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 X-Spam-Level: X-Spam-Status: No, score=-9.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FORGED_FROMDOMAIN,FREEMAIL_FROM, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY, SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id CBFC6C4727D for ; Fri, 25 Sep 2020 14:09:29 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 93EEC20BED for ; Fri, 25 Sep 2020 14:09:29 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="VAE/JCw4" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729044AbgIYOJ2 (ORCPT ); Fri, 25 Sep 2020 10:09:28 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:50588 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728466AbgIYOJ1 (ORCPT ); Fri, 25 Sep 2020 10:09:27 -0400 Received: from mail-ua1-x944.google.com (mail-ua1-x944.google.com [IPv6:2607:f8b0:4864:20::944]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 36250C0613CE for ; Fri, 25 Sep 2020 07:09:27 -0700 (PDT) Received: by mail-ua1-x944.google.com with SMTP id z1so973765uaa.6 for ; Fri, 25 Sep 2020 07:09:27 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=W4ybNDBj3JkG7xwtC+pQm/GVM9Ki73UbHYGcb1DNgoE=; b=VAE/JCw4Kc2YKn1yJ8L/Q9+LADI9BPVrPzMbIS0mOQDh1e+IitPGEnQLZfakaeotSX fJpnesPgh6fudFH4hBFOTUzOvja6IHah1EG+sAlo5mDeRe30T7WE5Z+EmmrqyLBVjPT2 Y6MypsxJU5ecTFL5KTTnudzV3iHFKzBhkicIF7nAwXtRPFhwdOi9g9/Y/qXRPNUiOlZ/ S833Agy78MEYaybNf9Tk+M+dZyKRMHoBkxs25nGIJGGXWhSky07MsU9xI+7dnENL4I7l bH2CDxH5hj1vhP5g8v/h/6Mm80YH+iZTF4wuYNk0h7j6DAwbRzp57D1NpHpN8KHRNu/G TlSA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=W4ybNDBj3JkG7xwtC+pQm/GVM9Ki73UbHYGcb1DNgoE=; b=RpzCW7SLiTW4PGoB60gmjMv4BXz+kKYt+gmDkZ5nBB4LqBn2ikg9oc0Hu7vTNxmT1F kwVnrH4Hu7JwTv5bRNQ+H8S7JD45EURS5XuYvqBe5yYZiXEmtGeaqt9K635zOkYXv/Hu ww2ibWqmvZTbatB2l0Q6WNeiK7Qpy6pbCLKG6SYXZ1m2kQtz99yN6174MQodna8BLCEh R3iJxrM4TBVLn1oBtg14FJYjK0mICMNJ/OY396RT8kGKm1SbLihuTMCzzICXS87GkMts 605qXGuTp73OhiTSPJdwkewUwTM9pmau8qfMOyZu3nAGezScJ2S1vnEAABXUM7PYTb9N rm/g== X-Gm-Message-State: AOAM532nnVBMHS7rhhPqEBx0QW9+RddUedCiK3pd6mgrv0d0mEeLEYJk /LqSOi8TAyDaQ23ottcKqzfUpLWIrtdnN/PCUJY= X-Google-Smtp-Source: ABdhPJy0IxUr4anWia0Qm7ztGxjztU5H4LAKNIvZHwQZc9rZ+mSfXU53xRXTbzCYoiyClMt8qe//fl65KD/2aNPSkzU= X-Received: by 2002:ab0:6f91:: with SMTP id f17mr2536965uav.129.1601042966264; Fri, 25 Sep 2020 07:09:26 -0700 (PDT) MIME-Version: 1.0 References: <20200924135853.875294-1-hch@lst.de> <20200924135853.875294-9-hch@lst.de> In-Reply-To: <20200924135853.875294-9-hch@lst.de> From: Matthew Auld Date: Fri, 25 Sep 2020 15:08:59 +0100 Message-ID: Subject: Re: [Intel-gfx] [PATCH 08/11] drm/i915: use vmap in i915_gem_object_map To: Christoph Hellwig Cc: Andrew Morton , Juergen Gross , Stefano Stabellini , Matthew Wilcox , ML dri-devel , linux-mm@kvack.org, Peter Zijlstra , kernel list , Intel Graphics Development , x86@kernel.org, Chris Wilson , Minchan Kim , Matthew Auld , xen-devel@lists.xenproject.org, Boris Ostrovsky , Nitin Gupta Content-Type: text/plain; charset="UTF-8" Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 24 Sep 2020 at 14:59, Christoph Hellwig wrote: > > i915_gem_object_map implements fairly low-level vmap functionality in > a driver. Split it into two helpers, one for remapping kernel memory > which can use vmap, and one for I/O memory that uses vmap_pfn. > > The only practical difference is that alloc_vm_area prefeaults the > vmalloc area PTEs, which doesn't seem to be required here for the > kernel memory case (and could be added to vmap using a flag if actually > required). > > Signed-off-by: Christoph Hellwig > --- > drivers/gpu/drm/i915/Kconfig | 1 + > drivers/gpu/drm/i915/gem/i915_gem_pages.c | 126 ++++++++++------------ > 2 files changed, 59 insertions(+), 68 deletions(-) > > diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig > index 9afa5c4a6bf006..1e1cb245fca778 100644 > --- a/drivers/gpu/drm/i915/Kconfig > +++ b/drivers/gpu/drm/i915/Kconfig > @@ -25,6 +25,7 @@ config DRM_I915 > select CRC32 > select SND_HDA_I915 if SND_HDA_CORE > select CEC_CORE if CEC_NOTIFIER > + select VMAP_PFN > help > Choose this option if you have a system that has "Intel Graphics > Media Accelerator" or "HD Graphics" integrated graphics, > diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c > index 6550c0bc824ea2..b519417667eb4b 100644 > --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c > +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c > @@ -232,34 +232,21 @@ int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj) > return err; > } > > -static inline pte_t iomap_pte(resource_size_t base, > - dma_addr_t offset, > - pgprot_t prot) > -{ > - return pte_mkspecial(pfn_pte((base + offset) >> PAGE_SHIFT, prot)); > -} > - > /* The 'mapping' part of i915_gem_object_pin_map() below */ > -static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > - enum i915_map_type type) > +static void *i915_gem_object_map_page(struct drm_i915_gem_object *obj, > + enum i915_map_type type) > { > - unsigned long n_pte = obj->base.size >> PAGE_SHIFT; > - struct sg_table *sgt = obj->mm.pages; > - pte_t *stack[32], **mem; > - struct vm_struct *area; > + unsigned long n_pages = obj->base.size >> PAGE_SHIFT, i; > + struct page *stack[32], **pages = stack, *page; > + struct sgt_iter iter; > pgprot_t pgprot; > + void *vaddr; > > - if (!i915_gem_object_has_struct_page(obj) && type != I915_MAP_WC) > - return NULL; > - > - if (GEM_WARN_ON(type == I915_MAP_WC && > - !static_cpu_has(X86_FEATURE_PAT))) > - return NULL; > - > - /* A single page can always be kmapped */ > - if (n_pte == 1 && type == I915_MAP_WB) { > - struct page *page = sg_page(sgt->sgl); > - > + switch (type) { > + default: > + MISSING_CASE(type); > + fallthrough; /* to use PAGE_KERNEL anyway */ > + case I915_MAP_WB: > /* > * On 32b, highmem using a finite set of indirect PTE (i.e. > * vmap) to provide virtual mappings of the high pages. > @@ -277,30 +264,8 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > * So if the page is beyond the 32b boundary, make an explicit > * vmap. > */ > - if (!PageHighMem(page)) > - return page_address(page); > - } > - > - mem = stack; > - if (n_pte > ARRAY_SIZE(stack)) { > - /* Too big for stack -- allocate temporary array instead */ > - mem = kvmalloc_array(n_pte, sizeof(*mem), GFP_KERNEL); > - if (!mem) > - return NULL; > - } > - > - area = alloc_vm_area(obj->base.size, mem); > - if (!area) { > - if (mem != stack) > - kvfree(mem); > - return NULL; > - } > - > - switch (type) { > - default: > - MISSING_CASE(type); > - fallthrough; /* to use PAGE_KERNEL anyway */ > - case I915_MAP_WB: > + if (n_pages == 1 && !PageHighMem(sg_page(obj->mm.pages->sgl))) > + return page_address(sg_page(obj->mm.pages->sgl)); > pgprot = PAGE_KERNEL; > break; > case I915_MAP_WC: > @@ -308,30 +273,49 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > break; > } > > - if (i915_gem_object_has_struct_page(obj)) { > - struct sgt_iter iter; > - struct page *page; > - pte_t **ptes = mem; > + if (n_pages > ARRAY_SIZE(stack)) { > + /* Too big for stack -- allocate temporary array instead */ > + pages = kvmalloc_array(n_pages, sizeof(*pages), GFP_KERNEL); > + if (!pages) > + return NULL; > + } > > - for_each_sgt_page(page, iter, sgt) > - **ptes++ = mk_pte(page, pgprot); > - } else { > - resource_size_t iomap; > - struct sgt_iter iter; > - pte_t **ptes = mem; > - dma_addr_t addr; > + i = 0; > + for_each_sgt_page(page, iter, obj->mm.pages) > + pages[i++] = page; > + vaddr = vmap(pages, n_pages, 0, pgprot); > + if (pages != stack) > + kvfree(pages); > + return vaddr; > +} > > - iomap = obj->mm.region->iomap.base; > - iomap -= obj->mm.region->region.start; > +static void *i915_gem_object_map_pfn(struct drm_i915_gem_object *obj, > + enum i915_map_type type) > +{ > + resource_size_t iomap = obj->mm.region->iomap.base - > + obj->mm.region->region.start; > + unsigned long n_pfn = obj->base.size >> PAGE_SHIFT; > + unsigned long stack[32], *pfns = stack, i; > + struct sgt_iter iter; > + dma_addr_t addr; > + void *vaddr; > + > + if (type != I915_MAP_WC) > + return NULL; > > - for_each_sgt_daddr(addr, iter, sgt) > - **ptes++ = iomap_pte(iomap, addr, pgprot); > + if (n_pfn > ARRAY_SIZE(stack)) { > + /* Too big for stack -- allocate temporary array instead */ > + pfns = kvmalloc_array(n_pfn, sizeof(*pfns), GFP_KERNEL); > + if (!pfns) > + return NULL; > } > > - if (mem != stack) > - kvfree(mem); > - > - return area->addr; > + for_each_sgt_daddr(addr, iter, obj->mm.pages) > + pfns[i++] = (iomap + addr) >> PAGE_SHIFT; Missing the i = 0 fix from Dan? > + vaddr = vmap_pfn(pfns, n_pfn, pgprot_writecombine(PAGE_KERNEL_IO)); > + if (pfns != stack) > + kvfree(pfns); > + return vaddr; > } > > /* get, pin, and map the pages of the object into kernel space */ > @@ -383,7 +367,13 @@ void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj, > } > > if (!ptr) { > - ptr = i915_gem_object_map(obj, type); > + if (GEM_WARN_ON(type == I915_MAP_WC && > + !static_cpu_has(X86_FEATURE_PAT))) > + ptr = NULL; > + else if (i915_gem_object_has_struct_page(obj)) > + ptr = i915_gem_object_map_page(obj, type); > + else > + ptr = i915_gem_object_map_pfn(obj, type); > if (!ptr) { > err = -ENOMEM; > goto err_unpin; > -- > 2.28.0 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx 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 X-Spam-Level: X-Spam-Status: No, score=-9.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FORGED_FROMDOMAIN,FREEMAIL_FROM, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY, SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 7744BC4363D for ; Fri, 25 Sep 2020 14:09:29 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id E61CC20BED for ; Fri, 25 Sep 2020 14:09:28 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="VAE/JCw4" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org E61CC20BED Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=gmail.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id 570666B005D; Fri, 25 Sep 2020 10:09:28 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 520C36B0062; Fri, 25 Sep 2020 10:09:28 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 435D96B0068; Fri, 25 Sep 2020 10:09:28 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0216.hostedemail.com [216.40.44.216]) by kanga.kvack.org (Postfix) with ESMTP id 2CFAF6B005D for ; Fri, 25 Sep 2020 10:09:28 -0400 (EDT) Received: from smtpin23.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay03.hostedemail.com (Postfix) with ESMTP id DC9C88249980 for ; Fri, 25 Sep 2020 14:09:27 +0000 (UTC) X-FDA: 77301766374.23.wrist91_5712c0727168 Received: from filter.hostedemail.com (10.5.16.251.rfc1918.com [10.5.16.251]) by smtpin23.hostedemail.com (Postfix) with ESMTP id B74EE37606 for ; Fri, 25 Sep 2020 14:09:27 +0000 (UTC) X-HE-Tag: wrist91_5712c0727168 X-Filterd-Recvd-Size: 11454 Received: from mail-ua1-f68.google.com (mail-ua1-f68.google.com [209.85.222.68]) by imf22.hostedemail.com (Postfix) with ESMTP for ; Fri, 25 Sep 2020 14:09:27 +0000 (UTC) Received: by mail-ua1-f68.google.com with SMTP id n26so968140uao.8 for ; Fri, 25 Sep 2020 07:09:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=W4ybNDBj3JkG7xwtC+pQm/GVM9Ki73UbHYGcb1DNgoE=; b=VAE/JCw4Kc2YKn1yJ8L/Q9+LADI9BPVrPzMbIS0mOQDh1e+IitPGEnQLZfakaeotSX fJpnesPgh6fudFH4hBFOTUzOvja6IHah1EG+sAlo5mDeRe30T7WE5Z+EmmrqyLBVjPT2 Y6MypsxJU5ecTFL5KTTnudzV3iHFKzBhkicIF7nAwXtRPFhwdOi9g9/Y/qXRPNUiOlZ/ S833Agy78MEYaybNf9Tk+M+dZyKRMHoBkxs25nGIJGGXWhSky07MsU9xI+7dnENL4I7l bH2CDxH5hj1vhP5g8v/h/6Mm80YH+iZTF4wuYNk0h7j6DAwbRzp57D1NpHpN8KHRNu/G TlSA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=W4ybNDBj3JkG7xwtC+pQm/GVM9Ki73UbHYGcb1DNgoE=; b=sLU6TvF0AyrnOUYJXIlNbeu0y91SIJNvHbsWbMRl3YIiic0+ldnEiuiBQ7WgKFXMEp 9JgH8yDADVj+s/Naj6BGwZQqRqokHSo7Ip7c5BGVK0pKenglmRPDrsbY0blQd2jPb55Y kf5c63pyviwm2JqDnrzWmIh9EkDFSuf4CoRZLv6so6TT51HWMFgQ1f2X7dAFUrEYIcHN EpzoDH0t12jJWLQfjU/kD+RwQbF3FfZJMyLahEFxPx7EMMdbIgl5KDOe51QgBZaaIagb phj35jOx6EwCEvbbkV/Az6aP2YNMGAV3vpILpju7Z9m9hILH1c4tM7+xm77ldoVbaoqA 4aUg== X-Gm-Message-State: AOAM530q0PmtdGU6awUIBc/JfHGNgu8qKl5Wuui/wjaDrmfnYd0W4KaJ SnfHkMU+GT4R+lX3Z1Xc1EYi3f/4s4mi+lqEw1s= X-Google-Smtp-Source: ABdhPJy0IxUr4anWia0Qm7ztGxjztU5H4LAKNIvZHwQZc9rZ+mSfXU53xRXTbzCYoiyClMt8qe//fl65KD/2aNPSkzU= X-Received: by 2002:ab0:6f91:: with SMTP id f17mr2536965uav.129.1601042966264; Fri, 25 Sep 2020 07:09:26 -0700 (PDT) MIME-Version: 1.0 References: <20200924135853.875294-1-hch@lst.de> <20200924135853.875294-9-hch@lst.de> In-Reply-To: <20200924135853.875294-9-hch@lst.de> From: Matthew Auld Date: Fri, 25 Sep 2020 15:08:59 +0100 Message-ID: Subject: Re: [Intel-gfx] [PATCH 08/11] drm/i915: use vmap in i915_gem_object_map To: Christoph Hellwig Cc: Andrew Morton , Juergen Gross , Stefano Stabellini , Matthew Wilcox , ML dri-devel , linux-mm@kvack.org, Peter Zijlstra , kernel list , Intel Graphics Development , x86@kernel.org, Chris Wilson , Minchan Kim , Matthew Auld , xen-devel@lists.xenproject.org, Boris Ostrovsky , Nitin Gupta Content-Type: text/plain; charset="UTF-8" X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: On Thu, 24 Sep 2020 at 14:59, Christoph Hellwig wrote: > > i915_gem_object_map implements fairly low-level vmap functionality in > a driver. Split it into two helpers, one for remapping kernel memory > which can use vmap, and one for I/O memory that uses vmap_pfn. > > The only practical difference is that alloc_vm_area prefeaults the > vmalloc area PTEs, which doesn't seem to be required here for the > kernel memory case (and could be added to vmap using a flag if actually > required). > > Signed-off-by: Christoph Hellwig > --- > drivers/gpu/drm/i915/Kconfig | 1 + > drivers/gpu/drm/i915/gem/i915_gem_pages.c | 126 ++++++++++------------ > 2 files changed, 59 insertions(+), 68 deletions(-) > > diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig > index 9afa5c4a6bf006..1e1cb245fca778 100644 > --- a/drivers/gpu/drm/i915/Kconfig > +++ b/drivers/gpu/drm/i915/Kconfig > @@ -25,6 +25,7 @@ config DRM_I915 > select CRC32 > select SND_HDA_I915 if SND_HDA_CORE > select CEC_CORE if CEC_NOTIFIER > + select VMAP_PFN > help > Choose this option if you have a system that has "Intel Graphics > Media Accelerator" or "HD Graphics" integrated graphics, > diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c > index 6550c0bc824ea2..b519417667eb4b 100644 > --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c > +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c > @@ -232,34 +232,21 @@ int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj) > return err; > } > > -static inline pte_t iomap_pte(resource_size_t base, > - dma_addr_t offset, > - pgprot_t prot) > -{ > - return pte_mkspecial(pfn_pte((base + offset) >> PAGE_SHIFT, prot)); > -} > - > /* The 'mapping' part of i915_gem_object_pin_map() below */ > -static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > - enum i915_map_type type) > +static void *i915_gem_object_map_page(struct drm_i915_gem_object *obj, > + enum i915_map_type type) > { > - unsigned long n_pte = obj->base.size >> PAGE_SHIFT; > - struct sg_table *sgt = obj->mm.pages; > - pte_t *stack[32], **mem; > - struct vm_struct *area; > + unsigned long n_pages = obj->base.size >> PAGE_SHIFT, i; > + struct page *stack[32], **pages = stack, *page; > + struct sgt_iter iter; > pgprot_t pgprot; > + void *vaddr; > > - if (!i915_gem_object_has_struct_page(obj) && type != I915_MAP_WC) > - return NULL; > - > - if (GEM_WARN_ON(type == I915_MAP_WC && > - !static_cpu_has(X86_FEATURE_PAT))) > - return NULL; > - > - /* A single page can always be kmapped */ > - if (n_pte == 1 && type == I915_MAP_WB) { > - struct page *page = sg_page(sgt->sgl); > - > + switch (type) { > + default: > + MISSING_CASE(type); > + fallthrough; /* to use PAGE_KERNEL anyway */ > + case I915_MAP_WB: > /* > * On 32b, highmem using a finite set of indirect PTE (i.e. > * vmap) to provide virtual mappings of the high pages. > @@ -277,30 +264,8 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > * So if the page is beyond the 32b boundary, make an explicit > * vmap. > */ > - if (!PageHighMem(page)) > - return page_address(page); > - } > - > - mem = stack; > - if (n_pte > ARRAY_SIZE(stack)) { > - /* Too big for stack -- allocate temporary array instead */ > - mem = kvmalloc_array(n_pte, sizeof(*mem), GFP_KERNEL); > - if (!mem) > - return NULL; > - } > - > - area = alloc_vm_area(obj->base.size, mem); > - if (!area) { > - if (mem != stack) > - kvfree(mem); > - return NULL; > - } > - > - switch (type) { > - default: > - MISSING_CASE(type); > - fallthrough; /* to use PAGE_KERNEL anyway */ > - case I915_MAP_WB: > + if (n_pages == 1 && !PageHighMem(sg_page(obj->mm.pages->sgl))) > + return page_address(sg_page(obj->mm.pages->sgl)); > pgprot = PAGE_KERNEL; > break; > case I915_MAP_WC: > @@ -308,30 +273,49 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > break; > } > > - if (i915_gem_object_has_struct_page(obj)) { > - struct sgt_iter iter; > - struct page *page; > - pte_t **ptes = mem; > + if (n_pages > ARRAY_SIZE(stack)) { > + /* Too big for stack -- allocate temporary array instead */ > + pages = kvmalloc_array(n_pages, sizeof(*pages), GFP_KERNEL); > + if (!pages) > + return NULL; > + } > > - for_each_sgt_page(page, iter, sgt) > - **ptes++ = mk_pte(page, pgprot); > - } else { > - resource_size_t iomap; > - struct sgt_iter iter; > - pte_t **ptes = mem; > - dma_addr_t addr; > + i = 0; > + for_each_sgt_page(page, iter, obj->mm.pages) > + pages[i++] = page; > + vaddr = vmap(pages, n_pages, 0, pgprot); > + if (pages != stack) > + kvfree(pages); > + return vaddr; > +} > > - iomap = obj->mm.region->iomap.base; > - iomap -= obj->mm.region->region.start; > +static void *i915_gem_object_map_pfn(struct drm_i915_gem_object *obj, > + enum i915_map_type type) > +{ > + resource_size_t iomap = obj->mm.region->iomap.base - > + obj->mm.region->region.start; > + unsigned long n_pfn = obj->base.size >> PAGE_SHIFT; > + unsigned long stack[32], *pfns = stack, i; > + struct sgt_iter iter; > + dma_addr_t addr; > + void *vaddr; > + > + if (type != I915_MAP_WC) > + return NULL; > > - for_each_sgt_daddr(addr, iter, sgt) > - **ptes++ = iomap_pte(iomap, addr, pgprot); > + if (n_pfn > ARRAY_SIZE(stack)) { > + /* Too big for stack -- allocate temporary array instead */ > + pfns = kvmalloc_array(n_pfn, sizeof(*pfns), GFP_KERNEL); > + if (!pfns) > + return NULL; > } > > - if (mem != stack) > - kvfree(mem); > - > - return area->addr; > + for_each_sgt_daddr(addr, iter, obj->mm.pages) > + pfns[i++] = (iomap + addr) >> PAGE_SHIFT; Missing the i = 0 fix from Dan? > + vaddr = vmap_pfn(pfns, n_pfn, pgprot_writecombine(PAGE_KERNEL_IO)); > + if (pfns != stack) > + kvfree(pfns); > + return vaddr; > } > > /* get, pin, and map the pages of the object into kernel space */ > @@ -383,7 +367,13 @@ void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj, > } > > if (!ptr) { > - ptr = i915_gem_object_map(obj, type); > + if (GEM_WARN_ON(type == I915_MAP_WC && > + !static_cpu_has(X86_FEATURE_PAT))) > + ptr = NULL; > + else if (i915_gem_object_has_struct_page(obj)) > + ptr = i915_gem_object_map_page(obj, type); > + else > + ptr = i915_gem_object_map_pfn(obj, type); > if (!ptr) { > err = -ENOMEM; > goto err_unpin; > -- > 2.28.0 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx 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 X-Spam-Level: X-Spam-Status: No, score=-9.5 required=3.0 tests=BAYES_00,DKIM_ADSP_CUSTOM_MED, DKIM_INVALID,DKIM_SIGNED,FREEMAIL_FORGED_FROMDOMAIN,FREEMAIL_FROM, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY, SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 69F82C4741F for ; Fri, 25 Sep 2020 14:09:30 +0000 (UTC) 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 mail.kernel.org (Postfix) with ESMTPS id 1690921D7A for ; Fri, 25 Sep 2020 14:09:30 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=fail reason="signature verification failed" (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="VAE/JCw4" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 1690921D7A Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=gmail.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=dri-devel-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id EF5BD6ECDD; Fri, 25 Sep 2020 14:09:28 +0000 (UTC) Received: from mail-ua1-x944.google.com (mail-ua1-x944.google.com [IPv6:2607:f8b0:4864:20::944]) by gabe.freedesktop.org (Postfix) with ESMTPS id 3AAB76ECDD; Fri, 25 Sep 2020 14:09:27 +0000 (UTC) Received: by mail-ua1-x944.google.com with SMTP id h15so976710uab.3; Fri, 25 Sep 2020 07:09:27 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=W4ybNDBj3JkG7xwtC+pQm/GVM9Ki73UbHYGcb1DNgoE=; b=VAE/JCw4Kc2YKn1yJ8L/Q9+LADI9BPVrPzMbIS0mOQDh1e+IitPGEnQLZfakaeotSX fJpnesPgh6fudFH4hBFOTUzOvja6IHah1EG+sAlo5mDeRe30T7WE5Z+EmmrqyLBVjPT2 Y6MypsxJU5ecTFL5KTTnudzV3iHFKzBhkicIF7nAwXtRPFhwdOi9g9/Y/qXRPNUiOlZ/ S833Agy78MEYaybNf9Tk+M+dZyKRMHoBkxs25nGIJGGXWhSky07MsU9xI+7dnENL4I7l bH2CDxH5hj1vhP5g8v/h/6Mm80YH+iZTF4wuYNk0h7j6DAwbRzp57D1NpHpN8KHRNu/G TlSA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=W4ybNDBj3JkG7xwtC+pQm/GVM9Ki73UbHYGcb1DNgoE=; b=iT+rGlEO8i49daiE7JlxXd5aVJhQ1exTWoxm46UeUdAZebvdC/es70NBQQiBD+kqih 63RNhn887ffcA6O3+BXbcNoVDj8CECE+iUAxVsq4dlmzDHjHbcA1mTPT9T/9fvffjRAD mw5A/Nu53iZVNeBnzluU/+HYUvc1Ub1exWLOVpI588VkLPiyakUE9QdpPAVot5TjNc97 hevOMa+FcPwMLgDd2pIzAGiaNs/5JZDdJIo0qBuF4bARnUgGmVMU9i9gNA+4AmL1Xi5/ F0+wITliEA207dds+og9lg1W7S/Xz4M8EYaNHjio06dmwK1zOQxAwbftz7jjDhAZWL7f 9w8A== X-Gm-Message-State: AOAM5324kCxQ0zqiAPqBIe8hOssvJMFq10SIMb8sD57iaS1Ihu6OlTSJ 8gS7AZ2YnqNfBEgM7WQkMjr+QrB1cnna9cIUeds= X-Google-Smtp-Source: ABdhPJy0IxUr4anWia0Qm7ztGxjztU5H4LAKNIvZHwQZc9rZ+mSfXU53xRXTbzCYoiyClMt8qe//fl65KD/2aNPSkzU= X-Received: by 2002:ab0:6f91:: with SMTP id f17mr2536965uav.129.1601042966264; Fri, 25 Sep 2020 07:09:26 -0700 (PDT) MIME-Version: 1.0 References: <20200924135853.875294-1-hch@lst.de> <20200924135853.875294-9-hch@lst.de> In-Reply-To: <20200924135853.875294-9-hch@lst.de> From: Matthew Auld Date: Fri, 25 Sep 2020 15:08:59 +0100 Message-ID: Subject: Re: [Intel-gfx] [PATCH 08/11] drm/i915: use vmap in i915_gem_object_map To: Christoph Hellwig X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Juergen Gross , Stefano Stabellini , Minchan Kim , Peter Zijlstra , Intel Graphics Development , x86@kernel.org, kernel list , ML dri-devel , Chris Wilson , linux-mm@kvack.org, Matthew Wilcox , xen-devel@lists.xenproject.org, Andrew Morton , Boris Ostrovsky , Nitin Gupta , Matthew Auld Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" On Thu, 24 Sep 2020 at 14:59, Christoph Hellwig wrote: > > i915_gem_object_map implements fairly low-level vmap functionality in > a driver. Split it into two helpers, one for remapping kernel memory > which can use vmap, and one for I/O memory that uses vmap_pfn. > > The only practical difference is that alloc_vm_area prefeaults the > vmalloc area PTEs, which doesn't seem to be required here for the > kernel memory case (and could be added to vmap using a flag if actually > required). > > Signed-off-by: Christoph Hellwig > --- > drivers/gpu/drm/i915/Kconfig | 1 + > drivers/gpu/drm/i915/gem/i915_gem_pages.c | 126 ++++++++++------------ > 2 files changed, 59 insertions(+), 68 deletions(-) > > diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig > index 9afa5c4a6bf006..1e1cb245fca778 100644 > --- a/drivers/gpu/drm/i915/Kconfig > +++ b/drivers/gpu/drm/i915/Kconfig > @@ -25,6 +25,7 @@ config DRM_I915 > select CRC32 > select SND_HDA_I915 if SND_HDA_CORE > select CEC_CORE if CEC_NOTIFIER > + select VMAP_PFN > help > Choose this option if you have a system that has "Intel Graphics > Media Accelerator" or "HD Graphics" integrated graphics, > diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c > index 6550c0bc824ea2..b519417667eb4b 100644 > --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c > +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c > @@ -232,34 +232,21 @@ int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj) > return err; > } > > -static inline pte_t iomap_pte(resource_size_t base, > - dma_addr_t offset, > - pgprot_t prot) > -{ > - return pte_mkspecial(pfn_pte((base + offset) >> PAGE_SHIFT, prot)); > -} > - > /* The 'mapping' part of i915_gem_object_pin_map() below */ > -static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > - enum i915_map_type type) > +static void *i915_gem_object_map_page(struct drm_i915_gem_object *obj, > + enum i915_map_type type) > { > - unsigned long n_pte = obj->base.size >> PAGE_SHIFT; > - struct sg_table *sgt = obj->mm.pages; > - pte_t *stack[32], **mem; > - struct vm_struct *area; > + unsigned long n_pages = obj->base.size >> PAGE_SHIFT, i; > + struct page *stack[32], **pages = stack, *page; > + struct sgt_iter iter; > pgprot_t pgprot; > + void *vaddr; > > - if (!i915_gem_object_has_struct_page(obj) && type != I915_MAP_WC) > - return NULL; > - > - if (GEM_WARN_ON(type == I915_MAP_WC && > - !static_cpu_has(X86_FEATURE_PAT))) > - return NULL; > - > - /* A single page can always be kmapped */ > - if (n_pte == 1 && type == I915_MAP_WB) { > - struct page *page = sg_page(sgt->sgl); > - > + switch (type) { > + default: > + MISSING_CASE(type); > + fallthrough; /* to use PAGE_KERNEL anyway */ > + case I915_MAP_WB: > /* > * On 32b, highmem using a finite set of indirect PTE (i.e. > * vmap) to provide virtual mappings of the high pages. > @@ -277,30 +264,8 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > * So if the page is beyond the 32b boundary, make an explicit > * vmap. > */ > - if (!PageHighMem(page)) > - return page_address(page); > - } > - > - mem = stack; > - if (n_pte > ARRAY_SIZE(stack)) { > - /* Too big for stack -- allocate temporary array instead */ > - mem = kvmalloc_array(n_pte, sizeof(*mem), GFP_KERNEL); > - if (!mem) > - return NULL; > - } > - > - area = alloc_vm_area(obj->base.size, mem); > - if (!area) { > - if (mem != stack) > - kvfree(mem); > - return NULL; > - } > - > - switch (type) { > - default: > - MISSING_CASE(type); > - fallthrough; /* to use PAGE_KERNEL anyway */ > - case I915_MAP_WB: > + if (n_pages == 1 && !PageHighMem(sg_page(obj->mm.pages->sgl))) > + return page_address(sg_page(obj->mm.pages->sgl)); > pgprot = PAGE_KERNEL; > break; > case I915_MAP_WC: > @@ -308,30 +273,49 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > break; > } > > - if (i915_gem_object_has_struct_page(obj)) { > - struct sgt_iter iter; > - struct page *page; > - pte_t **ptes = mem; > + if (n_pages > ARRAY_SIZE(stack)) { > + /* Too big for stack -- allocate temporary array instead */ > + pages = kvmalloc_array(n_pages, sizeof(*pages), GFP_KERNEL); > + if (!pages) > + return NULL; > + } > > - for_each_sgt_page(page, iter, sgt) > - **ptes++ = mk_pte(page, pgprot); > - } else { > - resource_size_t iomap; > - struct sgt_iter iter; > - pte_t **ptes = mem; > - dma_addr_t addr; > + i = 0; > + for_each_sgt_page(page, iter, obj->mm.pages) > + pages[i++] = page; > + vaddr = vmap(pages, n_pages, 0, pgprot); > + if (pages != stack) > + kvfree(pages); > + return vaddr; > +} > > - iomap = obj->mm.region->iomap.base; > - iomap -= obj->mm.region->region.start; > +static void *i915_gem_object_map_pfn(struct drm_i915_gem_object *obj, > + enum i915_map_type type) > +{ > + resource_size_t iomap = obj->mm.region->iomap.base - > + obj->mm.region->region.start; > + unsigned long n_pfn = obj->base.size >> PAGE_SHIFT; > + unsigned long stack[32], *pfns = stack, i; > + struct sgt_iter iter; > + dma_addr_t addr; > + void *vaddr; > + > + if (type != I915_MAP_WC) > + return NULL; > > - for_each_sgt_daddr(addr, iter, sgt) > - **ptes++ = iomap_pte(iomap, addr, pgprot); > + if (n_pfn > ARRAY_SIZE(stack)) { > + /* Too big for stack -- allocate temporary array instead */ > + pfns = kvmalloc_array(n_pfn, sizeof(*pfns), GFP_KERNEL); > + if (!pfns) > + return NULL; > } > > - if (mem != stack) > - kvfree(mem); > - > - return area->addr; > + for_each_sgt_daddr(addr, iter, obj->mm.pages) > + pfns[i++] = (iomap + addr) >> PAGE_SHIFT; Missing the i = 0 fix from Dan? > + vaddr = vmap_pfn(pfns, n_pfn, pgprot_writecombine(PAGE_KERNEL_IO)); > + if (pfns != stack) > + kvfree(pfns); > + return vaddr; > } > > /* get, pin, and map the pages of the object into kernel space */ > @@ -383,7 +367,13 @@ void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj, > } > > if (!ptr) { > - ptr = i915_gem_object_map(obj, type); > + if (GEM_WARN_ON(type == I915_MAP_WC && > + !static_cpu_has(X86_FEATURE_PAT))) > + ptr = NULL; > + else if (i915_gem_object_has_struct_page(obj)) > + ptr = i915_gem_object_map_page(obj, type); > + else > + ptr = i915_gem_object_map_pfn(obj, type); > if (!ptr) { > err = -ENOMEM; > goto err_unpin; > -- > 2.28.0 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx _______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel 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 X-Spam-Level: X-Spam-Status: No, score=-9.5 required=3.0 tests=BAYES_00,DKIM_ADSP_CUSTOM_MED, DKIM_INVALID,DKIM_SIGNED,FREEMAIL_FORGED_FROMDOMAIN,FREEMAIL_FROM, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY, SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 22EB6C4727C for ; Fri, 25 Sep 2020 14:09:30 +0000 (UTC) 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 mail.kernel.org (Postfix) with ESMTPS id D1DD120936 for ; Fri, 25 Sep 2020 14:09:29 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=fail reason="signature verification failed" (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="VAE/JCw4" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org D1DD120936 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=gmail.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 2C9F76ECE0; Fri, 25 Sep 2020 14:09:29 +0000 (UTC) Received: from mail-ua1-x944.google.com (mail-ua1-x944.google.com [IPv6:2607:f8b0:4864:20::944]) by gabe.freedesktop.org (Postfix) with ESMTPS id 3AAB76ECDD; Fri, 25 Sep 2020 14:09:27 +0000 (UTC) Received: by mail-ua1-x944.google.com with SMTP id h15so976710uab.3; Fri, 25 Sep 2020 07:09:27 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=W4ybNDBj3JkG7xwtC+pQm/GVM9Ki73UbHYGcb1DNgoE=; b=VAE/JCw4Kc2YKn1yJ8L/Q9+LADI9BPVrPzMbIS0mOQDh1e+IitPGEnQLZfakaeotSX fJpnesPgh6fudFH4hBFOTUzOvja6IHah1EG+sAlo5mDeRe30T7WE5Z+EmmrqyLBVjPT2 Y6MypsxJU5ecTFL5KTTnudzV3iHFKzBhkicIF7nAwXtRPFhwdOi9g9/Y/qXRPNUiOlZ/ S833Agy78MEYaybNf9Tk+M+dZyKRMHoBkxs25nGIJGGXWhSky07MsU9xI+7dnENL4I7l bH2CDxH5hj1vhP5g8v/h/6Mm80YH+iZTF4wuYNk0h7j6DAwbRzp57D1NpHpN8KHRNu/G TlSA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=W4ybNDBj3JkG7xwtC+pQm/GVM9Ki73UbHYGcb1DNgoE=; b=iT+rGlEO8i49daiE7JlxXd5aVJhQ1exTWoxm46UeUdAZebvdC/es70NBQQiBD+kqih 63RNhn887ffcA6O3+BXbcNoVDj8CECE+iUAxVsq4dlmzDHjHbcA1mTPT9T/9fvffjRAD mw5A/Nu53iZVNeBnzluU/+HYUvc1Ub1exWLOVpI588VkLPiyakUE9QdpPAVot5TjNc97 hevOMa+FcPwMLgDd2pIzAGiaNs/5JZDdJIo0qBuF4bARnUgGmVMU9i9gNA+4AmL1Xi5/ F0+wITliEA207dds+og9lg1W7S/Xz4M8EYaNHjio06dmwK1zOQxAwbftz7jjDhAZWL7f 9w8A== X-Gm-Message-State: AOAM5324kCxQ0zqiAPqBIe8hOssvJMFq10SIMb8sD57iaS1Ihu6OlTSJ 8gS7AZ2YnqNfBEgM7WQkMjr+QrB1cnna9cIUeds= X-Google-Smtp-Source: ABdhPJy0IxUr4anWia0Qm7ztGxjztU5H4LAKNIvZHwQZc9rZ+mSfXU53xRXTbzCYoiyClMt8qe//fl65KD/2aNPSkzU= X-Received: by 2002:ab0:6f91:: with SMTP id f17mr2536965uav.129.1601042966264; Fri, 25 Sep 2020 07:09:26 -0700 (PDT) MIME-Version: 1.0 References: <20200924135853.875294-1-hch@lst.de> <20200924135853.875294-9-hch@lst.de> In-Reply-To: <20200924135853.875294-9-hch@lst.de> From: Matthew Auld Date: Fri, 25 Sep 2020 15:08:59 +0100 Message-ID: To: Christoph Hellwig Subject: Re: [Intel-gfx] [PATCH 08/11] drm/i915: use vmap in i915_gem_object_map 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: Juergen Gross , Stefano Stabellini , Minchan Kim , Peter Zijlstra , Intel Graphics Development , x86@kernel.org, kernel list , ML dri-devel , Chris Wilson , linux-mm@kvack.org, Matthew Wilcox , xen-devel@lists.xenproject.org, Andrew Morton , Boris Ostrovsky , Nitin Gupta , Matthew Auld Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" On Thu, 24 Sep 2020 at 14:59, Christoph Hellwig wrote: > > i915_gem_object_map implements fairly low-level vmap functionality in > a driver. Split it into two helpers, one for remapping kernel memory > which can use vmap, and one for I/O memory that uses vmap_pfn. > > The only practical difference is that alloc_vm_area prefeaults the > vmalloc area PTEs, which doesn't seem to be required here for the > kernel memory case (and could be added to vmap using a flag if actually > required). > > Signed-off-by: Christoph Hellwig > --- > drivers/gpu/drm/i915/Kconfig | 1 + > drivers/gpu/drm/i915/gem/i915_gem_pages.c | 126 ++++++++++------------ > 2 files changed, 59 insertions(+), 68 deletions(-) > > diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig > index 9afa5c4a6bf006..1e1cb245fca778 100644 > --- a/drivers/gpu/drm/i915/Kconfig > +++ b/drivers/gpu/drm/i915/Kconfig > @@ -25,6 +25,7 @@ config DRM_I915 > select CRC32 > select SND_HDA_I915 if SND_HDA_CORE > select CEC_CORE if CEC_NOTIFIER > + select VMAP_PFN > help > Choose this option if you have a system that has "Intel Graphics > Media Accelerator" or "HD Graphics" integrated graphics, > diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c > index 6550c0bc824ea2..b519417667eb4b 100644 > --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c > +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c > @@ -232,34 +232,21 @@ int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj) > return err; > } > > -static inline pte_t iomap_pte(resource_size_t base, > - dma_addr_t offset, > - pgprot_t prot) > -{ > - return pte_mkspecial(pfn_pte((base + offset) >> PAGE_SHIFT, prot)); > -} > - > /* The 'mapping' part of i915_gem_object_pin_map() below */ > -static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > - enum i915_map_type type) > +static void *i915_gem_object_map_page(struct drm_i915_gem_object *obj, > + enum i915_map_type type) > { > - unsigned long n_pte = obj->base.size >> PAGE_SHIFT; > - struct sg_table *sgt = obj->mm.pages; > - pte_t *stack[32], **mem; > - struct vm_struct *area; > + unsigned long n_pages = obj->base.size >> PAGE_SHIFT, i; > + struct page *stack[32], **pages = stack, *page; > + struct sgt_iter iter; > pgprot_t pgprot; > + void *vaddr; > > - if (!i915_gem_object_has_struct_page(obj) && type != I915_MAP_WC) > - return NULL; > - > - if (GEM_WARN_ON(type == I915_MAP_WC && > - !static_cpu_has(X86_FEATURE_PAT))) > - return NULL; > - > - /* A single page can always be kmapped */ > - if (n_pte == 1 && type == I915_MAP_WB) { > - struct page *page = sg_page(sgt->sgl); > - > + switch (type) { > + default: > + MISSING_CASE(type); > + fallthrough; /* to use PAGE_KERNEL anyway */ > + case I915_MAP_WB: > /* > * On 32b, highmem using a finite set of indirect PTE (i.e. > * vmap) to provide virtual mappings of the high pages. > @@ -277,30 +264,8 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > * So if the page is beyond the 32b boundary, make an explicit > * vmap. > */ > - if (!PageHighMem(page)) > - return page_address(page); > - } > - > - mem = stack; > - if (n_pte > ARRAY_SIZE(stack)) { > - /* Too big for stack -- allocate temporary array instead */ > - mem = kvmalloc_array(n_pte, sizeof(*mem), GFP_KERNEL); > - if (!mem) > - return NULL; > - } > - > - area = alloc_vm_area(obj->base.size, mem); > - if (!area) { > - if (mem != stack) > - kvfree(mem); > - return NULL; > - } > - > - switch (type) { > - default: > - MISSING_CASE(type); > - fallthrough; /* to use PAGE_KERNEL anyway */ > - case I915_MAP_WB: > + if (n_pages == 1 && !PageHighMem(sg_page(obj->mm.pages->sgl))) > + return page_address(sg_page(obj->mm.pages->sgl)); > pgprot = PAGE_KERNEL; > break; > case I915_MAP_WC: > @@ -308,30 +273,49 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > break; > } > > - if (i915_gem_object_has_struct_page(obj)) { > - struct sgt_iter iter; > - struct page *page; > - pte_t **ptes = mem; > + if (n_pages > ARRAY_SIZE(stack)) { > + /* Too big for stack -- allocate temporary array instead */ > + pages = kvmalloc_array(n_pages, sizeof(*pages), GFP_KERNEL); > + if (!pages) > + return NULL; > + } > > - for_each_sgt_page(page, iter, sgt) > - **ptes++ = mk_pte(page, pgprot); > - } else { > - resource_size_t iomap; > - struct sgt_iter iter; > - pte_t **ptes = mem; > - dma_addr_t addr; > + i = 0; > + for_each_sgt_page(page, iter, obj->mm.pages) > + pages[i++] = page; > + vaddr = vmap(pages, n_pages, 0, pgprot); > + if (pages != stack) > + kvfree(pages); > + return vaddr; > +} > > - iomap = obj->mm.region->iomap.base; > - iomap -= obj->mm.region->region.start; > +static void *i915_gem_object_map_pfn(struct drm_i915_gem_object *obj, > + enum i915_map_type type) > +{ > + resource_size_t iomap = obj->mm.region->iomap.base - > + obj->mm.region->region.start; > + unsigned long n_pfn = obj->base.size >> PAGE_SHIFT; > + unsigned long stack[32], *pfns = stack, i; > + struct sgt_iter iter; > + dma_addr_t addr; > + void *vaddr; > + > + if (type != I915_MAP_WC) > + return NULL; > > - for_each_sgt_daddr(addr, iter, sgt) > - **ptes++ = iomap_pte(iomap, addr, pgprot); > + if (n_pfn > ARRAY_SIZE(stack)) { > + /* Too big for stack -- allocate temporary array instead */ > + pfns = kvmalloc_array(n_pfn, sizeof(*pfns), GFP_KERNEL); > + if (!pfns) > + return NULL; > } > > - if (mem != stack) > - kvfree(mem); > - > - return area->addr; > + for_each_sgt_daddr(addr, iter, obj->mm.pages) > + pfns[i++] = (iomap + addr) >> PAGE_SHIFT; Missing the i = 0 fix from Dan? > + vaddr = vmap_pfn(pfns, n_pfn, pgprot_writecombine(PAGE_KERNEL_IO)); > + if (pfns != stack) > + kvfree(pfns); > + return vaddr; > } > > /* get, pin, and map the pages of the object into kernel space */ > @@ -383,7 +367,13 @@ void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj, > } > > if (!ptr) { > - ptr = i915_gem_object_map(obj, type); > + if (GEM_WARN_ON(type == I915_MAP_WC && > + !static_cpu_has(X86_FEATURE_PAT))) > + ptr = NULL; > + else if (i915_gem_object_has_struct_page(obj)) > + ptr = i915_gem_object_map_page(obj, type); > + else > + ptr = i915_gem_object_map_pfn(obj, type); > if (!ptr) { > err = -ENOMEM; > goto err_unpin; > -- > 2.28.0 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx 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 X-Spam-Level: X-Spam-Status: No, score=-9.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FORGED_FROMDOMAIN,FREEMAIL_FROM, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY, SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id AD586C4363D for ; Fri, 25 Sep 2020 14:10:42 +0000 (UTC) Received: from lists.xenproject.org (lists.xenproject.org [192.237.175.120]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 528FD20936 for ; Fri, 25 Sep 2020 14:10:42 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="VAE/JCw4" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 528FD20936 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=gmail.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=xen-devel-bounces@lists.xenproject.org Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1kLoQd-00041G-QC; Fri, 25 Sep 2020 14:10:27 +0000 Received: from us1-rack-iad1.inumbo.com ([172.99.69.81]) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1kLoPf-0003Ka-OL for xen-devel@lists.xenproject.org; Fri, 25 Sep 2020 14:09:27 +0000 X-Inumbo-ID: 27089a99-8e6d-49b2-bcb3-17c2d5968536 Received: from mail-ua1-x941.google.com (unknown [2607:f8b0:4864:20::941]) by us1-rack-iad1.inumbo.com (Halon) with ESMTPS id 27089a99-8e6d-49b2-bcb3-17c2d5968536; Fri, 25 Sep 2020 14:09:26 +0000 (UTC) Received: by mail-ua1-x941.google.com with SMTP id j12so971066ual.7 for ; Fri, 25 Sep 2020 07:09:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=W4ybNDBj3JkG7xwtC+pQm/GVM9Ki73UbHYGcb1DNgoE=; b=VAE/JCw4Kc2YKn1yJ8L/Q9+LADI9BPVrPzMbIS0mOQDh1e+IitPGEnQLZfakaeotSX fJpnesPgh6fudFH4hBFOTUzOvja6IHah1EG+sAlo5mDeRe30T7WE5Z+EmmrqyLBVjPT2 Y6MypsxJU5ecTFL5KTTnudzV3iHFKzBhkicIF7nAwXtRPFhwdOi9g9/Y/qXRPNUiOlZ/ S833Agy78MEYaybNf9Tk+M+dZyKRMHoBkxs25nGIJGGXWhSky07MsU9xI+7dnENL4I7l bH2CDxH5hj1vhP5g8v/h/6Mm80YH+iZTF4wuYNk0h7j6DAwbRzp57D1NpHpN8KHRNu/G TlSA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=W4ybNDBj3JkG7xwtC+pQm/GVM9Ki73UbHYGcb1DNgoE=; b=TbnO3r0valohidIm64MHCOhBCOdRKjXtuj4e4fLtkKq5KluTMYFPsPXsdd57XnDhPZ VKfxcBQnEFLnY7tL5d+R/adM1YLnLkLLfS98vyi4BG2UQKvJVORGMWzZRYzUeRSXI/9B nW7PeppAyH5Hn/UuB1bAw/IFXJFs4hceptYm6IDGYT2KqkeMN4BAfuJOWNjQnGjmJc1f lao/G4/Rw4120aTy/v/6hymEJ52pIF3/BC4or03VELcrLdizsLyk6ldqg8YWJeazzz3u R3raKpGTvLMeIz5DTVvnG6Xlz4wB6IYeGz/yYZpOG/r3QF7PaH9P40WkxPcK4gZjej2a AWVw== X-Gm-Message-State: AOAM530N9qaWhOX6T+/7ewvVBpP+yZB8TtP6GDhEJDN3kB7Y7Baoj/wF WfTro7WXhVN+MzVCrqBCo4rjIl1Ifa4vfFnNzGs= X-Google-Smtp-Source: ABdhPJy0IxUr4anWia0Qm7ztGxjztU5H4LAKNIvZHwQZc9rZ+mSfXU53xRXTbzCYoiyClMt8qe//fl65KD/2aNPSkzU= X-Received: by 2002:ab0:6f91:: with SMTP id f17mr2536965uav.129.1601042966264; Fri, 25 Sep 2020 07:09:26 -0700 (PDT) MIME-Version: 1.0 References: <20200924135853.875294-1-hch@lst.de> <20200924135853.875294-9-hch@lst.de> In-Reply-To: <20200924135853.875294-9-hch@lst.de> From: Matthew Auld Date: Fri, 25 Sep 2020 15:08:59 +0100 Message-ID: Subject: Re: [Intel-gfx] [PATCH 08/11] drm/i915: use vmap in i915_gem_object_map To: Christoph Hellwig Cc: Andrew Morton , Juergen Gross , Stefano Stabellini , Matthew Wilcox , ML dri-devel , linux-mm@kvack.org, Peter Zijlstra , kernel list , Intel Graphics Development , x86@kernel.org, Chris Wilson , Minchan Kim , Matthew Auld , xen-devel@lists.xenproject.org, Boris Ostrovsky , Nitin Gupta Content-Type: text/plain; charset="UTF-8" X-Mailman-Approved-At: Fri, 25 Sep 2020 14:10:26 +0000 X-BeenThere: xen-devel@lists.xenproject.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Errors-To: xen-devel-bounces@lists.xenproject.org Sender: "Xen-devel" On Thu, 24 Sep 2020 at 14:59, Christoph Hellwig wrote: > > i915_gem_object_map implements fairly low-level vmap functionality in > a driver. Split it into two helpers, one for remapping kernel memory > which can use vmap, and one for I/O memory that uses vmap_pfn. > > The only practical difference is that alloc_vm_area prefeaults the > vmalloc area PTEs, which doesn't seem to be required here for the > kernel memory case (and could be added to vmap using a flag if actually > required). > > Signed-off-by: Christoph Hellwig > --- > drivers/gpu/drm/i915/Kconfig | 1 + > drivers/gpu/drm/i915/gem/i915_gem_pages.c | 126 ++++++++++------------ > 2 files changed, 59 insertions(+), 68 deletions(-) > > diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig > index 9afa5c4a6bf006..1e1cb245fca778 100644 > --- a/drivers/gpu/drm/i915/Kconfig > +++ b/drivers/gpu/drm/i915/Kconfig > @@ -25,6 +25,7 @@ config DRM_I915 > select CRC32 > select SND_HDA_I915 if SND_HDA_CORE > select CEC_CORE if CEC_NOTIFIER > + select VMAP_PFN > help > Choose this option if you have a system that has "Intel Graphics > Media Accelerator" or "HD Graphics" integrated graphics, > diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c > index 6550c0bc824ea2..b519417667eb4b 100644 > --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c > +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c > @@ -232,34 +232,21 @@ int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj) > return err; > } > > -static inline pte_t iomap_pte(resource_size_t base, > - dma_addr_t offset, > - pgprot_t prot) > -{ > - return pte_mkspecial(pfn_pte((base + offset) >> PAGE_SHIFT, prot)); > -} > - > /* The 'mapping' part of i915_gem_object_pin_map() below */ > -static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > - enum i915_map_type type) > +static void *i915_gem_object_map_page(struct drm_i915_gem_object *obj, > + enum i915_map_type type) > { > - unsigned long n_pte = obj->base.size >> PAGE_SHIFT; > - struct sg_table *sgt = obj->mm.pages; > - pte_t *stack[32], **mem; > - struct vm_struct *area; > + unsigned long n_pages = obj->base.size >> PAGE_SHIFT, i; > + struct page *stack[32], **pages = stack, *page; > + struct sgt_iter iter; > pgprot_t pgprot; > + void *vaddr; > > - if (!i915_gem_object_has_struct_page(obj) && type != I915_MAP_WC) > - return NULL; > - > - if (GEM_WARN_ON(type == I915_MAP_WC && > - !static_cpu_has(X86_FEATURE_PAT))) > - return NULL; > - > - /* A single page can always be kmapped */ > - if (n_pte == 1 && type == I915_MAP_WB) { > - struct page *page = sg_page(sgt->sgl); > - > + switch (type) { > + default: > + MISSING_CASE(type); > + fallthrough; /* to use PAGE_KERNEL anyway */ > + case I915_MAP_WB: > /* > * On 32b, highmem using a finite set of indirect PTE (i.e. > * vmap) to provide virtual mappings of the high pages. > @@ -277,30 +264,8 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > * So if the page is beyond the 32b boundary, make an explicit > * vmap. > */ > - if (!PageHighMem(page)) > - return page_address(page); > - } > - > - mem = stack; > - if (n_pte > ARRAY_SIZE(stack)) { > - /* Too big for stack -- allocate temporary array instead */ > - mem = kvmalloc_array(n_pte, sizeof(*mem), GFP_KERNEL); > - if (!mem) > - return NULL; > - } > - > - area = alloc_vm_area(obj->base.size, mem); > - if (!area) { > - if (mem != stack) > - kvfree(mem); > - return NULL; > - } > - > - switch (type) { > - default: > - MISSING_CASE(type); > - fallthrough; /* to use PAGE_KERNEL anyway */ > - case I915_MAP_WB: > + if (n_pages == 1 && !PageHighMem(sg_page(obj->mm.pages->sgl))) > + return page_address(sg_page(obj->mm.pages->sgl)); > pgprot = PAGE_KERNEL; > break; > case I915_MAP_WC: > @@ -308,30 +273,49 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj, > break; > } > > - if (i915_gem_object_has_struct_page(obj)) { > - struct sgt_iter iter; > - struct page *page; > - pte_t **ptes = mem; > + if (n_pages > ARRAY_SIZE(stack)) { > + /* Too big for stack -- allocate temporary array instead */ > + pages = kvmalloc_array(n_pages, sizeof(*pages), GFP_KERNEL); > + if (!pages) > + return NULL; > + } > > - for_each_sgt_page(page, iter, sgt) > - **ptes++ = mk_pte(page, pgprot); > - } else { > - resource_size_t iomap; > - struct sgt_iter iter; > - pte_t **ptes = mem; > - dma_addr_t addr; > + i = 0; > + for_each_sgt_page(page, iter, obj->mm.pages) > + pages[i++] = page; > + vaddr = vmap(pages, n_pages, 0, pgprot); > + if (pages != stack) > + kvfree(pages); > + return vaddr; > +} > > - iomap = obj->mm.region->iomap.base; > - iomap -= obj->mm.region->region.start; > +static void *i915_gem_object_map_pfn(struct drm_i915_gem_object *obj, > + enum i915_map_type type) > +{ > + resource_size_t iomap = obj->mm.region->iomap.base - > + obj->mm.region->region.start; > + unsigned long n_pfn = obj->base.size >> PAGE_SHIFT; > + unsigned long stack[32], *pfns = stack, i; > + struct sgt_iter iter; > + dma_addr_t addr; > + void *vaddr; > + > + if (type != I915_MAP_WC) > + return NULL; > > - for_each_sgt_daddr(addr, iter, sgt) > - **ptes++ = iomap_pte(iomap, addr, pgprot); > + if (n_pfn > ARRAY_SIZE(stack)) { > + /* Too big for stack -- allocate temporary array instead */ > + pfns = kvmalloc_array(n_pfn, sizeof(*pfns), GFP_KERNEL); > + if (!pfns) > + return NULL; > } > > - if (mem != stack) > - kvfree(mem); > - > - return area->addr; > + for_each_sgt_daddr(addr, iter, obj->mm.pages) > + pfns[i++] = (iomap + addr) >> PAGE_SHIFT; Missing the i = 0 fix from Dan? > + vaddr = vmap_pfn(pfns, n_pfn, pgprot_writecombine(PAGE_KERNEL_IO)); > + if (pfns != stack) > + kvfree(pfns); > + return vaddr; > } > > /* get, pin, and map the pages of the object into kernel space */ > @@ -383,7 +367,13 @@ void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj, > } > > if (!ptr) { > - ptr = i915_gem_object_map(obj, type); > + if (GEM_WARN_ON(type == I915_MAP_WC && > + !static_cpu_has(X86_FEATURE_PAT))) > + ptr = NULL; > + else if (i915_gem_object_has_struct_page(obj)) > + ptr = i915_gem_object_map_page(obj, type); > + else > + ptr = i915_gem_object_map_pfn(obj, type); > if (!ptr) { > err = -ENOMEM; > goto err_unpin; > -- > 2.28.0 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx