<div dir="ltr">On 7 November 2013 14:46, Chris Forbes <<a href="mailto:chrisf@ijw.co.nz" target="_blank">chrisf@ijw.co.nz</a>> wrote: <div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> The current plan I think is just to drop the extra state in brw, and just reference the indirect bo directly when we need it [see Eric's comment on 09/15] </blockquote><div> </div><div>Ah, ok, that makes sense. Thanks. </div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div class="HOEnZb"><div class="h5"> On Fri, Nov 8, 2013 at 11:41 AM, Paul Berry <<a href="mailto:stereotype441@gmail.com">stereotype441@gmail.com</a>> wrote: > On 6 November 2013 23:06, Chris Forbes <<a href="mailto:chrisf@ijw.co.nz">chrisf@ijw.co.nz</a>> wrote: >> >> Signed-off-by: Chris Forbes <<a href="mailto:chrisf@ijw.co.nz">chrisf@ijw.co.nz</a>> >> --- >> src/mesa/drivers/dri/i965/brw_draw.c | 8 ++++++-- >> 1 file changed, 6 insertions(+), 2 deletions(-) >> >> diff --git a/src/mesa/drivers/dri/i965/brw_draw.c >> b/src/mesa/drivers/dri/i965/brw_draw.c >> index e6c3508..9f8ad45 100644 >> --- a/src/mesa/drivers/dri/i965/brw_draw.c >> +++ b/src/mesa/drivers/dri/i965/brw_draw.c >> @@ -323,7 +323,8 @@ static bool brw_try_draw_prims( struct gl_context >> *ctx, >> GLuint nr_prims, >> const struct _mesa_index_buffer *ib, >> GLuint min_index, >> - GLuint max_index ) >> + GLuint max_index, >> + const struct gl_buffer_object >> *indirect) >> { >> struct brw_context *brw = brw_context(ctx); >> bool retval = true; >> @@ -377,6 +378,9 @@ static bool brw_try_draw_prims( struct gl_context >> *ctx, >> brw->vb.max_index = max_index; >> brw->state.dirty.brw |= BRW_NEW_VERTICES; >> >> + brw->indirect_buffer.indb = indirect; >> + brw->state.dirty.brw |= BRW_NEW_INDIRECT_BUFFER; >> + > > > It seems wasteful to flag this bit on every draw call. Couldn't we do > something like this? > > if (brw->indirect_buffer.indb != NULL || indirect != NULL) { > brw->indirect_buffer.indb = indirect; > brw->state.dirty.brw |= BRW_NEW_INDIRECT_BUFFER; > } > > Assuming that's addressed, this patch is: > > Reviewed-by: Paul Berry <<a href="mailto:stereotype441@gmail.com">stereotype441@gmail.com</a>> > >> >> for (i = 0; i < nr_prims; i++) { >> int estimated_max_prim_size; >> >> @@ -506,7 +510,7 @@ void brw_draw_prims( struct gl_context *ctx, >> * manage it. swrast doesn't support our featureset, so we can't fall >> back >> * to it. >> */ >> - brw_try_draw_prims(ctx, arrays, prims, nr_prims, ib, min_index, >> max_index); >> + brw_try_draw_prims(ctx, arrays, prims, nr_prims, ib, min_index, >> max_index, indirect); >> } >> >> void brw_draw_init( struct brw_context *brw ) >> -- >> 1.8.4.2 >> >> _______________________________________________ >> mesa-dev mailing list >> <a href="mailto:mesa-dev@lists.freedesktop.org">mesa-dev@lists.freedesktop.org</a> >> <a href="http://lists.freedesktop.org/mailman/listinfo/mesa-dev" target="_blank">http://lists.freedesktop.org/mailman/listinfo/mesa-dev</a> > > </div></div></blockquote></div> </div></div>