AMD General Please ignore will send another v2
-----Original Message----- From: Kamal, Asad <[email protected]> Sent: Tuesday, June 23, 2026 8:11 PM To: [email protected] Cc: Lazar, Lijo <[email protected]>; Zhang, Hawking <[email protected]>; Ma, Le <[email protected]>; Zhang, Morris <[email protected]>; Deucher, Alexander <[email protected]>; Wang, Yang(Kevin) <[email protected]>; Kamal, Asad <[email protected]>; Koenig, Christian <[email protected]> Subject: [PATCH v2] drm/amdgpu: fix aperture mapping leak 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") Signed-off-by: Asad Kamal <[email protected]> Reviewed-by: Christian König <[email protected]> --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 -- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 39 +++++++++------------- 2 files changed, 15 insertions(+), 26 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 5442a1fc1c37..70d07ca187a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4190,8 +4190,6 @@ 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 */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index f80371bc74d7..352ea6820fcb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -2118,20 +2118,20 @@ 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) - 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); -#endif + 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 (IS_ENABLED(CONFIG_64BIT) && !adev->gmc.is_app_apu) { + 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; + } amdgpu_ttm_init_vram_resv_regions(adev); @@ -2247,8 +2247,6 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) */ void amdgpu_ttm_fini(struct amdgpu_device *adev) { - int idx; - if (!adev->mman.initialized) return; @@ -2271,14 +2269,7 @@ 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); -- 2.46.0
