Zachary Cook

When enabled this forces MaskEvalTA true, needs a runtime check of g->GPU.MaskEvalAND for each function that has a MaskEvalTA template like you did with pgxp_runtime. When GPU.MaskEvalAND is true command->func[][]...

PlotPixel, PlotNativePixel, DrawSpan, DrawSprite look like they change behavior with MaskEval_TA forced to 1 with these changes. I definitely agree with reverting my fast compile hack with those runtime performance...

There is also the rsx_intf_push_* hw renderer stuff that uses MaskEval_TA too. Does removing pgxp from the command_drawpolygon template slow down runtime perf of the user build too like MaskEval_TA...

I think now that I know the impact of removing the templates, that we don't really need this fast compile option, it will only make things harder to understand, maintain,...

You need to build with HAVE_LIGHTREC=0 gnu-lightning would need support for webassembly to have a chance at working If HAVE_LIGHTREC=0 works the Makefile should be changed to disable it for...

I can reproduce this depending on which Dynarec Code Invalidation setting and the Dynarec DMA number, can you try lowering Dynarec DMA to 128 and set Dynarec Code Invalidation to...

@ds22x now that #599 has been merged can you re-test those crashes and glitches

@DonelBueno PGXP should work with dynarec as of e5e83cde, not too much faster than interpreter with PGXP, but better than ignoring the setting.

I'm working on that now, but not sure that performance will be improved much.

> **Tekken 2 (USA)** on ARMv8 (RPI4 in **32-bit** mode): > quickly crashed when I enabled dynarec during opening cinematic: > > ``` > ERROR: No more registers! Abandon ship!...