January 2023 - Linux-stable-mirror

FAILED: patch "[PATCH] drm/amd/display: Fix invalid DPIA AUX reply causing system" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

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 <stable(a)vger.kernel.org>. Possible dependencies: a5d313b4165c ("drm/amd/display: Fix invalid DPIA AUX reply causing system hang") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From a5d313b4165ca08686d5d41ee08b0a8cab2737ca Mon Sep 17 00:00:00 2001 From: Stylon Wang <stylon.wang(a)amd.com> Date: Wed, 26 Oct 2022 21:00:40 +0800 Subject: [PATCH] drm/amd/display: Fix invalid DPIA AUX reply causing system hang [Why] Some DPIA AUX replies have incorrect data length from original request. This could lead to overwriting of destination buffer if reply length is larger, which could cause invalid access to stack since many destination buffers are declared as local variables. [How] Check for invalid length from DPIA AUX replies and trigger a retry if reply length is not the same as original request. A DRM_WARN() dmesg log is also produced. Reviewed-by: Roman Li <Roman.Li(a)amd.com> Acked-by: Tom Chung <chiahsuan.chung(a)amd.com> Signed-off-by: Stylon Wang <stylon.wang(a)amd.com> Tested-by: Daniel Wheeler <daniel.wheeler(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org # 6.0.x diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 2dc75b1b005b..fc7da97b6c00 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -147,6 +147,14 @@ MODULE_FIRMWARE(FIRMWARE_NAVI12_DMCU); /* Number of bytes in PSP footer for firmware. */ #define PSP_FOOTER_BYTES 0x100 +/* + * DMUB Async to Sync Mechanism Status + */ +#define DMUB_ASYNC_TO_SYNC_ACCESS_FAIL 1 +#define DMUB_ASYNC_TO_SYNC_ACCESS_TIMEOUT 2 +#define DMUB_ASYNC_TO_SYNC_ACCESS_SUCCESS 3 +#define DMUB_ASYNC_TO_SYNC_ACCESS_INVALID 4 + /** * DOC: overview * @@ -10176,6 +10184,8 @@ static int amdgpu_dm_set_dmub_async_sync_status(bool is_cmd_aux, *operation_result = AUX_RET_ERROR_TIMEOUT; } else if (status_type == DMUB_ASYNC_TO_SYNC_ACCESS_FAIL) { *operation_result = AUX_RET_ERROR_ENGINE_ACQUIRE; + } else if (status_type == DMUB_ASYNC_TO_SYNC_ACCESS_INVALID) { + *operation_result = AUX_RET_ERROR_INVALID_REPLY; } else { *operation_result = AUX_RET_ERROR_UNKNOWN; } @@ -10223,6 +10233,16 @@ int amdgpu_dm_process_dmub_aux_transfer_sync(bool is_cmd_aux, struct dc_context payload->reply[0] = adev->dm.dmub_notify->aux_reply.command; if (!payload->write && adev->dm.dmub_notify->aux_reply.length && payload->reply[0] == AUX_TRANSACTION_REPLY_AUX_ACK) { + + if (payload->length != adev->dm.dmub_notify->aux_reply.length) { + DRM_WARN("invalid read from DPIA AUX %x(%d) got length %d!\n", + payload->address, payload->length, + adev->dm.dmub_notify->aux_reply.length); + return amdgpu_dm_set_dmub_async_sync_status(is_cmd_aux, ctx, + DMUB_ASYNC_TO_SYNC_ACCESS_INVALID, + (uint32_t *)operation_result); + } + memcpy(payload->data, adev->dm.dmub_notify->aux_reply.data, adev->dm.dmub_notify->aux_reply.length); } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index b618b2586e7b..83436ef3b26b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -50,12 +50,6 @@ #define AMDGPU_DMUB_NOTIFICATION_MAX 5 -/* - * DMUB Async to Sync Mechanism Status - */ -#define DMUB_ASYNC_TO_SYNC_ACCESS_FAIL 1 -#define DMUB_ASYNC_TO_SYNC_ACCESS_TIMEOUT 2 -#define DMUB_ASYNC_TO_SYNC_ACCESS_SUCCESS 3 /* #include "include/amdgpu_dal_power_if.h" #include "amdgpu_dm_irq.h"

2 years, 1 month

1
0
0 0

FAILED: patch "[PATCH] drm/amdgpu: fix userptr HMM range handling v2" failed to apply to 6.0-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.0-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 <stable(a)vger.kernel.org>. Possible dependencies: fec8fdb54e8f ("drm/amdgpu: fix userptr HMM range handling v2") 4953b6b22ab9 ("drm/amdgpu: cleanup error handling in amdgpu_cs_parser_bos") 736ec9fadd7a ("drm/amdgpu: move setting the job resources") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From fec8fdb54e8f74d88951c9f998f47bf4f2031fe0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig(a)amd.com> Date: Thu, 10 Nov 2022 12:31:41 +0100 Subject: [PATCH] drm/amdgpu: fix userptr HMM range handling v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The basic problem here is that it's not allowed to page fault while holding the reservation lock. So it can happen that multiple processes try to validate an userptr at the same time. Work around that by putting the HMM range object into the mutex protected bo list for now. v2: make sure range is set to NULL in case of an error Signed-off-by: Christian König <christian.koenig(a)amd.com> Reviewed-by: Alex Deucher <alexander.deucher(a)amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling(a)amd.com> CC: stable(a)vger.kernel.org Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 68741b157153..e44d740022bc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -938,6 +938,7 @@ static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr, struct amdkfd_process_info *process_info = mem->process_info; struct amdgpu_bo *bo = mem->bo; struct ttm_operation_ctx ctx = { true, false }; + struct hmm_range *range; int ret = 0; mutex_lock(&process_info->lock); @@ -967,7 +968,7 @@ static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr, return 0; } - ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages); + ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages, &range); if (ret) { pr_err("%s: Failed to get user pages: %d\n", __func__, ret); goto unregister_out; @@ -985,7 +986,7 @@ static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr, amdgpu_bo_unreserve(bo); release_out: - amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, range); unregister_out: if (ret) amdgpu_mn_unregister(bo); @@ -2317,6 +2318,8 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info, /* Go through userptr_inval_list and update any invalid user_pages */ list_for_each_entry(mem, &process_info->userptr_inval_list, validate_list.head) { + struct hmm_range *range; + invalid = atomic_read(&mem->invalid); if (!invalid) /* BO hasn't been invalidated since the last @@ -2327,7 +2330,8 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info, bo = mem->bo; /* Get updated user pages */ - ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages); + ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages, + &range); if (ret) { pr_debug("Failed %d to get user pages\n", ret); @@ -2346,7 +2350,7 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info, * FIXME: Cannot ignore the return code, must hold * notifier_lock */ - amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, range); } /* Mark the BO as valid unless it was invalidated diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c index 2168163aad2d..252a876b0725 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c @@ -209,6 +209,7 @@ void amdgpu_bo_list_get_list(struct amdgpu_bo_list *list, list_add_tail(&e->tv.head, &bucket[priority]); e->user_pages = NULL; + e->range = NULL; } /* Connect the sorted buckets in the output list. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h index 9caea1688fc3..e4d78491bcc7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h @@ -26,6 +26,8 @@ #include <drm/ttm/ttm_execbuf_util.h> #include <drm/amdgpu_drm.h> +struct hmm_range; + struct amdgpu_device; struct amdgpu_bo; struct amdgpu_bo_va; @@ -36,6 +38,7 @@ struct amdgpu_bo_list_entry { struct amdgpu_bo_va *bo_va; uint32_t priority; struct page **user_pages; + struct hmm_range *range; bool user_invalidated; }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 255d545e5524..275da612cd87 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -912,7 +912,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, goto out_free_user_pages; } - r = amdgpu_ttm_tt_get_user_pages(bo, e->user_pages); + r = amdgpu_ttm_tt_get_user_pages(bo, e->user_pages, &e->range); if (r) { kvfree(e->user_pages); e->user_pages = NULL; @@ -990,9 +990,10 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, if (!e->user_pages) continue; - amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, e->range); kvfree(e->user_pages); e->user_pages = NULL; + e->range = NULL; } mutex_unlock(&p->bo_list->bo_list_mutex); return r; @@ -1267,7 +1268,8 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, amdgpu_bo_list_for_each_userptr_entry(e, p->bo_list) { struct amdgpu_bo *bo = ttm_to_amdgpu_bo(e->tv.bo); - r |= !amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + r |= !amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, e->range); + e->range = NULL; } if (r) { r = -EAGAIN; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 111484ceb47d..91571b1324f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -378,6 +378,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, struct amdgpu_device *adev = drm_to_adev(dev); struct drm_amdgpu_gem_userptr *args = data; struct drm_gem_object *gobj; + struct hmm_range *range; struct amdgpu_bo *bo; uint32_t handle; int r; @@ -418,7 +419,8 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, goto release_object; if (args->flags & AMDGPU_GEM_USERPTR_VALIDATE) { - r = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages); + r = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages, + &range); if (r) goto release_object; @@ -441,7 +443,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, user_pages_done: if (args->flags & AMDGPU_GEM_USERPTR_VALIDATE) - amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, range); release_object: drm_gem_object_put(gobj); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 02a8f25ae134..965808f4f682 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -643,9 +643,6 @@ struct amdgpu_ttm_tt { struct task_struct *usertask; uint32_t userflags; bool bound; -#if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR) - struct hmm_range *range; -#endif }; #define ttm_to_amdgpu_ttm_tt(ptr) container_of(ptr, struct amdgpu_ttm_tt, ttm) @@ -658,7 +655,8 @@ struct amdgpu_ttm_tt { * Calling function must call amdgpu_ttm_tt_userptr_range_done() once and only * once afterwards to stop HMM tracking */ -int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) +int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages, + struct hmm_range **range) { struct ttm_tt *ttm = bo->tbo.ttm; struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); @@ -668,16 +666,15 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) bool readonly; int r = 0; + /* Make sure get_user_pages_done() can cleanup gracefully */ + *range = NULL; + mm = bo->notifier.mm; if (unlikely(!mm)) { DRM_DEBUG_DRIVER("BO is not registered?\n"); return -EFAULT; } - /* Another get_user_pages is running at the same time?? */ - if (WARN_ON(gtt->range)) - return -EFAULT; - if (!mmget_not_zero(mm)) /* Happens during process shutdown */ return -ESRCH; @@ -695,7 +692,7 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) readonly = amdgpu_ttm_tt_is_readonly(ttm); r = amdgpu_hmm_range_get_pages(&bo->notifier, mm, pages, start, - ttm->num_pages, &gtt->range, readonly, + ttm->num_pages, range, readonly, true, NULL); out_unlock: mmap_read_unlock(mm); @@ -713,30 +710,24 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) * * Returns: true if pages are still valid */ -bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm) +bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm, + struct hmm_range *range) { struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); - bool r = false; - if (!gtt || !gtt->userptr) + if (!gtt || !gtt->userptr || !range) return false; DRM_DEBUG_DRIVER("user_pages_done 0x%llx pages 0x%x\n", gtt->userptr, ttm->num_pages); - WARN_ONCE(!gtt->range || !gtt->range->hmm_pfns, - "No user pages to check\n"); + WARN_ONCE(!range->hmm_pfns, "No user pages to check\n"); - if (gtt->range) { - /* - * FIXME: Must always hold notifier_lock for this, and must - * not ignore the return code. - */ - r = amdgpu_hmm_range_get_pages_done(gtt->range); - gtt->range = NULL; - } - - return !r; + /* + * FIXME: Must always hold notifier_lock for this, and must + * not ignore the return code. + */ + return !amdgpu_hmm_range_get_pages_done(range); } #endif @@ -813,20 +804,6 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_device *bdev, /* unmap the pages mapped to the device */ dma_unmap_sgtable(adev->dev, ttm->sg, direction, 0); sg_free_table(ttm->sg); - -#if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR) - if (gtt->range) { - unsigned long i; - - for (i = 0; i < ttm->num_pages; i++) { - if (ttm->pages[i] != - hmm_pfn_to_page(gtt->range->hmm_pfns[i])) - break; - } - - WARN((i == ttm->num_pages), "Missing get_user_page_done\n"); - } -#endif } static void amdgpu_ttm_gart_bind(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index 7c38843f411e..b391c8d076ff 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -39,6 +39,8 @@ #define AMDGPU_POISON 0xd0bed0be +struct hmm_range; + struct amdgpu_gtt_mgr { struct ttm_resource_manager manager; struct drm_mm mm; @@ -154,15 +156,19 @@ void amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo); uint64_t amdgpu_ttm_domain_start(struct amdgpu_device *adev, uint32_t type); #if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR) -int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages); -bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm); +int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages, + struct hmm_range **range); +bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm, + struct hmm_range *range); #else static inline int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, - struct page **pages) + struct page **pages, + struct hmm_range **range) { return -EPERM; } -static inline bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm) +static inline bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm, + struct hmm_range *range) { return false; }

2 years, 1 month

1
0
0 0

FAILED: patch "[PATCH] drm/amdgpu: fix userptr HMM range handling v2" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

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 <stable(a)vger.kernel.org>. Possible dependencies: fec8fdb54e8f ("drm/amdgpu: fix userptr HMM range handling v2") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From fec8fdb54e8f74d88951c9f998f47bf4f2031fe0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig(a)amd.com> Date: Thu, 10 Nov 2022 12:31:41 +0100 Subject: [PATCH] drm/amdgpu: fix userptr HMM range handling v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The basic problem here is that it's not allowed to page fault while holding the reservation lock. So it can happen that multiple processes try to validate an userptr at the same time. Work around that by putting the HMM range object into the mutex protected bo list for now. v2: make sure range is set to NULL in case of an error Signed-off-by: Christian König <christian.koenig(a)amd.com> Reviewed-by: Alex Deucher <alexander.deucher(a)amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling(a)amd.com> CC: stable(a)vger.kernel.org Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 68741b157153..e44d740022bc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -938,6 +938,7 @@ static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr, struct amdkfd_process_info *process_info = mem->process_info; struct amdgpu_bo *bo = mem->bo; struct ttm_operation_ctx ctx = { true, false }; + struct hmm_range *range; int ret = 0; mutex_lock(&process_info->lock); @@ -967,7 +968,7 @@ static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr, return 0; } - ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages); + ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages, &range); if (ret) { pr_err("%s: Failed to get user pages: %d\n", __func__, ret); goto unregister_out; @@ -985,7 +986,7 @@ static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr, amdgpu_bo_unreserve(bo); release_out: - amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, range); unregister_out: if (ret) amdgpu_mn_unregister(bo); @@ -2317,6 +2318,8 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info, /* Go through userptr_inval_list and update any invalid user_pages */ list_for_each_entry(mem, &process_info->userptr_inval_list, validate_list.head) { + struct hmm_range *range; + invalid = atomic_read(&mem->invalid); if (!invalid) /* BO hasn't been invalidated since the last @@ -2327,7 +2330,8 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info, bo = mem->bo; /* Get updated user pages */ - ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages); + ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages, + &range); if (ret) { pr_debug("Failed %d to get user pages\n", ret); @@ -2346,7 +2350,7 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info, * FIXME: Cannot ignore the return code, must hold * notifier_lock */ - amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, range); } /* Mark the BO as valid unless it was invalidated diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c index 2168163aad2d..252a876b0725 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c @@ -209,6 +209,7 @@ void amdgpu_bo_list_get_list(struct amdgpu_bo_list *list, list_add_tail(&e->tv.head, &bucket[priority]); e->user_pages = NULL; + e->range = NULL; } /* Connect the sorted buckets in the output list. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h index 9caea1688fc3..e4d78491bcc7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h @@ -26,6 +26,8 @@ #include <drm/ttm/ttm_execbuf_util.h> #include <drm/amdgpu_drm.h> +struct hmm_range; + struct amdgpu_device; struct amdgpu_bo; struct amdgpu_bo_va; @@ -36,6 +38,7 @@ struct amdgpu_bo_list_entry { struct amdgpu_bo_va *bo_va; uint32_t priority; struct page **user_pages; + struct hmm_range *range; bool user_invalidated; }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 255d545e5524..275da612cd87 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -912,7 +912,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, goto out_free_user_pages; } - r = amdgpu_ttm_tt_get_user_pages(bo, e->user_pages); + r = amdgpu_ttm_tt_get_user_pages(bo, e->user_pages, &e->range); if (r) { kvfree(e->user_pages); e->user_pages = NULL; @@ -990,9 +990,10 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, if (!e->user_pages) continue; - amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, e->range); kvfree(e->user_pages); e->user_pages = NULL; + e->range = NULL; } mutex_unlock(&p->bo_list->bo_list_mutex); return r; @@ -1267,7 +1268,8 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, amdgpu_bo_list_for_each_userptr_entry(e, p->bo_list) { struct amdgpu_bo *bo = ttm_to_amdgpu_bo(e->tv.bo); - r |= !amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + r |= !amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, e->range); + e->range = NULL; } if (r) { r = -EAGAIN; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 111484ceb47d..91571b1324f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -378,6 +378,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, struct amdgpu_device *adev = drm_to_adev(dev); struct drm_amdgpu_gem_userptr *args = data; struct drm_gem_object *gobj; + struct hmm_range *range; struct amdgpu_bo *bo; uint32_t handle; int r; @@ -418,7 +419,8 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, goto release_object; if (args->flags & AMDGPU_GEM_USERPTR_VALIDATE) { - r = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages); + r = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages, + &range); if (r) goto release_object; @@ -441,7 +443,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, user_pages_done: if (args->flags & AMDGPU_GEM_USERPTR_VALIDATE) - amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, range); release_object: drm_gem_object_put(gobj); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 02a8f25ae134..965808f4f682 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -643,9 +643,6 @@ struct amdgpu_ttm_tt { struct task_struct *usertask; uint32_t userflags; bool bound; -#if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR) - struct hmm_range *range; -#endif }; #define ttm_to_amdgpu_ttm_tt(ptr) container_of(ptr, struct amdgpu_ttm_tt, ttm) @@ -658,7 +655,8 @@ struct amdgpu_ttm_tt { * Calling function must call amdgpu_ttm_tt_userptr_range_done() once and only * once afterwards to stop HMM tracking */ -int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) +int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages, + struct hmm_range **range) { struct ttm_tt *ttm = bo->tbo.ttm; struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); @@ -668,16 +666,15 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) bool readonly; int r = 0; + /* Make sure get_user_pages_done() can cleanup gracefully */ + *range = NULL; + mm = bo->notifier.mm; if (unlikely(!mm)) { DRM_DEBUG_DRIVER("BO is not registered?\n"); return -EFAULT; } - /* Another get_user_pages is running at the same time?? */ - if (WARN_ON(gtt->range)) - return -EFAULT; - if (!mmget_not_zero(mm)) /* Happens during process shutdown */ return -ESRCH; @@ -695,7 +692,7 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) readonly = amdgpu_ttm_tt_is_readonly(ttm); r = amdgpu_hmm_range_get_pages(&bo->notifier, mm, pages, start, - ttm->num_pages, &gtt->range, readonly, + ttm->num_pages, range, readonly, true, NULL); out_unlock: mmap_read_unlock(mm); @@ -713,30 +710,24 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) * * Returns: true if pages are still valid */ -bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm) +bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm, + struct hmm_range *range) { struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); - bool r = false; - if (!gtt || !gtt->userptr) + if (!gtt || !gtt->userptr || !range) return false; DRM_DEBUG_DRIVER("user_pages_done 0x%llx pages 0x%x\n", gtt->userptr, ttm->num_pages); - WARN_ONCE(!gtt->range || !gtt->range->hmm_pfns, - "No user pages to check\n"); + WARN_ONCE(!range->hmm_pfns, "No user pages to check\n"); - if (gtt->range) { - /* - * FIXME: Must always hold notifier_lock for this, and must - * not ignore the return code. - */ - r = amdgpu_hmm_range_get_pages_done(gtt->range); - gtt->range = NULL; - } - - return !r; + /* + * FIXME: Must always hold notifier_lock for this, and must + * not ignore the return code. + */ + return !amdgpu_hmm_range_get_pages_done(range); } #endif @@ -813,20 +804,6 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_device *bdev, /* unmap the pages mapped to the device */ dma_unmap_sgtable(adev->dev, ttm->sg, direction, 0); sg_free_table(ttm->sg); - -#if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR) - if (gtt->range) { - unsigned long i; - - for (i = 0; i < ttm->num_pages; i++) { - if (ttm->pages[i] != - hmm_pfn_to_page(gtt->range->hmm_pfns[i])) - break; - } - - WARN((i == ttm->num_pages), "Missing get_user_page_done\n"); - } -#endif } static void amdgpu_ttm_gart_bind(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index 7c38843f411e..b391c8d076ff 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -39,6 +39,8 @@ #define AMDGPU_POISON 0xd0bed0be +struct hmm_range; + struct amdgpu_gtt_mgr { struct ttm_resource_manager manager; struct drm_mm mm; @@ -154,15 +156,19 @@ void amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo); uint64_t amdgpu_ttm_domain_start(struct amdgpu_device *adev, uint32_t type); #if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR) -int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages); -bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm); +int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages, + struct hmm_range **range); +bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm, + struct hmm_range *range); #else static inline int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, - struct page **pages) + struct page **pages, + struct hmm_range **range) { return -EPERM; } -static inline bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm) +static inline bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm, + struct hmm_range *range) { return false; }

2 years, 1 month

1
0
0 0

FAILED: patch "[PATCH] drm/amdgpu: always register an MMU notifier for userptr" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

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 <stable(a)vger.kernel.org>. Possible dependencies: 631945e04e1e ("drm/amdgpu: always register an MMU notifier for userptr") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 631945e04e1e243a503089f4487cad797476e8ca Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig(a)amd.com> Date: Wed, 9 Nov 2022 12:14:44 +0100 Subject: [PATCH] drm/amdgpu: always register an MMU notifier for userptr MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since switching to HMM we always need that because we no longer grab references to the pages. Signed-off-by: Christian König <christian.koenig(a)amd.com> Reviewed-by: Alex Deucher <alexander.deucher(a)amd.com> Acked-by: Felix Kuehling <Felix.Kuehling(a)amd.com> CC: stable(a)vger.kernel.org Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 8ef31d687ef3..111484ceb47d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -413,11 +413,9 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, if (r) goto release_object; - if (args->flags & AMDGPU_GEM_USERPTR_REGISTER) { - r = amdgpu_mn_register(bo, args->addr); - if (r) - goto release_object; - } + r = amdgpu_mn_register(bo, args->addr); + if (r) + goto release_object; if (args->flags & AMDGPU_GEM_USERPTR_VALIDATE) { r = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages);

2 years, 1 month

1
0
0 0

FAILED: patch "[PATCH] drm/amdgpu: always register an MMU notifier for userptr" failed to apply to 6.0-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.0-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 <stable(a)vger.kernel.org>. Possible dependencies: 631945e04e1e ("drm/amdgpu: always register an MMU notifier for userptr") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 631945e04e1e243a503089f4487cad797476e8ca Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig(a)amd.com> Date: Wed, 9 Nov 2022 12:14:44 +0100 Subject: [PATCH] drm/amdgpu: always register an MMU notifier for userptr MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since switching to HMM we always need that because we no longer grab references to the pages. Signed-off-by: Christian König <christian.koenig(a)amd.com> Reviewed-by: Alex Deucher <alexander.deucher(a)amd.com> Acked-by: Felix Kuehling <Felix.Kuehling(a)amd.com> CC: stable(a)vger.kernel.org Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 8ef31d687ef3..111484ceb47d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -413,11 +413,9 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, if (r) goto release_object; - if (args->flags & AMDGPU_GEM_USERPTR_REGISTER) { - r = amdgpu_mn_register(bo, args->addr); - if (r) - goto release_object; - } + r = amdgpu_mn_register(bo, args->addr); + if (r) + goto release_object; if (args->flags & AMDGPU_GEM_USERPTR_VALIDATE) { r = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages);

2 years, 1 month

1
0
0 0

FAILED: patch "[PATCH] drm/amd/dc/dce120: Fix audio register mapping, stop" failed to apply to 6.0-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.0-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 <stable(a)vger.kernel.org>. Possible dependencies: 6110a80f31b7 ("drm/amd/dc/dce120: Fix audio register mapping, stop triggering KASAN") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 6110a80f31b74773a28394259695513e0ac38406 Mon Sep 17 00:00:00 2001 From: Lyude Paul <lyude(a)redhat.com> Date: Mon, 14 Nov 2022 17:20:45 -0500 Subject: [PATCH] drm/amd/dc/dce120: Fix audio register mapping, stop triggering KASAN There's been a very long running bug that seems to have been neglected for a while, where amdgpu consistently triggers a KASAN error at start: BUG: KASAN: global-out-of-bounds in read_indirect_azalia_reg+0x1d4/0x2a0 [amdgpu] Read of size 4 at addr ffffffffc2274b28 by task modprobe/1889 After digging through amd's rather creative method for accessing registers, I eventually discovered the problem likely has to do with the fact that on my dce120 GPU there are supposedly 7 sets of audio registers. But we only define a register mapping for 6 sets. So, fix this and fix the KASAN warning finally. Signed-off-by: Lyude Paul <lyude(a)redhat.com> Cc: stable(a)vger.kernel.org Reviewed-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c index 1b70b78e2fa1..af631085e88c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c @@ -359,7 +359,8 @@ static const struct dce_audio_registers audio_regs[] = { audio_regs(2), audio_regs(3), audio_regs(4), - audio_regs(5) + audio_regs(5), + audio_regs(6), }; #define DCE120_AUD_COMMON_MASK_SH_LIST(mask_sh)\

2 years, 1 month

1
0
0 0

FAILED: patch "[PATCH] drm/amd/dc/dce120: Fix audio register mapping, stop" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

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 <stable(a)vger.kernel.org>. Possible dependencies: 6110a80f31b7 ("drm/amd/dc/dce120: Fix audio register mapping, stop triggering KASAN") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 6110a80f31b74773a28394259695513e0ac38406 Mon Sep 17 00:00:00 2001 From: Lyude Paul <lyude(a)redhat.com> Date: Mon, 14 Nov 2022 17:20:45 -0500 Subject: [PATCH] drm/amd/dc/dce120: Fix audio register mapping, stop triggering KASAN There's been a very long running bug that seems to have been neglected for a while, where amdgpu consistently triggers a KASAN error at start: BUG: KASAN: global-out-of-bounds in read_indirect_azalia_reg+0x1d4/0x2a0 [amdgpu] Read of size 4 at addr ffffffffc2274b28 by task modprobe/1889 After digging through amd's rather creative method for accessing registers, I eventually discovered the problem likely has to do with the fact that on my dce120 GPU there are supposedly 7 sets of audio registers. But we only define a register mapping for 6 sets. So, fix this and fix the KASAN warning finally. Signed-off-by: Lyude Paul <lyude(a)redhat.com> Cc: stable(a)vger.kernel.org Reviewed-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c index 1b70b78e2fa1..af631085e88c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c @@ -359,7 +359,8 @@ static const struct dce_audio_registers audio_regs[] = { audio_regs(2), audio_regs(3), audio_regs(4), - audio_regs(5) + audio_regs(5), + audio_regs(6), }; #define DCE120_AUD_COMMON_MASK_SH_LIST(mask_sh)\

2 years, 1 month

1
0
0 0

FAILED: patch "[PATCH] drm/amd/pm: fulfill SMU13.0.0 cstate control interface" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

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 <stable(a)vger.kernel.org>. Possible dependencies: c6863be23179 ("drm/amd/pm: fulfill SMU13.0.0 cstate control interface") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From c6863be231791fa80dca203b9295780488671195 Mon Sep 17 00:00:00 2001 From: Evan Quan <evan.quan(a)amd.com> Date: Thu, 29 Sep 2022 10:24:51 +0800 Subject: [PATCH] drm/amd/pm: fulfill SMU13.0.0 cstate control interface Fulfill the functionality for cstate control. Signed-off-by: Evan Quan <evan.quan(a)amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang(a)amd.com> Reviewed-by: Lijo Lazar <lijo.lazar(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org # 6.0.x diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 1d454485e0d9..29529328152d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -119,6 +119,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_0_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(NotifyPowerSource, PPSMC_MSG_NotifyPowerSource, 0), MSG_MAP(Mode1Reset, PPSMC_MSG_Mode1Reset, 0), MSG_MAP(PrepareMp1ForUnload, PPSMC_MSG_PrepareMp1ForUnload, 0), + MSG_MAP(DFCstateControl, PPSMC_MSG_SetExternalClientDfCstateAllow, 0), }; static struct cmn2asic_mapping smu_v13_0_0_clk_map[SMU_CLK_COUNT] = { @@ -1753,6 +1754,15 @@ static int smu_v13_0_0_set_mp1_state(struct smu_context *smu, return ret; } +static int smu_v13_0_0_set_df_cstate(struct smu_context *smu, + enum pp_df_cstate state) +{ + return smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_DFCstateControl, + state, + NULL); +} + static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_0_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_0_set_default_dpm_table, @@ -1822,6 +1832,7 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .mode1_reset_is_support = smu_v13_0_0_is_mode1_reset_supported, .mode1_reset = smu_v13_0_mode1_reset, .set_mp1_state = smu_v13_0_0_set_mp1_state, + .set_df_cstate = smu_v13_0_0_set_df_cstate, }; void smu_v13_0_0_set_ppt_funcs(struct smu_context *smu)

2 years, 1 month

1
0
0 0

FAILED: patch "[PATCH] drm/amd/pm: fulfill SMU13.0.0 cstate control interface" failed to apply to 6.0-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.0-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 <stable(a)vger.kernel.org>. Possible dependencies: c6863be23179 ("drm/amd/pm: fulfill SMU13.0.0 cstate control interface") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From c6863be231791fa80dca203b9295780488671195 Mon Sep 17 00:00:00 2001 From: Evan Quan <evan.quan(a)amd.com> Date: Thu, 29 Sep 2022 10:24:51 +0800 Subject: [PATCH] drm/amd/pm: fulfill SMU13.0.0 cstate control interface Fulfill the functionality for cstate control. Signed-off-by: Evan Quan <evan.quan(a)amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang(a)amd.com> Reviewed-by: Lijo Lazar <lijo.lazar(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org # 6.0.x diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 1d454485e0d9..29529328152d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -119,6 +119,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_0_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(NotifyPowerSource, PPSMC_MSG_NotifyPowerSource, 0), MSG_MAP(Mode1Reset, PPSMC_MSG_Mode1Reset, 0), MSG_MAP(PrepareMp1ForUnload, PPSMC_MSG_PrepareMp1ForUnload, 0), + MSG_MAP(DFCstateControl, PPSMC_MSG_SetExternalClientDfCstateAllow, 0), }; static struct cmn2asic_mapping smu_v13_0_0_clk_map[SMU_CLK_COUNT] = { @@ -1753,6 +1754,15 @@ static int smu_v13_0_0_set_mp1_state(struct smu_context *smu, return ret; } +static int smu_v13_0_0_set_df_cstate(struct smu_context *smu, + enum pp_df_cstate state) +{ + return smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_DFCstateControl, + state, + NULL); +} + static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_0_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_0_set_default_dpm_table, @@ -1822,6 +1832,7 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .mode1_reset_is_support = smu_v13_0_0_is_mode1_reset_supported, .mode1_reset = smu_v13_0_mode1_reset, .set_mp1_state = smu_v13_0_0_set_mp1_state, + .set_df_cstate = smu_v13_0_0_set_df_cstate, }; void smu_v13_0_0_set_ppt_funcs(struct smu_context *smu)

2 years, 1 month

1
0
0 0

FAILED: patch "[PATCH] drm/amd/pm: fulfill SMU13.0.7 cstate control interface" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

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 <stable(a)vger.kernel.org>. Possible dependencies: 3cf377ee8df7 ("drm/amd/pm: fulfill SMU13.0.7 cstate control interface") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 3cf377ee8df7dc4ae5e543c37833ae5a5b2a78d3 Mon Sep 17 00:00:00 2001 From: Evan Quan <evan.quan(a)amd.com> Date: Thu, 29 Sep 2022 10:30:01 +0800 Subject: [PATCH] drm/amd/pm: fulfill SMU13.0.7 cstate control interface Fulfill the functionality for cstate control. Signed-off-by: Evan Quan <evan.quan(a)amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang(a)amd.com> Reviewed-by: Lijo Lazar <lijo.lazar(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org # 6.0.x diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index c422bf8a09b1..c4102cfb734c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -121,6 +121,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_7_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(Mode1Reset, PPSMC_MSG_Mode1Reset, 0), MSG_MAP(PrepareMp1ForUnload, PPSMC_MSG_PrepareMp1ForUnload, 0), MSG_MAP(SetMGpuFanBoostLimitRpm, PPSMC_MSG_SetMGpuFanBoostLimitRpm, 0), + MSG_MAP(DFCstateControl, PPSMC_MSG_SetExternalClientDfCstateAllow, 0), }; static struct cmn2asic_mapping smu_v13_0_7_clk_map[SMU_CLK_COUNT] = { @@ -1587,6 +1588,16 @@ static bool smu_v13_0_7_is_mode1_reset_supported(struct smu_context *smu) return true; } + +static int smu_v13_0_7_set_df_cstate(struct smu_context *smu, + enum pp_df_cstate state) +{ + return smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_DFCstateControl, + state, + NULL); +} + static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_7_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_7_set_default_dpm_table, @@ -1649,6 +1660,7 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .mode1_reset_is_support = smu_v13_0_7_is_mode1_reset_supported, .mode1_reset = smu_v13_0_mode1_reset, .set_mp1_state = smu_v13_0_7_set_mp1_state, + .set_df_cstate = smu_v13_0_7_set_df_cstate, }; void smu_v13_0_7_set_ppt_funcs(struct smu_context *smu)

2 years, 1 month

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror January 2023