From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 037CDC10F16 for ; Tue, 30 Apr 2024 12:02:29 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 84CA310E8F7; Tue, 30 Apr 2024 12:02:29 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="fipyfLRI"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [198.175.65.16]) by gabe.freedesktop.org (Postfix) with ESMTPS id 0A25210F4C5 for ; Tue, 30 Apr 2024 12:02:27 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1714478548; x=1746014548; h=from:to:cc:subject:date:message-id:mime-version: content-transfer-encoding; bh=stBnyHrPISNhZXjKqJnlvY0LS+aesrmFYq+32BCXVaM=; b=fipyfLRIxAN4Nj0PtWDCjvNG6+sopJA/9qKazZ2BuaPtFaxJveJEbqeA nvfN1vylL5Ywr2N3iccy3Iimm4UBC9Mzyon3aczIjgnaVsW0Ep/kPlNCn TZLYC/4Awc0g4IDUMUdX7lUBddKJ1fz1rd+saIhDFzC6RIJ/1p3Pqje7b 5PElYyS2nONfzSxspEEQfjUSMKceV9JXp876QaWX7gDjITmpIztsBl1F7 A5WPBmvhYw4oXJLTNyAyPmhennBc54TByTay7tXn2F0ewtGtYYv19pzM4 +s1abtPtBdFdT5HyqRwoQJnadSgN6rGoghBACcpkIEIIEgzDmPDGFWWeF Q==; X-CSE-ConnectionGUID: ibEYAJ4pSGanW1I5Ag8OQw== X-CSE-MsgGUID: qFMmHfUzQ3GY8yU6pYKbCQ== X-IronPort-AV: E=McAfee;i="6600,9927,11059"; a="10302960" X-IronPort-AV: E=Sophos;i="6.07,242,1708416000"; d="scan'208";a="10302960" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by orvoesa108.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Apr 2024 05:02:27 -0700 X-CSE-ConnectionGUID: pmw1w57/QmawqxacEcF3Iw== X-CSE-MsgGUID: Z5vKJEOnTwubcg2m+XZT2A== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.07,242,1708416000"; d="scan'208";a="57620308" Received: from isokolov-mobl.ger.corp.intel.com (HELO fedora..) ([10.249.254.37]) by smtpauth.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Apr 2024 05:02:27 -0700 From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= To: intel-xe@lists.freedesktop.org Cc: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= Subject: [PATCH] drm/xe: Perform dma_map when moving system buffer objects to TT Date: Tue, 30 Apr 2024 14:02:14 +0200 Message-ID: <20240430120214.5372-1-thomas.hellstrom@linux.intel.com> X-Mailer: git-send-email 2.44.0 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-BeenThere: intel-xe@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel Xe graphics driver List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-xe-bounces@lists.freedesktop.org Sender: "Intel-xe" Currently we dma_map on ttm_tt population and dma_unmap when the pages are released in ttm_tt unpopulate. Strictly, the dma_map is not needed until the bo is moved to the XE_PL_TT placement, so perform the dma_mapping on such moves instead, and remove the dma_mappig when moving to XE_PL_SYSTEM. This is desired for the upcoming shrinker series where shrinking of a ttm_tt might fail. That would lead to an odd construct where we first dma_unmap, then shrink and if shrinking fails dma_map again. If dma_mapping instead is performed on move like this, shrinking does not need to care at all about dma mapping. Finally, where a ttm_tt is destroyed while bound to a different memory type than XE_PL_SYSTEM, we keep the dma_unmap in unpopulate(). Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_bo.c | 45 ++++++++++++++++++++++++-------------- 1 file changed, 29 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index bc1f794e3e61..4c1dd67a4588 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -302,6 +302,18 @@ static int xe_tt_map_sg(struct ttm_tt *tt) return 0; } +static void xe_tt_unmap_sg(struct ttm_tt *tt) +{ + struct xe_ttm_tt *xe_tt = container_of(tt, struct xe_ttm_tt, ttm); + + if (xe_tt->sg) { + dma_unmap_sgtable(xe_tt->dev, xe_tt->sg, + DMA_BIDIRECTIONAL, 0); + sg_free_table(xe_tt->sg); + xe_tt->sg = NULL; + } +} + struct sg_table *xe_bo_sg(struct xe_bo *bo) { struct ttm_tt *tt = bo->ttm.ttm; @@ -377,27 +389,15 @@ static int xe_ttm_tt_populate(struct ttm_device *ttm_dev, struct ttm_tt *tt, if (err) return err; - /* A follow up may move this xe_bo_move when BO is moved to XE_PL_TT */ - err = xe_tt_map_sg(tt); - if (err) - ttm_pool_free(&ttm_dev->pool, tt); - return err; } static void xe_ttm_tt_unpopulate(struct ttm_device *ttm_dev, struct ttm_tt *tt) { - struct xe_ttm_tt *xe_tt = container_of(tt, struct xe_ttm_tt, ttm); - if (tt->page_flags & TTM_TT_FLAG_EXTERNAL) return; - if (xe_tt->sg) { - dma_unmap_sgtable(xe_tt->dev, xe_tt->sg, - DMA_BIDIRECTIONAL, 0); - sg_free_table(xe_tt->sg); - xe_tt->sg = NULL; - } + xe_tt_unmap_sg(tt); return ttm_pool_free(&ttm_dev->pool, tt); } @@ -628,10 +628,14 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, bool handle_system_ccs = (!IS_DGFX(xe) && xe_bo_needs_ccs_pages(bo) && ttm && ttm_tt_is_populated(ttm)) ? true : false; int ret = 0; + /* Bo creation path, moving to system or TT. */ if ((!old_mem && ttm) && !handle_system_ccs) { - ttm_bo_move_null(ttm_bo, new_mem); - return 0; + if (new_mem->mem_type == XE_PL_TT) + ret = xe_tt_map_sg(ttm); + if (!ret) + ttm_bo_move_null(ttm_bo, new_mem); + goto out; } if (ttm_bo->type == ttm_bo_type_sg) { @@ -650,6 +654,12 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, needs_clear = (ttm && ttm->page_flags & TTM_TT_FLAG_ZERO_ALLOC) || (!ttm && ttm_bo->type == ttm_bo_type_device); + if (new_mem->mem_type == XE_PL_TT) { + ret = xe_tt_map_sg(ttm); + if (ret) + goto out; + } + if ((move_lacks_source && !needs_clear)) { ttm_bo_move_null(ttm_bo, new_mem); goto out; @@ -786,8 +796,11 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, xe_pm_runtime_put(xe); out: - return ret; + if ((!ttm_bo->resource || ttm_bo->resource->mem_type == XE_PL_SYSTEM) && + ttm_bo->ttm) + xe_tt_unmap_sg(ttm_bo->ttm); + return ret; } /** -- 2.44.0