From 68132cc6d1bcbc78ade524c6c6c226de42139f0e Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Fri, 11 Aug 2023 04:50:20 +1000 Subject: nouveau/u_memcpya: use vmemdup_user I think there are limit checks in place for most things but the new uAPI wants to not have them. Add a limit check and use the vmemdup_user helper instead. Signed-off-by: Dave Airlie Reviewed-by: Danilo Krummrich Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20230810185020.231135-1-airlied@gmail.com --- drivers/gpu/drm/nouveau/nouveau_drv.h | 19 +++++-------------- 1 file changed, 5 insertions(+), 14 deletions(-) (limited to 'drivers/gpu/drm/nouveau') diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h index 1fe17ff95f5e..3666a7403e47 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drv.h +++ b/drivers/gpu/drm/nouveau/nouveau_drv.h @@ -189,21 +189,12 @@ u_free(void *addr) static inline void * u_memcpya(uint64_t user, unsigned int nmemb, unsigned int size) { - void *mem; - void __user *userptr = (void __force __user *)(uintptr_t)user; + void __user *userptr = u64_to_user_ptr(user); + size_t bytes; - size *= nmemb; - - mem = kvmalloc(size, GFP_KERNEL); - if (!mem) - return ERR_PTR(-ENOMEM); - - if (copy_from_user(mem, userptr, size)) { - u_free(mem); - return ERR_PTR(-EFAULT); - } - - return mem; + if (unlikely(check_mul_overflow(nmemb, size, &bytes))) + return NULL; + return vmemdup_user(userptr, bytes); } #include -- cgit v1.2.3 From 7baf605564133405443556b415692d3c7aa54351 Mon Sep 17 00:00:00 2001 From: Danilo Krummrich Date: Fri, 11 Aug 2023 03:06:25 +0200 Subject: drm/nouveau: sched: avoid job races between entities If a sched job depends on a dma-fence from a job from the same GPU scheduler instance, but a different scheduler entity, the GPU scheduler does only wait for the particular job to be scheduled, rather than for the job to fully complete. This is due to the GPU scheduler assuming that there is a scheduler instance per ring. However, the current implementation, in order to avoid arbitrary amounts of kthreads, has a single scheduler instance while scheduler entities represent rings. As a workaround, set the DRM_SCHED_FENCE_DONT_PIPELINE for all out-fences in order to force the scheduler to wait for full job completion for dependent jobs from different entities and same scheduler instance. There is some work in progress [1] to address the issues of firmware schedulers; once it is in-tree the scheduler topology in Nouveau should be re-worked accordingly. [1] https://lore.kernel.org/dri-devel/20230801205103.627779-1-matthew.brost@intel.com/ Signed-off-by: Danilo Krummrich Reviewed-by: Faith Ekstrand Link: https://patchwork.freedesktop.org/patch/msgid/20230811010632.2473-1-dakr@redhat.com --- drivers/gpu/drm/nouveau/nouveau_sched.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'drivers/gpu/drm/nouveau') diff --git a/drivers/gpu/drm/nouveau/nouveau_sched.c b/drivers/gpu/drm/nouveau/nouveau_sched.c index 3424a1bf6af3..88217185e0f3 100644 --- a/drivers/gpu/drm/nouveau/nouveau_sched.c +++ b/drivers/gpu/drm/nouveau/nouveau_sched.c @@ -292,6 +292,28 @@ nouveau_job_submit(struct nouveau_job *job) if (job->sync) done_fence = dma_fence_get(job->done_fence); + /* If a sched job depends on a dma-fence from a job from the same GPU + * scheduler instance, but a different scheduler entity, the GPU + * scheduler does only wait for the particular job to be scheduled, + * rather than for the job to fully complete. This is due to the GPU + * scheduler assuming that there is a scheduler instance per ring. + * However, the current implementation, in order to avoid arbitrary + * amounts of kthreads, has a single scheduler instance while scheduler + * entities represent rings. + * + * As a workaround, set the DRM_SCHED_FENCE_DONT_PIPELINE for all + * out-fences in order to force the scheduler to wait for full job + * completion for dependent jobs from different entities and same + * scheduler instance. + * + * There is some work in progress [1] to address the issues of firmware + * schedulers; once it is in-tree the scheduler topology in Nouveau + * should be re-worked accordingly. + * + * [1] https://lore.kernel.org/dri-devel/20230801205103.627779-1-matthew.brost@intel.com/ + */ + set_bit(DRM_SCHED_FENCE_DONT_PIPELINE, &job->done_fence->flags); + if (job->ops->armed_submit) job->ops->armed_submit(job); -- cgit v1.2.3 From b4e9fa933551e51459c634dc4396171dc65284a6 Mon Sep 17 00:00:00 2001 From: Danilo Krummrich Date: Mon, 21 Aug 2023 00:29:16 +0200 Subject: drm/nouveau: uvmm: fix unset region pointer on remap Transfer the region pointer of a uvma to the new uvma(s) on re-map to prevent potential shader faults when the re-mapped uvma(s) are unmapped. Signed-off-by: Danilo Krummrich Reviewed-by: Dave Airlie Link: https://patchwork.freedesktop.org/patch/msgid/20230820222920.2344-1-dakr@redhat.com --- drivers/gpu/drm/nouveau/nouveau_uvmm.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers/gpu/drm/nouveau') diff --git a/drivers/gpu/drm/nouveau/nouveau_uvmm.c b/drivers/gpu/drm/nouveau/nouveau_uvmm.c index 3a1e8538f205..aae780e4a4aa 100644 --- a/drivers/gpu/drm/nouveau/nouveau_uvmm.c +++ b/drivers/gpu/drm/nouveau/nouveau_uvmm.c @@ -639,6 +639,7 @@ nouveau_uvmm_sm_prepare(struct nouveau_uvmm *uvmm, struct drm_gpuva *va = r->unmap->va; struct uvmm_map_args remap_args = { .kind = uvma_from_va(va)->kind, + .region = uvma_from_va(va)->region, }; u64 ustart = va->va.addr; u64 urange = va->va.range; -- cgit v1.2.3 From b715dcd3db4a9a57b3fbe7820db37cae930f0867 Mon Sep 17 00:00:00 2001 From: Danilo Krummrich Date: Wed, 23 Aug 2023 20:15:34 +0200 Subject: drm/nouveau: uapi: don't pass NO_PREFETCH flag implicitly Currently, NO_PREFETCH is passed implicitly through drm_nouveau_gem_pushbuf_push::length and drm_nouveau_exec_push::va_len. Since this is a direct representation of how the HW is programmed it isn't really future proof for a uAPI. Hence, fix this up for the new uAPI and split up the va_len field of struct drm_nouveau_exec_push, such that we keep 32bit for va_len and 32bit for flags. For drm_nouveau_gem_pushbuf_push::length at least provide NOUVEAU_GEM_PUSHBUF_NO_PREFETCH to indicate the bit shift. While at it, fix up nv50_dma_push() as well, such that the caller doesn't need to encode the NO_PREFETCH flag into the length parameter. Signed-off-by: Danilo Krummrich Reviewed-by: Faith Ekstrand Link: https://patchwork.freedesktop.org/patch/msgid/20230823181746.3446-1-dakr@redhat.com --- drivers/gpu/drm/nouveau/nouveau_dma.c | 7 +++++-- drivers/gpu/drm/nouveau/nouveau_dma.h | 8 ++++++-- drivers/gpu/drm/nouveau/nouveau_exec.c | 19 ++++++++++++++++--- drivers/gpu/drm/nouveau/nouveau_gem.c | 6 ++++-- include/uapi/drm/nouveau_drm.h | 8 +++++++- 5 files changed, 38 insertions(+), 10 deletions(-) (limited to 'drivers/gpu/drm/nouveau') diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.c b/drivers/gpu/drm/nouveau/nouveau_dma.c index b90cac6d5772..b01c029f3a90 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dma.c +++ b/drivers/gpu/drm/nouveau/nouveau_dma.c @@ -69,16 +69,19 @@ READ_GET(struct nouveau_channel *chan, uint64_t *prev_get, int *timeout) } void -nv50_dma_push(struct nouveau_channel *chan, u64 offset, int length) +nv50_dma_push(struct nouveau_channel *chan, u64 offset, u32 length, + bool no_prefetch) { struct nvif_user *user = &chan->drm->client.device.user; struct nouveau_bo *pb = chan->push.buffer; int ip = (chan->dma.ib_put * 2) + chan->dma.ib_base; BUG_ON(chan->dma.ib_free < 1); + WARN_ON(length > NV50_DMA_PUSH_MAX_LENGTH); nouveau_bo_wr32(pb, ip++, lower_32_bits(offset)); - nouveau_bo_wr32(pb, ip++, upper_32_bits(offset) | length << 8); + nouveau_bo_wr32(pb, ip++, upper_32_bits(offset) | length << 8 | + (no_prefetch ? (1 << 31) : 0)); chan->dma.ib_put = (chan->dma.ib_put + 1) & chan->dma.ib_max; diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.h b/drivers/gpu/drm/nouveau/nouveau_dma.h index 035a709c7be1..1744d95b233e 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dma.h +++ b/drivers/gpu/drm/nouveau/nouveau_dma.h @@ -31,7 +31,8 @@ #include "nouveau_chan.h" int nouveau_dma_wait(struct nouveau_channel *, int slots, int size); -void nv50_dma_push(struct nouveau_channel *, u64 addr, int length); +void nv50_dma_push(struct nouveau_channel *, u64 addr, u32 length, + bool no_prefetch); /* * There's a hw race condition where you can't jump to your PUT offset, @@ -45,6 +46,9 @@ void nv50_dma_push(struct nouveau_channel *, u64 addr, int length); */ #define NOUVEAU_DMA_SKIPS (128 / 4) +/* Maximum push buffer size. */ +#define NV50_DMA_PUSH_MAX_LENGTH 0x7fffff + /* Object handles - for stuff that's doesn't use handle == oclass. */ enum { NvDmaFB = 0x80000002, @@ -89,7 +93,7 @@ FIRE_RING(struct nouveau_channel *chan) if (chan->dma.ib_max) { nv50_dma_push(chan, chan->push.addr + (chan->dma.put << 2), - (chan->dma.cur - chan->dma.put) << 2); + (chan->dma.cur - chan->dma.put) << 2, false); } else { WRITE_PUT(chan->dma.cur); } diff --git a/drivers/gpu/drm/nouveau/nouveau_exec.c b/drivers/gpu/drm/nouveau/nouveau_exec.c index 0f927adda4ed..a90c4cd8cbb2 100644 --- a/drivers/gpu/drm/nouveau/nouveau_exec.c +++ b/drivers/gpu/drm/nouveau/nouveau_exec.c @@ -164,8 +164,10 @@ nouveau_exec_job_run(struct nouveau_job *job) } for (i = 0; i < exec_job->push.count; i++) { - nv50_dma_push(chan, exec_job->push.s[i].va, - exec_job->push.s[i].va_len); + struct drm_nouveau_exec_push *p = &exec_job->push.s[i]; + bool no_prefetch = p->flags & DRM_NOUVEAU_EXEC_PUSH_NO_PREFETCH; + + nv50_dma_push(chan, p->va, p->va_len, no_prefetch); } ret = nouveau_fence_emit(fence, chan); @@ -223,7 +225,18 @@ nouveau_exec_job_init(struct nouveau_exec_job **pjob, { struct nouveau_exec_job *job; struct nouveau_job_args args = {}; - int ret; + int i, ret; + + for (i = 0; i < __args->push.count; i++) { + struct drm_nouveau_exec_push *p = &__args->push.s[i]; + + if (unlikely(p->va_len > NV50_DMA_PUSH_MAX_LENGTH)) { + NV_PRINTK(err, nouveau_cli(__args->file_priv), + "pushbuf size exceeds limit: 0x%x max 0x%x\n", + p->va_len, NV50_DMA_PUSH_MAX_LENGTH); + return -EINVAL; + } + } job = *pjob = kzalloc(sizeof(*job), GFP_KERNEL); if (!job) diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c index f39360870c70..c0b10d8d3d03 100644 --- a/drivers/gpu/drm/nouveau/nouveau_gem.c +++ b/drivers/gpu/drm/nouveau/nouveau_gem.c @@ -856,9 +856,11 @@ revalidate: for (i = 0; i < req->nr_push; i++) { struct nouveau_vma *vma = (void *)(unsigned long) bo[push[i].bo_index].user_priv; + u64 addr = vma->addr + push[i].offset; + u32 length = push[i].length & ~NOUVEAU_GEM_PUSHBUF_NO_PREFETCH; + bool no_prefetch = push[i].length & NOUVEAU_GEM_PUSHBUF_NO_PREFETCH; - nv50_dma_push(chan, vma->addr + push[i].offset, - push[i].length); + nv50_dma_push(chan, addr, length, no_prefetch); } } else if (drm->client.device.info.chipset >= 0x25) { diff --git a/include/uapi/drm/nouveau_drm.h b/include/uapi/drm/nouveau_drm.h index b1ad9d5ffce8..8d7402c13e56 100644 --- a/include/uapi/drm/nouveau_drm.h +++ b/include/uapi/drm/nouveau_drm.h @@ -138,6 +138,7 @@ struct drm_nouveau_gem_pushbuf_push { __u32 pad; __u64 offset; __u64 length; +#define NOUVEAU_GEM_PUSHBUF_NO_PREFETCH (1 << 23) }; struct drm_nouveau_gem_pushbuf { @@ -338,7 +339,12 @@ struct drm_nouveau_exec_push { /** * @va_len: the length of the push buffer mapping */ - __u64 va_len; + __u32 va_len; + /** + * @flags: the flags for this push buffer mapping + */ + __u32 flags; +#define DRM_NOUVEAU_EXEC_PUSH_NO_PREFETCH 0x1 }; /** -- cgit v1.2.3 From 5aa1dfcdf0a429e4941e2eef75b006a8c7a8ac49 Mon Sep 17 00:00:00 2001 From: Wayne Lin Date: Mon, 7 Aug 2023 10:56:38 +0800 Subject: drm/mst: Refactor the flow for payload allocation/removement [Why] Today, the allocation/deallocation steps and status is a bit unclear. For instance, payload->vc_start_slot = -1 stands for "the failure of updating DPCD payload ID table" and can also represent as "payload is not allocated yet". These two cases should be handled differently and hence better to distinguish them for better understanding. [How] Define enumeration - ALLOCATION_LOCAL, ALLOCATION_DFP and ALLOCATION_REMOTE to distinguish different allocation status. Adjust the code to handle different status accordingly for better understanding the sequence of payload allocation and payload removement. For payload creation, the procedure should look like this: DRM part 1: * step 1 - update sw mst mgr variables to add a new payload * step 2 - add payload at immediate DFP DPCD payload table Driver: * Add new payload in HW and sync up with DFP by sending ACT DRM Part 2: * Send ALLOCATE_PAYLOAD sideband message to allocate bandwidth along the virtual channel. And as for payload removement, the procedure should look like this: DRM part 1: * step 1 - Send ALLOCATE_PAYLOAD sideband message to release bandwidth along the virtual channel * step 2 - Clear payload allocation at immediate DFP DPCD payload table Driver: * Remove the payload in HW and sync up with DFP by sending ACT DRM part 2: * update sw mst mgr variables to remove the payload Note that it's fine to fail when communicate with the branch device connected at immediate downstrean-facing port, but updating variables of SW mst mgr and HW configuration should be conducted anyway. That's because it's under commit_tail and we need to complete the HW programming. Changes since v1: * Remove the set but not use variable 'old_payload' in function 'nv50_msto_prepare'. Catched by kernel test robot Signed-off-by: Wayne Lin Signed-off-by: Lyude Paul Link: https://patchwork.freedesktop.org/patch/msgid/20230807025639.1612361-3-Wayne.Lin@amd.com --- .../drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c | 20 +-- drivers/gpu/drm/display/drm_dp_mst_topology.c | 159 +++++++++++++-------- drivers/gpu/drm/i915/display/intel_dp_mst.c | 18 ++- drivers/gpu/drm/nouveau/dispnv50/disp.c | 21 +-- include/drm/display/drm_dp_mst_helper.h | 23 ++- 5 files changed, 153 insertions(+), 88 deletions(-) (limited to 'drivers/gpu/drm/nouveau') diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index d9a482908380..9ad509279b0a 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -219,7 +219,7 @@ static void dm_helpers_construct_old_payload( /* Set correct time_slots/PBN of old payload. * other fields (delete & dsc_enabled) in * struct drm_dp_mst_atomic_payload are don't care fields - * while calling drm_dp_remove_payload() + * while calling drm_dp_remove_payload_part2() */ for (i = 0; i < current_link_table.stream_count; i++) { dc_alloc = @@ -262,13 +262,12 @@ bool dm_helpers_dp_mst_write_payload_allocation_table( mst_mgr = &aconnector->mst_root->mst_mgr; mst_state = to_drm_dp_mst_topology_state(mst_mgr->base.state); - - /* It's OK for this to fail */ new_payload = drm_atomic_get_mst_payload_state(mst_state, aconnector->mst_output_port); if (enable) { target_payload = new_payload; + /* It's OK for this to fail */ drm_dp_add_payload_part1(mst_mgr, mst_state, new_payload); } else { /* construct old payload by VCPI*/ @@ -276,7 +275,7 @@ bool dm_helpers_dp_mst_write_payload_allocation_table( new_payload, &old_payload); target_payload = &old_payload; - drm_dp_remove_payload(mst_mgr, mst_state, &old_payload, new_payload); + drm_dp_remove_payload_part1(mst_mgr, mst_state, new_payload); } /* mst_mgr->->payloads are VC payload notify MST branch using DPCD or @@ -342,7 +341,7 @@ bool dm_helpers_dp_mst_send_payload_allocation( struct amdgpu_dm_connector *aconnector; struct drm_dp_mst_topology_state *mst_state; struct drm_dp_mst_topology_mgr *mst_mgr; - struct drm_dp_mst_atomic_payload *payload; + struct drm_dp_mst_atomic_payload *new_payload, *old_payload; enum mst_progress_status set_flag = MST_ALLOCATE_NEW_PAYLOAD; enum mst_progress_status clr_flag = MST_CLEAR_ALLOCATED_PAYLOAD; int ret = 0; @@ -355,15 +354,20 @@ bool dm_helpers_dp_mst_send_payload_allocation( mst_mgr = &aconnector->mst_root->mst_mgr; mst_state = to_drm_dp_mst_topology_state(mst_mgr->base.state); - payload = drm_atomic_get_mst_payload_state(mst_state, aconnector->mst_output_port); + new_payload = drm_atomic_get_mst_payload_state(mst_state, aconnector->mst_output_port); if (!enable) { set_flag = MST_CLEAR_ALLOCATED_PAYLOAD; clr_flag = MST_ALLOCATE_NEW_PAYLOAD; } - if (enable) - ret = drm_dp_add_payload_part2(mst_mgr, mst_state->base.state, payload); + if (enable) { + ret = drm_dp_add_payload_part2(mst_mgr, mst_state->base.state, new_payload); + } else { + dm_helpers_construct_old_payload(stream->link, mst_state->pbn_div, + new_payload, old_payload); + drm_dp_remove_payload_part2(mst_mgr, mst_state, old_payload, new_payload); + } if (ret) { amdgpu_dm_set_mst_status(&aconnector->mst_status, diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gpu/drm/display/drm_dp_mst_topology.c index 4d80426757ab..e04f87ff755a 100644 --- a/drivers/gpu/drm/display/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c @@ -3255,15 +3255,15 @@ out_get_port: } EXPORT_SYMBOL(drm_dp_send_query_stream_enc_status); -static int drm_dp_create_payload_step1(struct drm_dp_mst_topology_mgr *mgr, - struct drm_dp_mst_atomic_payload *payload) +static int drm_dp_create_payload_at_dfp(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_atomic_payload *payload) { return drm_dp_dpcd_write_payload(mgr, payload->vcpi, payload->vc_start_slot, payload->time_slots); } -static int drm_dp_create_payload_step2(struct drm_dp_mst_topology_mgr *mgr, - struct drm_dp_mst_atomic_payload *payload) +static int drm_dp_create_payload_to_remote(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_atomic_payload *payload) { int ret; struct drm_dp_mst_port *port = drm_dp_mst_topology_get_port_validated(mgr, payload->port); @@ -3276,17 +3276,20 @@ static int drm_dp_create_payload_step2(struct drm_dp_mst_topology_mgr *mgr, return ret; } -static int drm_dp_destroy_payload_step1(struct drm_dp_mst_topology_mgr *mgr, - struct drm_dp_mst_topology_state *mst_state, - struct drm_dp_mst_atomic_payload *payload) +static void drm_dp_destroy_payload_at_remote_and_dfp(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_topology_state *mst_state, + struct drm_dp_mst_atomic_payload *payload) { drm_dbg_kms(mgr->dev, "\n"); /* it's okay for these to fail */ - drm_dp_payload_send_msg(mgr, payload->port, payload->vcpi, 0); - drm_dp_dpcd_write_payload(mgr, payload->vcpi, payload->vc_start_slot, 0); + if (payload->payload_allocation_status == DRM_DP_MST_PAYLOAD_ALLOCATION_REMOTE) { + drm_dp_payload_send_msg(mgr, payload->port, payload->vcpi, 0); + payload->payload_allocation_status = DRM_DP_MST_PAYLOAD_ALLOCATION_DFP; + } - return 0; + if (payload->payload_allocation_status == DRM_DP_MST_PAYLOAD_ALLOCATION_DFP) + drm_dp_dpcd_write_payload(mgr, payload->vcpi, payload->vc_start_slot, 0); } /** @@ -3296,81 +3299,105 @@ static int drm_dp_destroy_payload_step1(struct drm_dp_mst_topology_mgr *mgr, * @payload: The payload to write * * Determines the starting time slot for the given payload, and programs the VCPI for this payload - * into hardware. After calling this, the driver should generate ACT and payload packets. + * into the DPCD of DPRX. After calling this, the driver should generate ACT and payload packets. * - * Returns: 0 on success, error code on failure. In the event that this fails, - * @payload.vc_start_slot will also be set to -1. + * Returns: 0 on success, error code on failure. */ int drm_dp_add_payload_part1(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_topology_state *mst_state, struct drm_dp_mst_atomic_payload *payload) { struct drm_dp_mst_port *port; - int ret; + int ret = 0; + bool allocate = true; + + /* Update mst mgr info */ + if (mgr->payload_count == 0) + mgr->next_start_slot = mst_state->start_slot; + + payload->vc_start_slot = mgr->next_start_slot; + + mgr->payload_count++; + mgr->next_start_slot += payload->time_slots; + /* Allocate payload to immediate downstream facing port */ port = drm_dp_mst_topology_get_port_validated(mgr, payload->port); if (!port) { drm_dbg_kms(mgr->dev, - "VCPI %d for port %p not in topology, not creating a payload\n", + "VCPI %d for port %p not in topology, not creating a payload to remote\n", payload->vcpi, payload->port); - payload->vc_start_slot = -1; - return 0; + allocate = false; } - if (mgr->payload_count == 0) - mgr->next_start_slot = mst_state->start_slot; - - payload->vc_start_slot = mgr->next_start_slot; + if (allocate) { + ret = drm_dp_create_payload_at_dfp(mgr, payload); + if (ret < 0) + drm_warn(mgr->dev, "Failed to create MST payload for port %p: %d\n", + payload->port, ret); - ret = drm_dp_create_payload_step1(mgr, payload); - drm_dp_mst_topology_put_port(port); - if (ret < 0) { - drm_warn(mgr->dev, "Failed to create MST payload for port %p: %d\n", - payload->port, ret); - payload->vc_start_slot = -1; - return ret; } - mgr->payload_count++; - mgr->next_start_slot += payload->time_slots; + payload->payload_allocation_status = + (!allocate || ret < 0) ? DRM_DP_MST_PAYLOAD_ALLOCATION_LOCAL : + DRM_DP_MST_PAYLOAD_ALLOCATION_DFP; - return 0; + drm_dp_mst_topology_put_port(port); + + return ret; } EXPORT_SYMBOL(drm_dp_add_payload_part1); /** - * drm_dp_remove_payload() - Remove an MST payload + * drm_dp_remove_payload_part1() - Remove an MST payload along the virtual channel * @mgr: Manager to use. * @mst_state: The MST atomic state - * @old_payload: The payload with its old state - * @new_payload: The payload to write + * @payload: The payload to remove * - * Removes a payload from an MST topology if it was successfully assigned a start slot. Also updates - * the starting time slots of all other payloads which would have been shifted towards the start of - * the VC table as a result. After calling this, the driver should generate ACT and payload packets. + * Removes a payload along the virtual channel if it was successfully allocated. + * After calling this, the driver should set HW to generate ACT and then switch to new + * payload allocation state. */ -void drm_dp_remove_payload(struct drm_dp_mst_topology_mgr *mgr, - struct drm_dp_mst_topology_state *mst_state, - const struct drm_dp_mst_atomic_payload *old_payload, - struct drm_dp_mst_atomic_payload *new_payload) +void drm_dp_remove_payload_part1(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_topology_state *mst_state, + struct drm_dp_mst_atomic_payload *payload) { - struct drm_dp_mst_atomic_payload *pos; + /* Remove remote payload allocation */ bool send_remove = false; - /* We failed to make the payload, so nothing to do */ - if (new_payload->vc_start_slot == -1) - return; - mutex_lock(&mgr->lock); - send_remove = drm_dp_mst_port_downstream_of_branch(new_payload->port, mgr->mst_primary); + send_remove = drm_dp_mst_port_downstream_of_branch(payload->port, mgr->mst_primary); mutex_unlock(&mgr->lock); if (send_remove) - drm_dp_destroy_payload_step1(mgr, mst_state, new_payload); + drm_dp_destroy_payload_at_remote_and_dfp(mgr, mst_state, payload); else drm_dbg_kms(mgr->dev, "Payload for VCPI %d not in topology, not sending remove\n", - new_payload->vcpi); + payload->vcpi); + + payload->payload_allocation_status = DRM_DP_MST_PAYLOAD_ALLOCATION_LOCAL; +} +EXPORT_SYMBOL(drm_dp_remove_payload_part1); +/** + * drm_dp_remove_payload_part2() - Remove an MST payload locally + * @mgr: Manager to use. + * @mst_state: The MST atomic state + * @old_payload: The payload with its old state + * @new_payload: The payload with its latest state + * + * Updates the starting time slots of all other payloads which would have been shifted towards + * the start of the payload ID table as a result of removing a payload. Driver should call this + * function whenever it removes a payload in its HW. It's independent to the result of payload + * allocation/deallocation at branch devices along the virtual channel. + */ +void drm_dp_remove_payload_part2(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_topology_state *mst_state, + const struct drm_dp_mst_atomic_payload *old_payload, + struct drm_dp_mst_atomic_payload *new_payload) +{ + struct drm_dp_mst_atomic_payload *pos; + + /* Remove local payload allocation */ list_for_each_entry(pos, &mst_state->payloads, next) { if (pos != new_payload && pos->vc_start_slot > new_payload->vc_start_slot) pos->vc_start_slot -= old_payload->time_slots; @@ -3382,9 +3409,10 @@ void drm_dp_remove_payload(struct drm_dp_mst_topology_mgr *mgr, if (new_payload->delete) drm_dp_mst_put_port_malloc(new_payload->port); -} -EXPORT_SYMBOL(drm_dp_remove_payload); + new_payload->payload_allocation_status = DRM_DP_MST_PAYLOAD_ALLOCATION_NONE; +} +EXPORT_SYMBOL(drm_dp_remove_payload_part2); /** * drm_dp_add_payload_part2() - Execute payload update part 2 * @mgr: Manager to use. @@ -3403,17 +3431,19 @@ int drm_dp_add_payload_part2(struct drm_dp_mst_topology_mgr *mgr, int ret = 0; /* Skip failed payloads */ - if (payload->vc_start_slot == -1) { - drm_dbg_kms(mgr->dev, "Part 1 of payload creation for %s failed, skipping part 2\n", + if (payload->payload_allocation_status != DRM_DP_MST_PAYLOAD_ALLOCATION_DFP) { + drm_dbg_kms(state->dev, "Part 1 of payload creation for %s failed, skipping part 2\n", payload->port->connector->name); return -EIO; } - ret = drm_dp_create_payload_step2(mgr, payload); - if (ret < 0) { + /* Allocate payload to remote end */ + ret = drm_dp_create_payload_to_remote(mgr, payload); + if (ret < 0) drm_err(mgr->dev, "Step 2 of creating MST payload for %p failed: %d\n", payload->port, ret); - } + else + payload->payload_allocation_status = DRM_DP_MST_PAYLOAD_ALLOCATION_REMOTE; return ret; } @@ -4324,6 +4354,7 @@ int drm_dp_atomic_find_time_slots(struct drm_atomic_state *state, drm_dp_mst_get_port_malloc(port); payload->port = port; payload->vc_start_slot = -1; + payload->payload_allocation_status = DRM_DP_MST_PAYLOAD_ALLOCATION_NONE; list_add(&payload->next, &topology_state->payloads); } payload->time_slots = req_slots; @@ -4493,7 +4524,7 @@ void drm_dp_mst_atomic_wait_for_dependencies(struct drm_atomic_state *state) } /* Now that previous state is committed, it's safe to copy over the start slot - * assignments + * and allocation status assignments */ list_for_each_entry(old_payload, &old_mst_state->payloads, next) { if (old_payload->delete) @@ -4502,6 +4533,8 @@ void drm_dp_mst_atomic_wait_for_dependencies(struct drm_atomic_state *state) new_payload = drm_atomic_get_mst_payload_state(new_mst_state, old_payload->port); new_payload->vc_start_slot = old_payload->vc_start_slot; + new_payload->payload_allocation_status = + old_payload->payload_allocation_status; } } } @@ -4818,6 +4851,13 @@ void drm_dp_mst_dump_topology(struct seq_file *m, struct drm_dp_mst_atomic_payload *payload; int i, ret; + static const char *const status[] = { + "None", + "Local", + "DFP", + "Remote", + }; + mutex_lock(&mgr->lock); if (mgr->mst_primary) drm_dp_mst_dump_mstb(m, mgr->mst_primary); @@ -4834,7 +4874,7 @@ void drm_dp_mst_dump_topology(struct seq_file *m, seq_printf(m, "payload_mask: %x, max_payloads: %d, start_slot: %u, pbn_div: %d\n", state->payload_mask, mgr->max_payloads, state->start_slot, state->pbn_div); - seq_printf(m, "\n| idx | port | vcpi | slots | pbn | dsc | sink name |\n"); + seq_printf(m, "\n| idx | port | vcpi | slots | pbn | dsc | status | sink name |\n"); for (i = 0; i < mgr->max_payloads; i++) { list_for_each_entry(payload, &state->payloads, next) { char name[14]; @@ -4843,7 +4883,7 @@ void drm_dp_mst_dump_topology(struct seq_file *m, continue; fetch_monitor_name(mgr, payload->port, name, sizeof(name)); - seq_printf(m, " %5d %6d %6d %02d - %02d %5d %5s %19s\n", + seq_printf(m, " %5d %6d %6d %02d - %02d %5d %5s %8s %19s\n", i, payload->port->port_num, payload->vcpi, @@ -4851,6 +4891,7 @@ void drm_dp_mst_dump_topology(struct seq_file *m, payload->vc_start_slot + payload->time_slots - 1, payload->pbn, payload->dsc_enabled ? "Y" : "N", + status[payload->payload_allocation_status], (*name != 0) ? name : "Unknown"); } } diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index e3f176a093d2..5f73cdabe7a1 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -557,12 +557,8 @@ static void intel_mst_disable_dp(struct intel_atomic_state *state, struct intel_dp *intel_dp = &dig_port->dp; struct intel_connector *connector = to_intel_connector(old_conn_state->connector); - struct drm_dp_mst_topology_state *old_mst_state = - drm_atomic_get_old_mst_topology_state(&state->base, &intel_dp->mst_mgr); struct drm_dp_mst_topology_state *new_mst_state = drm_atomic_get_new_mst_topology_state(&state->base, &intel_dp->mst_mgr); - const struct drm_dp_mst_atomic_payload *old_payload = - drm_atomic_get_mst_payload_state(old_mst_state, connector->port); struct drm_dp_mst_atomic_payload *new_payload = drm_atomic_get_mst_payload_state(new_mst_state, connector->port); struct drm_i915_private *i915 = to_i915(connector->base.dev); @@ -572,8 +568,7 @@ static void intel_mst_disable_dp(struct intel_atomic_state *state, intel_hdcp_disable(intel_mst->connector); - drm_dp_remove_payload(&intel_dp->mst_mgr, new_mst_state, - old_payload, new_payload); + drm_dp_remove_payload_part1(&intel_dp->mst_mgr, new_mst_state, new_payload); intel_audio_codec_disable(encoder, old_crtc_state, old_conn_state); } @@ -588,6 +583,14 @@ static void intel_mst_post_disable_dp(struct intel_atomic_state *state, struct intel_dp *intel_dp = &dig_port->dp; struct intel_connector *connector = to_intel_connector(old_conn_state->connector); + struct drm_dp_mst_topology_state *old_mst_state = + drm_atomic_get_old_mst_topology_state(&state->base, &intel_dp->mst_mgr); + struct drm_dp_mst_topology_state *new_mst_state = + drm_atomic_get_new_mst_topology_state(&state->base, &intel_dp->mst_mgr); + const struct drm_dp_mst_atomic_payload *old_payload = + drm_atomic_get_mst_payload_state(old_mst_state, connector->port); + struct drm_dp_mst_atomic_payload *new_payload = + drm_atomic_get_mst_payload_state(new_mst_state, connector->port); struct drm_i915_private *dev_priv = to_i915(connector->base.dev); bool last_mst_stream; @@ -608,6 +611,9 @@ static void intel_mst_post_disable_dp(struct intel_atomic_state *state, wait_for_act_sent(encoder, old_crtc_state); + drm_dp_remove_payload_part2(&intel_dp->mst_mgr, new_mst_state, + old_payload, new_payload); + intel_ddi_disable_transcoder_func(old_crtc_state); if (DISPLAY_VER(dev_priv) >= 9) diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c index d18e24b6f1eb..39e9ba4139c7 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -882,21 +882,26 @@ struct nouveau_encoder *nv50_real_outp(struct drm_encoder *encoder) static void nv50_msto_cleanup(struct drm_atomic_state *state, - struct drm_dp_mst_topology_state *mst_state, + struct drm_dp_mst_topology_state *new_mst_state, struct drm_dp_mst_topology_mgr *mgr, struct nv50_msto *msto) { struct nouveau_drm *drm = nouveau_drm(msto->encoder.dev); - struct drm_dp_mst_atomic_payload *payload = - drm_atomic_get_mst_payload_state(mst_state, msto->mstc->port); + struct drm_dp_mst_atomic_payload *new_payload = + drm_atomic_get_mst_payload_state(new_mst_state, msto->mstc->port); + struct drm_dp_mst_topology_state *old_mst_state = + drm_atomic_get_old_mst_topology_state(state, mgr); + const struct drm_dp_mst_atomic_payload *old_payload = + drm_atomic_get_mst_payload_state(old_mst_state, msto->mstc->port); NV_ATOMIC(drm, "%s: msto cleanup\n", msto->encoder.name); if (msto->disabled) { msto->mstc = NULL; msto->disabled = false; + drm_dp_remove_payload_part2(mgr, new_mst_state, old_payload, new_payload); } else if (msto->enabled) { - drm_dp_add_payload_part2(mgr, state, payload); + drm_dp_add_payload_part2(mgr, state, new_payload); msto->enabled = false; } } @@ -910,19 +915,15 @@ nv50_msto_prepare(struct drm_atomic_state *state, struct nouveau_drm *drm = nouveau_drm(msto->encoder.dev); struct nv50_mstc *mstc = msto->mstc; struct nv50_mstm *mstm = mstc->mstm; - struct drm_dp_mst_topology_state *old_mst_state; - struct drm_dp_mst_atomic_payload *payload, *old_payload; + struct drm_dp_mst_atomic_payload *payload; NV_ATOMIC(drm, "%s: msto prepare\n", msto->encoder.name); - old_mst_state = drm_atomic_get_old_mst_topology_state(state, mgr); - payload = drm_atomic_get_mst_payload_state(mst_state, mstc->port); - old_payload = drm_atomic_get_mst_payload_state(old_mst_state, mstc->port); // TODO: Figure out if we want to do a better job of handling VCPI allocation failures here? if (msto->disabled) { - drm_dp_remove_payload(mgr, mst_state, old_payload, payload); + drm_dp_remove_payload_part1(mgr, mst_state, payload); nvif_outp_dp_mst_vcpi(&mstm->outp->outp, msto->head->base.index, 0, 0, 0, 0); } else { diff --git a/include/drm/display/drm_dp_mst_helper.h b/include/drm/display/drm_dp_mst_helper.h index ed5c9660563c..4429d3b1745b 100644 --- a/include/drm/display/drm_dp_mst_helper.h +++ b/include/drm/display/drm_dp_mst_helper.h @@ -46,6 +46,13 @@ struct drm_dp_mst_topology_ref_history { }; #endif /* IS_ENABLED(CONFIG_DRM_DEBUG_DP_MST_TOPOLOGY_REFS) */ +enum drm_dp_mst_payload_allocation { + DRM_DP_MST_PAYLOAD_ALLOCATION_NONE, + DRM_DP_MST_PAYLOAD_ALLOCATION_LOCAL, + DRM_DP_MST_PAYLOAD_ALLOCATION_DFP, + DRM_DP_MST_PAYLOAD_ALLOCATION_REMOTE, +}; + struct drm_dp_mst_branch; /** @@ -537,7 +544,7 @@ struct drm_dp_mst_atomic_payload { * drm_dp_mst_atomic_wait_for_dependencies() has been called, which will ensure the * previous MST states payload start slots have been copied over to the new state. Note * that a new start slot won't be assigned/removed from this payload until - * drm_dp_add_payload_part1()/drm_dp_remove_payload() have been called. + * drm_dp_add_payload_part1()/drm_dp_remove_payload_part2() have been called. * * Acquire the MST modesetting lock, and then wait for any pending MST-related commits to * get committed to hardware by calling drm_crtc_commit_wait() on each of the * &drm_crtc_commit structs in &drm_dp_mst_topology_state.commit_deps. @@ -564,6 +571,9 @@ struct drm_dp_mst_atomic_payload { /** @dsc_enabled: Whether or not this payload has DSC enabled */ bool dsc_enabled : 1; + /** @payload_allocation_status: The allocation status of this payload */ + enum drm_dp_mst_payload_allocation payload_allocation_status; + /** @next: The list node for this payload */ struct list_head next; }; @@ -842,10 +852,13 @@ int drm_dp_add_payload_part1(struct drm_dp_mst_topology_mgr *mgr, int drm_dp_add_payload_part2(struct drm_dp_mst_topology_mgr *mgr, struct drm_atomic_state *state, struct drm_dp_mst_atomic_payload *payload); -void drm_dp_remove_payload(struct drm_dp_mst_topology_mgr *mgr, - struct drm_dp_mst_topology_state *mst_state, - const struct drm_dp_mst_atomic_payload *old_payload, - struct drm_dp_mst_atomic_payload *new_payload); +void drm_dp_remove_payload_part1(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_topology_state *mst_state, + struct drm_dp_mst_atomic_payload *payload); +void drm_dp_remove_payload_part2(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_topology_state *mst_state, + const struct drm_dp_mst_atomic_payload *old_payload, + struct drm_dp_mst_atomic_payload *new_payload); int drm_dp_check_act_status(struct drm_dp_mst_topology_mgr *mgr); -- cgit v1.2.3