From: Yong Wu <yong.wu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org> To: Robin Murphy <robin.murphy-5wv7dgnIgG8@public.gmane.org> Cc: srv_heupstream-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org, joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org, will.deacon-5wv7dgnIgG8@public.gmane.org, dianders-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org, iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, linux-mediatek-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org, linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org Subject: Re: [PATCH v2] iommu/dma: Finish optimising higher-order allocations Date: Thu, 21 Apr 2016 13:47:24 +0800 [thread overview] Message-ID: <1461217644.19705.10.camel@mhfsdcap03> (raw) In-Reply-To: <3e4572cb0a175061c1c4b436e3806ba9d7b9f199.1460563676.git.robin.murphy-5wv7dgnIgG8@public.gmane.org> On Wed, 2016-04-13 at 17:29 +0100, Robin Murphy wrote: > Now that we know exactly which page sizes our caller wants to use in the > given domain, we can restrict higher-order allocation attempts to just > those sizes, if any, and avoid wasting any time or effort on other sizes > which offer no benefit. In the same vein, this also lets us accommodate > a minimum order greater than 0 for special cases. > > Signed-off-by: Robin Murphy <robin.murphy-5wv7dgnIgG8@public.gmane.org> Hi Robin, Thanks very much for this patch. It works well on our MT8173. Tested-by: Yong Wu <yong.wu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org> > --- > > Just throwing this out as a quick solo update as I'm still expecting > discussion on the rest of the series. > [...] > while (count) { > struct page *page = NULL; > - int j; > + unsigned int order_size; > > /* > * Higher-order allocations are a convenience rather > * than a necessity, hence using __GFP_NORETRY until > - * falling back to single-page allocations. > + * falling back to minimum-order allocations. > */ > - for (order = min_t(unsigned int, order, __fls(count)); > - order > 0; order--) { > - page = alloc_pages(gfp | __GFP_NORETRY, order); > + for (order_mask &= (2U << __fls(count)) - 1; > + order_mask; order_mask &= ~order_size) { > + unsigned int order = __fls(order_mask); > + > + order_size = 1U << order; > + page = alloc_pages((order_mask - order_size) ? > + gfp | __GFP_NORETRY : gfp, order); > if (!page) > continue; > - if (PageCompound(page)) { > - if (!split_huge_page(page)) > - break; > - __free_pages(page, order); > - } else { > + if (!order) > + break; I also added this "if" in my old code. I don't know much about PageCompound and split_page, but from Will's suggestion[1], this "if" is unnecessary. [1]:http://lists.linuxfoundation.org/pipermail/iommu/2016-April/016422.html > + if (!PageCompound(page)) { > split_page(page, order); > break; > + } else if (!split_huge_page(page)) { > + break; > } > + __free_pages(page, order); > } > - if (!page) > - page = alloc_page(gfp); > if (!page) { > __iommu_dma_free_pages(pages, i); > return NULL; > } > - j = 1 << order; > - count -= j; > - while (j--) > + count -= order_size; > + while (order_size--) > pages[i++] = page++; > } [...]
WARNING: multiple messages have this Message-ID (diff)
From: yong.wu@mediatek.com (Yong Wu) To: linux-arm-kernel@lists.infradead.org Subject: [PATCH v2] iommu/dma: Finish optimising higher-order allocations Date: Thu, 21 Apr 2016 13:47:24 +0800 [thread overview] Message-ID: <1461217644.19705.10.camel@mhfsdcap03> (raw) In-Reply-To: <3e4572cb0a175061c1c4b436e3806ba9d7b9f199.1460563676.git.robin.murphy@arm.com> On Wed, 2016-04-13 at 17:29 +0100, Robin Murphy wrote: > Now that we know exactly which page sizes our caller wants to use in the > given domain, we can restrict higher-order allocation attempts to just > those sizes, if any, and avoid wasting any time or effort on other sizes > which offer no benefit. In the same vein, this also lets us accommodate > a minimum order greater than 0 for special cases. > > Signed-off-by: Robin Murphy <robin.murphy@arm.com> Hi Robin, Thanks very much for this patch. It works well on our MT8173. Tested-by: Yong Wu <yong.wu@mediatek.com> > --- > > Just throwing this out as a quick solo update as I'm still expecting > discussion on the rest of the series. > [...] > while (count) { > struct page *page = NULL; > - int j; > + unsigned int order_size; > > /* > * Higher-order allocations are a convenience rather > * than a necessity, hence using __GFP_NORETRY until > - * falling back to single-page allocations. > + * falling back to minimum-order allocations. > */ > - for (order = min_t(unsigned int, order, __fls(count)); > - order > 0; order--) { > - page = alloc_pages(gfp | __GFP_NORETRY, order); > + for (order_mask &= (2U << __fls(count)) - 1; > + order_mask; order_mask &= ~order_size) { > + unsigned int order = __fls(order_mask); > + > + order_size = 1U << order; > + page = alloc_pages((order_mask - order_size) ? > + gfp | __GFP_NORETRY : gfp, order); > if (!page) > continue; > - if (PageCompound(page)) { > - if (!split_huge_page(page)) > - break; > - __free_pages(page, order); > - } else { > + if (!order) > + break; I also added this "if" in my old code. I don't know much about PageCompound and split_page, but from Will's suggestion[1], this "if" is unnecessary. [1]:http://lists.linuxfoundation.org/pipermail/iommu/2016-April/016422.html > + if (!PageCompound(page)) { > split_page(page, order); > break; > + } else if (!split_huge_page(page)) { > + break; > } > + __free_pages(page, order); > } > - if (!page) > - page = alloc_page(gfp); > if (!page) { > __iommu_dma_free_pages(pages, i); > return NULL; > } > - j = 1 << order; > - count -= j; > - while (j--) > + count -= order_size; > + while (order_size--) > pages[i++] = page++; > } [...]
next prev parent reply other threads:[~2016-04-21 5:47 UTC|newest] Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top 2016-04-07 17:42 [PATCH 0/5] Introduce per-domain page sizes Robin Murphy 2016-04-07 17:42 ` Robin Murphy [not found] ` <cover.1460048991.git.robin.murphy-5wv7dgnIgG8@public.gmane.org> 2016-04-07 17:42 ` [PATCH 1/5] iommu: remove unused priv field from struct iommu_ops Robin Murphy 2016-04-07 17:42 ` Robin Murphy 2016-04-07 17:42 ` [PATCH 2/5] iommu: of: enforce const-ness of " Robin Murphy 2016-04-07 17:42 ` Robin Murphy 2016-04-07 17:42 ` [PATCH 3/5] iommu: Allow selecting page sizes per domain Robin Murphy 2016-04-07 17:42 ` Robin Murphy 2016-04-07 17:42 ` [PATCH 4/5] iommu/dma: Finish optimising higher-order allocations Robin Murphy 2016-04-07 17:42 ` Robin Murphy [not found] ` <89763f6b1ac684c3d8712e38760bec55b7885e3b.1460048991.git.robin.murphy-5wv7dgnIgG8@public.gmane.org> 2016-04-08 5:32 ` Yong Wu 2016-04-08 5:32 ` Yong Wu 2016-04-08 16:33 ` Robin Murphy 2016-04-08 16:33 ` Robin Murphy 2016-04-13 16:29 ` [PATCH v2] " Robin Murphy 2016-04-13 16:29 ` Robin Murphy [not found] ` <3e4572cb0a175061c1c4b436e3806ba9d7b9f199.1460563676.git.robin.murphy-5wv7dgnIgG8@public.gmane.org> 2016-04-21 5:47 ` Yong Wu [this message] 2016-04-21 5:47 ` Yong Wu 2016-04-07 17:42 ` [PATCH 5/5] iommu/arm-smmu: Use per-domain page sizes Robin Murphy 2016-04-07 17:42 ` Robin Murphy 2016-04-21 16:38 ` [PATCH 0/5] Introduce " Will Deacon 2016-04-21 16:38 ` Will Deacon 2016-05-09 11:21 ` Joerg Roedel 2016-05-09 11:21 ` Joerg Roedel [not found] ` <20160509112138.GB13275-zLv9SwRftAIdnm+yROfE0A@public.gmane.org> 2016-05-09 11:45 ` Robin Murphy 2016-05-09 11:45 ` Robin Murphy [not found] ` <57307863.1070706-5wv7dgnIgG8@public.gmane.org> 2016-05-09 14:51 ` Joerg Roedel 2016-05-09 14:51 ` Joerg Roedel [not found] ` <20160509145157.GD13971-zLv9SwRftAIdnm+yROfE0A@public.gmane.org> 2016-05-09 15:18 ` Robin Murphy 2016-05-09 15:18 ` Robin Murphy 2016-05-09 15:50 ` Joerg Roedel 2016-05-09 15:50 ` Joerg Roedel [not found] ` <ea520b8c72b5a72a1731bd35f6e3e50872fe6764.1460048991.git.robin.murphy-5wv7dgnIgG8@public.gmane.org> 2016-05-09 16:20 ` [PATCH v2] iommu/arm-smmu: Use " Robin Murphy 2016-05-09 16:20 ` Robin Murphy [not found] ` <112fc0e5f9bbe08007778b8438b35025d8e876a4.1462810410.git.robin.murphy-5wv7dgnIgG8@public.gmane.org> 2016-05-10 9:45 ` Joerg Roedel 2016-05-10 9:45 ` Joerg Roedel
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=1461217644.19705.10.camel@mhfsdcap03 \ --to=yong.wu-nus5lvnupcjwk0htik3j/w@public.gmane.org \ --cc=dianders-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org \ --cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \ --cc=joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org \ --cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \ --cc=linux-mediatek-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \ --cc=robin.murphy-5wv7dgnIgG8@public.gmane.org \ --cc=srv_heupstream-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org \ --cc=will.deacon-5wv7dgnIgG8@public.gmane.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.