From patchwork Mon Jul 10 08:39:14 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Yicong Yang X-Patchwork-Id: 13306473 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 kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id ACD97C001DD for ; Mon, 10 Jul 2023 08:41:21 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 924FD6B0078; Mon, 10 Jul 2023 04:41:20 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 8D5056B007B; Mon, 10 Jul 2023 04:41:20 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 7768E6B007D; Mon, 10 Jul 2023 04:41:20 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0011.hostedemail.com [216.40.44.11]) by kanga.kvack.org (Postfix) with ESMTP id 6932B6B0078 for ; Mon, 10 Jul 2023 04:41:20 -0400 (EDT) Received: from smtpin08.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay08.hostedemail.com (Postfix) with ESMTP id 3ECDE1402E3 for ; Mon, 10 Jul 2023 08:41:20 +0000 (UTC) X-FDA: 80995057920.08.7F7FF78 Received: from szxga08-in.huawei.com (szxga08-in.huawei.com [45.249.212.255]) by imf19.hostedemail.com (Postfix) with ESMTP id 9CBE31A0011 for ; Mon, 10 Jul 2023 08:41:17 +0000 (UTC) Authentication-Results: imf19.hostedemail.com; dkim=none; spf=pass (imf19.hostedemail.com: domain of yangyicong@huawei.com designates 45.249.212.255 as permitted sender) smtp.mailfrom=yangyicong@huawei.com; dmarc=pass (policy=quarantine) header.from=huawei.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1688978478; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=yxOksWJmC1E9HL0ukTnBshvE/Q+99aLXO6wPG8oDNMM=; b=bH7huDljdk/Untld8uHjQ8t9G0YjtLf+PmL+IPJryekqjD9ROMDacOLemA+J14pHCIFKPl YyF0r7jAMXd80/sJWbCgMYz/Gl1tvH+mid7NLyRFu9CoJHz+MgoMmtw+8nU+eZP3aFPvl4 5ExGzWqIiDGztvhI+QVWYV/zSTKPLdo= ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1688978478; a=rsa-sha256; cv=none; b=kL14TSREHyZ000IbG/wTx4VqEO3yBNkyKFIjDdPOxepxSZQ1t1iLNek6or23IN9ZaDjGX7 brnbhyTxTeeSnLfbGr3VfpW5FwTsH5MtglLDBzpcp6Z92A/9VgJmJGDqhixqxFjRQw74z6 rtaIm+J8UFAO4bhWyFCR3r2C7VJ1puc= ARC-Authentication-Results: i=1; imf19.hostedemail.com; dkim=none; spf=pass (imf19.hostedemail.com: domain of yangyicong@huawei.com designates 45.249.212.255 as permitted sender) smtp.mailfrom=yangyicong@huawei.com; dmarc=pass (policy=quarantine) header.from=huawei.com Received: from canpemm500009.china.huawei.com (unknown [172.30.72.57]) by szxga08-in.huawei.com (SkyGuard) with ESMTP id 4QzyBd1pnCz1FDnF; Mon, 10 Jul 2023 16:40:41 +0800 (CST) Received: from localhost.localdomain (10.50.163.32) by canpemm500009.china.huawei.com (7.192.105.203) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2507.27; Mon, 10 Jul 2023 16:41:12 +0800 From: Yicong Yang To: , , , , , , , , , CC: , , , , , , , , , , , , , , , , , Barry Song <21cnbao@gmail.com>, , , , , Barry Song , Nadav Amit , Mel Gorman Subject: [PATCH v10 4/4] arm64: support batched/deferred tlb shootdown during page reclamation/migration Date: Mon, 10 Jul 2023 16:39:14 +0800 Message-ID: <20230710083914.18336-5-yangyicong@huawei.com> X-Mailer: git-send-email 2.31.0 In-Reply-To: <20230710083914.18336-1-yangyicong@huawei.com> References: <20230710083914.18336-1-yangyicong@huawei.com> MIME-Version: 1.0 X-Originating-IP: [10.50.163.32] X-ClientProxiedBy: dggems704-chm.china.huawei.com (10.3.19.181) To canpemm500009.china.huawei.com (7.192.105.203) X-CFilter-Loop: Reflected X-Rspamd-Queue-Id: 9CBE31A0011 X-Rspam-User: X-Rspamd-Server: rspam11 X-Stat-Signature: neppfbfeoo8jesf5ysjoteiwqm4xd41m X-HE-Tag: 1688978477-931373 X-HE-Meta: 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 3LIIAkEO cpPZ5aXtsQvki0M82cs8GhZwB9dLiWgHqYR7kTlhoXOsVmS3kyJQtVg1V47eEHDUn+EdKztJD6aJW7AADswGG3H6/lzuaXz701KY5RtTT5m9Vx14PSp1RR1jovA7H02DIbMfXGqqeO++28gg/nGl2i6SnscwC6yE4yyslNyDakFvmMybHeSu+wgzyRI3M7Cuqd2LwysDDz6SREjqt5Tf7bxpqw8y5Q8DqL0P/NMJ9Vip7gqWXVIltC36aWIGK13h7JTcf4FgR2oUQYy+ExhgamRBAdyy38kxukOe7ALNCsds/T7bWipe3vU7Gmo1T29iSzoZQXYFJDFcDiCERx667f/rY2oif/KHrSq7P7c3KcHRdd2Gh848MYCmQNyQQNihBC38Xngw+Dg2RYfohyZmd7ANeC3QMVwKWuuwdq7do39H0waLNGmVmG0cJUoGHr8e2rFdC/50Dr+lAZdsNzInNOl6XLDaJIyA8OFt4nxKu5sNLBoI= 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: From: Barry Song on x86, batched and deferred tlb shootdown has lead to 90% performance increase on tlb shootdown. on arm64, HW can do tlb shootdown without software IPI. But sync tlbi is still quite expensive. Even running a simplest program which requires swapout can prove this is true, #include #include #include #include int main() { #define SIZE (1 * 1024 * 1024) volatile unsigned char *p = mmap(NULL, SIZE, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0); memset(p, 0x88, SIZE); for (int k = 0; k < 10000; k++) { /* swap in */ for (int i = 0; i < SIZE; i += 4096) { (void)p[i]; } /* swap out */ madvise(p, SIZE, MADV_PAGEOUT); } } Perf result on snapdragon 888 with 8 cores by using zRAM as the swap block device. ~ # perf record taskset -c 4 ./a.out [ perf record: Woken up 10 times to write data ] [ perf record: Captured and wrote 2.297 MB perf.data (60084 samples) ] ~ # perf report # To display the perf.data header info, please use --header/--header-only options. # To display the perf.data header info, please use --header/--header-only options. # # # Total Lost Samples: 0 # # Samples: 60K of event 'cycles' # Event count (approx.): 35706225414 # # Overhead Command Shared Object Symbol # ........ ....... ................. ...... # 21.07% a.out [kernel.kallsyms] [k] _raw_spin_unlock_irq 8.23% a.out [kernel.kallsyms] [k] _raw_spin_unlock_irqrestore 6.67% a.out [kernel.kallsyms] [k] filemap_map_pages 6.16% a.out [kernel.kallsyms] [k] __zram_bvec_write 5.36% a.out [kernel.kallsyms] [k] ptep_clear_flush 3.71% a.out [kernel.kallsyms] [k] _raw_spin_lock 3.49% a.out [kernel.kallsyms] [k] memset64 1.63% a.out [kernel.kallsyms] [k] clear_page 1.42% a.out [kernel.kallsyms] [k] _raw_spin_unlock 1.26% a.out [kernel.kallsyms] [k] mod_zone_state.llvm.8525150236079521930 1.23% a.out [kernel.kallsyms] [k] xas_load 1.15% a.out [kernel.kallsyms] [k] zram_slot_lock ptep_clear_flush() takes 5.36% CPU in the micro-benchmark swapping in/out a page mapped by only one process. If the page is mapped by multiple processes, typically, like more than 100 on a phone, the overhead would be much higher as we have to run tlb flush 100 times for one single page. Plus, tlb flush overhead will increase with the number of CPU cores due to the bad scalability of tlb shootdown in HW, so those ARM64 servers should expect much higher overhead. Further perf annonate shows 95% cpu time of ptep_clear_flush is actually used by the final dsb() to wait for the completion of tlb flush. This provides us a very good chance to leverage the existing batched tlb in kernel. The minimum modification is that we only send async tlbi in the first stage and we send dsb while we have to sync in the second stage. With the above simplest micro benchmark, collapsed time to finish the program decreases around 5%. Typical collapsed time w/o patch: ~ # time taskset -c 4 ./a.out 0.21user 14.34system 0:14.69elapsed w/ patch: ~ # time taskset -c 4 ./a.out 0.22user 13.45system 0:13.80elapsed Also tested with benchmark in the commit on Kunpeng920 arm64 server and observed an improvement around 12.5% with command `time ./swap_bench`. w/o w/ real 0m13.460s 0m11.771s user 0m0.248s 0m0.279s sys 0m12.039s 0m11.458s Originally it's noticed a 16.99% overhead of ptep_clear_flush() which has been eliminated by this patch: [root@localhost yang]# perf record -- ./swap_bench && perf report [...] 16.99% swap_bench [kernel.kallsyms] [k] ptep_clear_flush It is tested on 4,8,128 CPU platforms and shows to be beneficial on large systems but may not have improvement on small systems like on a 4 CPU platform. So make this depends on EXPERT at this stage for tests on more small platforms. Also this patch improve the performance of page migration. Using pmbench and tries to migrate the pages of pmbench between node 0 and node 1 for 100 times for 1G memory, this patch decrease the time used around 20% (prev 18.338318910 sec after 13.981866350 sec) and saved the time used by ptep_clear_flush(). Cc: Anshuman Khandual Cc: Jonathan Corbet Cc: Nadav Amit Cc: Mel Gorman Tested-by: Yicong Yang Tested-by: Xin Hao Tested-by: Punit Agrawal Signed-off-by: Barry Song Signed-off-by: Yicong Yang Reviewed-by: Kefeng Wang Reviewed-by: Xin Hao Reviewed-by: Anshuman Khandual --- .../features/vm/TLB/arch-support.txt | 2 +- arch/arm64/Kconfig | 1 + arch/arm64/include/asm/tlbbatch.h | 12 +++++ arch/arm64/include/asm/tlbflush.h | 48 +++++++++++++++++-- 4 files changed, 59 insertions(+), 4 deletions(-) create mode 100644 arch/arm64/include/asm/tlbbatch.h diff --git a/Documentation/features/vm/TLB/arch-support.txt b/Documentation/features/vm/TLB/arch-support.txt index 7f049c251a79..76208db88f3b 100644 --- a/Documentation/features/vm/TLB/arch-support.txt +++ b/Documentation/features/vm/TLB/arch-support.txt @@ -9,7 +9,7 @@ | alpha: | TODO | | arc: | TODO | | arm: | TODO | - | arm64: | N/A | + | arm64: | ok | | csky: | TODO | | hexagon: | TODO | | ia64: | TODO | diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 7856c3a3e35a..f0ce8208c57f 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -96,6 +96,7 @@ config ARM64 select ARCH_SUPPORTS_NUMA_BALANCING select ARCH_SUPPORTS_PAGE_TABLE_CHECK select ARCH_SUPPORTS_PER_VMA_LOCK + select ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH if EXPERT select ARCH_WANT_COMPAT_IPC_PARSE_VERSION if COMPAT select ARCH_WANT_DEFAULT_BPF_JIT select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT diff --git a/arch/arm64/include/asm/tlbbatch.h b/arch/arm64/include/asm/tlbbatch.h new file mode 100644 index 000000000000..fedb0b87b8db --- /dev/null +++ b/arch/arm64/include/asm/tlbbatch.h @@ -0,0 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ARCH_ARM64_TLBBATCH_H +#define _ARCH_ARM64_TLBBATCH_H + +struct arch_tlbflush_unmap_batch { + /* + * For arm64, HW can do tlb shootdown, so we don't + * need to record cpumask for sending IPI + */ +}; + +#endif /* _ARCH_ARM64_TLBBATCH_H */ diff --git a/arch/arm64/include/asm/tlbflush.h b/arch/arm64/include/asm/tlbflush.h index 412a3b9a3c25..4bb9cec62e26 100644 --- a/arch/arm64/include/asm/tlbflush.h +++ b/arch/arm64/include/asm/tlbflush.h @@ -254,17 +254,23 @@ static inline void flush_tlb_mm(struct mm_struct *mm) dsb(ish); } -static inline void flush_tlb_page_nosync(struct vm_area_struct *vma, - unsigned long uaddr) +static inline void __flush_tlb_page_nosync(struct mm_struct *mm, + unsigned long uaddr) { unsigned long addr; dsb(ishst); - addr = __TLBI_VADDR(uaddr, ASID(vma->vm_mm)); + addr = __TLBI_VADDR(uaddr, ASID(mm)); __tlbi(vale1is, addr); __tlbi_user(vale1is, addr); } +static inline void flush_tlb_page_nosync(struct vm_area_struct *vma, + unsigned long uaddr) +{ + return __flush_tlb_page_nosync(vma->vm_mm, uaddr); +} + static inline void flush_tlb_page(struct vm_area_struct *vma, unsigned long uaddr) { @@ -272,6 +278,42 @@ static inline void flush_tlb_page(struct vm_area_struct *vma, dsb(ish); } +#ifdef CONFIG_ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH + +static inline bool arch_tlbbatch_should_defer(struct mm_struct *mm) +{ +#ifdef CONFIG_ARM64_WORKAROUND_REPEAT_TLBI + /* + * TLB flush deferral is not required on systems, which are affected with + * ARM64_WORKAROUND_REPEAT_TLBI, as __tlbi()/__tlbi_user() implementation + * will have two consecutive TLBI instructions with a dsb(ish) in between + * defeating the purpose (i.e save overall 'dsb ish' cost). + */ + if (unlikely(cpus_have_const_cap(ARM64_WORKAROUND_REPEAT_TLBI))) + return false; +#endif + return true; +} + +static inline void arch_tlbbatch_add_pending(struct arch_tlbflush_unmap_batch *batch, + struct mm_struct *mm, + unsigned long uaddr) +{ + __flush_tlb_page_nosync(mm, uaddr); +} + +static inline void arch_flush_tlb_batched_pending(struct mm_struct *mm) +{ + dsb(ish); +} + +static inline void arch_tlbbatch_flush(struct arch_tlbflush_unmap_batch *batch) +{ + dsb(ish); +} + +#endif /* CONFIG_ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH */ + /* * This is meant to avoid soft lock-ups on large TLB flushing ranges and not * necessarily a performance improvement.