From: Peter Zijlstra <peterz@infradead.org> To: Nicholas Piggin <npiggin@gmail.com> Cc: Will Deacon <will.deacon@arm.com>, linux-kernel@vger.kernel.org, benh@au1.ibm.com, torvalds@linux-foundation.org, catalin.marinas@arm.com, linux-arm-kernel@lists.infradead.org Subject: Re: [RFC PATCH 08/11] asm-generic/tlb: Track freeing of page-table directories in struct mmu_gather Date: Tue, 28 Aug 2018 15:48:42 +0200 [thread overview] Message-ID: <20180828134842.GQ24142@hirez.programming.kicks-ass.net> (raw) In-Reply-To: <20180828134638.GK24082@hirez.programming.kicks-ass.net> On Tue, Aug 28, 2018 at 03:46:38PM +0200, Peter Zijlstra wrote: > On Mon, Aug 27, 2018 at 02:44:57PM +1000, Nicholas Piggin wrote: > > > powerpc may be able to use the unmap granule thing to improve > > its page size dependent flushes, but it might prefer to go > > a different way and track start-end for different page sizes. > > I don't really see how tracking multiple ranges would help much with > THP. The ranges would end up being almost the same if there is a good > mix of page sizes. > > But something like: > > void tlb_flush_one(struct mmu_gather *tlb, unsigned long addr) > { > if (tlb->cleared_ptes && (addr << BITS_PER_LONG - PAGE_SHIFT)) > tblie_pte(addr); > if (tlb->cleared_pmds && (addr << BITS_PER_LONG - PMD_SHIFT)) > tlbie_pmd(addr); > if (tlb->cleared_puds && (addr << BITS_PER_LONG - PUD_SHIFT)) > tlbie_pud(addr); > } Sorry, those all should (of course) be !(addr << ...). > void tlb_flush_range(struct mmu_gather *tlb) > { > unsigned long stride = 1UL << tlb_get_unmap_shift(tlb); > unsigned long addr; > > for (addr = tlb->start; addr < tlb->end; addr += stride) > tlb_flush_one(tlb, addr); > > ptesync(); > } And one could; like x86 has; add a threshold above which you just kill the complete mm.
WARNING: multiple messages have this Message-ID (diff)
From: peterz@infradead.org (Peter Zijlstra) To: linux-arm-kernel@lists.infradead.org Subject: [RFC PATCH 08/11] asm-generic/tlb: Track freeing of page-table directories in struct mmu_gather Date: Tue, 28 Aug 2018 15:48:42 +0200 [thread overview] Message-ID: <20180828134842.GQ24142@hirez.programming.kicks-ass.net> (raw) In-Reply-To: <20180828134638.GK24082@hirez.programming.kicks-ass.net> On Tue, Aug 28, 2018 at 03:46:38PM +0200, Peter Zijlstra wrote: > On Mon, Aug 27, 2018 at 02:44:57PM +1000, Nicholas Piggin wrote: > > > powerpc may be able to use the unmap granule thing to improve > > its page size dependent flushes, but it might prefer to go > > a different way and track start-end for different page sizes. > > I don't really see how tracking multiple ranges would help much with > THP. The ranges would end up being almost the same if there is a good > mix of page sizes. > > But something like: > > void tlb_flush_one(struct mmu_gather *tlb, unsigned long addr) > { > if (tlb->cleared_ptes && (addr << BITS_PER_LONG - PAGE_SHIFT)) > tblie_pte(addr); > if (tlb->cleared_pmds && (addr << BITS_PER_LONG - PMD_SHIFT)) > tlbie_pmd(addr); > if (tlb->cleared_puds && (addr << BITS_PER_LONG - PUD_SHIFT)) > tlbie_pud(addr); > } Sorry, those all should (of course) be !(addr << ...). > void tlb_flush_range(struct mmu_gather *tlb) > { > unsigned long stride = 1UL << tlb_get_unmap_shift(tlb); > unsigned long addr; > > for (addr = tlb->start; addr < tlb->end; addr += stride) > tlb_flush_one(tlb, addr); > > ptesync(); > } And one could; like x86 has; add a threshold above which you just kill the complete mm.
next prev parent reply other threads:[~2018-08-28 13:48 UTC|newest] Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top 2018-08-24 15:52 [RFC PATCH 00/11] Avoid synchronous TLB invalidation for intermediate page-table entries on arm64 Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-24 15:52 ` [RFC PATCH 01/11] arm64: tlb: Use last-level invalidation in flush_tlb_kernel_range() Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-24 15:52 ` [RFC PATCH 02/11] arm64: tlb: Add DSB ISHST prior to TLBI in __flush_tlb_[kernel_]pgtable() Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-24 17:56 ` Peter Zijlstra 2018-08-24 17:56 ` Peter Zijlstra 2018-08-28 13:03 ` Will Deacon 2018-08-28 13:03 ` Will Deacon 2018-08-24 15:52 ` [RFC PATCH 03/11] arm64: pgtable: Implement p[mu]d_valid() and check in set_p[mu]d() Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-24 16:15 ` Linus Torvalds 2018-08-24 16:15 ` Linus Torvalds 2018-08-28 12:49 ` Will Deacon 2018-08-28 12:49 ` Will Deacon 2018-08-24 15:52 ` [RFC PATCH 04/11] arm64: tlb: Justify non-leaf invalidation in flush_tlb_range() Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-24 15:52 ` [RFC PATCH 05/11] arm64: tlbflush: Allow stride to be specified for __flush_tlb_range() Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-24 15:52 ` [RFC PATCH 06/11] arm64: tlb: Remove redundant !CONFIG_HAVE_RCU_TABLE_FREE code Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-24 15:52 ` [RFC PATCH 07/11] asm-generic/tlb: Guard with #ifdef CONFIG_MMU Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-24 15:52 ` [RFC PATCH 08/11] asm-generic/tlb: Track freeing of page-table directories in struct mmu_gather Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-27 4:44 ` Nicholas Piggin 2018-08-27 4:44 ` Nicholas Piggin 2018-08-28 13:46 ` Peter Zijlstra 2018-08-28 13:46 ` Peter Zijlstra 2018-08-28 13:48 ` Peter Zijlstra [this message] 2018-08-28 13:48 ` Peter Zijlstra 2018-08-28 14:12 ` Nicholas Piggin 2018-08-28 14:12 ` Nicholas Piggin 2018-08-24 15:52 ` [RFC PATCH 09/11] asm-generic/tlb: Track which levels of the page tables have been cleared Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-27 7:53 ` Peter Zijlstra 2018-08-27 7:53 ` Peter Zijlstra 2018-08-28 13:12 ` Will Deacon 2018-08-28 13:12 ` Will Deacon 2018-08-24 15:52 ` [RFC PATCH 10/11] arm64: tlb: Adjust stride and type of TLBI according to mmu_gather Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-24 15:52 ` [RFC PATCH 11/11] arm64: tlb: Avoid synchronous TLBIs when freeing page tables Will Deacon 2018-08-24 15:52 ` Will Deacon 2018-08-24 16:20 ` [RFC PATCH 00/11] Avoid synchronous TLB invalidation for intermediate page-table entries on arm64 Linus Torvalds 2018-08-24 16:20 ` Linus Torvalds 2018-08-26 10:56 ` Peter Zijlstra 2018-08-26 10:56 ` Peter Zijlstra 2018-09-04 18:38 ` Jon Masters 2018-09-04 18:38 ` Jon Masters 2018-09-05 12:28 ` Will Deacon 2018-09-05 12:28 ` Will Deacon 2018-09-07 6:36 ` Jon Masters 2018-09-07 6:36 ` Jon Masters 2018-09-13 15:53 ` Will Deacon 2018-09-13 15:53 ` Will Deacon 2018-09-13 16:53 ` Jon Masters 2018-09-13 16:53 ` Jon Masters
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=20180828134842.GQ24142@hirez.programming.kicks-ass.net \ --to=peterz@infradead.org \ --cc=benh@au1.ibm.com \ --cc=catalin.marinas@arm.com \ --cc=linux-arm-kernel@lists.infradead.org \ --cc=linux-kernel@vger.kernel.org \ --cc=npiggin@gmail.com \ --cc=torvalds@linux-foundation.org \ --cc=will.deacon@arm.com \ /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.