Merge tag 'amd-drm-next-5.19-2022-05-18' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-5.19-2022-05-18:

amdgpu:
- Misc code cleanups
- Additional SMU 13.x enablement
- Smartshift fixes
- GFX11 fixes
- Support for SMU 13.0.4
- SMU mutex fix
- Suspend/resume fix

amdkfd:
- static checker fix
- Doorbell/MMIO resource handling fix

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220518205621.5741-1-alexander.deucher@amd.com

Author: airlied

drivers/gpu/drm/amd/amdgpu/Makefile

@@ -58,7 +58,7 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \
 	amdgpu_vm_sdma.o amdgpu_discovery.o amdgpu_ras_eeprom.o amdgpu_nbio.o \
 	amdgpu_umc.o smu_v11_0_i2c.o amdgpu_fru_eeprom.o amdgpu_rap.o \
 	amdgpu_fw_attestation.o amdgpu_securedisplay.o \
-	amdgpu_eeprom.o amdgpu_mca.o amdgpu_psp_ta.o
+	amdgpu_eeprom.o amdgpu_mca.o amdgpu_psp_ta.o amdgpu_lsdma.o
 
 amdgpu-$(CONFIG_PROC_FS) += amdgpu_fdinfo.o
 
@@ -74,7 +74,8 @@ 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 nbio_v7_4.o nbio_v2_3.o nv.o arct_reg_init.o mxgpu_nv.o \
-	nbio_v7_2.o hdp_v4_0.o hdp_v5_0.o aldebaran_reg_init.o aldebaran.o soc21.o
+	nbio_v7_2.o hdp_v4_0.o hdp_v5_0.o aldebaran_reg_init.o aldebaran.o soc21.o \
+	nbio_v4_3.o hdp_v6_0.o nbio_v7_7.o hdp_v5_2.o lsdma_v6_0.o
 
 # add DF block
 amdgpu-y += \
@@ -87,7 +88,7 @@ amdgpu-y += \
 	gmc_v8_0.o \
 	gfxhub_v1_0.o mmhub_v1_0.o gmc_v9_0.o gfxhub_v1_1.o mmhub_v9_4.o \
 	gfxhub_v2_0.o mmhub_v2_0.o gmc_v10_0.o gfxhub_v2_1.o mmhub_v2_3.o \
-	mmhub_v1_7.o
+	mmhub_v1_7.o gfxhub_v3_0.o mmhub_v3_0.o mmhub_v3_0_2.o gmc_v11_0.o
 
 # add UMC block
 amdgpu-y += \
@@ -102,7 +103,8 @@ amdgpu-y += \
 	cz_ih.o \
 	vega10_ih.o \
 	vega20_ih.o \
-	navi10_ih.o
+	navi10_ih.o \
+	ih_v6_0.o
 
 # add PSP block
 amdgpu-y += \
@@ -128,7 +130,9 @@ amdgpu-y += \
 	gfx_v9_0.o \
 	gfx_v9_4.o \
 	gfx_v9_4_2.o \
-	gfx_v10_0.o
+	gfx_v10_0.o \
+	imu_v11_0.o \
+	gfx_v11_0.o
 
 # add async DMA block
 amdgpu-y += \
@@ -138,11 +142,14 @@ amdgpu-y += \
 	sdma_v4_0.o \
 	sdma_v4_4.o \
 	sdma_v5_0.o \
-	sdma_v5_2.o
+	sdma_v5_2.o \
+	sdma_v6_0.o
 
 # add MES block
 amdgpu-y += \
-	mes_v10_1.o
+	amdgpu_mes.o \
+	mes_v10_1.o \
+	mes_v11_0.o
 
 # add UVD block
 amdgpu-y += \
@@ -160,28 +167,33 @@ amdgpu-y += \
 # add VCN and JPEG block
 amdgpu-y += \
 	amdgpu_vcn.o \
+	vcn_sw_ring.o \
 	vcn_v1_0.o \
 	vcn_v2_0.o \
 	vcn_v2_5.o \
 	vcn_v3_0.o \
+	vcn_v4_0.o \
 	amdgpu_jpeg.o \
 	jpeg_v1_0.o \
 	jpeg_v2_0.o \
 	jpeg_v2_5.o \
-	jpeg_v3_0.o
+	jpeg_v3_0.o \
+	jpeg_v4_0.o
 
 # add ATHUB block
 amdgpu-y += \
 	athub_v1_0.o \
 	athub_v2_0.o \
-	athub_v2_1.o
+	athub_v2_1.o \
+	athub_v3_0.o
 
 # add SMUIO block
 amdgpu-y += \
 	smuio_v9_0.o \
 	smuio_v11_0.o \
 	smuio_v11_0_6.o \
-	smuio_v13_0.o
+	smuio_v13_0.o \
+	smuio_v13_0_6.o
 
 # add reset block
 amdgpu-y += \
@@ -207,7 +219,8 @@ amdgpu-y += \
 	amdgpu_amdkfd_arcturus.o \
 	amdgpu_amdkfd_aldebaran.o \
 	amdgpu_amdkfd_gfx_v10.o \
-	amdgpu_amdkfd_gfx_v10_3.o
+	amdgpu_amdkfd_gfx_v10_3.o \
+	amdgpu_amdkfd_gfx_v11.o
 
 ifneq ($(CONFIG_DRM_AMDGPU_CIK),)
 amdgpu-y += amdgpu_amdkfd_gfx_v7.o

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

@@ -86,11 +86,13 @@
 #include "amdgpu_gmc.h"
 #include "amdgpu_gfx.h"
 #include "amdgpu_sdma.h"
+#include "amdgpu_lsdma.h"
 #include "amdgpu_nbio.h"
 #include "amdgpu_hdp.h"
 #include "amdgpu_dm.h"
 #include "amdgpu_virt.h"
 #include "amdgpu_csa.h"
+#include "amdgpu_mes_ctx.h"
 #include "amdgpu_gart.h"
 #include "amdgpu_debugfs.h"
 #include "amdgpu_job.h"
@@ -207,6 +209,7 @@ extern int amdgpu_async_gfx_ring;
 extern int amdgpu_mcbp;
 extern int amdgpu_discovery;
 extern int amdgpu_mes;
+extern int amdgpu_mes_kiq;
 extern int amdgpu_noretry;
 extern int amdgpu_force_asic_type;
 extern int amdgpu_smartshift_bias;
@@ -641,6 +644,7 @@ enum amd_hw_ip_block_type {
 	SDMA5_HWIP,
 	SDMA6_HWIP,
 	SDMA7_HWIP,
+	LSDMA_HWIP,
 	MMHUB_HWIP,
 	ATHUB_HWIP,
 	NBIO_HWIP,
@@ -720,6 +724,26 @@ struct ip_discovery_top;
 					  (rid == 0x01) || \
 					  (rid == 0x10))))
 
+struct amdgpu_mqd_prop {
+	uint64_t mqd_gpu_addr;
+	uint64_t hqd_base_gpu_addr;
+	uint64_t rptr_gpu_addr;
+	uint64_t wptr_gpu_addr;
+	uint32_t queue_size;
+	bool use_doorbell;
+	uint32_t doorbell_index;
+	uint64_t eop_gpu_addr;
+	uint32_t hqd_pipe_priority;
+	uint32_t hqd_queue_priority;
+	bool hqd_active;
+};
+
+struct amdgpu_mqd {
+	unsigned mqd_size;
+	int (*init_mqd)(struct amdgpu_device *adev, void *mqd,
+			struct amdgpu_mqd_prop *p);
+};
+
 #define AMDGPU_RESET_MAGIC_NUM 64
 #define AMDGPU_MAX_DF_PERFMONS 4
 #define AMDGPU_PRODUCT_NAME_LEN 64
@@ -887,6 +911,9 @@ struct amdgpu_device {
 	/* sdma */
 	struct amdgpu_sdma		sdma;
 
+	/* lsdma */
+	struct amdgpu_lsdma		lsdma;
+
 	/* uvd */
 	struct amdgpu_uvd		uvd;
 
@@ -919,7 +946,9 @@ struct amdgpu_device {
 
 	/* mes */
 	bool                            enable_mes;
+	bool                            enable_mes_kiq;
 	struct amdgpu_mes               mes;
+	struct amdgpu_mqd               mqds[AMDGPU_HW_IP_NUM];
 
 	/* df */
 	struct amdgpu_df                df;
@@ -981,10 +1010,10 @@ struct amdgpu_device {
 	bool                            runpm;
 	bool                            in_runpm;
 	bool                            has_pr3;
-	bool                            is_fw_fb;
 
 	bool                            pm_sysfs_en;
 	bool                            ucode_sysfs_en;
+	bool                            psp_sysfs_en;
 
 	/* Chip product information */
 	char				product_number[16];
@@ -1016,6 +1045,9 @@ struct amdgpu_device {
 	/* reset dump register */
 	uint32_t                        *reset_dump_reg_list;
 	int                             num_regs;
+
+	bool                            scpm_enabled;
+	uint32_t                        scpm_status;
 };
 
 static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev)
@@ -1188,7 +1220,8 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 #define amdgpu_asic_flush_hdp(adev, r) \
 	((adev)->asic_funcs->flush_hdp ? (adev)->asic_funcs->flush_hdp((adev), (r)) : (adev)->hdp.funcs->flush_hdp((adev), (r)))
 #define amdgpu_asic_invalidate_hdp(adev, r) \
-	((adev)->asic_funcs->invalidate_hdp ? (adev)->asic_funcs->invalidate_hdp((adev), (r)) : (adev)->hdp.funcs->invalidate_hdp((adev), (r)))
+	((adev)->asic_funcs->invalidate_hdp ? (adev)->asic_funcs->invalidate_hdp((adev), (r)) : \
+	 ((adev)->hdp.funcs->invalidate_hdp ? (adev)->hdp.funcs->invalidate_hdp((adev), (r)) : 0))
 #define amdgpu_asic_need_full_reset(adev) (adev)->asic_funcs->need_full_reset((adev))
 #define amdgpu_asic_init_doorbell_index(adev) (adev)->asic_funcs->init_doorbell_index((adev))
 #define amdgpu_asic_get_pcie_usage(adev, cnt0, cnt1) ((adev)->asic_funcs->get_pcie_usage((adev), (cnt0), (cnt1)))
@@ -1345,9 +1378,11 @@ static inline int amdgpu_acpi_smart_shift_update(struct drm_device *dev,
 
 #if defined(CONFIG_ACPI) && defined(CONFIG_SUSPEND)
 bool amdgpu_acpi_is_s3_active(struct amdgpu_device *adev);
+bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev);
 bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev);
 #else
 static inline bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev) { return false; }
+static inline bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev) { return false; }
 static inline bool amdgpu_acpi_is_s3_active(struct amdgpu_device *adev) { return false; }
 #endif
 

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

@@ -1045,6 +1045,20 @@ bool amdgpu_acpi_is_s3_active(struct amdgpu_device *adev)
 		(pm_suspend_target_state == PM_SUSPEND_MEM);
 }
 
+/**
+ * amdgpu_acpi_should_gpu_reset
+ *
+ * @adev: amdgpu_device_pointer
+ *
+ * returns true if should reset GPU, false if not
+ */
+bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev)
+{
+	if (adev->flags & AMD_IS_APU)
+		return false;
+	return pm_suspend_target_state != PM_SUSPEND_TO_IDLE;
+}
+
 /**
  * amdgpu_acpi_is_s0ix_active
  *

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

@@ -100,7 +100,18 @@ static void amdgpu_doorbell_get_kfd_info(struct amdgpu_device *adev,
 	 * The first num_doorbells are used by amdgpu.
 	 * amdkfd takes whatever's left in the aperture.
 	 */
-	if (adev->doorbell.size > adev->doorbell.num_doorbells * sizeof(u32)) {
+	if (adev->enable_mes) {
+		/*
+		 * With MES enabled, we only need to initialize
+		 * the base address. The size and offset are
+		 * not initialized as AMDGPU manages the whole
+		 * doorbell space.
+		 */
+		*aperture_base = adev->doorbell.base;
+		*aperture_size = 0;
+		*start_offset = 0;
+	} else if (adev->doorbell.size > adev->doorbell.num_doorbells *
+						sizeof(u32)) {
 		*aperture_base = adev->doorbell.base;
 		*aperture_size = adev->doorbell.size;
 		*start_offset = adev->doorbell.num_doorbells * sizeof(u32);
@@ -128,7 +139,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 					  AMDGPU_GMC_HOLE_START),
 			.drm_render_minor = adev_to_drm(adev)->render->index,
 			.sdma_doorbell_idx = adev->doorbell_index.sdma_engine,
-
+			.enable_mes = adev->enable_mes,
 		};
 
 		/* this is going to have a few of the MSBs set that we need to