Merge tag 'amd-drm-next-6.7-2023-10-27' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.7-2023-10-27:

amdgpu:
- RAS fixes
- Seamless boot fixes
- NBIO 7.7 fix
- SMU 14.0 fixes
- GC 11.5 fixes
- DML2 fixes
- ASPM fixes
- VPE fixes
- Misc code cleanups
- SRIOV fixes
- Add some missing copyright notices
- DCN 3.5 fixes
- FAMS fixes
- Backlight fix
- S/G display fix
- fdinfo cleanups
- EXT_COHERENT fixes for APU and NUMA systems

amdkfd:
- Misc fixes
- Misc code cleanups
- SVM fixes

Signed-off-by: Dave Airlie <[email protected]>

From: Alex Deucher <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

Author: airlied

drivers/gpu/drm/amd/amdgpu/Makefile

@@ -104,7 +104,8 @@ amdgpu-y += \
 amdgpu-y += \
 	df_v1_7.o \
 	df_v3_6.o \
-	df_v4_3.o
+	df_v4_3.o \
+	df_v4_6_2.o
 
 # add GMC block
 amdgpu-y += \

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

@@ -1119,6 +1119,13 @@ static inline uint32_t amdgpu_ip_version(const struct amdgpu_device *adev,
 	return adev->ip_versions[ip][inst] & ~0xFFU;
 }
 
+static inline uint32_t amdgpu_ip_version_full(const struct amdgpu_device *adev,
+					      uint8_t ip, uint8_t inst)
+{
+	/* This returns full version - major/minor/rev/variant/subrevision */
+	return adev->ip_versions[ip][inst];
+}
+
 static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev)
 {
 	return container_of(ddev, struct amdgpu_device, ddev);
@@ -1333,9 +1340,7 @@ void amdgpu_device_pci_config_reset(struct amdgpu_device *adev);
 int amdgpu_device_pci_reset(struct amdgpu_device *adev);
 bool amdgpu_device_need_post(struct amdgpu_device *adev);
 bool amdgpu_device_seamless_boot_supported(struct amdgpu_device *adev);
-bool amdgpu_device_pcie_dynamic_switching_supported(void);
 bool amdgpu_device_should_use_aspm(struct amdgpu_device *adev);
-bool amdgpu_device_aspm_support_quirk(void);
 
 void amdgpu_cs_report_moved_bytes(struct amdgpu_device *adev, u64 num_bytes,
 				  u64 num_vis_bytes);

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

@@ -68,7 +68,7 @@ struct amdgpu_acpi_xcc_info {
 struct amdgpu_acpi_dev_info {
 	struct list_head list;
 	struct list_head xcc_list;
-	uint16_t bdf;
+	uint32_t sbdf;
 	uint16_t supp_xcp_mode;
 	uint16_t xcp_mode;
 	uint16_t mem_mode;
@@ -927,22 +927,22 @@ static acpi_status amdgpu_acpi_get_node_id(acpi_handle handle,
 #endif
 }
 
-static struct amdgpu_acpi_dev_info *amdgpu_acpi_get_dev(u16 bdf)
+static struct amdgpu_acpi_dev_info *amdgpu_acpi_get_dev(u32 sbdf)
 {
 	struct amdgpu_acpi_dev_info *acpi_dev;
 
 	if (list_empty(&amdgpu_acpi_dev_list))
 		return NULL;
 
 	list_for_each_entry(acpi_dev, &amdgpu_acpi_dev_list, list)
-		if (acpi_dev->bdf == bdf)
+		if (acpi_dev->sbdf == sbdf)
 			return acpi_dev;
 
 	return NULL;
 }
 
 static int amdgpu_acpi_dev_init(struct amdgpu_acpi_dev_info **dev_info,
-				struct amdgpu_acpi_xcc_info *xcc_info, u16 bdf)
+				struct amdgpu_acpi_xcc_info *xcc_info, u32 sbdf)
 {
 	struct amdgpu_acpi_dev_info *tmp;
 	union acpi_object *obj;
@@ -955,7 +955,7 @@ static int amdgpu_acpi_dev_init(struct amdgpu_acpi_dev_info **dev_info,
 
 	INIT_LIST_HEAD(&tmp->xcc_list);
 	INIT_LIST_HEAD(&tmp->list);
-	tmp->bdf = bdf;
+	tmp->sbdf = sbdf;
 
 	obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0,
 				      AMD_XCC_DSM_GET_SUPP_MODE, NULL,
@@ -1007,7 +1007,7 @@ static int amdgpu_acpi_dev_init(struct amdgpu_acpi_dev_info **dev_info,
 
 	DRM_DEBUG_DRIVER(
 		"New dev(%x): Supported xcp mode: %x curr xcp_mode : %x mem mode : %x, tmr base: %llx tmr size: %llx  ",
-		tmp->bdf, tmp->supp_xcp_mode, tmp->xcp_mode, tmp->mem_mode,
+		tmp->sbdf, tmp->supp_xcp_mode, tmp->xcp_mode, tmp->mem_mode,
 		tmp->tmr_base, tmp->tmr_size);
 	list_add_tail(&tmp->list, &amdgpu_acpi_dev_list);
 	*dev_info = tmp;
@@ -1023,7 +1023,7 @@ static int amdgpu_acpi_dev_init(struct amdgpu_acpi_dev_info **dev_info,
 }
 
 static int amdgpu_acpi_get_xcc_info(struct amdgpu_acpi_xcc_info *xcc_info,
-				    u16 *bdf)
+				    u32 *sbdf)
 {
 	union acpi_object *obj;
 	acpi_status status;
@@ -1054,8 +1054,10 @@ static int amdgpu_acpi_get_xcc_info(struct amdgpu_acpi_xcc_info *xcc_info,
 	xcc_info->phy_id = (obj->integer.value >> 32) & 0xFF;
 	/* xcp node of this xcc [47:40] */
 	xcc_info->xcp_node = (obj->integer.value >> 40) & 0xFF;
+	/* PF domain of this xcc [31:16] */
+	*sbdf = (obj->integer.value) & 0xFFFF0000;
 	/* PF bus/dev/fn of this xcc [63:48] */
-	*bdf = (obj->integer.value >> 48) & 0xFFFF;
+	*sbdf |= (obj->integer.value >> 48) & 0xFFFF;
 	ACPI_FREE(obj);
 	obj = NULL;
 
@@ -1079,7 +1081,7 @@ static int amdgpu_acpi_enumerate_xcc(void)
 	struct acpi_device *acpi_dev;
 	char hid[ACPI_ID_LEN];
 	int ret, id;
-	u16 bdf;
+	u32 sbdf;
 
 	INIT_LIST_HEAD(&amdgpu_acpi_dev_list);
 	xa_init(&numa_info_xa);
@@ -1107,16 +1109,16 @@ static int amdgpu_acpi_enumerate_xcc(void)
 		xcc_info->handle = acpi_device_handle(acpi_dev);
 		acpi_dev_put(acpi_dev);
 
-		ret = amdgpu_acpi_get_xcc_info(xcc_info, &bdf);
+		ret = amdgpu_acpi_get_xcc_info(xcc_info, &sbdf);
 		if (ret) {
 			kfree(xcc_info);
 			continue;
 		}
 
-		dev_info = amdgpu_acpi_get_dev(bdf);
+		dev_info = amdgpu_acpi_get_dev(sbdf);
 
 		if (!dev_info)
-			ret = amdgpu_acpi_dev_init(&dev_info, xcc_info, bdf);
+			ret = amdgpu_acpi_dev_init(&dev_info, xcc_info, sbdf);
 
 		if (ret == -ENOMEM)
 			return ret;
@@ -1136,13 +1138,14 @@ int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev, u64 *tmr_offset,
 			     u64 *tmr_size)
 {
 	struct amdgpu_acpi_dev_info *dev_info;
-	u16 bdf;
+	u32 sbdf;
 
 	if (!tmr_offset || !tmr_size)
 		return -EINVAL;
 
-	bdf = pci_dev_id(adev->pdev);
-	dev_info = amdgpu_acpi_get_dev(bdf);
+	sbdf = (pci_domain_nr(adev->pdev->bus) << 16);
+	sbdf |= pci_dev_id(adev->pdev);
+	dev_info = amdgpu_acpi_get_dev(sbdf);
 	if (!dev_info)
 		return -ENOENT;
 
@@ -1157,13 +1160,14 @@ int amdgpu_acpi_get_mem_info(struct amdgpu_device *adev, int xcc_id,
 {
 	struct amdgpu_acpi_dev_info *dev_info;
 	struct amdgpu_acpi_xcc_info *xcc_info;
-	u16 bdf;
+	u32 sbdf;
 
 	if (!numa_info)
 		return -EINVAL;
 
-	bdf = pci_dev_id(adev->pdev);
-	dev_info = amdgpu_acpi_get_dev(bdf);
+	sbdf = (pci_domain_nr(adev->pdev->bus) << 16);
+	sbdf |= pci_dev_id(adev->pdev);
+	dev_info = amdgpu_acpi_get_dev(sbdf);
 	if (!dev_info)
 		return -ENOENT;
 

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

@@ -748,6 +748,9 @@ static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf,
 	ssize_t result = 0;
 	int r;
 
+	if (!adev->smc_rreg)
+		return -EPERM;
+
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
@@ -804,6 +807,9 @@ static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user *
 	ssize_t result = 0;
 	int r;
 
+	if (!adev->smc_wreg)
+		return -EPERM;
+
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 

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

@@ -1456,14 +1456,14 @@ bool amdgpu_device_seamless_boot_supported(struct amdgpu_device *adev)
 }
 
 /*
- * Intel hosts such as Raptor Lake and Sapphire Rapids don't support dynamic
- * speed switching. Until we have confirmation from Intel that a specific host
- * supports it, it's safer that we keep it disabled for all.
+ * Intel hosts such as Rocket Lake, Alder Lake, Raptor Lake and Sapphire Rapids
+ * don't support dynamic speed switching. Until we have confirmation from Intel
+ * that a specific host supports it, it's safer that we keep it disabled for all.
  *
  * https://edc.intel.com/content/www/us/en/design/products/platforms/details/raptor-lake-s/13th-generation-core-processors-datasheet-volume-1-of-2/005/pci-express-support/
  * https://gitlab.freedesktop.org/drm/amd/-/issues/2663
  */
-bool amdgpu_device_pcie_dynamic_switching_supported(void)
+static bool amdgpu_device_pcie_dynamic_switching_supported(void)
 {
 #if IS_ENABLED(CONFIG_X86)
 	struct cpuinfo_x86 *c = &cpu_data(0);
@@ -1496,20 +1496,13 @@ bool amdgpu_device_should_use_aspm(struct amdgpu_device *adev)
 	default:
 		return false;
 	}
+	if (adev->flags & AMD_IS_APU)
+		return false;
+	if (!(adev->pm.pp_feature & PP_PCIE_DPM_MASK))
+		return false;
 	return pcie_aspm_enabled(adev->pdev);
 }
 
-bool amdgpu_device_aspm_support_quirk(void)
-{
-#if IS_ENABLED(CONFIG_X86)
-	struct cpuinfo_x86 *c = &cpu_data(0);
-
-	return !(c->x86 == 6 && c->x86_model == INTEL_FAM6_ALDERLAKE);
-#else
-	return true;
-#endif
-}
-
 /* if we get transitioned to only one device, take VGA back */
 /**
  * amdgpu_device_vga_set_decode - enable/disable vga decode
@@ -2315,6 +2308,8 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 		adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
 	if (amdgpu_sriov_vf(adev) && adev->asic_type == CHIP_SIENNA_CICHLID)
 		adev->pm.pp_feature &= ~PP_OVERDRIVE_MASK;
+	if (!amdgpu_device_pcie_dynamic_switching_supported())
+		adev->pm.pp_feature &= ~PP_PCIE_DPM_MASK;
 
 	total = true;
 	for (i = 0; i < adev->num_ip_blocks; i++) {

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

@@ -35,6 +35,7 @@
 #include "df_v1_7.h"
 #include "df_v3_6.h"
 #include "df_v4_3.h"
+#include "df_v4_6_2.h"
 #include "nbio_v6_1.h"
 #include "nbio_v7_0.h"
 #include "nbio_v7_4.h"
@@ -1487,7 +1488,7 @@ static int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev)
 		adev->gfx.config.num_sc_per_sh = le32_to_cpu(gc_info->v2.gc_num_sc_per_se) /
 			le32_to_cpu(gc_info->v2.gc_num_sh_per_se);
 		adev->gfx.config.num_packer_per_sc = le32_to_cpu(gc_info->v2.gc_num_packer_per_sc);
-		if (le16_to_cpu(gc_info->v2.header.version_minor == 1)) {
+		if (le16_to_cpu(gc_info->v2.header.version_minor) == 1) {
 			adev->gfx.config.gc_num_tcp_per_sa = le32_to_cpu(gc_info->v2_1.gc_num_tcp_per_sh);
 			adev->gfx.config.gc_tcp_size_per_cu = le32_to_cpu(gc_info->v2_1.gc_tcp_size_per_cu);
 			adev->gfx.config.gc_num_sdp_interface = le32_to_cpu(gc_info->v2_1.gc_num_sdp_interface); /* per XCD */
@@ -2559,6 +2560,9 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 	case IP_VERSION(4, 3, 0):
 		adev->df.funcs = &df_v4_3_funcs;
 		break;
+	case IP_VERSION(4, 6, 2):
+		adev->df.funcs = &df_v4_6_2_funcs;
+		break;
 	default:
 		break;
 	}

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

@@ -56,21 +56,15 @@ static const char *amdgpu_ip_name[AMDGPU_HW_IP_NUM] = {
 
 void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file)
 {
-	struct amdgpu_device *adev = drm_to_adev(file->minor->dev);
 	struct amdgpu_fpriv *fpriv = file->driver_priv;
 	struct amdgpu_vm *vm = &fpriv->vm;
 
 	struct amdgpu_mem_stats stats;
 	ktime_t usage[AMDGPU_HW_IP_NUM];
-	uint32_t bus, dev, fn, domain;
 	unsigned int hw_ip;
 	int ret;
 
 	memset(&stats, 0, sizeof(stats));
-	bus = adev->pdev->bus->number;
-	domain = pci_domain_nr(adev->pdev->bus);
-	dev = PCI_SLOT(adev->pdev->devfn);
-	fn = PCI_FUNC(adev->pdev->devfn);
 
 	ret = amdgpu_bo_reserve(vm->root.bo, false);
 	if (ret)
@@ -88,9 +82,6 @@ void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file)
 	 */
 
 	drm_printf(p, "pasid:\t%u\n", fpriv->vm.pasid);
-	drm_printf(p, "drm-driver:\t%s\n", file->minor->dev->driver->name);
-	drm_printf(p, "drm-pdev:\t%04x:%02x:%02x.%d\n", domain, bus, dev, fn);
-	drm_printf(p, "drm-client-id:\t%llu\n", vm->immediate.fence_context);
 	drm_printf(p, "drm-memory-vram:\t%llu KiB\n", stats.vram/1024UL);
 	drm_printf(p, "drm-memory-gtt: \t%llu KiB\n", stats.gtt/1024UL);
 	drm_printf(p, "drm-memory-cpu: \t%llu KiB\n", stats.cpu/1024UL);

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

@@ -786,6 +786,7 @@ void amdgpu_gmc_tmz_set(struct amdgpu_device *adev)
 	/* YELLOW_CARP*/
 	case IP_VERSION(10, 3, 3):
 	case IP_VERSION(11, 0, 4):
+	case IP_VERSION(11, 5, 0):
 		/* Don't enable it by default yet.
 		 */
 		if (amdgpu_tmz < 1) {