From: Marek Olšák Date: Tue, 19 May 2020 22:31:55 +0000 (-0400) Subject: Revert "radeonsi: don't wait for idle at the end of gfx IBs" X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=56af131f330f06bf6075681c528fe70f221fbbe2;p=mesa.git Revert "radeonsi: don't wait for idle at the end of gfx IBs" This reverts commit 266fec1307b26a544007423582afd8618791893c. The kernel doesn't wait for idle as part of implicit sync. Fixes: 266fec1307b26a544007423582afd8618791893c Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/2950 Reviewed-by: Bas Nieuwenhuizen Reviewed-by: Pierre-Eric Pelloux-Prayer Part-of: --- diff --git a/src/gallium/drivers/radeonsi/si_gfx_cs.c b/src/gallium/drivers/radeonsi/si_gfx_cs.c index d6baf1ec9d4..3efa1731ee8 100644 --- a/src/gallium/drivers/radeonsi/si_gfx_cs.c +++ b/src/gallium/drivers/radeonsi/si_gfx_cs.c @@ -78,19 +78,6 @@ void si_flush_gfx_cs(struct si_context *ctx, unsigned flags, struct pipe_fence_h if (ctx->gfx_flush_in_progress) return; - /* The amdgpu kernel driver always synchronizes execution for shared DMABUFs - * between processes, so we don't have to wait at the end of IBs to make sure - * everything is idle. - * - * The amdgpu winsys synchronizes execution for buffers shared by different - * contexts within the same process. - * - * Interop with AMDVLK, RADV, or OpenCL within the same process requires - * explicit fences or glFinish. - */ - if (ctx->screen->info.is_amdgpu) - flags |= RADEON_FLUSH_START_NEXT_GFX_IB_NOW; - if (!ctx->screen->info.kernel_flushes_tc_l2_after_ib) { wait_flags |= wait_ps_cs | SI_CONTEXT_INV_L2; } else if (ctx->chip_class == GFX6) {