Skip to contentSkip to conversation

FAILED: patch "[PATCH] drm/amdgpu: fix aperture mapping leak" failed to apply to 6.1-stable tree

PATCH [email protected] 1 participant 1 message Jul 29

Subsystems (3)
lore
Share & tools
lore

[email protected]

author maintainer
#0 lore raw

The patch below does not apply to the 6.1-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y
git checkout FETCH_HEAD
git cherry-pick -x ea772a440d56b285f4d491affac50ecd41f6b402
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2026072923-squeak-tying-6839@gregkh' --subject-prefix 'PATCH 6.1.y' 'HEAD^..'

Possible dependencies:

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From ea772a440d56b285f4d491affac50ecd41f6b402 Mon Sep 17 00:00:00 2001
From: Asad Kamal <[email protected]>
Date: Sun, 14 Jun 2026 12:50:28 +0800
Subject: [PATCH] drm/amdgpu: fix aperture mapping leak
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

amdgpu_pci_remove() calls drm_dev_unplug() before invoking the driver fini routines. This causes drm_dev_enter() in amdgpu_ttm_fini() to always return false, so iounmap(aper_base_kaddr) never runs on normal driver unload, leaving an orphaned entry in the x86 PAT interval tree.

On connected_to_cpu hardware, the aperture is mapped write-back (WB) via ioremap_cache(). On reload, IP discovery calls memremap(..., MEMREMAP_WC) over the same range. The WC vs WB conflict causes:

ioremap error for 0x..., requested 0x1, got 0x0
amdgpu: discovery failed: -2

Fix by switching to devres-managed mappings so cleanup is guaranteed regardless of drm_dev_enter() state:

- connected_to_cpu path: devm_memremap(MEMREMAP_WB). For IORESOURCE_SYSTEM_RAM ranges this takes the try_ram_remap() shortcut, returning __va(offset) from the existing kernel direct map. No new ioremap VA or PAT entry is created, so there is nothing to orphan.

- dGPU path: devm_ioremap_wc() registers iounmap() as a devres action, guaranteeing cleanup at device_del() time.

Also remove iounmap(aper_base_kaddr) from amdgpu_device_unmap_mmio() since the mapping is now devres-owned.

v2: Remove redundant x86_64 guard (Lijo)

Fixes 9d0af8b4def0 ("drm/amdgpu: pre-map device buffer as cached for A+A config")
(cherry picked from commit d871e99879cb5fd1fa798b006b4888887e63a17a)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +0 -2
index 211d30f03d25..8d6502a94306 100644
@@ -4184,8 +4184,6 @@ static void amdgpu_device_unmap_mmio(struct amdgpu_device *adev)
 
 	iounmap(adev->rmmio);
 	adev->rmmio = NULL;
-	if (adev->mman.aper_base_kaddr)
-		iounmap(adev->mman.aper_base_kaddr);
 	adev->mman.aper_base_kaddr = NULL;
 
 	/* Memory manager related */
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +16 -20
index 16c060badaee..00b5317f77f8 100644
@@ -2118,18 +2118,23 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	/* Change the size here instead of the init above so only lpfn is affected */
 	amdgpu_ttm_disable_buffer_funcs(adev);
 #ifdef CONFIG_64BIT
-#ifdef CONFIG_X86
-	if (adev->gmc.xgmi.connected_to_cpu)
-		adev->mman.aper_base_kaddr = ioremap_cache(adev->gmc.aper_base,
-				adev->gmc.visible_vram_size);
-
-	else if (adev->gmc.is_app_apu)
+	if (adev->gmc.xgmi.connected_to_cpu) {
+		void *kaddr = devm_memremap(adev->dev, adev->gmc.aper_base,
+					    adev->gmc.visible_vram_size,
+					    MEMREMAP_WB);
+		if (IS_ERR(kaddr))
+			return PTR_ERR(kaddr);
+		adev->mman.aper_base_kaddr = (__force void __iomem *)kaddr;
+	} else if (adev->gmc.is_app_apu) {
 		DRM_DEBUG_DRIVER(
 			"No need to ioremap when real vram size is 0\n");
-	else
-#endif
-		adev->mman.aper_base_kaddr = ioremap_wc(adev->gmc.aper_base,
-				adev->gmc.visible_vram_size);
+	} else {
+		adev->mman.aper_base_kaddr = devm_ioremap_wc(adev->dev,
+							     adev->gmc.aper_base,
+							     adev->gmc.visible_vram_size);
+		if (!adev->mman.aper_base_kaddr)
+			return -ENOMEM;
+	}
 #endif
 
 	amdgpu_ttm_init_vram_resv_regions(adev);
@@ -2246,8 +2251,6 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
  */
 void amdgpu_ttm_fini(struct amdgpu_device *adev)
 {
-	int idx;
-
 	if (!adev->mman.initialized)
 		return;
 
@@ -2270,14 +2273,7 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 	amdgpu_ttm_unmark_vram_reserved(adev, AMDGPU_RESV_FW_VRAM_USAGE);
 	amdgpu_ttm_unmark_vram_reserved(adev, AMDGPU_RESV_DRV_VRAM_USAGE);
 
-	if (drm_dev_enter(adev_to_drm(adev), &idx)) {
-
-		if (adev->mman.aper_base_kaddr)
-			iounmap(adev->mman.aper_base_kaddr);
-		adev->mman.aper_base_kaddr = NULL;
-
-		drm_dev_exit(idx);
-	}
+	adev->mman.aper_base_kaddr = NULL;
 
 	if (!adev->gmc.is_app_apu)
 		amdgpu_vram_mgr_fini(adev);