[question] [Fermi] Is there a way to accumulate buffer offset after transform feedback (aka streamout)

[imirkin]

[question] [Fermi] Is there a way to accumulate buffer offset after transform feedback (aka streamout)

[imirkin]

After staring at the blob driver's traces, it seems to be doing the exact same things nouveau is, but is getting the correct results (or at least the test passes). Which means either there's something it does slightly differently which is causing the hardware to behave properly (I notice it turns TF on/off left and right and uses a short query rather than a long one, and has slightly different synchronize/etc behavior, although attempting to do the same in nouveau did not improve things), or it's some grctx setting which controls it.

So... is there some GR bit which makes that query accumulate on top of the existing buffer offset? Or any other advice on making this work properly?

[imirkin]

Based on Ben's suggestion, I set bit 0 of 0x50405c and it all magically started working. Looks like in rnndb, this was previously documented as

$ lookup -a gf100 0x50405c
PGRAPH.GPC[0].TPC[0].POLY.TFB_UNFUCKUP_OFFSET_QUERIES => 0

So I guess someone knew at some point :) But then it was forgotten.

[Gnurou]

Sorry for not having come with this answer before you found out. Well it wasn't trivial so not sure we would have thought of this. Closing this issue.