/**
* amdgpu_ib_schedule - schedule an IB (Indirect Buffer) on the ring
*
* @adev: amdgpu_device pointer
* @num_ibs: number of IBs to schedule
* @ibs: IB objects to schedule
* @f: fence created during this submission
*
* Schedule an IB on the associated ring (all asics).
* Returns 0 on success, error on failure.
*
* On SI, there are two parallel engines fed from the primary ring,
* the CE (Constant Engine) and the DE (Drawing Engine). Since
* resource descriptors have moved to memory, the CE allows you to
* prime the caches while the DE is updating register state so that
* the resource descriptors will be already in cache when the draw is
* processed. To accomplish this, the userspace driver submits two
* IBs, one for the CE and one for the DE. If there is a CE IB (called
* a CONST_IB), it will be put on the ring prior to the DE IB. Prior
* to SI there was just a DE IB.
*/
int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
struct amdgpu_ib *ibs, struct amdgpu_job *job,
struct dma_fence **f)
{
struct amdgpu_device *adev = ring->adev;
struct amdgpu_ib *ib = &ibs[0];
struct dma_fence *tmp = NULL;
bool skip_preamble, need_ctx_switch;
unsigned patch_offset = ~0;
struct amdgpu_vm *vm;
uint64_t fence_ctx;
uint32_t status = 0, alloc_size;
unsigned fence_flags = 0;
unsigned i;
int r = 0;
bool need_pipe_sync = false;
if (num_ibs == 0)
return -EINVAL;
/* ring tests don't use a job */
if (job) {
vm = job->vm;
fence_ctx = job->base.s_fence->scheduled.context;
} else {
vm = NULL;
fence_ctx = 0;
}
if (!ring->sched.ready) {
dev_err(adev->dev, "couldn't schedule ib on ring <%s>\n", ring->name);
return -EINVAL;
}
if (vm && !job->vmid) {
dev_err(adev->dev, "VM IB without ID\n");
return -EINVAL;
}
alloc_size = ring->funcs->emit_frame_size + num_ibs *
ring->funcs->emit_ib_size;
r = amdgpu_ring_alloc(ring, alloc_size);
if (r) {
dev_err(adev->dev, "scheduling IB failed (%d).\n", r);
return r;
}
need_ctx_switch = ring->current_ctx != fence_ctx;
if (ring->funcs->emit_pipeline_sync && job &&
((tmp = amdgpu_sync_get_fence(&job->sched_sync, NULL)) ||
(amdgpu_sriov_vf(adev) && need_ctx_switch) ||
amdgpu_vm_need_pipeline_sync(ring, job))) {
need_pipe_sync = true;
if (tmp)
trace_amdgpu_ib_pipe_sync(job, tmp);
dma_fence_put(tmp);
}
if (ring->funcs->insert_start)
ring->funcs->insert_start(ring);
if (job) {
r = amdgpu_vm_flush(ring, job, need_pipe_sync);
if (r) {
amdgpu_ring_undo(ring);
return r;
}
}
if (job && ring->funcs->init_cond_exec)
patch_offset = amdgpu_ring_init_cond_exec(ring);
#ifdef CONFIG_X86_64
if (!(adev->flags & AMD_IS_APU))
#endif
{
if (ring->funcs->emit_hdp_flush)
amdgpu_ring_emit_hdp_flush(ring);
else
amdgpu_asic_flush_hdp(adev, ring);
}
if (need_ctx_switch)
status |= AMDGPU_HAVE_CTX_SWITCH;
skip_preamble = ring->current_ctx == fence_ctx;
if (job && ring->funcs->emit_cntxcntl) {
status |= job->preamble_status;
amdgpu_ring_emit_cntxcntl(ring, status);
}
for (i = 0; i < num_ibs; ++i) {
ib = &ibs[i];
/* drop preamble IBs if we don't have a context switch */
if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) &&
skip_preamble &&
!(status & AMDGPU_PREAMBLE_IB_PRESENT_FIRST) &&
!amdgpu_sriov_vf(adev)) /* for SRIOV preemption, Preamble CE ib must be inserted anyway */
continue;
amdgpu_ring_emit_ib(ring, job, ib, status);
status &= ~AMDGPU_HAVE_CTX_SWITCH;
}
if (ring->funcs->emit_tmz)
amdgpu_ring_emit_tmz(ring, false);
#ifdef CONFIG_X86_64
if (!(adev->flags & AMD_IS_APU))
#endif
amdgpu_asic_invalidate_hdp(adev, ring);
if (ib->flags & AMDGPU_IB_FLAG_TC_WB_NOT_INVALIDATE)
fence_flags |= AMDGPU_FENCE_FLAG_TC_WB_ONLY;
/* wrap the last IB with fence */
if (job && job->uf_addr) {
amdgpu_ring_emit_fence(ring, job->uf_addr, job->uf_sequence,
fence_flags | AMDGPU_FENCE_FLAG_64BIT);
}
r = amdgpu_fence_emit(ring, f, fence_flags);
if (r) {
dev_err(adev->dev, "failed to emit fence (%d)\n", r);
if (job && job->vmid)
amdgpu_vmid_reset(adev, ring->funcs->vmhub, job->vmid);
amdgpu_ring_undo(ring);
return r;
}
if (ring->funcs->insert_end)
ring->funcs->insert_end(ring);
if (patch_offset != ~0 && ring->funcs->patch_cond_exec)
amdgpu_ring_patch_cond_exec(ring, patch_offset);
ring->current_ctx = fence_ctx;
if (vm && ring->funcs->emit_switch_buffer)
amdgpu_ring_emit_switch_buffer(ring);
amdgpu_ring_commit(ring);
return 0;
}
/**
* amdgpu_ib_schedule - schedule an IB (Indirect Buffer) on the ring
*
* @adev: amdgpu_device pointer
* @num_ibs: number of IBs to schedule
* @ibs: IB objects to schedule
* @f: fence created during this submission
*
* Schedule an IB on the associated ring (all asics).
* Returns 0 on success, error on failure.
*
* On SI, there are two parallel engines fed from the primary ring,
* the CE (Constant Engine) and the DE (Drawing Engine). Since
* resource descriptors have moved to memory, the CE allows you to
* prime the caches while the DE is updating register state so that
* the resource descriptors will be already in cache when the draw is
* processed. To accomplish this, the userspace driver submits two
* IBs, one for the CE and one for the DE. If there is a CE IB (called
* a CONST_IB), it will be put on the ring prior to the DE IB. Prior
* to SI there was just a DE IB.
*/
int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
struct amdgpu_ib *ibs, struct fence *last_vm_update,
struct amdgpu_job *job, struct fence **f)
{
struct amdgpu_device *adev = ring->adev;
struct amdgpu_ib *ib = &ibs[0];
bool skip_preamble, need_ctx_switch;
unsigned patch_offset = ~0;
struct amdgpu_vm *vm;
struct fence *hwf;
uint64_t ctx;
unsigned i;
int r = 0;
if (num_ibs == 0)
return -EINVAL;
/* ring tests don't use a job */
if (job) {
vm = job->vm;
ctx = job->ctx;
} else {
vm = NULL;
ctx = 0;
}
if (!ring->ready) {
dev_err(adev->dev, "couldn't schedule ib\n");
return -EINVAL;
}
if (vm && !job->vm_id) {
dev_err(adev->dev, "VM IB without ID\n");
return -EINVAL;
}
r = amdgpu_ring_alloc(ring, 256 * num_ibs);
if (r) {
dev_err(adev->dev, "scheduling IB failed (%d).\n", r);
return r;
}
if (ring->type == AMDGPU_RING_TYPE_SDMA && ring->funcs->init_cond_exec)
patch_offset = amdgpu_ring_init_cond_exec(ring);
if (vm) {
r = amdgpu_vm_flush(ring, job->vm_id, job->vm_pd_addr,
job->gds_base, job->gds_size,
job->gws_base, job->gws_size,
job->oa_base, job->oa_size);
if (r) {
amdgpu_ring_undo(ring);
return r;
}
}
if (ring->funcs->emit_hdp_flush)
amdgpu_ring_emit_hdp_flush(ring);
/* always set cond_exec_polling to CONTINUE */
*ring->cond_exe_cpu_addr = 1;
skip_preamble = ring->current_ctx == ctx;
need_ctx_switch = ring->current_ctx != ctx;
for (i = 0; i < num_ibs; ++i) {
ib = &ibs[i];
/* drop preamble IBs if we don't have a context switch */
if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && skip_preamble)
continue;
amdgpu_ring_emit_ib(ring, ib, job ? job->vm_id : 0,
need_ctx_switch);
need_ctx_switch = false;
}
if (ring->funcs->emit_hdp_invalidate)
amdgpu_ring_emit_hdp_invalidate(ring);
r = amdgpu_fence_emit(ring, &hwf);
if (r) {
dev_err(adev->dev, "failed to emit fence (%d)\n", r);
if (job && job->vm_id)
amdgpu_vm_reset_id(adev, job->vm_id);
amdgpu_ring_undo(ring);
return r;
}
/* wrap the last IB with fence */
if (job && job->uf_bo) {
uint64_t addr = amdgpu_bo_gpu_offset(job->uf_bo);
addr += job->uf_offset;
amdgpu_ring_emit_fence(ring, addr, job->uf_sequence,
AMDGPU_FENCE_FLAG_64BIT);
}
if (f)
*f = fence_get(hwf);
if (patch_offset != ~0 && ring->funcs->patch_cond_exec)
amdgpu_ring_patch_cond_exec(ring, patch_offset);
ring->current_ctx = ctx;
amdgpu_ring_commit(ring);
return 0;
}