 val = 0;
-	u32 count = 0;
-
-	if ((tile  < ACP_TILE_P1) || (tile > ACP_TILE_DSP2)) {
-		pr_err("Invalid ACP tile to resume\n");
-		return -1;
-	}
-
-	val = cgs_read_register(cgs_dev, mmACP_PGFSM_READ_REG_0 + tile);
-	val = val & ACP_TILE_ON_MASK;
-
-	if (val != 0x0) {
-		cgs_write_register(cgs_dev, mmACP_PGFSM_CONFIG_REG,
-					0x600 + tile);
-		count = ACP_TIMEOUT_LOOP;
-		while (true) {
-			val = cgs_read_register(cgs_dev, mmACP_PGFSM_READ_REG_0
-							+ tile);
-			val = val & ACP_TILE_ON_MASK;
-			if (val == 0x0)
-				break;
-			if (--count == 0) {
-				pr_err("Timeout reading ACP PGFSM status\n");
-				return -ETIMEDOUT;
-			}
-			udelay(100);
-		}
-		val = cgs_read_register(cgs_dev, mmACP_PGFSM_RETAIN_REG);
-		if (tile == ACP_TILE_P1)
-			val = val & (ACP_TILE_P1_MASK);
-		else if (tile == ACP_TILE_P2)
-			val = val & (ACP_TILE_P2_MASK);
-
-		cgs_write_register(cgs_dev, mmACP_PGFSM_RETAIN_REG, val);
-	}
-	return 0;
-}
-
 struct acp_pm_domain {
-	void *cgs_dev;
+	void *adev;
 	struct generic_pm_domain gpd;
 };
 
 static int acp_poweroff(struct generic_pm_domain *genpd)
 {
-	int i, ret;
 	struct acp_pm_domain *apd;
+	struct amdgpu_device *adev;
 
 	apd = container_of(genpd, struct acp_pm_domain, gpd);
 	if (apd != NULL) {
-		/* Donot return abruptly if any of power tile fails to suspend.
-		 * Log it and continue powering off other tile
-		 */
-		for (i = 4; i >= 0 ; i--) {
-			ret = acp_suspend_tile(apd->cgs_dev, ACP_TILE_P1 + i);
-			if (ret)
-				pr_err("ACP tile %d tile suspend failed\n", i);
-		}
+		adev = apd->adev;
+	/* call smu to POWER GATE ACP block
+	 * smu will
+	 * 1. turn off the acp clock
+	 * 2. power off the acp tiles
+	 * 3. check and enter ulv state
+	 */
+		if (adev->powerplay.pp_funcs->set_powergating_by_smu)
+			amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, true);
 	}
 	return 0;
 }
 
 static int acp_poweron(struct generic_pm_domain *genpd)
 {
-	int i, ret;
 	struct acp_pm_domain *apd;
+	struct amdgpu_device *adev;
 
 	apd = container_of(genpd, struct acp_pm_domain, gpd);
 	if (apd != NULL) {
-		for (i = 0; i < 2; i++) {
-			ret = acp_resume_tile(apd->cgs_dev, ACP_TILE_P1 + i);
-			if (ret) {
-				pr_err("ACP tile %d resume failed\n", i);
-				break;
-			}
-		}
-
-		/* Disable DSPs which are not going to be used */
-		for (i = 0; i < 3; i++) {
-			ret = acp_suspend_tile(apd->cgs_dev, ACP_TILE_DSP0 + i);
-			/* Continue suspending other DSP, even if one fails */
-			if (ret)
-				pr_err("ACP DSP %d suspend failed\n", i);
-		}
+		adev = apd->adev;
+	/* call smu to UNGATE ACP block
+	 * smu will
+	 * 1. exit ulv
+	 * 2. turn on acp clock
+	 * 3. power on acp tiles
+	 */
+		if (adev->powerplay.pp_funcs->set_powergating_by_smu)
+			amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, false);
 	}
 	return 0;
 }
@@ -289,30 +200,31 @@ static int acp_hw_init(void *handle)
 	r = amd_acp_hw_init(adev->acp.cgs_device,
 			    ip_block->version->major, ip_block->version->minor);
 	/* -ENODEV means board uses AZ rather than ACP */
-	if (r == -ENODEV)
+	if (r == -ENODEV) {
+		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, true);
 		return 0;
-	else if (r)
+	} else if (r) {
 		return r;
+	}
 
 	if (adev->rmmio_size == 0 || adev->rmmio_size < 0x5289)
 		return -EINVAL;
 
 	acp_base = adev->rmmio_base;
 
-	if (adev->asic_type != CHIP_STONEY) {
-		adev->acp.acp_genpd = kzalloc(sizeof(struct acp_pm_domain), GFP_KERNEL);
-		if (adev->acp.acp_genpd == NULL)
-			return -ENOMEM;
 
-		adev->acp.acp_genpd->gpd.name = "ACP_AUDIO";
-		adev->acp.acp_genpd->gpd.power_off = acp_poweroff;
-		adev->acp.acp_genpd->gpd.power_on = acp_poweron;
+	adev->acp.acp_genpd = kzalloc(sizeof(struct acp_pm_domain), GFP_KERNEL);
+	if (adev->acp.acp_genpd == NULL)
+		return -ENOMEM;
+
+	adev->acp.acp_genpd->gpd.name = "ACP_AUDIO";
+	adev->acp.acp_genpd->gpd.power_off = acp_poweroff;
+	adev->acp.acp_genpd->gpd.power_on = acp_poweron;
 
 
-		adev->acp.acp_genpd->cgs_dev = adev->acp.cgs_device;
+	adev->acp.acp_genpd->adev = adev;
 
-		pm_genpd_init(&adev->acp.acp_genpd->gpd, NULL, false);
-	}
+	pm_genpd_init(&adev->acp.acp_genpd->gpd, NULL, false);
 
 	adev->acp.acp_cell = kcalloc(ACP_DEVS, sizeof(struct mfd_cell),
 							GFP_KERNEL);
@@ -429,17 +341,16 @@ static int acp_hw_init(void *handle)
 	if (r)
 		return r;
 
-	if (adev->asic_type != CHIP_STONEY) {
-		for (i = 0; i < ACP_DEVS ; i++) {
-			dev = get_mfd_cell_dev(adev->acp.acp_cell[i].name, i);
-			r = pm_genpd_add_device(&adev->acp.acp_genpd->gpd, dev);
-			if (r) {
-				dev_err(dev, "Failed to add dev to genpd\n");
-				return r;
-			}
+	for (i = 0; i < ACP_DEVS ; i++) {
+		dev = get_mfd_cell_dev(adev->acp.acp_cell[i].name, i);
+		r = pm_genpd_add_device(&adev->acp.acp_genpd->gpd, dev);
+		if (r) {
+			dev_err(dev, "Failed to add dev to genpd\n");
+			return r;
 		}
 	}
 
+
 	/* Assert Soft reset of ACP */
 	val = cgs_read_register(adev->acp.cgs_device, mmACP_SOFT_RESET);
 
@@ -497,8 +408,10 @@ static int acp_hw_fini(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	/* return early if no ACP */
-	if (!adev->acp.acp_cell)
+	if (!adev->acp.acp_genpd) {
+		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, false);
 		return 0;
+	}
 
 	/* Assert Soft reset of ACP */
 	val = cgs_read_register(adev->acp.cgs_device, mmACP_SOFT_RESET);
@@ -536,19 +449,17 @@ static int acp_hw_fini(void *handle)
 		udelay(100);
 	}
 
-	if (adev->acp.acp_genpd) {
-		for (i = 0; i < ACP_DEVS ; i++) {
-			dev = get_mfd_cell_dev(adev->acp.acp_cell[i].name, i);
-			ret = pm_genpd_remove_device(dev);
-			/* If removal fails, dont giveup and try rest */
-			if (ret)
-				dev_err(dev, "remove dev from genpd failed\n");
-		}
-		kfree(adev->acp.acp_genpd);
+	for (i = 0; i < ACP_DEVS ; i++) {
+		dev = get_mfd_cell_dev(adev->acp.acp_cell[i].name, i);
+		ret = pm_genpd_remove_device(dev);
+		/* If removal fails, dont giveup and try rest */
+		if (ret)
+			dev_err(dev, "remove dev from genpd failed\n");
 	}
 
 	mfd_remove_devices(adev->acp.parent);
 	kfree(adev->acp.acp_res);
+	kfree(adev->acp.acp_genpd);
 	kfree(adev->acp.acp_cell);
 
 	return 0;
@@ -556,11 +467,21 @@ static int acp_hw_fini(void *handle)
 
 static int acp_suspend(void *handle)
 {
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	/* power up on suspend */
+	if (!adev->acp.acp_cell)
+		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, false);
 	return 0;
 }
 
 static int acp_resume(void *handle)
 {
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	/* power down again on resume */
+	if (!adev->acp.acp_cell)
+		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, true);
 	return 0;
 }
 
@@ -593,6 +514,12 @@ static int acp_set_clockgating_state(void *handle,
 static int acp_set_powergating_state(void *handle,
 				     enum amd_powergating_state state)
 {
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	bool enable = state == AMD_PG_STATE_GATE ? true : false;
+
+	if (adev->powerplay.pp_funcs->set_powergating_by_smu)
+		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ACP, enable);
+
 	return 0;
 }
 

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

@@ -31,6 +31,7 @@
 #include <drm/drm_crtc_helper.h>
 #include "amdgpu.h"
 #include "amdgpu_pm.h"
+#include "amdgpu_display.h"
 #include "amd_acpi.h"
 #include "atom.h"
 

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

@@ -28,7 +28,6 @@
 #include <linux/module.h>
 
 const struct kgd2kfd_calls *kgd2kfd;
-bool (*kgd2kfd_init_p)(unsigned int, const struct kgd2kfd_calls**);
 
 static const unsigned int compute_vmid_bitmap = 0xFF00;
 
@@ -36,45 +35,23 @@ int amdgpu_amdkfd_init(void)
 {
 	int ret;
 
-#if defined(CONFIG_HSA_AMD_MODULE)
-	int (*kgd2kfd_init_p)(unsigned int, const struct kgd2kfd_calls**);
-
-	kgd2kfd_init_p = symbol_request(kgd2kfd_init);
-
-	if (kgd2kfd_init_p == NULL)
-		return -ENOENT;
-
-	ret = kgd2kfd_init_p(KFD_INTERFACE_VERSION, &kgd2kfd);
-	if (ret) {
-		symbol_put(kgd2kfd_init);
-		kgd2kfd = NULL;
-	}
-
-
-#elif defined(CONFIG_HSA_AMD)
-
+#ifdef CONFIG_HSA_AMD
 	ret = kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd);
 	if (ret)
 		kgd2kfd = NULL;
-
+	amdgpu_amdkfd_gpuvm_init_mem_limits();
 #else
 	kgd2kfd = NULL;
 	ret = -ENOENT;
 #endif
 
-#if defined(CONFIG_HSA_AMD_MODULE) || defined(CONFIG_HSA_AMD)
-	amdgpu_amdkfd_gpuvm_init_mem_limits();
-#endif
-
 	return ret;
 }
 
 void amdgpu_amdkfd_fini(void)
 {
-	if (kgd2kfd) {
+	if (kgd2kfd)
 		kgd2kfd->exit();
-		symbol_put(kgd2kfd_init);
-	}
 }
 
 void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev)
@@ -155,7 +132,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 			.num_queue_per_pipe = adev->gfx.mec.num_queue_per_pipe,
 			.gpuvm_size = min(adev->vm_manager.max_pfn
 					  << AMDGPU_GPU_PAGE_SHIFT,
-					  AMDGPU_VA_HOLE_START),
+					  AMDGPU_GMC_HOLE_START),
 			.drm_render_minor = adev->ddev->render->index
 		};
 
@@ -267,7 +244,8 @@ void amdgpu_amdkfd_gpu_reset(struct kgd_dev *kgd)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
 
-	amdgpu_device_gpu_recover(adev, NULL, false);
+	if (amdgpu_device_should_recover_gpu(adev))
+		amdgpu_device_gpu_recover(adev, NULL);
 }
 
 int alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
@@ -433,6 +411,13 @@ uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd)
 	return amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
 }
 
+uint64_t amdgpu_amdkfd_get_hive_id(struct kgd_dev *kgd)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
+
+	return adev->gmc.xgmi.hive_id;
+}
+
 int amdgpu_amdkfd_submit_ib(struct kgd_dev *kgd, enum kgd_engine_type engine,
 				uint32_t vmid, uint64_t gpu_addr,
 				uint32_t *ib_cmd, uint32_t ib_len)
@@ -506,7 +491,7 @@ bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid)
 	return false;
 }
 
-#if !defined(CONFIG_HSA_AMD_MODULE) && !defined(CONFIG_HSA_AMD)
+#ifndef CONFIG_HSA_AMD
 bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm)
 {
 	return false;

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

@@ -145,6 +145,7 @@ uint64_t get_gpu_clock_counter(struct kgd_dev *kgd);
 uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd);
 void get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info);
 uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd);
+uint64_t amdgpu_amdkfd_get_hive_id(struct kgd_dev *kgd);
 
 #define read_user_wptr(mmptr, wptr, dst)				\
 	({								\
@@ -162,16 +163,17 @@ uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd);
 	})
 
 /* GPUVM API */
-int amdgpu_amdkfd_gpuvm_create_process_vm(struct kgd_dev *kgd, void **vm,
-					void **process_info,
+int amdgpu_amdkfd_gpuvm_create_process_vm(struct kgd_dev *kgd, unsigned int pasid,
+					void **vm, void **process_info,
 					struct dma_fence **ef);
 int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct kgd_dev *kgd,
-					struct file *filp,
+					struct file *filp, unsigned int pasid,
 					void **vm, void **process_info,
 					struct dma_fence **ef);
 void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev,
 				struct amdgpu_vm *vm);
 void amdgpu_amdkfd_gpuvm_destroy_process_vm(struct kgd_dev *kgd, void *vm);
+void amdgpu_amdkfd_gpuvm_release_process_vm(struct kgd_dev *kgd, void *vm);
 uint32_t amdgpu_amdkfd_gpuvm_get_process_page_dir(void *vm);
 int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 		struct kgd_dev *kgd, uint64_t va, uint64_t size,

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c

@@ -205,6 +205,7 @@ static const struct kfd2kgd_calls kfd2kgd = {
 	.create_process_vm = amdgpu_amdkfd_gpuvm_create_process_vm,
 	.acquire_process_vm = amdgpu_amdkfd_gpuvm_acquire_process_vm,
 	.destroy_process_vm = amdgpu_amdkfd_gpuvm_destroy_process_vm,
+	.release_process_vm = amdgpu_amdkfd_gpuvm_release_process_vm,
 	.get_process_page_dir = amdgpu_amdkfd_gpuvm_get_process_page_dir,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
 	.alloc_memory_of_gpu = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu,

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c

@@ -164,6 +164,7 @@ static const struct kfd2kgd_calls kfd2kgd = {
 	.create_process_vm = amdgpu_amdkfd_gpuvm_create_process_vm,
 	.acquire_process_vm = amdgpu_amdkfd_gpuvm_acquire_process_vm,
 	.destroy_process_vm = amdgpu_amdkfd_gpuvm_destroy_process_vm,
+	.release_process_vm = amdgpu_amdkfd_gpuvm_release_process_vm,
 	.get_process_page_dir = amdgpu_amdkfd_gpuvm_get_process_page_dir,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
 	.alloc_memory_of_gpu = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu,

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c

@@ -201,6 +201,7 @@ static const struct kfd2kgd_calls kfd2kgd = {
 	.create_process_vm = amdgpu_amdkfd_gpuvm_create_process_vm,
 	.acquire_process_vm = amdgpu_amdkfd_gpuvm_acquire_process_vm,
 	.destroy_process_vm = amdgpu_amdkfd_gpuvm_destroy_process_vm,
+	.release_process_vm = amdgpu_amdkfd_gpuvm_release_process_vm,
 	.get_process_page_dir = amdgpu_amdkfd_gpuvm_get_process_page_dir,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
 	.alloc_memory_of_gpu = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu,
@@ -214,7 +215,8 @@ static const struct kfd2kgd_calls kfd2kgd = {
 	.invalidate_tlbs_vmid = invalidate_tlbs_vmid,
 	.submit_ib = amdgpu_amdkfd_submit_ib,
 	.gpu_recover = amdgpu_amdkfd_gpu_reset,
-	.set_compute_idle = amdgpu_amdkfd_set_compute_idle
+	.set_compute_idle = amdgpu_amdkfd_set_compute_idle,
+	.get_hive_id = amdgpu_amdkfd_get_hive_id,
 };
 
 struct kfd2kgd_calls *amdgpu_amdkfd_gfx_9_0_get_functions(void)

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

@@ -364,7 +364,6 @@ static int vm_validate_pt_pd_bos(struct amdgpu_vm *vm)
 	struct amdgpu_bo *pd = vm->root.base.bo;
 	struct amdgpu_device *adev = amdgpu_ttm_adev(pd->tbo.bdev);
 	struct amdgpu_vm_parser param;
-	uint64_t addr, flags = AMDGPU_PTE_VALID;
 	int ret;
 
 	param.domain = AMDGPU_GEM_DOMAIN_VRAM;
@@ -383,9 +382,7 @@ static int vm_validate_pt_pd_bos(struct amdgpu_vm *vm)
 		return ret;
 	}
 
-	addr = amdgpu_bo_gpu_offset(vm->root.base.bo);
-	amdgpu_gmc_get_vm_pde(adev, -1, &addr, &flags);
-	vm->pd_phys_addr = addr;
+	vm->pd_phys_addr = amdgpu_gmc_pd_addr(vm->root.base.bo);
 
 	if (vm->use_cpu_for_update) {
 		ret = amdgpu_bo_kmap(pd, NULL);
@@ -678,7 +675,6 @@ static int reserve_bo_and_vm(struct kgd_mem *mem,
 	if (!ctx->vm_pd)
 		return -ENOMEM;
 
-	ctx->kfd_bo.robj = bo;
 	ctx->kfd_bo.priority = 0;
 	ctx->kfd_bo.tv.bo = &bo->tbo;
 	ctx->kfd_bo.tv.shared = true;
@@ -743,7 +739,6 @@ static int reserve_bo_and_cond_vms(struct kgd_mem *mem,
 			return -ENOMEM;
 	}
 
-	ctx->kfd_bo.robj = bo;
 	ctx->kfd_bo.priority = 0;
 	ctx->kfd_bo.tv.bo = &bo->tbo;
 	ctx->kfd_bo.tv.shared = true;
@@ -1003,8 +998,8 @@ static int init_kfd_vm(struct amdgpu_vm *vm, void **process_info,
 	return ret;
 }
 
-int amdgpu_amdkfd_gpuvm_create_process_vm(struct kgd_dev *kgd, void **vm,
-					  void **process_info,
+int amdgpu_amdkfd_gpuvm_create_process_vm(struct kgd_dev *kgd, unsigned int pasid,
+					  void **vm, void **process_info,
 					  struct dma_fence **ef)
 {
 	struct amdgpu_device *adev = get_amdgpu_device(kgd);
@@ -1016,7 +1011,7 @@ int amdgpu_amdkfd_gpuvm_create_process_vm(struct kgd_dev *kgd, void **vm,
 		return -ENOMEM;
 
 	/* Initialize AMDGPU part of the VM */
-	ret = amdgpu_vm_init(adev, new_vm, AMDGPU_VM_CONTEXT_COMPUTE, 0);
+	ret = amdgpu_vm_init(adev, new_vm, AMDGPU_VM_CONTEXT_COMPUTE, pasid);
 	if (ret) {
 		pr_err("Failed init vm ret %d\n", ret);
 		goto amdgpu_vm_init_fail;
@@ -1039,7 +1034,7 @@ int amdgpu_amdkfd_gpuvm_create_process_vm(struct kgd_dev *kgd, void **vm,
 }
 
 int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct kgd_dev *kgd,
-					   struct file *filp,
+					   struct file *filp, unsigned int pasid,
 					   void **vm, void **process_info,
 					   struct dma_fence **ef)
 {
@@ -1054,7 +1049,7 @@ int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct kgd_dev *kgd,
 		return -EINVAL;
 
 	/* Convert VM into a compute VM */
-	ret = amdgpu_vm_make_compute(adev, avm);
+	ret = amdgpu_vm_make_compute(adev, avm, pasid);
 	if (ret)
 		return ret;
 
@@ -1117,6 +1112,25 @@ void amdgpu_amdkfd_gpuvm_destroy_process_vm(struct kgd_dev *kgd, void *vm)
 	kfree(vm);
 }
 
+void amdgpu_amdkfd_gpuvm_release_process_vm(struct kgd_dev *kgd, void *vm)
+{
+	struct amdgpu_device *adev = get_amdgpu_device(kgd);
+        struct amdgpu_vm *avm = (struct amdgpu_vm *)vm;
+
+	if (WARN_ON(!kgd || !vm))
+                return;
+
+        pr_debug("Releasing process vm %p\n", vm);
+
+        /* The original pasid of amdgpu vm has already been
+         * released during making a amdgpu vm to a compute vm
+         * The current pasid is managed by kfd and will be
+         * released on kfd process destroy. Set amdgpu pasid
+         * to 0 to avoid duplicate release.
+         */
+	amdgpu_vm_release_compute(adev, avm);
+}
+
 uint32_t amdgpu_amdkfd_gpuvm_get_process_page_dir(void *vm)
 {
 	struct amdgpu_vm *avm = (struct amdgpu_vm *)vm;

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

@@ -29,6 +29,7 @@
 #include "amdgpu_atombios.h"
 #include "amdgpu_atomfirmware.h"
 #include "amdgpu_i2c.h"
+#include "amdgpu_display.h"
 
 #include "atom.h"
 #include "atom-bits.h"

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

@@ -117,6 +117,10 @@ union igp_info {
 union umc_info {
 	struct atom_umc_info_v3_1 v31;
 };
+
+union vram_info {
+	struct atom_vram_info_header_v2_3 v23;
+};
 /*
  * Return vram width from integrated system info table, if available,
  * or 0 if not.
@@ -174,7 +178,7 @@ static int convert_atom_mem_type_to_vram_type (struct amdgpu_device *adev,
 		case ATOM_DGPU_VRAM_TYPE_GDDR5:
 			vram_type = AMDGPU_VRAM_TYPE_GDDR5;
 			break;
-		case ATOM_DGPU_VRAM_TYPE_HBM:
+		case ATOM_DGPU_VRAM_TYPE_HBM2:
 			vram_type = AMDGPU_VRAM_TYPE_HBM;
 			break;
 		default:
@@ -195,7 +199,7 @@ int amdgpu_atomfirmware_get_vram_type(struct amdgpu_device *adev)
 	int index;
 	u16 data_offset, size;
 	union igp_info *igp_info;
-	union umc_info *umc_info;
+	union vram_info *vram_info;
 	u8 frev, crev;
 	u8 mem_type;
 
@@ -204,7 +208,7 @@ int amdgpu_atomfirmware_get_vram_type(struct amdgpu_device *adev)
 						    integratedsysteminfo);
 	else
 		index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
-						    umc_info);
+						    vram_info);
 	if (amdgpu_atom_parse_data_header(mode_info->atom_context,
 					  index, &size,
 					  &frev, &crev, &data_offset)) {
@@ -219,11 +223,11 @@ int amdgpu_atomfirmware_get_vram_type(struct amdgpu_device *adev)
 				return 0;
 			}
 		} else {
-			umc_info = (union umc_info *)
+			vram_info = (union vram_info *)
 				(mode_info->atom_context->bios + data_offset);
 			switch (crev) {
-			case 1:
-				mem_type = umc_info->v31.vram_type;
+			case 3:
+				mem_type = vram_info->v23.vram_module[0].memory_type;
 				return convert_atom_mem_type_to_vram_type(adev, mem_type);
 			default:
 				return 0;

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

@@ -49,8 +49,11 @@ static void amdgpu_bo_list_free(struct kref *ref)
 						   refcount);
 	struct amdgpu_bo_list_entry *e;
 
-	amdgpu_bo_list_for_each_entry(e, list)
-		amdgpu_bo_unref(&e->robj);
+	amdgpu_bo_list_for_each_entry(e, list) {
+		struct amdgpu_bo *bo = ttm_to_amdgpu_bo(e->tv.bo);
+
+		amdgpu_bo_unref(&bo);
+	}
 
 	call_rcu(&list->rhead, amdgpu_bo_list_free_rcu);
 }
@@ -67,7 +70,8 @@ int amdgpu_bo_list_create(struct amdgpu_device *adev, struct drm_file *filp,
 	unsigned i;
 	int r;
 
-	if (num_entries > SIZE_MAX / sizeof(struct amdgpu_bo_list_entry))
+	if (num_entries > (SIZE_MAX - sizeof(struct amdgpu_bo_list))
+				/ sizeof(struct amdgpu_bo_list_entry))
 		return -EINVAL;
 
 	size = sizeof(struct amdgpu_bo_list);
@@ -111,21 +115,20 @@ int amdgpu_bo_list_create(struct amdgpu_device *adev, struct drm_file *filp,
 			entry = &array[last_entry++];
 		}
 
-		entry->robj = bo;
 		entry->priority = min(info[i].bo_priority,
 				      AMDGPU_BO_LIST_MAX_PRIORITY);
-		entry->tv.bo = &entry->robj->tbo;
-		entry->tv.shared = !entry->robj->prime_shared_count;
-
-		if (entry->robj->preferred_domains == AMDGPU_GEM_DOMAIN_GDS)
-			list->gds_obj = entry->robj;
-		if (entry->robj->preferred_domains == AMDGPU_GEM_DOMAIN_GWS)
-			list->gws_obj = entry->robj;
-		if (entry->robj->preferred_domains == AMDGPU_GEM_DOMAIN_OA)
-			list->oa_obj = entry->robj;
-
-		total_size += amdgpu_bo_size(entry->robj);
-		trace_amdgpu_bo_list_set(list, entry->robj);
+		entry->tv.bo = &bo->tbo;
+		entry->tv.shared = !bo->prime_shared_count;
+
+		if (bo->preferred_domains == AMDGPU_GEM_DOMAIN_GDS)
+			list->gds_obj = bo;
+		if (bo->preferred_domains == AMDGPU_GEM_DOMAIN_GWS)
+			list->gws_obj = bo;
+		if (bo->preferred_domains == AMDGPU_GEM_DOMAIN_OA)
+			list->oa_obj = bo;
+
+		total_size += amdgpu_bo_size(bo);
+		trace_amdgpu_bo_list_set(list, bo);
 	}
 
 	list->first_userptr = first_userptr;
@@ -137,8 +140,11 @@ int amdgpu_bo_list_create(struct amdgpu_device *adev, struct drm_file *filp,
 	return 0;
 
 error_free:
-	while (i--)
-		amdgpu_bo_unref(&array[i].robj);
+	while (i--) {
+		struct amdgpu_bo *bo = ttm_to_amdgpu_bo(array[i].tv.bo);
+
+		amdgpu_bo_unref(&bo);
+	}
 	kvfree(list);
 	return r;
 
@@ -190,9 +196,10 @@ void amdgpu_bo_list_get_list(struct amdgpu_bo_list *list,
 	 * with the same priority, i.e. it must be stable.
 	 */
 	amdgpu_bo_list_for_each_entry(e, list) {
+		struct amdgpu_bo *bo = ttm_to_amdgpu_bo(e->tv.bo);
 		unsigned priority = e->priority;
 
-		if (!e->robj->parent)
+		if (!bo->parent)
 			list_add_tail(&e->tv.head, &bucket[priority]);
 
 		e->user_pages = NULL;

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

@@ -32,7 +32,6 @@ struct amdgpu_bo_va;
 struct amdgpu_fpriv;
 
 struct amdgpu_bo_list_entry {
-	struct amdgpu_bo		*robj;
 	struct ttm_validate_buffer	tv;
 	struct amdgpu_bo_va		*bo_va;
 	uint32_t			priority;

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

@@ -34,6 +34,7 @@
 #include "atombios_dp.h"
 #include "amdgpu_connectors.h"
 #include "amdgpu_i2c.h"
+#include "amdgpu_display.h"
 
 #include <linux/pm_runtime.h>
 

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

+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __AMDGPU_CTX_H__
+#define __AMDGPU_CTX_H__
+
+#include "amdgpu_ring.h"
+
+struct drm_device;
+struct drm_file;
+struct amdgpu_fpriv;
+
+struct amdgpu_ctx_entity {
+	uint64_t		sequence;
+	struct dma_fence	**fences;
+	struct drm_sched_entity	entity;
+};
+
+struct amdgpu_ctx {
+	struct kref			refcount;
+	struct amdgpu_device		*adev;
+	unsigned			reset_counter;
+	unsigned			reset_counter_query;
+	uint32_t			vram_lost_counter;
+	spinlock_t			ring_lock;
+	struct dma_fence		**fences;
+	struct amdgpu_ctx_entity	*entities[AMDGPU_HW_IP_NUM];
+	bool				preamble_presented;
+	enum drm_sched_priority		init_priority;
+	enum drm_sched_priority		override_priority;
+	struct mutex			lock;
+	atomic_t			guilty;
+};
+
+struct amdgpu_ctx_mgr {
+	struct amdgpu_device	*adev;
+	struct mutex		lock;
+	/* protected by lock */
+	struct idr		ctx_handles;
+};
+
+extern const unsigned int amdgpu_ctx_num_entities[AMDGPU_HW_IP_NUM];
+
+struct amdgpu_ctx *amdgpu_ctx_get(struct amdgpu_fpriv *fpriv, uint32_t id);
+int amdgpu_ctx_put(struct amdgpu_ctx *ctx);
+
+int amdgpu_ctx_get_entity(struct amdgpu_ctx *ctx, u32 hw_ip, u32 instance,
+			  u32 ring, struct drm_sched_entity **entity);
+void amdgpu_ctx_add_fence(struct amdgpu_ctx *ctx,
+			  struct drm_sched_entity *entity,
+			  struct dma_fence *fence, uint64_t *seq);
+struct dma_fence *amdgpu_ctx_get_fence(struct amdgpu_ctx *ctx,
+				       struct drm_sched_entity *entity,
+				       uint64_t seq);
+void amdgpu_ctx_priority_override(struct amdgpu_ctx *ctx,
+				  enum drm_sched_priority priority);
+
+int amdgpu_ctx_ioctl(struct drm_device *dev, void *data,
+		     struct drm_file *filp);
+
+int amdgpu_ctx_wait_prev_fence(struct amdgpu_ctx *ctx,
+			       struct drm_sched_entity *entity);
+
+void amdgpu_ctx_mgr_init(struct amdgpu_ctx_mgr *mgr);
+void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr);
+void amdgpu_ctx_mgr_entity_flush(struct amdgpu_ctx_mgr *mgr);
+void amdgpu_ctx_mgr_fini(struct amdgpu_ctx_mgr *mgr);
+
+#endif

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

@@ -23,6 +23,21 @@
 #ifndef __AMDGPU_DISPLAY_H__
 #define __AMDGPU_DISPLAY_H__
 
+#define amdgpu_display_vblank_get_counter(adev, crtc) (adev)->mode_info.funcs->vblank_get_counter((adev), (crtc))
+#define amdgpu_display_backlight_set_level(adev, e, l) (adev)->mode_info.funcs->backlight_set_level((e), (l))
+#define amdgpu_display_backlight_get_level(adev, e) (adev)->mode_info.funcs->backlight_get_level((e))
+#define amdgpu_display_hpd_sense(adev, h) (adev)->mode_info.funcs->hpd_sense((adev), (h))
+#define amdgpu_display_hpd_set_polarity(adev, h) (adev)->mode_info.funcs->hpd_set_polarity((adev), (h))
+#define amdgpu_display_hpd_get_gpio_reg(adev) (adev)->mode_info.funcs->hpd_get_gpio_reg((adev))
+#define amdgpu_display_bandwidth_update(adev) (adev)->mode_info.funcs->bandwidth_update((adev))
+#define amdgpu_display_page_flip(adev, crtc, base, async) (adev)->mode_info.funcs->page_flip((adev), (crtc), (base), (async))
+#define amdgpu_display_page_flip_get_scanoutpos(adev, crtc, vbl, pos) (adev)->mode_info.funcs->page_flip_get_scanoutpos((adev), (crtc), (vbl), (pos))
+#define amdgpu_display_add_encoder(adev, e, s, c) (adev)->mode_info.funcs->add_encoder((adev), (e), (s), (c))
+#define amdgpu_display_add_connector(adev, ci, sd, ct, ib, coi, h, r) (adev)->mode_info.funcs->add_connector((adev), (ci), (sd), (ct), (ib), (coi), (h), (r))
+
+int amdgpu_display_freesync_ioctl(struct drm_device *dev, void *data,
+				  struct drm_file *filp);
+void amdgpu_display_update_priority(struct amdgpu_device *adev);
 uint32_t amdgpu_display_supported_domains(struct amdgpu_device *adev);
 struct drm_framebuffer *
 amdgpu_display_user_framebuffer_create(struct drm_device *dev,

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

@@ -36,6 +36,7 @@
 
 #include "amdgpu.h"
 #include "amdgpu_irq.h"
+#include "amdgpu_gem.h"
 
 #include "amdgpu_amdkfd.h"
 
@@ -113,8 +114,8 @@ uint amdgpu_pg_mask = 0xffffffff;
 uint amdgpu_sdma_phase_quantum = 32;
 char *amdgpu_disable_cu = NULL;
 char *amdgpu_virtual_display = NULL;
-/* OverDrive(bit 14),gfxoff(bit 15),stutter mode(bit 17) disabled by default*/
-uint amdgpu_pp_feature_mask = 0xfffd3fff;
+/* OverDrive(bit 14) disabled by default*/
+uint amdgpu_pp_feature_mask = 0xffffbfff;
 int amdgpu_ngg = 0;
 int amdgpu_prim_buf_per_se = 0;
 int amdgpu_pos_buf_per_se = 0;
@@ -531,6 +532,102 @@ MODULE_PARM_DESC(smu_memory_pool_size,
 		"0x1 = 256Mbyte, 0x2 = 512Mbyte, 0x4 = 1 Gbyte, 0x8 = 2GByte");
 module_param_named(smu_memory_pool_size, amdgpu_smu_memory_pool_size, uint, 0444);
 
+#ifdef CONFIG_HSA_AMD
+/**
+ * DOC: sched_policy (int)
+ * Set scheduling policy. Default is HWS(hardware scheduling) with over-subscription.
+ * Setting 1 disables over-subscription. Setting 2 disables HWS and statically
+ * assigns queues to HQDs.
+ */
+int sched_policy = KFD_SCHED_POLICY_HWS;
+module_param(sched_policy, int, 0444);
+MODULE_PARM_DESC(sched_policy,
+	"Scheduling policy (0 = HWS (Default), 1 = HWS without over-subscription, 2 = Non-HWS (Used for debugging only)");
+
+/**
+ * DOC: hws_max_conc_proc (int)
+ * Maximum number of processes that HWS can schedule concurrently. The maximum is the
+ * number of VMIDs assigned to the HWS, which is also the default.
+ */
+int hws_max_conc_proc = 8;
+module_param(hws_max_conc_proc, int, 0444);
+MODULE_PARM_DESC(hws_max_conc_proc,
+	"Max # processes HWS can execute concurrently when sched_policy=0 (0 = no concurrency, #VMIDs for KFD = Maximum(default))");
+
+/**
+ * DOC: cwsr_enable (int)
+ * CWSR(compute wave store and resume) allows the GPU to preempt shader execution in
+ * the middle of a compute wave. Default is 1 to enable this feature. Setting 0
+ * disables it.
+ */
+int cwsr_enable = 1;
+module_param(cwsr_enable, int, 0444);
+MODULE_PARM_DESC(cwsr_enable, "CWSR enable (0 = Off, 1 = On (Default))");
+
+/**
+ * DOC: max_num_of_queues_per_device (int)
+ * Maximum number of queues per device. Valid setting is between 1 and 4096. Default
+ * is 4096.
+ */
+int max_num_of_queues_per_device = KFD_MAX_NUM_OF_QUEUES_PER_DEVICE_DEFAULT;
+module_param(max_num_of_queues_per_device, int, 0444);
+MODULE_PARM_DESC(max_num_of_queues_per_device,
+	"Maximum number of supported queues per device (1 = Minimum, 4096 = default)");
+
+/**
+ * DOC: send_sigterm (int)
+ * Send sigterm to HSA process on unhandled exceptions. Default is not to send sigterm
+ * but just print errors on dmesg. Setting 1 enables sending sigterm.
+ */
+int send_sigterm;
+module_param(send_sigterm, int, 0444);
+MODULE_PARM_DESC(send_sigterm,
+	"Send sigterm to HSA process on unhandled exception (0 = disable, 1 = enable)");
+
+/**
+ * DOC: debug_largebar (int)
+ * Set debug_largebar as 1 to enable simulating large-bar capability on non-large bar
+ * system. This limits the VRAM size reported to ROCm applications to the visible
+ * size, usually 256MB.
+ * Default value is 0, diabled.
+ */
+int debug_largebar;
+module_param(debug_largebar, int, 0444);
+MODULE_PARM_DESC(debug_largebar,
+	"Debug large-bar flag used to simulate large-bar capability on non-large bar machine (0 = disable, 1 = enable)");
+
+/**
+ * DOC: ignore_crat (int)
+ * Ignore CRAT table during KFD initialization. By default, KFD uses the ACPI CRAT
+ * table to get information about AMD APUs. This option can serve as a workaround on
+ * systems with a broken CRAT table.
+ */
+int ignore_crat;
+module_param(ignore_crat, int, 0444);
+MODULE_PARM_DESC(ignore_crat,
+	"Ignore CRAT table during KFD initialization (0 = use CRAT (default), 1 = ignore CRAT)");
+
+/**
+ * DOC: noretry (int)
+ * This parameter sets sh_mem_config.retry_disable. Default value, 0, enables retry.
+ * Setting 1 disables retry.
+ * Retry is needed for recoverable page faults.
+ */
+int noretry;
+module_param(noretry, int, 0644);
+MODULE_PARM_DESC(noretry,
+	"Set sh_mem_config.retry_disable on Vega10 (0 = retry enabled (default), 1 = retry disabled)");
+
+/**
+ * DOC: halt_if_hws_hang (int)
+ * Halt if HWS hang is detected. Default value, 0, disables the halt on hang.
+ * Setting 1 enables halt on hang.
+ */
+int halt_if_hws_hang;
+module_param(halt_if_hws_hang, int, 0644);
+MODULE_PARM_DESC(halt_if_hws_hang, "Halt if HWS hang is detected (0 = off (default), 1 = on)");
+#endif
+
 static const struct pci_device_id pciidlist[] = {
 #ifdef  CONFIG_DRM_AMDGPU_SI
 	{0x1002, 0x6780, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI},
@@ -769,14 +866,15 @@ static const struct pci_device_id pciidlist[] = {
 	{0x1002, 0x69A3, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA12},
 	{0x1002, 0x69AF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA12},
 	/* Vega 20 */
-	{0x1002, 0x66A0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT},
-	{0x1002, 0x66A1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT},
-	{0x1002, 0x66A2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT},
-	{0x1002, 0x66A3, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT},
-	{0x1002, 0x66A7, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT},
-	{0x1002, 0x66AF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20|AMD_EXP_HW_SUPPORT},
+	{0x1002, 0x66A0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20},
+	{0x1002, 0x66A1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20},
+	{0x1002, 0x66A2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20},
+	{0x1002, 0x66A3, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20},
+	{0x1002, 0x66A7, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20},
+	{0x1002, 0x66AF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20},
 	/* Raven */
 	{0x1002, 0x15dd, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RAVEN|AMD_IS_APU},
+	{0x1002, 0x15d8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RAVEN|AMD_IS_APU},
 
 	{0, 0, 0}
 };
@@ -803,14 +901,6 @@ static int amdgpu_pci_probe(struct pci_dev *pdev,
 		return -ENODEV;
 	}
 
-	/*
-	 * Initialize amdkfd before starting radeon. If it was not loaded yet,
-	 * defer radeon probing
-	 */
-	ret = amdgpu_amdkfd_init();
-	if (ret == -EPROBE_DEFER)
-		return ret;
-
 	/* Get rid of things like offb */
 	ret = drm_fb_helper_remove_conflicting_pci_framebuffers(pdev, 0, "amdgpudrmfb");
 	if (ret)
@@ -855,8 +945,8 @@ amdgpu_pci_remove(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
 
-	drm_dev_unregister(dev);
-	drm_dev_put(dev);
+	DRM_ERROR("Device removal is currently not supported outside of fbcon\n");
+	drm_dev_unplug(dev);
 	pci_disable_device(pdev);
 	pci_set_drvdata(pdev, NULL);
 }
@@ -1151,6 +1241,10 @@ static int __init amdgpu_init(void)
 	pdriver = &amdgpu_kms_pci_driver;
 	driver->num_ioctls = amdgpu_max_kms_ioctl;
 	amdgpu_register_atpx_handler();
+
+	/* Ignore KFD init failures. Normal when CONFIG_HSA_AMD is not set. */
+	amdgpu_amdkfd_init();
+
 	/* let modprobe override vga console setting */
 	return pci_register_driver(pdriver);
 

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

@@ -28,6 +28,7 @@
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
 #include "amdgpu_connectors.h"
+#include "amdgpu_display.h"
 #include "atom.h"
 #include "atombios_encoders.h"
 

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

@@ -33,6 +33,7 @@
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
 #include "cikd.h"
+#include "amdgpu_gem.h"
 
 #include <drm/drm_fb_helper.h>
 

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

@@ -195,19 +195,6 @@ int amdgpu_fence_emit_polling(struct amdgpu_ring *ring, uint32_t *s)
 	return 0;
 }
 
-/**
- * amdgpu_fence_schedule_fallback - schedule fallback check
- *
- * @ring: pointer to struct amdgpu_ring
- *
- * Start a timer as fallback to our interrupts.
- */
-static void amdgpu_fence_schedule_fallback(struct amdgpu_ring *ring)
-{
-	mod_timer(&ring->fence_drv.fallback_timer,
-		  jiffies + AMDGPU_FENCE_JIFFIES_TIMEOUT);
-}
-
 /**
  * amdgpu_fence_process - check for fence activity
  *
@@ -229,9 +216,6 @@ void amdgpu_fence_process(struct amdgpu_ring *ring)
 
 	} while (atomic_cmpxchg(&drv->last_seq, last_seq, seq) != last_seq);
 
-	if (seq != ring->fence_drv.sync_seq)
-		amdgpu_fence_schedule_fallback(ring);
-
 	if (unlikely(seq == last_seq))
 		return;
 
@@ -262,21 +246,6 @@ void amdgpu_fence_process(struct amdgpu_ring *ring)
 	} while (last_seq != seq);
 }
 
-/**
- * amdgpu_fence_fallback - fallback for hardware interrupts
- *
- * @work: delayed work item
- *
- * Checks for fence activity.
- */
-static void amdgpu_fence_fallback(struct timer_list *t)
-{
-	struct amdgpu_ring *ring = from_timer(ring, t,
-					      fence_drv.fallback_timer);
-
-	amdgpu_fence_process(ring);
-}
-
 /**
  * amdgpu_fence_wait_empty - wait for all fences to signal
  *
@@ -424,8 +393,6 @@ int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring,
 	atomic_set(&ring->fence_drv.last_seq, 0);
 	ring->fence_drv.initialized = false;
 
-	timer_setup(&ring->fence_drv.fallback_timer, amdgpu_fence_fallback, 0);
-
 	ring->fence_drv.num_fences_mask = num_hw_submission * 2 - 1;
 	spin_lock_init(&ring->fence_drv.lock);
 	ring->fence_drv.fences = kcalloc(num_hw_submission * 2, sizeof(void *),
@@ -501,7 +468,6 @@ void amdgpu_fence_driver_fini(struct amdgpu_device *adev)
 		amdgpu_irq_put(adev, ring->fence_drv.irq_src,
 			       ring->fence_drv.irq_type);
 		drm_sched_fini(&ring->sched);
-		del_timer_sync(&ring->fence_drv.fallback_timer);
 		for (j = 0; j <= ring->fence_drv.num_fences_mask; ++j)
 			dma_fence_put(ring->fence_drv.fences[j]);
 		kfree(ring->fence_drv.fences);
@@ -594,27 +560,6 @@ static const char *amdgpu_fence_get_timeline_name(struct dma_fence *f)
 	return (const char *)fence->ring->name;
 }
 
-/**
- * amdgpu_fence_enable_signaling - enable signalling on fence
- * @fence: fence
- *
- * This function is called with fence_queue lock held, and adds a callback
- * to fence_queue that checks if this fence is signaled, and if so it
- * signals the fence and removes itself.
- */
-static bool amdgpu_fence_enable_signaling(struct dma_fence *f)
-{
-	struct amdgpu_fence *fence = to_amdgpu_fence(f);
-	struct amdgpu_ring *ring = fence->ring;
-
-	if (!timer_pending(&ring->fence_drv.fallback_timer))
-		amdgpu_fence_schedule_fallback(ring);
-
-	DMA_FENCE_TRACE(&fence->base, "armed on ring %i!\n", ring->idx);
-
-	return true;
-}
-
 /**
  * amdgpu_fence_free - free up the fence memory
  *
@@ -645,7 +590,6 @@ static void amdgpu_fence_release(struct dma_fence *f)
 static const struct dma_fence_ops amdgpu_fence_ops = {
 	.get_driver_name = amdgpu_fence_get_driver_name,
 	.get_timeline_name = amdgpu_fence_get_timeline_name,
-	.enable_signaling = amdgpu_fence_enable_signaling,
 	.release = amdgpu_fence_release,
 };
 
@@ -701,7 +645,7 @@ static int amdgpu_debugfs_gpu_recover(struct seq_file *m, void *data)
 	struct amdgpu_device *adev = dev->dev_private;
 
 	seq_printf(m, "gpu recover\n");
-	amdgpu_device_gpu_recover(adev, NULL, true);
+	amdgpu_device_gpu_recover(adev, NULL);
 
 	return 0;
 }

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

@@ -112,7 +112,7 @@ int amdgpu_gart_table_vram_alloc(struct amdgpu_device *adev)
 {
 	int r;
 
-	if (adev->gart.robj == NULL) {
+	if (adev->gart.bo == NULL) {
 		struct amdgpu_bo_param bp;
 
 		memset(&bp, 0, sizeof(bp));
@@ -123,7 +123,7 @@ int amdgpu_gart_table_vram_alloc(struct amdgpu_device *adev)
 			AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS;
 		bp.type = ttm_bo_type_kernel;
 		bp.resv = NULL;
-		r = amdgpu_bo_create(adev, &bp, &adev->gart.robj);
+		r = amdgpu_bo_create(adev, &bp, &adev->gart.bo);
 		if (r) {
 			return r;
 		}
@@ -145,19 +145,18 @@ int amdgpu_gart_table_vram_pin(struct amdgpu_device *adev)
 {
 	int r;
 
-	r = amdgpu_bo_reserve(adev->gart.robj, false);
+	r = amdgpu_bo_reserve(adev->gart.bo, false);
 	if (unlikely(r != 0))
 		return r;
-	r = amdgpu_bo_pin(adev->gart.robj, AMDGPU_GEM_DOMAIN_VRAM);
+	r = amdgpu_bo_pin(adev->gart.bo, AMDGPU_GEM_DOMAIN_VRAM);
 	if (r) {
-		amdgpu_bo_unreserve(adev->gart.robj);
+		amdgpu_bo_unreserve(adev->gart.bo);
 		return r;
 	}
-	r = amdgpu_bo_kmap(adev->gart.robj, &adev->gart.ptr);
+	r = amdgpu_bo_kmap(adev->gart.bo, &adev->gart.ptr);
 	if (r)
-		amdgpu_bo_unpin(adev->gart.robj);
-	amdgpu_bo_unreserve(adev->gart.robj);
-	adev->gart.table_addr = amdgpu_bo_gpu_offset(adev->gart.robj);
+		amdgpu_bo_unpin(adev->gart.bo);
+	amdgpu_bo_unreserve(adev->gart.bo);
 	return r;
 }
 
@@ -173,14 +172,14 @@ void amdgpu_gart_table_vram_unpin(struct amdgpu_device *adev)
 {
 	int r;
 
-	if (adev->gart.robj == NULL) {
+	if (adev->gart.bo == NULL) {
 		return;
 	}
-	r = amdgpu_bo_reserve(adev->gart.robj, true);
+	r = amdgpu_bo_reserve(adev->gart.bo, true);
 	if (likely(r == 0)) {
-		amdgpu_bo_kunmap(adev->gart.robj);
-		amdgpu_bo_unpin(adev->gart.robj);
-		amdgpu_bo_unreserve(adev->gart.robj);
+		amdgpu_bo_kunmap(adev->gart.bo);
+		amdgpu_bo_unpin(adev->gart.bo);
+		amdgpu_bo_unreserve(adev->gart.bo);
 		adev->gart.ptr = NULL;
 	}
 }
@@ -196,10 +195,10 @@ void amdgpu_gart_table_vram_unpin(struct amdgpu_device *adev)
  */
 void amdgpu_gart_table_vram_free(struct amdgpu_device *adev)
 {
-	if (adev->gart.robj == NULL) {
+	if (adev->gart.bo == NULL) {
 		return;
 	}
-	amdgpu_bo_unref(&adev->gart.robj);
+	amdgpu_bo_unref(&adev->gart.bo);
 }
 
 /*

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

@@ -40,8 +40,7 @@ struct amdgpu_bo;
 #define AMDGPU_GPU_PAGES_IN_CPU_PAGE (PAGE_SIZE / AMDGPU_GPU_PAGE_SIZE)
 
 struct amdgpu_gart {
-	u64				table_addr;
-	struct amdgpu_bo		*robj;
+	struct amdgpu_bo		*bo;
 	void				*ptr;
 	unsigned			num_gpu_pages;
 	unsigned			num_cpu_pages;

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

@@ -24,13 +24,6 @@
 #ifndef __AMDGPU_GDS_H__
 #define __AMDGPU_GDS_H__
 
-/* Because TTM request that alloacted buffer should be PAGE_SIZE aligned,
- * we should report GDS/GWS/OA size as PAGE_SIZE aligned
- * */
-#define AMDGPU_GDS_SHIFT	2
-#define AMDGPU_GWS_SHIFT	PAGE_SHIFT
-#define AMDGPU_OA_SHIFT		PAGE_SHIFT
-
 struct amdgpu_ring;
 struct amdgpu_bo;
 

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

@@ -244,16 +244,10 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 			return -EINVAL;
 		}
 		flags |= AMDGPU_GEM_CREATE_NO_CPU_ACCESS;
-		if (args->in.domains == AMDGPU_GEM_DOMAIN_GDS)
-			size = size << AMDGPU_GDS_SHIFT;
-		else if (args->in.domains == AMDGPU_GEM_DOMAIN_GWS)
-			size = size << AMDGPU_GWS_SHIFT;
-		else if (args->in.domains == AMDGPU_GEM_DOMAIN_OA)
-			size = size << AMDGPU_OA_SHIFT;
-		else
-			return -EINVAL;
+		/* GDS allocations must be DW aligned */
+		if (args->in.domains & AMDGPU_GEM_DOMAIN_GDS)
+			size = ALIGN(size, 4);
 	}
-	size = roundup(size, PAGE_SIZE);
 
 	if (flags & AMDGPU_GEM_CREATE_VM_ALWAYS_VALID) {
 		r = amdgpu_bo_reserve(vm->root.base.bo, false);
@@ -572,16 +566,16 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
-	if (args->va_address >= AMDGPU_VA_HOLE_START &&
-	    args->va_address < AMDGPU_VA_HOLE_END) {
+	if (args->va_address >= AMDGPU_GMC_HOLE_START &&
+	    args->va_address < AMDGPU_GMC_HOLE_END) {
 		dev_dbg(&dev->pdev->dev,
 			"va_address 0x%LX is in VA hole 0x%LX-0x%LX\n",
-			args->va_address, AMDGPU_VA_HOLE_START,
-			AMDGPU_VA_HOLE_END);
+			args->va_address, AMDGPU_GMC_HOLE_START,
+			AMDGPU_GMC_HOLE_END);
 		return -EINVAL;
 	}
 
-	args->va_address &= AMDGPU_VA_HOLE_MASK;
+	args->va_address &= AMDGPU_GMC_HOLE_MASK;
 
 	if ((args->flags & ~valid_flags) && (args->flags & ~prt_flags)) {
 		dev_dbg(&dev->pdev->dev, "invalid flags combination 0x%08X\n",

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

+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __AMDGPU_GEM_H__
+#define __AMDGPU_GEM_H__
+
+#include <drm/amdgpu_drm.h>
+#include <drm/drm_gem.h>
+
+/*
+ * GEM.
+ */
+
+#define AMDGPU_GEM_DOMAIN_MAX		0x3
+#define gem_to_amdgpu_bo(gobj) container_of((gobj), struct amdgpu_bo, gem_base)
+
+void amdgpu_gem_object_free(struct drm_gem_object *obj);
+int amdgpu_gem_object_open(struct drm_gem_object *obj,
+				struct drm_file *file_priv);
+void amdgpu_gem_object_close(struct drm_gem_object *obj,
+				struct drm_file *file_priv);
+unsigned long amdgpu_gem_timeout(uint64_t timeout_ns);
+struct sg_table *amdgpu_gem_prime_get_sg_table(struct drm_gem_object *obj);
+struct drm_gem_object *
+amdgpu_gem_prime_import_sg_table(struct drm_device *dev,
+				 struct dma_buf_attachment *attach,
+				 struct sg_table *sg);
+struct dma_buf *amdgpu_gem_prime_export(struct drm_device *dev,
+					struct drm_gem_object *gobj,
+					int flags);
+struct drm_gem_object *amdgpu_gem_prime_import(struct drm_device *dev,
+					    struct dma_buf *dma_buf);
+struct reservation_object *amdgpu_gem_prime_res_obj(struct drm_gem_object *);
+void *amdgpu_gem_prime_vmap(struct drm_gem_object *obj);
+void amdgpu_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
+int amdgpu_gem_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma);
+
+/*
+ * GEM objects.
+ */
+void amdgpu_gem_force_release(struct amdgpu_device *adev);
+int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
+			     int alignment, u32 initial_domain,
+			     u64 flags, enum ttm_bo_type type,
+			     struct reservation_object *resv,
+			     struct drm_gem_object **obj);
+
+int amdgpu_mode_dumb_create(struct drm_file *file_priv,
+			    struct drm_device *dev,
+			    struct drm_mode_create_dumb *args);
+int amdgpu_mode_dumb_mmap(struct drm_file *filp,
+			  struct drm_device *dev,
+			  uint32_t handle, uint64_t *offset_p);
+
+int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
+			    struct drm_file *filp);
+int amdgpu_gem_info_ioctl(struct drm_device *dev, void *data,
+			  struct drm_file *filp);
+int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
+			struct drm_file *filp);
+int amdgpu_gem_mmap_ioctl(struct drm_device *dev, void *data,
+			  struct drm_file *filp);
+int amdgpu_gem_wait_idle_ioctl(struct drm_device *dev, void *data,
+			      struct drm_file *filp);
+int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
+			  struct drm_file *filp);
+int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
+			struct drm_file *filp);
+
+int amdgpu_gem_metadata_ioctl(struct drm_device *dev, void *data,
+				struct drm_file *filp);
+
+#endif

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

@@ -26,9 +26,44 @@
 #include "amdgpu.h"
 #include "amdgpu_gfx.h"
 
+/* delay 0.1 second to enable gfx off feature */
+#define GFX_OFF_DELAY_ENABLE         msecs_to_jiffies(100)
+
 /*
- * GPU scratch registers helpers function.
+ * GPU GFX IP block helpers function.
  */
+
+int amdgpu_gfx_queue_to_bit(struct amdgpu_device *adev, int mec,
+			    int pipe, int queue)
+{
+	int bit = 0;
+
+	bit += mec * adev->gfx.mec.num_pipe_per_mec
+		* adev->gfx.mec.num_queue_per_pipe;
+	bit += pipe * adev->gfx.mec.num_queue_per_pipe;
+	bit += queue;
+
+	return bit;
+}
+
+void amdgpu_gfx_bit_to_queue(struct amdgpu_device *adev, int bit,
+			     int *mec, int *pipe, int *queue)
+{
+	*queue = bit % adev->gfx.mec.num_queue_per_pipe;
+	*pipe = (bit / adev->gfx.mec.num_queue_per_pipe)
+		% adev->gfx.mec.num_pipe_per_mec;
+	*mec = (bit / adev->gfx.mec.num_queue_per_pipe)
+	       / adev->gfx.mec.num_pipe_per_mec;
+
+}
+
+bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev,
+				     int mec, int pipe, int queue)
+{
+	return test_bit(amdgpu_gfx_queue_to_bit(adev, mec, pipe, queue),
+			adev->gfx.mec.queue_bitmap);
+}
+
 /**
  * amdgpu_gfx_scratch_get - Allocate a scratch register
  *
@@ -340,3 +375,40 @@ void amdgpu_gfx_compute_mqd_sw_fini(struct amdgpu_device *adev)
 			      &ring->mqd_gpu_addr,
 			      &ring->mqd_ptr);
 }
+
+/* amdgpu_gfx_off_ctrl - Handle gfx off feature enable/disable
+ *
+ * @adev: amdgpu_device pointer
+ * @bool enable true: enable gfx off feature, false: disable gfx off feature
+ *
+ * 1. gfx off feature will be enabled by gfx ip after gfx cg gp enabled.
+ * 2. other client can send request to disable gfx off feature, the request should be honored.
+ * 3. other client can cancel their request of disable gfx off feature
+ * 4. other client should not send request to enable gfx off feature before disable gfx off feature.
+ */
+
+void amdgpu_gfx_off_ctrl(struct amdgpu_device *adev, bool enable)
+{
+	if (!(adev->powerplay.pp_feature & PP_GFXOFF_MASK))
+		return;
+
+	if (!adev->powerplay.pp_funcs->set_powergating_by_smu)
+		return;
+
+
+	mutex_lock(&adev->gfx.gfx_off_mutex);
+
+	if (!enable)
+		adev->gfx.gfx_off_req_count++;
+	else if (adev->gfx.gfx_off_req_count > 0)
+		adev->gfx.gfx_off_req_count--;
+
+	if (enable && !adev->gfx.gfx_off_state && !adev->gfx.gfx_off_req_count) {
+		schedule_delayed_work(&adev->gfx.gfx_off_delay_work, GFX_OFF_DELAY_ENABLE);
+	} else if (!enable && adev->gfx.gfx_off_state) {
+		if (!amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_GFX, false))
+			adev->gfx.gfx_off_state = false;
+	}
+
+	mutex_unlock(&adev->gfx.gfx_off_mutex);
+}

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

+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the
+ * "Software"), to deal in the Software without restriction, including
+ * without limitation the rights to use, copy, modify, merge, publish,
+ * distribute, sub license, and/or sell copies of the Software, and to
+ * permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
+ * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+ * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+ * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * The above copyright notice and this permission notice (including the
+ * next paragraph) shall be included in all copies or substantial portions
+ * of the Software.
+ *
+ */
+
+#include "amdgpu.h"
+
+/**
+ * amdgpu_gmc_get_pde_for_bo - get the PDE for a BO
+ *
+ * @bo: the BO to get the PDE for
+ * @level: the level in the PD hirarchy
+ * @addr: resulting addr
+ * @flags: resulting flags
+ *
+ * Get the address and flags to be used for a PDE (Page Directory Entry).
+ */
+void amdgpu_gmc_get_pde_for_bo(struct amdgpu_bo *bo, int level,
+			       uint64_t *addr, uint64_t *flags)
+{
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+	struct ttm_dma_tt *ttm;
+
+	switch (bo->tbo.mem.mem_type) {
+	case TTM_PL_TT:
+		ttm = container_of(bo->tbo.ttm, struct ttm_dma_tt, ttm);
+		*addr = ttm->dma_address[0];
+		break;
+	case TTM_PL_VRAM:
+		*addr = amdgpu_bo_gpu_offset(bo);
+		break;
+	default:
+		*addr = 0;
+		break;
+	}
+	*flags = amdgpu_ttm_tt_pde_flags(bo->tbo.ttm, &bo->tbo.mem);
+	amdgpu_gmc_get_vm_pde(adev, level, addr, flags);
+}
+
+/**
+ * amdgpu_gmc_pd_addr - return the address of the root directory
+ *
+ */
+uint64_t amdgpu_gmc_pd_addr(struct amdgpu_bo *bo)
+{
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+	uint64_t pd_addr;
+
+	/* TODO: move that into ASIC specific code */
+	if (adev->asic_type >= CHIP_VEGA10) {
+		uint64_t flags = AMDGPU_PTE_VALID;
+
+		amdgpu_gmc_get_pde_for_bo(bo, -1, &pd_addr, &flags);
+		pd_addr |= flags;
+	} else {
+		pd_addr = amdgpu_bo_gpu_offset(bo);
+	}
+	return pd_addr;
+}
+
+/**
+ * amdgpu_gmc_agp_addr - return the address in the AGP address space
+ *
+ * @tbo: TTM BO which needs the address, must be in GTT domain
+ *
+ * Tries to figure out how to access the BO through the AGP aperture. Returns
+ * AMDGPU_BO_INVALID_OFFSET if that is not possible.
+ */
+uint64_t amdgpu_gmc_agp_addr(struct ttm_buffer_object *bo)
+{
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
+	struct ttm_dma_tt *ttm;
+
+	if (bo->num_pages != 1 || bo->ttm->caching_state == tt_cached)
+		return AMDGPU_BO_INVALID_OFFSET;
+
+	ttm = container_of(bo->ttm, struct ttm_dma_tt, ttm);
+	if (ttm->dma_address[0] + PAGE_SIZE >= adev->gmc.agp_size)
+		return AMDGPU_BO_INVALID_OFFSET;
+
+	return adev->gmc.agp_start + ttm->dma_address[0];
+}
+
+/**
+ * amdgpu_gmc_vram_location - try to find VRAM location
+ *
+ * @adev: amdgpu device structure holding all necessary informations
+ * @mc: memory controller structure holding memory informations
+ * @base: base address at which to put VRAM
+ *
+ * Function will try to place VRAM at base address provided
+ * as parameter.
+ */
+void amdgpu_gmc_vram_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc,
+			      u64 base)
+{
+	uint64_t limit = (uint64_t)amdgpu_vram_limit << 20;
+
+	mc->vram_start = base;
+	mc->vram_end = mc->vram_start + mc->mc_vram_size - 1;
+	if (limit && limit < mc->real_vram_size)
+		mc->real_vram_size = limit;
+
+	if (mc->xgmi.num_physical_nodes == 0) {
+		mc->fb_start = mc->vram_start;
+		mc->fb_end = mc->vram_end;
+	}
+	dev_info(adev->dev, "VRAM: %lluM 0x%016llX - 0x%016llX (%lluM used)\n",
+			mc->mc_vram_size >> 20, mc->vram_start,
+			mc->vram_end, mc->real_vram_size >> 20);
+}
+
+/**
+ * amdgpu_gmc_gart_location - try to find GART location
+ *
+ * @adev: amdgpu device structure holding all necessary informations
+ * @mc: memory controller structure holding memory informations
+ *
+ * Function will place try to place GART before or after VRAM.
+ *
+ * If GART size is bigger than space left then we ajust GART size.
+ * Thus function will never fails.
+ */
+void amdgpu_gmc_gart_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc)
+{
+	const uint64_t four_gb = 0x100000000ULL;
+	u64 size_af, size_bf;
+
+	mc->gart_size += adev->pm.smu_prv_buffer_size;
+
+	/* VCE doesn't like it when BOs cross a 4GB segment, so align
+	 * the GART base on a 4GB boundary as well.
+	 */
+	size_bf = mc->fb_start;
+	size_af = adev->gmc.mc_mask + 1 - ALIGN(mc->fb_end + 1, four_gb);
+
+	if (mc->gart_size > max(size_bf, size_af)) {
+		dev_warn(adev->dev, "limiting GART\n");
+		mc->gart_size = max(size_bf, size_af);
+	}
+
+	if ((size_bf >= mc->gart_size && size_bf < size_af) ||
+	    (size_af < mc->gart_size))
+		mc->gart_start = 0;
+	else
+		mc->gart_start = mc->mc_mask - mc->gart_size + 1;
+
+	mc->gart_start &= ~(four_gb - 1);
+	mc->gart_end = mc->gart_start + mc->gart_size - 1;
+	dev_info(adev->dev, "GART: %lluM 0x%016llX - 0x%016llX\n",
+			mc->gart_size >> 20, mc->gart_start, mc->gart_end);
+}
+
+/**
+ * amdgpu_gmc_agp_location - try to find AGP location
+ * @adev: amdgpu device structure holding all necessary informations
+ * @mc: memory controller structure holding memory informations
+ *
+ * Function will place try to find a place for the AGP BAR in the MC address
+ * space.
+ *
+ * AGP BAR will be assigned the largest available hole in the address space.
+ * Should be called after VRAM and GART locations are setup.
+ */
+void amdgpu_gmc_agp_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc)
+{
+	const uint64_t sixteen_gb = 1ULL << 34;
+	const uint64_t sixteen_gb_mask = ~(sixteen_gb - 1);
+	u64 size_af, size_bf;
+
+	if (mc->fb_start > mc->gart_start) {
+		size_bf = (mc->fb_start & sixteen_gb_mask) -
+			ALIGN(mc->gart_end + 1, sixteen_gb);
+		size_af = mc->mc_mask + 1 - ALIGN(mc->fb_end + 1, sixteen_gb);
+	} else {
+		size_bf = mc->fb_start & sixteen_gb_mask;
+		size_af = (mc->gart_start & sixteen_gb_mask) -
+			ALIGN(mc->fb_end + 1, sixteen_gb);
+	}
+
+	if (size_bf > size_af) {
+		mc->agp_start = mc->fb_start > mc->gart_start ?
+			mc->gart_end + 1 : 0;
+		mc->agp_size = size_bf;
+	} else {
+		mc->agp_start = (mc->fb_start > mc->gart_start ?
+			mc->fb_end : mc->gart_end) + 1,
+		mc->agp_size = size_af;
+	}
+
+	mc->agp_start = ALIGN(mc->agp_start, sixteen_gb);
+	mc->agp_end = mc->agp_start + mc->agp_size - 1;
+	dev_info(adev->dev, "AGP: %lluM 0x%016llX - 0x%016llX\n",
+			mc->agp_size >> 20, mc->agp_start, mc->agp_end);
+}

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

@@ -30,6 +30,19 @@
 
 #include "amdgpu_irq.h"
 
+/* VA hole for 48bit addresses on Vega10 */
+#define AMDGPU_GMC_HOLE_START	0x0000800000000000ULL
+#define AMDGPU_GMC_HOLE_END	0xffff800000000000ULL
+
+/*
+ * Hardware is programmed as if the hole doesn't exists with start and end
+ * address values.
+ *
+ * This mask is used to remove the upper 16bits of the VA and so come up with
+ * the linear addr value.
+ */
+#define AMDGPU_GMC_HOLE_MASK	0x0000ffffffffffffULL
+
 struct firmware;
 
 /*
@@ -74,6 +87,20 @@ struct amdgpu_gmc_funcs {
 			   u64 *dst, u64 *flags);
 };
 
+struct amdgpu_xgmi {
+	/* from psp */
+	u64 device_id;
+	u64 hive_id;
+	/* fixed per family */
+	u64 node_segment_size;
+	/* physical node (0-3) */
+	unsigned physical_node_id;
+	/* number of nodes (0-4) */
+	unsigned num_physical_nodes;
+	/* gpu list in the same hive */
+	struct list_head head;
+};
+
 struct amdgpu_gmc {
 	resource_size_t		aper_size;
 	resource_size_t		aper_base;
@@ -81,11 +108,22 @@ struct amdgpu_gmc {
 	 * about vram size near mc fb location */
 	u64			mc_vram_size;
 	u64			visible_vram_size;
+	u64			agp_size;
+	u64			agp_start;
+	u64			agp_end;
 	u64			gart_size;
 	u64			gart_start;
 	u64			gart_end;
 	u64			vram_start;
 	u64			vram_end;
+	/* FB region , it's same as local vram region in single GPU, in XGMI
+	 * configuration, this region covers all GPUs in the same hive ,
+	 * each GPU in the hive has the same view of this FB region .
+	 * GPU0's vram starts at offset (0 * segment size) ,
+	 * GPU1 starts at offset (1 * segment size), etc.
+	 */
+	u64			fb_start;
+	u64			fb_end;
 	unsigned		vram_width;
 	u64			real_vram_size;
 	int			vram_mtrr;
@@ -109,8 +147,17 @@ struct amdgpu_gmc {
 	atomic_t		vm_fault_info_updated;
 
 	const struct amdgpu_gmc_funcs	*gmc_funcs;
+
+	struct amdgpu_xgmi xgmi;
 };
 
+#define amdgpu_gmc_flush_gpu_tlb(adev, vmid) (adev)->gmc.gmc_funcs->flush_gpu_tlb((adev), (vmid))
+#define amdgpu_gmc_emit_flush_gpu_tlb(r, vmid, addr) (r)->adev->gmc.gmc_funcs->emit_flush_gpu_tlb((r), (vmid), (addr))
+#define amdgpu_gmc_emit_pasid_mapping(r, vmid, pasid) (r)->adev->gmc.gmc_funcs->emit_pasid_mapping((r), (vmid), (pasid))
+#define amdgpu_gmc_set_pte_pde(adev, pt, idx, addr, flags) (adev)->gmc.gmc_funcs->set_pte_pde((adev), (pt), (idx), (addr), (flags))
+#define amdgpu_gmc_get_vm_pde(adev, level, dst, flags) (adev)->gmc.gmc_funcs->get_vm_pde((adev), (level), (dst), (flags))
+#define amdgpu_gmc_get_pte_flags(adev, flags) (adev)->gmc.gmc_funcs->get_vm_pte_flags((adev),(flags))
+
 /**
  * amdgpu_gmc_vram_full_visible - Check if full VRAM is visible through the BAR
  *
@@ -126,4 +173,28 @@ static inline bool amdgpu_gmc_vram_full_visible(struct amdgpu_gmc *gmc)
 	return (gmc->real_vram_size == gmc->visible_vram_size);
 }
 
+/**
+ * amdgpu_gmc_sign_extend - sign extend the given gmc address
+ *
+ * @addr: address to extend
+ */
+static inline uint64_t amdgpu_gmc_sign_extend(uint64_t addr)
+{
+	if (addr >= AMDGPU_GMC_HOLE_START)
+		addr |= AMDGPU_GMC_HOLE_END;
+
+	return addr;
+}
+
+void amdgpu_gmc_get_pde_for_bo(struct amdgpu_bo *bo, int level,
+			       uint64_t *addr, uint64_t *flags);
+uint64_t amdgpu_gmc_pd_addr(struct amdgpu_bo *bo);
+uint64_t amdgpu_gmc_agp_addr(struct ttm_buffer_object *bo);
+void amdgpu_gmc_vram_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc,
+			      u64 base);
+void amdgpu_gmc_gart_location(struct amdgpu_device *adev,
+			      struct amdgpu_gmc *mc);
+void amdgpu_gmc_agp_location(struct amdgpu_device *adev,
+			     struct amdgpu_gmc *mc);
+
 #endif

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

@@ -32,6 +32,7 @@
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
 #include "atom.h"
+#include "amdgpu_trace.h"
 
 #define AMDGPU_IB_TEST_TIMEOUT	msecs_to_jiffies(1000)
 
@@ -170,6 +171,10 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 	     (amdgpu_sriov_vf(adev) && need_ctx_switch) ||
 	     amdgpu_vm_need_pipeline_sync(ring, job))) {
 		need_pipe_sync = true;
+
+		if (tmp)
+			trace_amdgpu_ib_pipe_sync(job, tmp);
+
 		dma_fence_put(tmp);
 	}
 

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

@@ -197,78 +197,3 @@ int amdgpu_ih_process(struct amdgpu_device *adev)
 	return IRQ_HANDLED;
 }
 
-/**
- * amdgpu_ih_add_fault - Add a page fault record
- *
- * @adev: amdgpu device pointer
- * @key: 64-bit encoding of PASID and address
- *
- * This should be called when a retry page fault interrupt is
- * received. If this is a new page fault, it will be added to a hash
- * table. The return value indicates whether this is a new fault, or
- * a fault that was already known and is already being handled.
- *
- * If there are too many pending page faults, this will fail. Retry
- * interrupts should be ignored in this case until there is enough
- * free space.
- *
- * Returns 0 if the fault was added, 1 if the fault was already known,
- * -ENOSPC if there are too many pending faults.
- */
-int amdgpu_ih_add_fault(struct amdgpu_device *adev, u64 key)
-{
-	unsigned long flags;
-	int r = -ENOSPC;
-
-	if (WARN_ON_ONCE(!adev->irq.ih.faults))
-		/* Should be allocated in <IP>_ih_sw_init on GPUs that
-		 * support retry faults and require retry filtering.
-		 */
-		return r;
-
-	spin_lock_irqsave(&adev->irq.ih.faults->lock, flags);
-
-	/* Only let the hash table fill up to 50% for best performance */
-	if (adev->irq.ih.faults->count >= (1 << (AMDGPU_PAGEFAULT_HASH_BITS-1)))
-		goto unlock_out;
-
-	r = chash_table_copy_in(&adev->irq.ih.faults->hash, key, NULL);
-	if (!r)
-		adev->irq.ih.faults->count++;
-
-	/* chash_table_copy_in should never fail unless we're losing count */
-	WARN_ON_ONCE(r < 0);
-
-unlock_out:
-	spin_unlock_irqrestore(&adev->irq.ih.faults->lock, flags);
-	return r;
-}
-
-/**
- * amdgpu_ih_clear_fault - Remove a page fault record
- *
- * @adev: amdgpu device pointer
- * @key: 64-bit encoding of PASID and address
- *
- * This should be called when a page fault has been handled. Any
- * future interrupt with this key will be processed as a new
- * page fault.
- */
-void amdgpu_ih_clear_fault(struct amdgpu_device *adev, u64 key)
-{
-	unsigned long flags;
-	int r;
-
-	if (!adev->irq.ih.faults)
-		return;
-
-	spin_lock_irqsave(&adev->irq.ih.faults->lock, flags);
-
-	r = chash_table_remove(&adev->irq.ih.faults->hash, key, NULL);
-	if (!WARN_ON_ONCE(r < 0)) {
-		adev->irq.ih.faults->count--;
-		WARN_ON_ONCE(adev->irq.ih.faults->count < 0);
-	}
-
-	spin_unlock_irqrestore(&adev->irq.ih.faults->lock, flags);
-}

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

@@ -24,7 +24,6 @@
 #ifndef __AMDGPU_IH_H__
 #define __AMDGPU_IH_H__
 
-#include <linux/chash.h>
 #include "soc15_ih_clientid.h"
 
 struct amdgpu_device;
@@ -32,13 +31,6 @@ struct amdgpu_device;
 #define AMDGPU_IH_CLIENTID_LEGACY 0
 #define AMDGPU_IH_CLIENTID_MAX SOC15_IH_CLIENTID_MAX
 
-#define AMDGPU_PAGEFAULT_HASH_BITS 8
-struct amdgpu_retryfault_hashtable {
-	DECLARE_CHASH_TABLE(hash, AMDGPU_PAGEFAULT_HASH_BITS, 8, 0);
-	spinlock_t	lock;
-	int		count;
-};
-
 /*
  * R6xx+ IH ring
  */
@@ -57,7 +49,6 @@ struct amdgpu_ih_ring {
 	bool			use_doorbell;
 	bool			use_bus_addr;
 	dma_addr_t		rb_dma_addr; /* only used when use_bus_addr = true */
-	struct amdgpu_retryfault_hashtable *faults;
 };
 
 #define AMDGPU_IH_SRC_DATA_MAX_SIZE_DW 4
@@ -76,11 +67,24 @@ struct amdgpu_iv_entry {
 	const uint32_t *iv_entry;
 };
 
+/* provided by the ih block */
+struct amdgpu_ih_funcs {
+	/* ring read/write ptr handling, called from interrupt context */
+	u32 (*get_wptr)(struct amdgpu_device *adev);
+	bool (*prescreen_iv)(struct amdgpu_device *adev);
+	void (*decode_iv)(struct amdgpu_device *adev,
+			  struct amdgpu_iv_entry *entry);
+	void (*set_rptr)(struct amdgpu_device *adev);
+};
+
+#define amdgpu_ih_get_wptr(adev) (adev)->irq.ih_funcs->get_wptr((adev))
+#define amdgpu_ih_prescreen_iv(adev) (adev)->irq.ih_funcs->prescreen_iv((adev))
+#define amdgpu_ih_decode_iv(adev, iv) (adev)->irq.ih_funcs->decode_iv((adev), (iv))
+#define amdgpu_ih_set_rptr(adev) (adev)->irq.ih_funcs->set_rptr((adev))
+
 int amdgpu_ih_ring_init(struct amdgpu_device *adev, unsigned ring_size,
 			bool use_bus_addr);
 void amdgpu_ih_ring_fini(struct amdgpu_device *adev);
 int amdgpu_ih_process(struct amdgpu_device *adev);
-int amdgpu_ih_add_fault(struct amdgpu_device *adev, u64 key);
-void amdgpu_ih_clear_fault(struct amdgpu_device *adev, u64 key);
 
 #endif

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

@@ -105,8 +105,8 @@ static void amdgpu_irq_reset_work_func(struct work_struct *work)
 	struct amdgpu_device *adev = container_of(work, struct amdgpu_device,
 						  reset_work);
 
-	if (!amdgpu_sriov_vf(adev))
-		amdgpu_device_gpu_recover(adev, NULL, false);
+	if (!amdgpu_sriov_vf(adev) && amdgpu_device_should_recover_gpu(adev))
+		amdgpu_device_gpu_recover(adev, NULL);
 }
 
 /**

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

@@ -33,11 +33,18 @@ static void amdgpu_job_timedout(struct drm_sched_job *s_job)
 	struct amdgpu_ring *ring = to_amdgpu_ring(s_job->sched);
 	struct amdgpu_job *job = to_amdgpu_job(s_job);
 
+	if (amdgpu_ring_soft_recovery(ring, job->vmid, s_job->s_fence->parent)) {
+		DRM_ERROR("ring %s timeout, but soft recovered\n",
+			  s_job->sched->name);
+		return;
+	}
+
 	DRM_ERROR("ring %s timeout, signaled seq=%u, emitted seq=%u\n",
 		  job->base.sched->name, atomic_read(&ring->fence_drv.last_seq),
 		  ring->fence_drv.sync_seq);
 
-	amdgpu_device_gpu_recover(ring->adev, job, false);
+	if (amdgpu_device_should_recover_gpu(ring->adev))
+		amdgpu_device_gpu_recover(ring->adev, job);
 }
 
 int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs,
@@ -66,6 +73,7 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs,
 	amdgpu_sync_create(&(*job)->sync);
 	amdgpu_sync_create(&(*job)->sched_sync);
 	(*job)->vram_lost_counter = atomic_read(&adev->vram_lost_counter);
+	(*job)->vm_pd_addr = AMDGPU_BO_INVALID_OFFSET;
 
 	return 0;
 }
@@ -82,8 +90,6 @@ int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size,
 	r = amdgpu_ib_get(adev, NULL, size, &(*job)->ibs[0]);
 	if (r)
 		kfree(*job);
-	else
-		(*job)->vm_pd_addr = adev->gart.table_addr;
 
 	return r;
 }

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

@@ -51,18 +51,6 @@
  *
  */
 
-static bool amdgpu_bo_need_backup(struct amdgpu_device *adev)
-{
-	if (adev->flags & AMD_IS_APU)
-		return false;
-
-	if (amdgpu_gpu_recovery == 0 ||
-	    (amdgpu_gpu_recovery == -1  && !amdgpu_sriov_vf(adev)))
-		return false;
-
-	return true;
-}
-
 /**
  * amdgpu_bo_subtract_pin_size - Remove BO from pin_size accounting
  *
@@ -163,10 +151,7 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 
 	if (domain & AMDGPU_GEM_DOMAIN_GTT) {
 		places[c].fpfn = 0;
-		if (flags & AMDGPU_GEM_CREATE_SHADOW)
-			places[c].lpfn = adev->gmc.gart_size >> PAGE_SHIFT;
-		else
-			places[c].lpfn = 0;
+		places[c].lpfn = 0;
 		places[c].flags = TTM_PL_FLAG_TT;
 		if (flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC)
 			places[c].flags |= TTM_PL_FLAG_WC |
@@ -253,6 +238,11 @@ int amdgpu_bo_create_reserved(struct amdgpu_device *adev,
 	bool free = false;
 	int r;
 
+	if (!size) {
+		amdgpu_bo_unref(bo_ptr);
+		return 0;
+	}
+
 	memset(&bp, 0, sizeof(bp));
 	bp.size = size;
 	bp.byte_align = align;
@@ -346,7 +336,8 @@ int amdgpu_bo_create_kernel(struct amdgpu_device *adev,
 	if (r)
 		return r;
 
-	amdgpu_bo_unreserve(*bo_ptr);
+	if (*bo_ptr)
+		amdgpu_bo_unreserve(*bo_ptr);
 
 	return 0;
 }
@@ -436,7 +427,11 @@ static int amdgpu_bo_do_create(struct amdgpu_device *adev,
 	int r;
 
 	page_align = roundup(bp->byte_align, PAGE_SIZE) >> PAGE_SHIFT;
-	size = ALIGN(size, PAGE_SIZE);
+	if (bp->domain & (AMDGPU_GEM_DOMAIN_GDS | AMDGPU_GEM_DOMAIN_GWS |
+			  AMDGPU_GEM_DOMAIN_OA))
+		size <<= PAGE_SHIFT;
+	else
+		size = ALIGN(size, PAGE_SIZE);
 
 	if (!amdgpu_bo_validate_size(adev, size, bp->domain))
 		return -ENOMEM;
@@ -451,7 +446,7 @@ static int amdgpu_bo_do_create(struct amdgpu_device *adev,
 		return -ENOMEM;
 	drm_gem_private_object_init(adev->ddev, &bo->gem_base, size);
 	INIT_LIST_HEAD(&bo->shadow_list);
-	INIT_LIST_HEAD(&bo->va);
+	bo->vm_bo = NULL;
 	bo->preferred_domains = bp->preferred_domain ? bp->preferred_domain :
 		bp->domain;
 	bo->allowed_domains = bo->preferred_domains;
@@ -541,7 +536,7 @@ static int amdgpu_bo_do_create(struct amdgpu_device *adev,
 }
 
 static int amdgpu_bo_create_shadow(struct amdgpu_device *adev,
-				   unsigned long size, int byte_align,
+				   unsigned long size,
 				   struct amdgpu_bo *bo)
 {
 	struct amdgpu_bo_param bp;
@@ -552,7 +547,6 @@ static int amdgpu_bo_create_shadow(struct amdgpu_device *adev,
 
 	memset(&bp, 0, sizeof(bp));
 	bp.size = size;
-	bp.byte_align = byte_align;
 	bp.domain = AMDGPU_GEM_DOMAIN_GTT;
 	bp.flags = AMDGPU_GEM_CREATE_CPU_GTT_USWC |
 		AMDGPU_GEM_CREATE_SHADOW;
@@ -563,7 +557,7 @@ static int amdgpu_bo_create_shadow(struct amdgpu_device *adev,
 	if (!r) {
 		bo->shadow->parent = amdgpu_bo_ref(bo);
 		mutex_lock(&adev->shadow_list_lock);
-		list_add_tail(&bo->shadow_list, &adev->shadow_list);
+		list_add_tail(&bo->shadow->shadow_list, &adev->shadow_list);
 		mutex_unlock(&adev->shadow_list_lock);
 	}
 
@@ -596,12 +590,12 @@ int amdgpu_bo_create(struct amdgpu_device *adev,
 	if (r)
 		return r;
 
-	if ((flags & AMDGPU_GEM_CREATE_SHADOW) && amdgpu_bo_need_backup(adev)) {
+	if ((flags & AMDGPU_GEM_CREATE_SHADOW) && !(adev->flags & AMD_IS_APU)) {
 		if (!bp->resv)
 			WARN_ON(reservation_object_lock((*bo_ptr)->tbo.resv,
 							NULL));
 
-		r = amdgpu_bo_create_shadow(adev, bp->size, bp->byte_align, (*bo_ptr));
+		r = amdgpu_bo_create_shadow(adev, bp->size, *bo_ptr);
 
 		if (!bp->resv)
 			reservation_object_unlock((*bo_ptr)->tbo.resv);
@@ -695,13 +689,10 @@ int amdgpu_bo_validate(struct amdgpu_bo *bo)
 }
 
 /**
- * amdgpu_bo_restore_from_shadow - restore an &amdgpu_bo buffer object
- * @adev: amdgpu device object
- * @ring: amdgpu_ring for the engine handling the buffer operations
- * @bo: &amdgpu_bo buffer to be restored
- * @resv: reservation object with embedded fence
+ * amdgpu_bo_restore_shadow - restore an &amdgpu_bo shadow
+ *
+ * @shadow: &amdgpu_bo shadow to be restored
  * @fence: dma_fence associated with the operation
- * @direct: whether to submit the job directly
  *
  * Copies a buffer object's shadow content back to the object.
  * This is used for recovering a buffer from its shadow in case of a gpu
@@ -710,36 +701,19 @@ int amdgpu_bo_validate(struct amdgpu_bo *bo)
  * Returns:
  * 0 for success or a negative error code on failure.
  */
-int amdgpu_bo_restore_from_shadow(struct amdgpu_device *adev,
-				  struct amdgpu_ring *ring,
-				  struct amdgpu_bo *bo,
-				  struct reservation_object *resv,
-				  struct dma_fence **fence,
-				  bool direct)
+int amdgpu_bo_restore_shadow(struct amdgpu_bo *shadow, struct dma_fence **fence)
 
 {
-	struct amdgpu_bo *shadow = bo->shadow;
-	uint64_t bo_addr, shadow_addr;
-	int r;
-
-	if (!shadow)
-		return -EINVAL;
+	struct amdgpu_device *adev = amdgpu_ttm_adev(shadow->tbo.bdev);
+	struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
+	uint64_t shadow_addr, parent_addr;
 
-	bo_addr = amdgpu_bo_gpu_offset(bo);
-	shadow_addr = amdgpu_bo_gpu_offset(bo->shadow);
+	shadow_addr = amdgpu_bo_gpu_offset(shadow);
+	parent_addr = amdgpu_bo_gpu_offset(shadow->parent);
 
-	r = reservation_object_reserve_shared(bo->tbo.resv);
-	if (r)
-		goto err;
-
-	r = amdgpu_copy_buffer(ring, shadow_addr, bo_addr,
-			       amdgpu_bo_size(bo), resv, fence,
-			       direct, false);
-	if (!r)
-		amdgpu_bo_fence(bo, *fence, true);
-
-err:
-	return r;
+	return amdgpu_copy_buffer(ring, shadow_addr, parent_addr,
+				  amdgpu_bo_size(shadow), NULL, fence,
+				  true, false);
 }
 
 /**
@@ -1019,10 +993,12 @@ int amdgpu_bo_unpin(struct amdgpu_bo *bo)
 int amdgpu_bo_evict_vram(struct amdgpu_device *adev)
 {
 	/* late 2.6.33 fix IGP hibernate - we need pm ops to do this correct */
-	if (0 && (adev->flags & AMD_IS_APU)) {
+#ifndef CONFIG_HIBERNATION
+	if (adev->flags & AMD_IS_APU) {
 		/* Useless to evict on IGP chips */
 		return 0;
 	}
+#endif
 	return ttm_bo_evict_mm(&adev->mman.bdev, TTM_PL_VRAM);
 }
 
@@ -1360,15 +1336,13 @@ void amdgpu_bo_fence(struct amdgpu_bo *bo, struct dma_fence *fence,
 u64 amdgpu_bo_gpu_offset(struct amdgpu_bo *bo)
 {
 	WARN_ON_ONCE(bo->tbo.mem.mem_type == TTM_PL_SYSTEM);
-	WARN_ON_ONCE(bo->tbo.mem.mem_type == TTM_PL_TT &&
-		     !amdgpu_gtt_mgr_has_gart_addr(&bo->tbo.mem));
 	WARN_ON_ONCE(!ww_mutex_is_locked(&bo->tbo.resv->lock) &&
-		     !bo->pin_count);
+		     !bo->pin_count && bo->tbo.type != ttm_bo_type_kernel);
 	WARN_ON_ONCE(bo->tbo.mem.start == AMDGPU_BO_INVALID_OFFSET);
 	WARN_ON_ONCE(bo->tbo.mem.mem_type == TTM_PL_VRAM &&
 		     !(bo->flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS));
 
-	return bo->tbo.offset;
+	return amdgpu_gmc_sign_extend(bo->tbo.offset);
 }
 
 /**

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

@@ -226,6 +226,8 @@ static bool amdgpu_sa_bo_next_hole(struct amdgpu_sa_manager *sa_manager,
 	for (i = 0; i < AMDGPU_SA_NUM_FENCE_LISTS; ++i) {
 		struct amdgpu_sa_bo *sa_bo;
 
+		fences[i] = NULL;
+
 		if (list_empty(&sa_manager->flist[i]))
 			continue;
 
@@ -296,10 +298,8 @@ int amdgpu_sa_bo_new(struct amdgpu_sa_manager *sa_manager,
 
 	spin_lock(&sa_manager->wq.lock);
 	do {
-		for (i = 0; i < AMDGPU_SA_NUM_FENCE_LISTS; ++i) {
-			fences[i] = NULL;
+		for (i = 0; i < AMDGPU_SA_NUM_FENCE_LISTS; ++i)
 			tries[i] = 0;
-		}
 
 		do {
 			amdgpu_sa_bo_try_free(sa_manager);