From patchwork Mon Apr 17 10:38:52 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: "Fabio M. De Francesco" X-Patchwork-Id: 13213680 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 782DAC77B77 for ; Mon, 17 Apr 2023 10:39:10 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 2D74210E3F8; Mon, 17 Apr 2023 10:39:06 +0000 (UTC) Received: from mail-wr1-x42b.google.com (mail-wr1-x42b.google.com [IPv6:2a00:1450:4864:20::42b]) by gabe.freedesktop.org (Postfix) with ESMTPS id 820FF10E3F9; Mon, 17 Apr 2023 10:39:04 +0000 (UTC) Received: by mail-wr1-x42b.google.com with SMTP id ffacd0b85a97d-2f46348728eso827551f8f.3; Mon, 17 Apr 2023 03:39:04 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20221208; t=1681727942; x=1684319942; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=mkh5X1oOnHmMtPThd/Ogb+Ei0/Iy+LdjudYL5nRHiA8=; b=XO/vfE0ttxJFtb+z+zDWyoXcM09x8Zs78N3GfyVo9gwDIuuHa3a1Gya8goc4sNmMJV vJR78/+//6TCZvI30CK/mDXRtQDDAPsnAizkwB7SubAmmPETo39N1bcxXg1o3cKdtQmJ iCoD6j4lKCLRHYwJI1MFIYJlEvuRAjgVkoOTZ/uX15pf4L+VkS92fyBzWSrLwGYY50At OR7OTpSJgtuz6ZDgg4F9K6kpnJv0D1ptjteBC4ZP3XgKXT3gyKF4/aYye3QW/JCfu8DK jN3oOf3Qkl4Z3pTExiIhOq34ri6WOzWaj3YJ2LKW9OBRVj7+1BrJHbhm4NXNHFDfBeZ8 h7ew== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1681727942; x=1684319942; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=mkh5X1oOnHmMtPThd/Ogb+Ei0/Iy+LdjudYL5nRHiA8=; b=UTzldRWNJPKX1IRSQubx2Td5gEocn6yGLuPiUG1B6yBNRQO7CFwGDc4uZ1mO61lLSb KN3E4YMuBQfv/KLwxpYtd1fJzvlWqAhLSMMuGyb1UQ+XevEgCS0knUyYT+hFsZoFP+MN elXyxj0gm7sRo2yp0h73nI6/um1iIuXZQYulgYVoAFKwJhVpnsu/zJWhIqNjEDuZ9GNg 0sDPU5ZoSS9QvOpSkeXYq0DRaTpOAq+n2S+60uEHOe+abiKRhI1Qfk4SUALmV0xnD+Yd y3JIZvyqQge6p8yPkXLV3XwUEuCGSGmiBqMp1/tiDhI4bRlDa3yj9pxSXpw0GHl9MB2L tSxw== X-Gm-Message-State: AAQBX9c++PWXppNq5unuRiOQLT0MxdiWuzJyDvPHYKN1FM0Z+8ai8bGs RM0ljD5ZLsXWFqGoovHBpR0= X-Google-Smtp-Source: AKy350ZWcTYKgg1QBjpNgdR9wjl3y2FJYThqP4aJuL8PsgT/C8+4jSADpdmdsMu3nAzGfM6xeTh5vg== X-Received: by 2002:a5d:6606:0:b0:2f8:dd25:f96e with SMTP id n6-20020a5d6606000000b002f8dd25f96emr4345092wru.6.1681727942023; Mon, 17 Apr 2023 03:39:02 -0700 (PDT) Received: from localhost.localdomain (host-87-19-108-254.retail.telecomitalia.it. [87.19.108.254]) by smtp.gmail.com with ESMTPSA id y12-20020a5d470c000000b002c70ce264bfsm10238018wrq.76.2023.04.17.03.39.00 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 17 Apr 2023 03:39:01 -0700 (PDT) From: "Fabio M. De Francesco" To: Jani Nikula , Joonas Lahtinen , Rodrigo Vivi , Tvrtko Ursulin , David Airlie , Daniel Vetter , Matt Roper , John Harrison , Daniele Ceraolo Spurio , Chris Wilson , intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org Date: Mon, 17 Apr 2023 12:38:52 +0200 Message-Id: <20230417103854.23333-2-fmdefrancesco@gmail.com> X-Mailer: git-send-email 2.40.0 In-Reply-To: <20230417103854.23333-1-fmdefrancesco@gmail.com> References: <20230417103854.23333-1-fmdefrancesco@gmail.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH v2 1/3] drm/i915: Replace kmap() with kmap_local_page() X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Ira Weiny , "Fabio M. De Francesco" Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" kmap() has been deprecated in favor of kmap_local_page(). There are two main problems with kmap(): (1) It comes with an overhead as mapping space is restricted and protected by a global lock for synchronization and (2) it also requires global TLB invalidation when the kmap’s pool wraps and it might block when the mapping space is fully utilized until a slot becomes available. With kmap_local_page() the mappings are per thread, CPU local, can take page faults, and can be called from any context (including interrupts). It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore, the tasks can be preempted and, when they are scheduled to run again, the kernel virtual addresses are restored and are still valid. Obviously, thread locality implies that the kernel voirtual addresses are valid only in the context of the callers. kmap_local_page() use in i915_gem.c doesn't break the above-mentioned constraint, so it should be preferred to kmap(). Therefore, replace kmap() with kmap_local_page() in i915_gem.c Suggested-by: Ira Weiny Signed-off-by: Fabio M. De Francesco --- drivers/gpu/drm/i915/i915_gem.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) -- 2.40.0 diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 35950fa91406..c35248555e42 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -212,14 +212,14 @@ shmem_pread(struct page *page, int offset, int len, char __user *user_data, char *vaddr; int ret; - vaddr = kmap(page); + vaddr = kmap_local_page(page); if (needs_clflush) drm_clflush_virt_range(vaddr + offset, len); ret = __copy_to_user(user_data, vaddr + offset, len); - kunmap(page); + kunmap_local(vaddr); return ret ? -EFAULT : 0; } @@ -643,7 +643,7 @@ shmem_pwrite(struct page *page, int offset, int len, char __user *user_data, char *vaddr; int ret; - vaddr = kmap(page); + vaddr = kmap_local_page(page); if (needs_clflush_before) drm_clflush_virt_range(vaddr + offset, len); @@ -652,7 +652,7 @@ shmem_pwrite(struct page *page, int offset, int len, char __user *user_data, if (!ret && needs_clflush_after) drm_clflush_virt_range(vaddr + offset, len); - kunmap(page); + kunmap_local(vaddr); return ret ? -EFAULT : 0; } From patchwork Mon Apr 17 10:38:53 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: "Fabio M. De Francesco" X-Patchwork-Id: 13213682 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 17896C77B70 for ; Mon, 17 Apr 2023 10:39:17 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 21BDB10E401; Mon, 17 Apr 2023 10:39:11 +0000 (UTC) Received: from mail-wm1-x330.google.com (mail-wm1-x330.google.com [IPv6:2a00:1450:4864:20::330]) by gabe.freedesktop.org (Postfix) with ESMTPS id 9D9F410E3F9; Mon, 17 Apr 2023 10:39:06 +0000 (UTC) Received: by mail-wm1-x330.google.com with SMTP id o9-20020a05600c510900b003f17012276fso2000813wms.4; Mon, 17 Apr 2023 03:39:06 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20221208; t=1681727944; x=1684319944; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=dAW7sB6+Z5ZOltjkFBMt2NbhGjStc/CgCoci8oN9nks=; b=lk90wuKk7kR9IX6uMik68yOAyW5/zZmQWcphsp7dCp9z3gtp/owgQL9h2aAbLKBZl4 fCT8G5zvkCrwE8UQwr3TI2qV7TEBSeuZSe9ChabtOXCucPLOySit4w+Ukcnt1MR6xcJD RBcU0cLyIawG1B8WldS9XwJKzvAn1uvF3XJQUQD4rxY6QsnOGA8yokADx7cMlcSkaxFs 5+cWcTxbRQ//yFlyW01nwMQQHtlmblF6pwjWClW/urNU4TC+6JZnbE5L8Bg4kfHflzxL /4UrOEAQM7KKog1/7dO7jGRKmFaB5tY3maCKiDRQFq5G039Z4m4zDMRjCuuJMMI16nKs eHFQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1681727944; x=1684319944; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=dAW7sB6+Z5ZOltjkFBMt2NbhGjStc/CgCoci8oN9nks=; b=LEaif8lI7lzBo4o5EVUxTFKfCplzJ081X2qKqr9/eNXBF5rpSyPAsE665KKsEqrUAX yYP+wW0pXkRzQfan8+z+nVvjWb7A67nJDeELWR/s8P09CuTWQn7w0R+I2/dMzSn1muwy xyUFBTFDwVP0cHAURN9mTsqJCryMFdoKJEjkVKv5267QmSqosfXPkrGWzU2iWDBO9ODj k8LUkCftrOObPdCMlLfv8ctQapIE3egoROZlPvFc3T9xcpJaY6K2VrDHZY2ImxJ8txCA +pFm4MzLg2Gug1jS3ts1Phe3+eY56MER6BsqzOZpq4FMpz1/UpBT7Dx/mAQcvesN4V6m FmfA== X-Gm-Message-State: AAQBX9f3J0PoJdKcob2wekRx1kKOxVn9SzmilZomRueXYWxGNpvo8q3D kksYf+EjctnmrhRo/ZzQhOE= X-Google-Smtp-Source: AKy350bqJ/bHsDjtLWvuVfbc6Uq/QInCoRKCm+KLyl/hHEccIaE3q+1++wO1znNLhcQOGVkqIM5BVw== X-Received: by 2002:a05:600c:4f46:b0:3f1:7443:6d6e with SMTP id m6-20020a05600c4f4600b003f174436d6emr1590807wmq.19.1681727943960; Mon, 17 Apr 2023 03:39:03 -0700 (PDT) Received: from localhost.localdomain (host-87-19-108-254.retail.telecomitalia.it. [87.19.108.254]) by smtp.gmail.com with ESMTPSA id y12-20020a5d470c000000b002c70ce264bfsm10238018wrq.76.2023.04.17.03.39.02 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 17 Apr 2023 03:39:03 -0700 (PDT) From: "Fabio M. De Francesco" To: Jani Nikula , Joonas Lahtinen , Rodrigo Vivi , Tvrtko Ursulin , David Airlie , Daniel Vetter , Matt Roper , John Harrison , Daniele Ceraolo Spurio , Chris Wilson , intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org Date: Mon, 17 Apr 2023 12:38:53 +0200 Message-Id: <20230417103854.23333-3-fmdefrancesco@gmail.com> X-Mailer: git-send-email 2.40.0 In-Reply-To: <20230417103854.23333-1-fmdefrancesco@gmail.com> References: <20230417103854.23333-1-fmdefrancesco@gmail.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH v2 2/3] drm/i915/gt: Replace kmap() with kmap_local_page() X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Ira Weiny , "Fabio M. De Francesco" Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" kmap() s been deprecated in favor of kmap_local_page(). There are two main problems with kmap(): (1) It comes with an overhead as mapping space is restricted and protected by a global lock for synchronization and (2) it also requires global TLB invalidation when the kmap’s pool wraps and it might block when the mapping space is fully utilized until a slot becomes available. With kmap_local_page() the mappings are per thread, CPU local, can take page faults, and can be called from any context (including interrupts). It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore, the tasks can be preempted and, when they are scheduled to run again, the kernel virtual addresses are restored and are still valid. Obviously, thread locality implies that the kernel virtual addresses are only valid in the context of the callers. The use of kmap_local_page() in i915/gt doesn't break the above-mentioned constraint, so it should be preferred to kmap(). Therefore, replace kmap() with kmap_local_page() in i915/gt. Suggested-by: Ira Weiny Signed-off-by: Fabio M. De Francesco --- drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c | 4 ++-- drivers/gpu/drm/i915/gt/shmem_utils.c | 11 ++++------- 2 files changed, 6 insertions(+), 9 deletions(-) -- 2.40.0 diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c b/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c index 37d0b0fe791d..89295c6921d6 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c @@ -749,7 +749,7 @@ static void swizzle_page(struct page *page) char *vaddr; int i; - vaddr = kmap(page); + vaddr = kmap_local_page(page); for (i = 0; i < PAGE_SIZE; i += 128) { memcpy(temp, &vaddr[i], 64); @@ -757,7 +757,7 @@ static void swizzle_page(struct page *page) memcpy(&vaddr[i + 64], temp, 64); } - kunmap(page); + kunmap_local(vaddr); } /** diff --git a/drivers/gpu/drm/i915/gt/shmem_utils.c b/drivers/gpu/drm/i915/gt/shmem_utils.c index 449c9ed44382..be809839a241 100644 --- a/drivers/gpu/drm/i915/gt/shmem_utils.c +++ b/drivers/gpu/drm/i915/gt/shmem_utils.c @@ -101,22 +101,19 @@ static int __shmem_rw(struct file *file, loff_t off, unsigned int this = min_t(size_t, PAGE_SIZE - offset_in_page(off), len); struct page *page; - void *vaddr; page = shmem_read_mapping_page_gfp(file->f_mapping, pfn, GFP_KERNEL); if (IS_ERR(page)) return PTR_ERR(page); - vaddr = kmap(page); if (write) { - memcpy(vaddr + offset_in_page(off), ptr, this); + memcpy_to_page(page, offset_in_page(off), ptr, this); set_page_dirty(page); } else { - memcpy(ptr, vaddr + offset_in_page(off), this); + memcpy_from_page(ptr, page, offset_in_page(off), this); } mark_page_accessed(page); - kunmap(page); put_page(page); len -= this; @@ -143,11 +140,11 @@ int shmem_read_to_iosys_map(struct file *file, loff_t off, if (IS_ERR(page)) return PTR_ERR(page); - vaddr = kmap(page); + vaddr = kmap_local_page(page); iosys_map_memcpy_to(map, map_off, vaddr + offset_in_page(off), this); mark_page_accessed(page); - kunmap(page); + kunmap_local(vaddr); put_page(page); len -= this; From patchwork Mon Apr 17 10:38:54 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: "Fabio M. De Francesco" X-Patchwork-Id: 13213681 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 465E9C77B77 for ; Mon, 17 Apr 2023 10:39:16 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 0727910E400; Mon, 17 Apr 2023 10:39:11 +0000 (UTC) Received: from mail-wr1-x42c.google.com (mail-wr1-x42c.google.com [IPv6:2a00:1450:4864:20::42c]) by gabe.freedesktop.org (Postfix) with ESMTPS id 3083610E3F9; Mon, 17 Apr 2023 10:39:08 +0000 (UTC) Received: by mail-wr1-x42c.google.com with SMTP id ffacd0b85a97d-2eed43bfa4bso1361038f8f.2; Mon, 17 Apr 2023 03:39:07 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20221208; t=1681727946; x=1684319946; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=cH79Swn8ZRJ2/YarOj91Bz0xxg30p0nVIyjLTbfZHEk=; b=YRKjuLMhTl40C+QqFYNyd1bRihczrfYWx/yGU36IRsJfmSPwWm5zftNVBsPia6yv+S Zxqe6js3OrRVHvL9eVEwsHAVrpNhJPpvZcA3/CuoUjouJePIvyzvpsD+QqepRoAlloZ/ 0fsrKdoUG/TsMKLWUBTOHqggXyBAQ2C+BlCWcOm9TvmHiaIgIBq5KhCWjFeagOpEmB8d 1DAQ0WsJbDJw9QEjshxivpMLqq89xEmLpV7P0ecn/QKP2xODoAfG9PcVZbGxJetEzBPJ tD5YqlokKpW5QqC7IQ0a6OqDD1KDVY5I3WNXvmUYrxVtT80Kmg/URg+3ZpE9rLMH6zR0 2NkA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1681727946; x=1684319946; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=cH79Swn8ZRJ2/YarOj91Bz0xxg30p0nVIyjLTbfZHEk=; b=QU0y8E6hlLiFSuuUIcs0v6xoBcXukG8qpEa369wJTm8QkAi4m9bOoWorUfJQAzOYnV WnXkU8FgGZto7lfmibeeHar+Cpddz8DE6vzQah9XQf2fEEu/T8QVGE0xOz7svLG3eVDT rbX6yxj2j6vpQIZPd1v2eEO5KX9WMuDyiSr9helFE1JBuxMmMT+/4yTH2EFhBKHuxP4Y bHuzXlx1OHJqmMJU7JAmOyKgbX8kngRRzeXDXEW0F0DX62+afyDlE1mzIOmarpkipiC+ G2RcS5YnbSY5BqJUboM0vslvrEEfknTFcWDNQgSry2r8nXVCrCD2DmdCpyNHtM61XUo1 FJyQ== X-Gm-Message-State: AAQBX9degkGvY8UuQ50DnfEQQfqFJxPiUm1fTnm8uiG0JPHWHs4N13eP cUdf2lKXvEcn+AuvxciBQIM= X-Google-Smtp-Source: AKy350ZHLsYNFXN5GyVB3dvmvIOZLfVbmy328/EyKvOC+gnAZGSMikYnaDOJgYq7gQ7O8a2gX7ryOA== X-Received: by 2002:adf:d085:0:b0:2fa:88f2:b04c with SMTP id y5-20020adfd085000000b002fa88f2b04cmr1956315wrh.20.1681727945938; Mon, 17 Apr 2023 03:39:05 -0700 (PDT) Received: from localhost.localdomain (host-87-19-108-254.retail.telecomitalia.it. [87.19.108.254]) by smtp.gmail.com with ESMTPSA id y12-20020a5d470c000000b002c70ce264bfsm10238018wrq.76.2023.04.17.03.39.04 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 17 Apr 2023 03:39:05 -0700 (PDT) From: "Fabio M. De Francesco" To: Jani Nikula , Joonas Lahtinen , Rodrigo Vivi , Tvrtko Ursulin , David Airlie , Daniel Vetter , Matt Roper , John Harrison , Daniele Ceraolo Spurio , Chris Wilson , intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org Date: Mon, 17 Apr 2023 12:38:54 +0200 Message-Id: <20230417103854.23333-4-fmdefrancesco@gmail.com> X-Mailer: git-send-email 2.40.0 In-Reply-To: <20230417103854.23333-1-fmdefrancesco@gmail.com> References: <20230417103854.23333-1-fmdefrancesco@gmail.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH v2 3/3] drm/i915/gem: Replace kmap() with kmap_local_page() X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Ira Weiny , "Fabio M. De Francesco" Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" kmap() s been deprecated in favor of kmap_local_page(). There are two main problems with kmap(): (1) It comes with an overhead as mapping space is restricted and protected by a global lock for synchronization and (2) it also requires global TLB invalidation when the kmap’s pool wraps and it might block when the mapping space is fully utilized until a slot becomes available. With kmap_local_page() the mappings are per thread, CPU local, can take page faults, and can be called from any context (including interrupts). It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore, the tasks can be preempted and, when they are scheduled to run again, the kernel virtual addresses are restored and are still valid. Obviously, thread locality implies that the kernel virtual addresses are only valid in the context of the callers. The kmap_local_page() use in i915/gem doesn't break the above-mentioned constraint, so it should be preferred to kmap(). Therefore, replace kmap() with kmap_local_page() in i915/gem and use memcpy_to_page() where it is possible to avoid the open coding of mapping + memcpy() + un-mapping. Suggested-by: Ira Weiny Signed-off-by: Fabio M. De Francesco --- drivers/gpu/drm/i915/gem/i915_gem_shmem.c | 6 ++---- drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c | 8 ++++---- 2 files changed, 6 insertions(+), 8 deletions(-) -- 2.40.0 diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c index 37d1efcd3ca6..8856a6409e83 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c @@ -657,16 +657,14 @@ i915_gem_object_create_shmem_from_data(struct drm_i915_private *dev_priv, do { unsigned int len = min_t(typeof(size), size, PAGE_SIZE); struct page *page; - void *pgdata, *vaddr; + void *pgdata; err = aops->write_begin(file, file->f_mapping, offset, len, &page, &pgdata); if (err < 0) goto fail; - vaddr = kmap(page); - memcpy(vaddr, data, len); - kunmap(page); + memcpy_to_page(page, 0, data, len); err = aops->write_end(file, file->f_mapping, offset, len, len, page, pgdata); diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c index 56279908ed30..5fd9e1ee2340 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c @@ -155,7 +155,7 @@ static int check_partial_mapping(struct drm_i915_gem_object *obj, intel_gt_flush_ggtt_writes(to_gt(i915)); p = i915_gem_object_get_page(obj, offset >> PAGE_SHIFT); - cpu = kmap(p) + offset_in_page(offset); + cpu = kmap_local_page(p) + offset_in_page(offset); drm_clflush_virt_range(cpu, sizeof(*cpu)); if (*cpu != (u32)page) { pr_err("Partial view for %lu [%u] (offset=%llu, size=%u [%llu, row size %u], fence=%d, tiling=%d, stride=%d) misalignment, expected write to page (%lu + %u [0x%lx]) of 0x%x, found 0x%x\n", @@ -173,7 +173,7 @@ static int check_partial_mapping(struct drm_i915_gem_object *obj, } *cpu = 0; drm_clflush_virt_range(cpu, sizeof(*cpu)); - kunmap(p); + kunmap_local(cpu); out: i915_gem_object_lock(obj, NULL); @@ -251,7 +251,7 @@ static int check_partial_mappings(struct drm_i915_gem_object *obj, intel_gt_flush_ggtt_writes(to_gt(i915)); p = i915_gem_object_get_page(obj, offset >> PAGE_SHIFT); - cpu = kmap(p) + offset_in_page(offset); + cpu = kmap_local_page(p) + offset_in_page(offset); drm_clflush_virt_range(cpu, sizeof(*cpu)); if (*cpu != (u32)page) { pr_err("Partial view for %lu [%u] (offset=%llu, size=%u [%llu, row size %u], fence=%d, tiling=%d, stride=%d) misalignment, expected write to page (%lu + %u [0x%lx]) of 0x%x, found 0x%x\n", @@ -269,7 +269,7 @@ static int check_partial_mappings(struct drm_i915_gem_object *obj, } *cpu = 0; drm_clflush_virt_range(cpu, sizeof(*cpu)); - kunmap(p); + kunmap_local(cpu); if (err) return err;