[PATCH 94/95] drm/amdgpu: add mtype_remote module parameter

[Alex Deucher <alexander.deucher@amd.com>]

From: Mukul Joshi <mukul.joshi@amd.com>

Add an amdgpu_mtype_remote module parameter so the MTYPE used for remote
memory accesses can be overridden. For now only MTYPE_NC and MTYPE_UC
are selectable (0 = MTYPE_NC, 1 = MTYPE_UC); it defaults to the
ASIC-dependent value. Currently, it is used only for GFX 12.1.

The MTYPEs resolved for both local and remote memory are logged once.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c |  7 +++++
 drivers/gpu/drm/amd/amdgpu/gmc_v12_1.c  | 34 +++++++++++++------------
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c    |  8 ++++--
 4 files changed, 32 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index f619d95a5ccfe..79b69d74eb2e5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -223,6 +223,7 @@ extern int amdgpu_force_asic_type;
 extern int amdgpu_smartshift_bias;
 extern int amdgpu_use_xgmi_p2p;
 extern int amdgpu_mtype_local;
+extern int amdgpu_mtype_remote;
 extern int amdgpu_enforce_isolation;
 extern uint amdgpu_debug_mask;
 #ifdef CONFIG_HSA_AMD
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 53738b40c97f6..04b21e456fbc0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -844,6 +844,13 @@ int amdgpu_mtype_local = -1;
 MODULE_PARM_DESC(mtype_local, "MTYPE for local memory (default: ASIC dependent, 0 = MTYPE_RW, 1 = MTYPE_NC, 2 = MTYPE_CC)");
 module_param_named_unsafe(mtype_local, amdgpu_mtype_local, int, 0444);
 
+/**
+ * DOC: mtype_remote (int)
+ */
+int amdgpu_mtype_remote = -1;
+MODULE_PARM_DESC(mtype_remote, "MTYPE for remote memory (default: ASIC dependent, 0 = MTYPE_NC, 1 = MTYPE_UC)");
+module_param_named_unsafe(mtype_remote, amdgpu_mtype_remote, int, 0444);
+
 /**
  * DOC: pcie_p2p (bool)
  * Enable PCIe P2P (requires large-BAR). Default value: true (on)
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_1.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_1.c
index a3b7e240abaed..4e7b6e2c87ab5 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_1.c
@@ -618,8 +618,8 @@ static void gmc_v12_1_get_npa_flags(struct amdgpu_device *adev,
 
 /*
  * Resolve the MTYPEs used for local and remote memory accesses on GFX 12.1.
- * Remote memory always uses MTYPE_UC; local memory depends on the AID stepping
- * and the amdgpu_mtype_local module parameter.
+ * Both default to an ASIC-dependent value that can be overridden through the
+ * amdgpu_mtype_local and amdgpu_mtype_remote module parameters.
  */
 static void gmc_v12_1_get_mtypes(struct amdgpu_device *adev,
 				 unsigned int *mtype_local,
@@ -627,23 +627,25 @@ static void gmc_v12_1_get_mtypes(struct amdgpu_device *adev,
 {
 	bool is_aid_a1 = (adev->rev_id & 0x10);
 
+	/* Local memory: ASIC default depends on the AID stepping. */
 	*mtype_local = is_aid_a1 ? MTYPE_RW : MTYPE_NC;
-	/* Remote memory always uses MTYPE_UC on GFX 12.1. */
-	*mtype_remote = MTYPE_UC;
-
-	if (amdgpu_mtype_local == 0) {
-		DRM_INFO_ONCE("Using MTYPE_RW for local memory\n");
+	if (amdgpu_mtype_local == 0)
 		*mtype_local = MTYPE_RW;
-	} else if (amdgpu_mtype_local == 1) {
-		DRM_INFO_ONCE("Using MTYPE_NC for local memory\n");
+	else if (amdgpu_mtype_local == 1)
 		*mtype_local = MTYPE_NC;
-	} else if (amdgpu_mtype_local == 2) {
-		DRM_INFO_ONCE("MTYPE_CC not supported, using %s for local memory\n",
-			      is_aid_a1 ? "MTYPE_RW" : "MTYPE_NC");
-	} else {
-		DRM_INFO_ONCE("Using %s for local memory and MTYPE_UC for remote memory\n",
-			      is_aid_a1 ? "MTYPE_RW" : "MTYPE_NC");
-	}
+	else if (amdgpu_mtype_local == 2)
+		DRM_INFO_ONCE("MTYPE_CC not supported for local memory\n");
+
+	/* Remote memory defaults to MTYPE_UC on GFX 12.1. */
+	*mtype_remote = MTYPE_UC;
+	if (amdgpu_mtype_remote == 0)
+		*mtype_remote = MTYPE_NC;
+	else if (amdgpu_mtype_remote == 1)
+		*mtype_remote = MTYPE_UC;
+
+	DRM_INFO_ONCE("Using %s for local memory and %s for remote memory\n",
+		      *mtype_local == MTYPE_RW ? "MTYPE_RW" : "MTYPE_NC",
+		      *mtype_remote == MTYPE_NC ? "MTYPE_NC" : "MTYPE_UC");
 }
 
 /*
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 64fb38400bd1a..c521ec07bf3c0 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1348,8 +1348,12 @@ svm_range_get_pte_flags(struct kfd_node *node, struct amdgpu_vm *vm,
 		mtype_local = amdgpu_mtype_local == 0 ? AMDGPU_VM_MTYPE_RW :
 				amdgpu_mtype_local == 1 ? AMDGPU_VM_MTYPE_NC :
 				is_aid_a1 ? AMDGPU_VM_MTYPE_RW : AMDGPU_VM_MTYPE_NC;
-		/* Remote memory always uses MTYPE_UC on GFX 12.1. */
-		mtype_remote = AMDGPU_VM_MTYPE_UC;
+		/* Remote memory defaults to MTYPE_UC on GFX 12.1 and can be
+		 * overridden through the amdgpu_mtype_remote module parameter
+		 * (0 = MTYPE_NC, 1 = MTYPE_UC).
+		 */
+		mtype_remote = amdgpu_mtype_remote == 0 ? AMDGPU_VM_MTYPE_NC :
+				AMDGPU_VM_MTYPE_UC;
 		snoop = true;
 
 		if (is_local) /* local HBM  */ {
-- 
2.55.0