search for: no_prefetch

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_nouvea...

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_nouvea...

On Tue, Aug 22, 2023 at 6:41?PM Danilo Krummrich <dakr at redhat.com> wrote: > 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 fi...

...e submitted for IB-mode DMA transfers, nor is it the maximum legal DMA transfer size (it is three bytes too large according to our reverse-engineering docs). 2) nouveau_gem_ioctl_pushbuf() still implicitly treats bit 23 of the pushbuffer buffer[].length field as NOUVEAU_GEM_PUSHBUF_NO_PREFETCH, despite previous commits moving the flag out of the length field for lower levels as this is an nv50-hardware-specific arrangement. It also conflicts with point 1 above as pushbuffers can be equal to or greater than 1<<23 in practice. (Does the ioctl code in question have...

On Sun, Dec 3, 2023 at 11:37?AM Monty Montgomery <xiphmont at gmail.com> wrote: > 1) The nv50 DMA code clearly believed push buffers could not (or > should not) exceed NV50_DMA_PUSH_MAX_LENGTH, when my analysis logging > shows that they regularly exceed this maximum by an order of magnitude > or more. Was the assumption originally true? If so, when/why did it > change? Is

...nce_new(&exec_job->fence); + /* Create a new fence, but do not emit yet. */ + ret = nouveau_fence_create(&exec_job->fence, exec_job->chan); if (ret) return ret; @@ -143,13 +144,17 @@ nouveau_exec_job_run(struct nouveau_job *job) nv50_dma_push(chan, p->va, p->va_len, no_prefetch); } - ret = nouveau_fence_emit(fence, chan); + ret = nouveau_fence_emit(fence); if (ret) { + nouveau_fence_unref(&exec_job->fence); NV_PRINTK(err, job->cli, "error fencing pushbuf: %d\n", ret); WIND_RING(chan); return ERR_PTR(ret); } + /* The fence was emitte...