From: Nicolai Hähnle Date: Mon, 1 Apr 2019 13:58:31 +0000 (+0200) Subject: radeonsi/gfx10: enable the workaround for unaligned vertex fetch X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=bcd2d2e1942ab7158dd46a5223130498cb0a8f44;p=mesa.git radeonsi/gfx10: enable the workaround for unaligned vertex fetch Yes, really. Note that non-format buffer loads are unaffected and work just fine with unaligned pointers (as long as SH_MEM_CONFIG is setup correctly, which amdgpu ensures). Fixes e.g. KHR-GL45.vertex_attrib_64bit.vao Acked-by: Bas Nieuwenhuizen --- diff --git a/src/gallium/drivers/radeonsi/si_state.c b/src/gallium/drivers/radeonsi/si_state.c index 35167f31fb1..68ba7d6403e 100644 --- a/src/gallium/drivers/radeonsi/si_state.c +++ b/src/gallium/drivers/radeonsi/si_state.c @@ -4970,7 +4970,9 @@ static void *si_create_vertex_elements(struct pipe_context *ctx, * into account would complicate the fast path (where everything * is nicely aligned). */ - bool check_alignment = log_hw_load_size >= 1 && sscreen->info.chip_class == GFX6; + bool check_alignment = + log_hw_load_size >= 1 && + (sscreen->info.chip_class == GFX6 || sscreen->info.chip_class == GFX10); bool opencode = sscreen->options.vs_fetch_always_opencode; if (check_alignment &&