From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by gabe.freedesktop.org (Postfix) with ESMTPS id B2EF210EA5C for ; Fri, 20 Jan 2023 07:35:12 +0000 (UTC) Message-ID: Date: Fri, 20 Jan 2023 08:35:08 +0100 Content-Language: en-US To: =?UTF-8?Q?Zbigniew_Kempczy=c5=84ski?= References: <20230119142929.127976-1-zbigniew.kempczynski@intel.com> From: Karolina Stolarek In-Reply-To: <20230119142929.127976-1-zbigniew.kempczynski@intel.com> Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 8bit MIME-Version: 1.0 Subject: Re: [igt-dev] [PATCH i-g-t] i915/gem_ccs: Use extended block-copy on MTL List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: igt-dev@lists.freedesktop.org Errors-To: igt-dev-bounces@lists.freedesktop.org Sender: "igt-dev" List-ID: On 19.01.2023 15:29, Zbigniew Kempczyński wrote: > Integrated platforms and DG1 (exception in discrete, doesn't have > flatccs) use shorter version of block-copy blit (12 dwords length, > extended command use 22 dwords). MTL is first integrated which uses > extended block-copy instruction. > > At the moment change enables using block-copy for uncompressed blits > on MTL only due to fact aux-ccs support for blitter is still unclear > for me. We haven't before code which uses aux-ccs for blitter so > enablement of this might occur in future changes. > > Signed-off-by: Zbigniew Kempczyński > Cc: Kamil Konieczny > --- > tests/i915/gem_ccs.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/tests/i915/gem_ccs.c b/tests/i915/gem_ccs.c > index ff28c6d889..6ae5ad85b8 100644 > --- a/tests/i915/gem_ccs.c > +++ b/tests/i915/gem_ccs.c > @@ -443,6 +443,7 @@ static void block_copy(int i915, > uint32_t run_id = mid_tiling; > uint32_t mid_region = region2, bb; > uint32_t width = param.width, height = param.height; > + uint32_t devid = intel_get_drm_devid(i915); > enum blt_compression mid_compression = config->compression; > int mid_compression_format = param.compression_format; > enum blt_compression_type comp_type = COMPRESSION_TYPE_3D; > @@ -451,7 +452,7 @@ static void block_copy(int i915, > > igt_assert(__gem_create_in_memory_regions(i915, &bb, &bb_size, region1) == 0); > > - if (!blt_supports_compression(i915)) > + if (!blt_supports_compression(i915) && !IS_METEORLAKE(devid)) This is fine as a temporary change -- I think we'll be able to handle this case with blt_tiling_info patches, but for now: Reviewed-by: Karolina Stolarek Thanks, Karolina > pext = NULL; > > src = create_object(i915, region1, width, height, bpp, uc_mocs,