Maarten Maathuis
2009-Dec-13 15:07 UTC
[Nouveau] [PATCH] nouveau: avoid running out of relocs (attempt 5)
- Added flush notify functions for NV30 and NV40. - NV30 and NV40 need testing (check for regressions). --- src/gallium/drivers/nouveau/nouveau_stateobj.h | 47 +++++++++++++++++++----- src/gallium/drivers/nv04/nv04_surface_2d.c | 9 +++-- src/gallium/drivers/nv30/nv30_context.c | 3 ++ src/gallium/drivers/nv30/nv30_context.h | 1 + src/gallium/drivers/nv30/nv30_state_emit.c | 10 +++++- src/gallium/drivers/nv40/nv40_context.c | 3 ++ src/gallium/drivers/nv40/nv40_context.h | 1 + src/gallium/drivers/nv40/nv40_state_emit.c | 10 +++++- src/gallium/drivers/nv50/nv50_query.c | 2 +- src/gallium/drivers/nv50/nv50_surface.c | 2 + src/gallium/drivers/nv50/nv50_transfer.c | 4 +- 11 files changed, 74 insertions(+), 18 deletions(-) diff --git a/src/gallium/drivers/nouveau/nouveau_stateobj.h b/src/gallium/drivers/nouveau/nouveau_stateobj.h index 62990f9..de045ef 100644 --- a/src/gallium/drivers/nouveau/nouveau_stateobj.h +++ b/src/gallium/drivers/nouveau/nouveau_stateobj.h @@ -112,19 +112,29 @@ so_emit(struct nouveau_channel *chan, struct nouveau_stateobj *so) { struct nouveau_pushbuf *pb = chan->pushbuf; unsigned nr, i; + int ret = 0; nr = so->cur - so->push; - if (pb->remaining < nr) - nouveau_pushbuf_flush(chan, nr); + /* This will flush if we need space. + * We don't actually need the marker. + */ + if ((ret = nouveau_pushbuf_marker_emit(chan, nr, so->cur_reloc))) { + debug_printf("so_emit failed marker emit with error %d\n", ret); + return; + } pb->remaining -= nr; memcpy(pb->cur, so->push, nr * 4); for (i = 0; i < so->cur_reloc; i++) { struct nouveau_stateobj_reloc *r = &so->reloc[i]; - nouveau_pushbuf_emit_reloc(chan, pb->cur + r->offset, + if ((ret = nouveau_pushbuf_emit_reloc(chan, pb->cur + r->offset, r->bo, r->data, 0, r->flags, - r->vor, r->tor); + r->vor, r->tor))) { + debug_printf("so_emit failed reloc with error %d\n", ret); + pb->remaining += nr; + return; + } } pb->cur += nr; } @@ -134,26 +144,43 @@ so_emit_reloc_markers(struct nouveau_channel *chan, struct nouveau_stateobj *so) { struct nouveau_pushbuf *pb = chan->pushbuf; unsigned i; + int ret = 0; if (!so) return; i = so->cur_reloc << 1; - if (pb->remaining < i) - nouveau_pushbuf_flush(chan, i); + /* This will flush if we need space. + * We don't actually need the marker. + */ + if ((ret = nouveau_pushbuf_marker_emit(chan, i, i))) { + debug_printf("so_emit_reloc_markers failed marker emit with \ + error %d\n", ret); + return; + } pb->remaining -= i; for (i = 0; i < so->cur_reloc; i++) { struct nouveau_stateobj_reloc *r = &so->reloc[i]; - nouveau_pushbuf_emit_reloc(chan, pb->cur++, r->bo, r->packet, 0, + if ((ret = nouveau_pushbuf_emit_reloc(chan, pb->cur++, r->bo, r->packet, 0, (r->flags & (NOUVEAU_BO_VRAM | NOUVEAU_BO_GART | NOUVEAU_BO_RDWR)) | - NOUVEAU_BO_DUMMY, 0, 0); - nouveau_pushbuf_emit_reloc(chan, pb->cur++, r->bo, r->data, 0, + NOUVEAU_BO_DUMMY, 0, 0))) { + debug_printf("so_emit_reloc_markers failed reloc \ + with error %d\n", ret); + pb->remaining += i; + return; + } + if ((ret = nouveau_pushbuf_emit_reloc(chan, pb->cur++, r->bo, r->data, 0, r->flags | NOUVEAU_BO_DUMMY, - r->vor, r->tor); + r->vor, r->tor))) { + debug_printf("so_emit_reloc_markers failed reloc \ + with error %d\n", ret); + pb->remaining += i; + return; + } } } diff --git a/src/gallium/drivers/nv04/nv04_surface_2d.c b/src/gallium/drivers/nv04/nv04_surface_2d.c index 932893e..3020806 100644 --- a/src/gallium/drivers/nv04/nv04_surface_2d.c +++ b/src/gallium/drivers/nv04/nv04_surface_2d.c @@ -133,6 +133,9 @@ nv04_surface_copy_swizzle(struct nv04_surface_2d *ctx, assert(sub_w == w || util_is_pot(sub_w)); assert(sub_h == h || util_is_pot(sub_h)); + MARK_RING (chan, 8 + ((w+sub_w)/sub_w)*((h+sub_h)/sub_h)*17, 2 + + ((w+sub_w)/sub_w)*((h+sub_h)/sub_h)*2); + BEGIN_RING(chan, swzsurf, NV04_SWIZZLED_SURFACE_DMA_IMAGE, 1); OUT_RELOCo(chan, dst_bo, NOUVEAU_BO_GART | NOUVEAU_BO_VRAM | NOUVEAU_BO_WR); @@ -202,7 +205,7 @@ nv04_surface_copy_m2mf(struct nv04_surface_2d *ctx, unsigned src_offset = src->offset + sy * src_pitch + sx * pf_get_blocksize(src->texture->format); - WAIT_RING (chan, 3 + ((h / 2047) + 1) * 9); + MARK_RING (chan, 3 + ((h / 2047) + 1) * 9, 2 + ((h / 2047) + 1) * 2); BEGIN_RING(chan, m2mf, NV04_MEMORY_TO_MEMORY_FORMAT_DMA_BUFFER_IN, 2); OUT_RELOCo(chan, src_bo, NOUVEAU_BO_GART | NOUVEAU_BO_VRAM | NOUVEAU_BO_RD); @@ -250,7 +253,7 @@ nv04_surface_copy_blit(struct nv04_surface_2d *ctx, struct pipe_surface *dst, if (format < 0) return 1; - WAIT_RING (chan, 12); + MARK_RING (chan, 12, 4); BEGIN_RING(chan, surf2d, NV04_CONTEXT_SURFACES_2D_DMA_IMAGE_SOURCE, 2); OUT_RELOCo(chan, src_bo, NOUVEAU_BO_VRAM | NOUVEAU_BO_RD); OUT_RELOCo(chan, dst_bo, NOUVEAU_BO_VRAM | NOUVEAU_BO_WR); @@ -315,7 +318,7 @@ nv04_surface_fill(struct nv04_surface_2d *ctx, struct pipe_surface *dst, gdirect_format = nv04_rect_format(dst->format); assert(gdirect_format >= 0); - WAIT_RING (chan, 16); + MARK_RING (chan, 16, 4); BEGIN_RING(chan, surf2d, NV04_CONTEXT_SURFACES_2D_DMA_IMAGE_SOURCE, 2); OUT_RELOCo(chan, dst_bo, NOUVEAU_BO_VRAM | NOUVEAU_BO_WR); OUT_RELOCo(chan, dst_bo, NOUVEAU_BO_VRAM | NOUVEAU_BO_WR); diff --git a/src/gallium/drivers/nv30/nv30_context.c b/src/gallium/drivers/nv30/nv30_context.c index d8300fd..46a821a 100644 --- a/src/gallium/drivers/nv30/nv30_context.c +++ b/src/gallium/drivers/nv30/nv30_context.c @@ -58,6 +58,9 @@ nv30_create(struct pipe_screen *pscreen, unsigned pctx_id) nv30->pipe.is_texture_referenced = nouveau_is_texture_referenced; nv30->pipe.is_buffer_referenced = nouveau_is_buffer_referenced; + screen->base.channel->user_private = nv30; + screen->base.channel->flush_notify = nv30_state_flush_notify; + nv30_init_query_functions(nv30); nv30_init_surface_functions(nv30); nv30_init_state_functions(nv30); diff --git a/src/gallium/drivers/nv30/nv30_context.h b/src/gallium/drivers/nv30/nv30_context.h index 8d49366..6f44b1c 100644 --- a/src/gallium/drivers/nv30/nv30_context.h +++ b/src/gallium/drivers/nv30/nv30_context.h @@ -184,6 +184,7 @@ extern void nv30_fragtex_bind(struct nv30_context *); /* nv30_state.c and friends */ extern boolean nv30_state_validate(struct nv30_context *nv30); extern void nv30_state_emit(struct nv30_context *nv30); +extern void nv30_state_flush_notify(struct nouveau_channel *chan); extern struct nv30_state_entry nv30_state_rasterizer; extern struct nv30_state_entry nv30_state_scissor; extern struct nv30_state_entry nv30_state_stipple; diff --git a/src/gallium/drivers/nv30/nv30_state_emit.c b/src/gallium/drivers/nv30/nv30_state_emit.c index 621b884..ac52d94 100644 --- a/src/gallium/drivers/nv30/nv30_state_emit.c +++ b/src/gallium/drivers/nv30/nv30_state_emit.c @@ -41,7 +41,7 @@ nv30_state_emit(struct nv30_context *nv30) struct nouveau_channel *chan = nv30->screen->base.channel; struct nv30_state *state = &nv30->state; struct nv30_screen *screen = nv30->screen; - unsigned i, samplers; + unsigned i; uint64_t states; if (nv30->pctx_id != screen->cur_pctx) { @@ -63,6 +63,14 @@ nv30_state_emit(struct nv30_context *nv30) } state->dirty = 0; +} + +void +nv30_state_flush_notify(struct nouveau_channel *chan) +{ + struct nv30_context *nv30 = chan->user_private; + struct nv30_state *state = &nv30->state; + unsigned i, samplers; so_emit_reloc_markers(chan, state->hw[NV30_STATE_FB]); for (i = 0, samplers = state->fp_samplers; i < 16 && samplers; i++) { diff --git a/src/gallium/drivers/nv40/nv40_context.c b/src/gallium/drivers/nv40/nv40_context.c index 7f00827..eb9cce4 100644 --- a/src/gallium/drivers/nv40/nv40_context.c +++ b/src/gallium/drivers/nv40/nv40_context.c @@ -58,6 +58,9 @@ nv40_create(struct pipe_screen *pscreen, unsigned pctx_id) nv40->pipe.is_texture_referenced = nouveau_is_texture_referenced; nv40->pipe.is_buffer_referenced = nouveau_is_buffer_referenced; + screen->base.channel->user_private = nv40; + screen->base.channel->flush_notify = nv40_state_flush_notify; + nv40_init_query_functions(nv40); nv40_init_surface_functions(nv40); nv40_init_state_functions(nv40); diff --git a/src/gallium/drivers/nv40/nv40_context.h b/src/gallium/drivers/nv40/nv40_context.h index a3d5941..cf33b64 100644 --- a/src/gallium/drivers/nv40/nv40_context.h +++ b/src/gallium/drivers/nv40/nv40_context.h @@ -204,6 +204,7 @@ extern void nv40_fragtex_bind(struct nv40_context *); extern boolean nv40_state_validate(struct nv40_context *nv40); extern boolean nv40_state_validate_swtnl(struct nv40_context *nv40); extern void nv40_state_emit(struct nv40_context *nv40); +extern void nv40_state_flush_notify(struct nouveau_channel *chan); extern struct nv40_state_entry nv40_state_rasterizer; extern struct nv40_state_entry nv40_state_scissor; extern struct nv40_state_entry nv40_state_stipple; diff --git a/src/gallium/drivers/nv40/nv40_state_emit.c b/src/gallium/drivers/nv40/nv40_state_emit.c index 1986929..ba0fbcb 100644 --- a/src/gallium/drivers/nv40/nv40_state_emit.c +++ b/src/gallium/drivers/nv40/nv40_state_emit.c @@ -57,7 +57,7 @@ nv40_state_emit(struct nv40_context *nv40) struct nouveau_channel *chan = nv40->screen->base.channel; struct nv40_state *state = &nv40->state; struct nv40_screen *screen = nv40->screen; - unsigned i, samplers; + unsigned i; uint64_t states; if (nv40->pctx_id != screen->cur_pctx) { @@ -87,6 +87,14 @@ nv40_state_emit(struct nv40_context *nv40) } state->dirty = 0; +} + +void +nv40_state_flush_notify(struct nouveau_channel *chan) +{ + struct nv40_context *nv40 = chan->user_private; + struct nv40_state *state = &nv40->state; + unsigned i, samplers; so_emit_reloc_markers(chan, state->hw[NV40_STATE_FB]); for (i = 0, samplers = state->fp_samplers; i < 16 && samplers; i++) { diff --git a/src/gallium/drivers/nv50/nv50_query.c b/src/gallium/drivers/nv50/nv50_query.c index 5305c93..268c982 100644 --- a/src/gallium/drivers/nv50/nv50_query.c +++ b/src/gallium/drivers/nv50/nv50_query.c @@ -93,7 +93,7 @@ nv50_query_end(struct pipe_context *pipe, struct pipe_query *pq) struct nouveau_grobj *tesla = nv50->screen->tesla; struct nv50_query *q = nv50_query(pq); - WAIT_RING (chan, 5); + MARK_RING (chan, 5, 2); /* flush on lack of space or relocs */ BEGIN_RING(chan, tesla, NV50TCL_QUERY_ADDRESS_HIGH, 4); OUT_RELOCh(chan, q->bo, 0, NOUVEAU_BO_VRAM | NOUVEAU_BO_WR); OUT_RELOCl(chan, q->bo, 0, NOUVEAU_BO_VRAM | NOUVEAU_BO_WR); diff --git a/src/gallium/drivers/nv50/nv50_surface.c b/src/gallium/drivers/nv50/nv50_surface.c index 6bf6f77..79655fc 100644 --- a/src/gallium/drivers/nv50/nv50_surface.c +++ b/src/gallium/drivers/nv50/nv50_surface.c @@ -62,6 +62,7 @@ nv50_surface_set(struct nv50_screen *screen, struct pipe_surface *ps, int dst) return 1; if (!bo->tile_flags) { + MARK_RING (chan, 9, 2); /* flush on lack of space or relocs */ BEGIN_RING(chan, eng2d, mthd, 2); OUT_RING (chan, format); OUT_RING (chan, 1); @@ -72,6 +73,7 @@ nv50_surface_set(struct nv50_screen *screen, struct pipe_surface *ps, int dst) OUT_RELOCh(chan, bo, ps->offset, flags); OUT_RELOCl(chan, bo, ps->offset, flags); } else { + MARK_RING (chan, 11, 2); /* flush on lack of space or relocs */ BEGIN_RING(chan, eng2d, mthd, 5); OUT_RING (chan, format); OUT_RING (chan, 0); diff --git a/src/gallium/drivers/nv50/nv50_transfer.c b/src/gallium/drivers/nv50/nv50_transfer.c index 4705f96..1b6c8d6 100644 --- a/src/gallium/drivers/nv50/nv50_transfer.c +++ b/src/gallium/drivers/nv50/nv50_transfer.c @@ -81,7 +81,7 @@ nv50_transfer_rect_m2mf(struct pipe_screen *pscreen, while (height) { int line_count = height > 2047 ? 2047 : height; - WAIT_RING (chan, 15); + MARK_RING (chan, 15, 4); /* flush on lack of space or relocs */ BEGIN_RING(chan, m2mf, NV50_MEMORY_TO_MEMORY_FORMAT_OFFSET_IN_HIGH, 2); OUT_RELOCh(chan, src_bo, src_offset, src_reloc); @@ -282,7 +282,7 @@ nv50_upload_sifc(struct nv50_context *nv50, reloc |= NOUVEAU_BO_WR; - WAIT_RING (chan, 32); + MARK_RING (chan, 32, 2); /* flush on lack of space or relocs */ if (bo->tile_flags) { BEGIN_RING(chan, eng2d, NV50_2D_DST_FORMAT, 5); -- 1.6.5.4
Patrice Mandin
2009-Dec-13 20:06 UTC
[Nouveau] [PATCH] nouveau: avoid running out of relocs (attempt 5)
Le Sun, 13 Dec 2009 16:07:37 +0100 Maarten Maathuis <madman2003 at gmail.com> a ?crit:> - Added flush notify functions for NV30 and NV40. > - NV30 and NV40 need testing (check for regressions).Just tested for NV30, no visible regression, just a minor performance penalty. So it's OK for me. Timedemo with ioquake3: mesa git head: 1260 frames 31.5 seconds 40.0 fps 9.0/25.0/50.0/8.0 ms 1260 frames 31.4 seconds 40.1 fps 9.0/24.9/50.0/7.9 ms with your patch, outofreloc.diff: 1260 frames 32.7 seconds 38.5 fps 9.0/26.0/52.0/8.2 ms 1260 frames 32.6 seconds 38.7 fps 9.0/25.8/51.0/8.1 ms -- Patrice Mandin WWW: http://pmandin.atari.org/ Programmeur Linux, Atari Sp?cialit?: D?veloppement, jeux
Xavier
2009-Dec-13 20:34 UTC
[Nouveau] [PATCH] nouveau: avoid running out of relocs (attempt 5)
On Sun, Dec 13, 2009 at 4:07 PM, Maarten Maathuis <madman2003 at gmail.com> wrote:> - Added flush notify functions for NV30 and NV40. > - NV30 and NV40 need testing (check for regressions). > ---Tested on NV35. No obvious regressions on xmoto and teeworlds (I didn't benchmark though, but the perf felt similar). I was wondering, is there any practical testsuite for this kind of regression testing ? It would be cool to just run a set of tests known to work on nv30 for example, and getting automatically a summary of the results. Then run again after a patch and compare. Any suggestions that get close to that are welcome :)
Christoph Bumiller
2009-Dec-14 16:51 UTC
[Nouveau] [PATCH] nouveau: avoid running out of relocs (attempt 5)
On 13.12.2009 16:07, Maarten Maathuis wrote:> - Added flush notify functions for NV30 and NV40. > - NV30 and NV40 need testing (check for regressions). > --- > > @@ -112,19 +112,29 @@ so_emit(struct nouveau_channel *chan, struct nouveau_stateobj *so) > { > struct nouveau_pushbuf *pb = chan->pushbuf; > unsigned nr, i; > + int ret = 0; > > nr = so->cur - so->push; > - if (pb->remaining < nr) > - nouveau_pushbuf_flush(chan, nr); > + /* This will flush if we need space. > + * We don't actually need the marker. > + */ > + if ((ret = nouveau_pushbuf_marker_emit(chan, nr, so->cur_reloc))) { > + debug_printf("so_emit failed marker emit with error %d\n", ret); > + return; > + } > pb->remaining -= nr; > >I'm not sure why flushing mid-frame would be bad as Ben said. After all, if there's lots of stuff done during a frame, we will inevitably flush at some point ...> - nouveau_pushbuf_emit_reloc(chan, pb->cur++, r->bo, r->data, 0, > + NOUVEAU_BO_DUMMY, 0, 0))) { > + debug_printf("so_emit_reloc_markers failed reloc \ > + with error %d\n", ret); >Now these debug_printf "will probably give you unwanted" "extra space the way they are" As for the interesting part, the too many relocs I got in vegastrike (and which would need a LOT of NOPs) are gone and nothing else seems to be broken. At some point we'll probably want to know if buffer relocation succeeded or not I guess, and fallback to software on failure. But that still doesn't have a high priority for me personally. Christoph
Maybe Matching Threads
- [PATCH] nouveau: avoid running out of relocs (attempt 4)
- [PATCH] nouveau: avoid running out of relocs (attempt 3)
- [PATCH 1/2] nv30-nv40: Rewrite primitive splitting and emission
- [PATCH] nv04-nv40: Rewrite and unify miptree and transfer code (v2)
- [PATCH] nv30-nv40: support unlimited queries (v2)