From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 5E6A07461 for ; Sun, 17 Sep 2023 20:17:52 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 59B02C433C7; Sun, 17 Sep 2023 20:17:52 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1694981872; bh=xvfI+eUOuRPJT8MJpoF94F7Wbo/9GCEio1Hd9nQXWKo=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=S0POz85UZxLT4srUbleYXuEFaJRbS+3z6QP1ymiurqwzZOGCU266Qv9v70pYjjX5a WLmMNAq63eFdHsy54r4ueI+f8ylnwTzwyjgAiqqF7sqeE9ipY1yXv/E9yFCaIMwMA4 UjtIhwS1HFAp4YAZwhcnoW9McVpFldtiqb2/17B8= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Jun Lei , Hamza Mahfooz , Gabe Teeger , Alex Deucher Subject: [PATCH 6.1 183/219] drm/amd/display: Remove wait while locked Date: Sun, 17 Sep 2023 21:15:10 +0200 Message-ID: <20230917191047.564563937@linuxfoundation.org> X-Mailer: git-send-email 2.42.0 In-Reply-To: <20230917191040.964416434@linuxfoundation.org> References: <20230917191040.964416434@linuxfoundation.org> User-Agent: quilt/0.67 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 6.1-stable review patch. If anyone has any objections, please let me know. ------------------ From: Gabe Teeger commit 5a3ccb1400339268c5e3dc1fa044a7f6c7f59a02 upstream. [Why] We wait for mpc idle while in a locked state, leading to potential deadlock. [What] Move the wait_for_idle call to outside of HW lock. This and a call to wait_drr_doublebuffer_pending_clear are moved added to a new static helper function called wait_for_outstanding_hw_updates, to make the interface clearer. Cc: stable@vger.kernel.org Fixes: 8f0d304d21b3 ("drm/amd/display: Do not commit pipe when updating DRR") Reviewed-by: Jun Lei Acked-by: Hamza Mahfooz Signed-off-by: Gabe Teeger Signed-off-by: Alex Deucher Signed-off-by: Greg Kroah-Hartman --- drivers/gpu/drm/amd/display/dc/Makefile | 1 drivers/gpu/drm/amd/display/dc/core/dc.c | 58 ++++++++++++++------- drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c | 11 --- 3 files changed, 42 insertions(+), 28 deletions(-) --- a/drivers/gpu/drm/amd/display/dc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/Makefile @@ -82,3 +82,4 @@ DC_EDID += dc_edid_parser.o AMD_DISPLAY_DMUB = $(addprefix $(AMDDALPATH)/dc/,$(DC_DMUB)) AMD_DISPLAY_EDID = $(addprefix $(AMDDALPATH)/dc/,$(DC_EDID)) AMD_DISPLAY_FILES += $(AMD_DISPLAY_DMUB) $(AMD_DISPLAY_EDID) + --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -3361,6 +3361,45 @@ void dc_dmub_update_dirty_rect(struct dc } } +static void wait_for_outstanding_hw_updates(struct dc *dc, const struct dc_state *dc_context) +{ +/* + * This function calls HWSS to wait for any potentially double buffered + * operations to complete. It should be invoked as a pre-amble prior + * to full update programming before asserting any HW locks. + */ + int pipe_idx; + int opp_inst; + int opp_count = dc->res_pool->pipe_count; + struct hubp *hubp; + int mpcc_inst; + const struct pipe_ctx *pipe_ctx; + + for (pipe_idx = 0; pipe_idx < dc->res_pool->pipe_count; pipe_idx++) { + pipe_ctx = &dc_context->res_ctx.pipe_ctx[pipe_idx]; + + if (!pipe_ctx->stream) + continue; + + if (pipe_ctx->stream_res.tg->funcs->wait_drr_doublebuffer_pending_clear) + pipe_ctx->stream_res.tg->funcs->wait_drr_doublebuffer_pending_clear(pipe_ctx->stream_res.tg); + + hubp = pipe_ctx->plane_res.hubp; + if (!hubp) + continue; + + mpcc_inst = hubp->inst; + // MPCC inst is equal to pipe index in practice + for (opp_inst = 0; opp_inst < opp_count; opp_inst++) { + if (dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst]) { + dc->res_pool->mpc->funcs->wait_for_idle(dc->res_pool->mpc, mpcc_inst); + dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst] = false; + break; + } + } + } +} + static void commit_planes_for_stream(struct dc *dc, struct dc_surface_update *srf_updates, int surface_count, @@ -3378,24 +3417,9 @@ static void commit_planes_for_stream(str // dc->current_state anymore, so we have to cache it before we apply // the new SubVP context subvp_prev_use = false; - - dc_z10_restore(dc); - - if (update_type == UPDATE_TYPE_FULL) { - /* wait for all double-buffer activity to clear on all pipes */ - int pipe_idx; - - for (pipe_idx = 0; pipe_idx < dc->res_pool->pipe_count; pipe_idx++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[pipe_idx]; - - if (!pipe_ctx->stream) - continue; - - if (pipe_ctx->stream_res.tg->funcs->wait_drr_doublebuffer_pending_clear) - pipe_ctx->stream_res.tg->funcs->wait_drr_doublebuffer_pending_clear(pipe_ctx->stream_res.tg); - } - } + if (update_type == UPDATE_TYPE_FULL) + wait_for_outstanding_hw_updates(dc, context); if (get_seamless_boot_stream_count(context) > 0 && surface_count > 0) { /* Optimize seamless boot flag keeps clocks and watermarks high until --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -1515,17 +1515,6 @@ static void dcn20_update_dchubp_dpp( || plane_state->update_flags.bits.global_alpha_change || plane_state->update_flags.bits.per_pixel_alpha_change) { // MPCC inst is equal to pipe index in practice - int mpcc_inst = hubp->inst; - int opp_inst; - int opp_count = dc->res_pool->pipe_count; - - for (opp_inst = 0; opp_inst < opp_count; opp_inst++) { - if (dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst]) { - dc->res_pool->mpc->funcs->wait_for_idle(dc->res_pool->mpc, mpcc_inst); - dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst] = false; - break; - } - } hws->funcs.update_mpcc(dc, pipe_ctx); }