From patchwork Fri Apr 12 06:47:50 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Kefeng Wang X-Patchwork-Id: 13626906 Received: from szxga06-in.huawei.com (szxga06-in.huawei.com [45.249.212.32]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id F18E02233B for ; Fri, 12 Apr 2024 06:20:49 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=45.249.212.32 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1712902852; cv=none; b=Ope1vhATa1lV8TzCqGKyJOHhyT0/R+MPrrHA7D5OSe9Pat0cTpZduJJbv4vYjCWzWi32ADmXABZ2z6bKXCOQzIERq78lqh/D6XbnoXmHdDF2056kuFWSsAw5e5/b4mQ5d9Y8nVa32A8Vl4Kv0gUKbLnwj/nwAKL/ogBabjRYoeE= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1712902852; c=relaxed/simple; bh=tgTu50zUtcu44+weOq7O3NyrqRzcEXudH59DnfTtzEs=; h=From:To:CC:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=EVfRWkAOBLMk9dLGl34rGn9gY5F4iNPyzEe4/+KYo6BXbtqT4Q+n59Q6Ko8KZbdZLdhauHY5rC7mZPs7jIQOxPJPF+GPX07Vdl7XqRBdKXcZcdzcLk+fhRYHOwpPe6cZ5eX6q8WFLf12vwMP2sHY/m1gnMgAUC9xBArx6MEAnAE= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=huawei.com; spf=pass smtp.mailfrom=huawei.com; arc=none smtp.client-ip=45.249.212.32 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=huawei.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=huawei.com Received: from mail.maildlp.com (unknown [172.19.88.163]) by szxga06-in.huawei.com (SkyGuard) with ESMTP id 4VG5y96MQFz1wrMk; Fri, 12 Apr 2024 14:19:45 +0800 (CST) Received: from dggpemm100001.china.huawei.com (unknown [7.185.36.93]) by mail.maildlp.com (Postfix) with ESMTPS id 23B0318001A; Fri, 12 Apr 2024 14:20:42 +0800 (CST) Received: from localhost.localdomain.localdomain (10.175.113.25) by dggpemm100001.china.huawei.com (7.185.36.93) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.1.2507.35; Fri, 12 Apr 2024 14:20:41 +0800 From: Kefeng Wang To: Andrew Morton CC: "Matthew Wilcox (Oracle)" , , , Kefeng Wang Subject: [PATCH v3 1/2] mm: move mm counter updating out of set_pte_range() Date: Fri, 12 Apr 2024 14:47:50 +0800 Message-ID: <20240412064751.119015-2-wangkefeng.wang@huawei.com> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20240412064751.119015-1-wangkefeng.wang@huawei.com> References: <20240412064751.119015-1-wangkefeng.wang@huawei.com> Precedence: bulk X-Mailing-List: linux-fsdevel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-ClientProxiedBy: dggems701-chm.china.huawei.com (10.3.19.178) To dggpemm100001.china.huawei.com (7.185.36.93) In order to support batch mm counter updating in filemap_map_pages(), move mm counter updating out of set_pte_range(), the folios are file from filemap, and distinguish folios by vmf->flags and vma->vm_flags from another caller finish_fault(). Signed-off-by: Kefeng Wang --- mm/filemap.c | 4 ++++ mm/memory.c | 8 +++++--- 2 files changed, 9 insertions(+), 3 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index 92e2d43e4c9d..04b813f0146c 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -3540,6 +3540,8 @@ static vm_fault_t filemap_map_folio_range(struct vm_fault *vmf, skip: if (count) { set_pte_range(vmf, folio, page, count, addr); + add_mm_counter(vmf->vma->vm_mm, mm_counter_file(folio), + count); folio_ref_add(folio, count); if (in_range(vmf->address, addr, count * PAGE_SIZE)) ret = VM_FAULT_NOPAGE; @@ -3554,6 +3556,7 @@ static vm_fault_t filemap_map_folio_range(struct vm_fault *vmf, if (count) { set_pte_range(vmf, folio, page, count, addr); + add_mm_counter(vmf->vma->vm_mm, mm_counter_file(folio), count); folio_ref_add(folio, count); if (in_range(vmf->address, addr, count * PAGE_SIZE)) ret = VM_FAULT_NOPAGE; @@ -3590,6 +3593,7 @@ static vm_fault_t filemap_map_order0_folio(struct vm_fault *vmf, ret = VM_FAULT_NOPAGE; set_pte_range(vmf, folio, page, 1, addr); + add_mm_counter(vmf->vma->vm_mm, mm_counter_file(folio), 1); folio_ref_inc(folio); return ret; diff --git a/mm/memory.c b/mm/memory.c index 78422d1c7381..fdfe965f32e4 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4685,12 +4685,10 @@ void set_pte_range(struct vm_fault *vmf, struct folio *folio, entry = pte_mkuffd_wp(entry); /* copy-on-write page */ if (write && !(vma->vm_flags & VM_SHARED)) { - add_mm_counter(vma->vm_mm, MM_ANONPAGES, nr); VM_BUG_ON_FOLIO(nr != 1, folio); folio_add_new_anon_rmap(folio, vma, addr); folio_add_lru_vma(folio, vma); } else { - add_mm_counter(vma->vm_mm, mm_counter_file(folio), nr); folio_add_file_rmap_ptes(folio, page, nr, vma); } set_ptes(vma->vm_mm, addr, vmf->pte, entry, nr); @@ -4727,9 +4725,11 @@ vm_fault_t finish_fault(struct vm_fault *vmf) struct vm_area_struct *vma = vmf->vma; struct page *page; vm_fault_t ret; + bool is_cow = (vmf->flags & FAULT_FLAG_WRITE) && + !(vma->vm_flags & VM_SHARED); /* Did we COW the page? */ - if ((vmf->flags & FAULT_FLAG_WRITE) && !(vma->vm_flags & VM_SHARED)) + if (is_cow) page = vmf->cow_page; else page = vmf->page; @@ -4765,8 +4765,10 @@ vm_fault_t finish_fault(struct vm_fault *vmf) /* Re-check under ptl */ if (likely(!vmf_pte_changed(vmf))) { struct folio *folio = page_folio(page); + int type = is_cow ? MM_ANONPAGES : mm_counter_file(folio); set_pte_range(vmf, folio, page, 1, vmf->address); + add_mm_counter(vma->vm_mm, type, 1); ret = 0; } else { update_mmu_tlb(vma, vmf->address, vmf->pte); From patchwork Fri Apr 12 06:47:51 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Kefeng Wang X-Patchwork-Id: 13626904 Received: from szxga04-in.huawei.com (szxga04-in.huawei.com [45.249.212.190]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id EA8853234 for ; Fri, 12 Apr 2024 06:20:44 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=45.249.212.190 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1712902848; cv=none; b=RXwB25HEbC0pG3jyVpb9Cqc8HXPkBPi8XitqvFH4kVT4gkS7nlHnW+tlHY5wUF0MJLu7de1ScKk+UgVORXtUBZRbCMdbcoTO6/vqgK/JCji8vysjzC5iH1opr7rD1zLCQBqdmpI6d/wXgXpvxCPNoKVbde+nSyctSCsE44U1ETo= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1712902848; c=relaxed/simple; bh=SRcZxI3A1FdS5Kz4sdiEuBlvhmd1re+y2WTrQM0qfAg=; h=From:To:CC:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=L8uGKNLij4ipnFSKQn+cLEns231/8OON7g3PYRDqPGO2wK/bqG7xv35BDvCWtj0DXC0sec0Z4+zrgde3xrWym43mMQbXynOrMA3bUuuZcp75NPQj5d0RIqQePaiRb3iOgbA3VQKQphSQvg5/KLFaAnkcmzdUgXueuzadwi+mlFc= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=huawei.com; spf=pass smtp.mailfrom=huawei.com; arc=none smtp.client-ip=45.249.212.190 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=huawei.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=huawei.com Received: from mail.maildlp.com (unknown [172.19.162.112]) by szxga04-in.huawei.com (SkyGuard) with ESMTP id 4VG5wc6D5cz1ynKD; Fri, 12 Apr 2024 14:18:24 +0800 (CST) Received: from dggpemm100001.china.huawei.com (unknown [7.185.36.93]) by mail.maildlp.com (Postfix) with ESMTPS id 85E5B140554; Fri, 12 Apr 2024 14:20:42 +0800 (CST) Received: from localhost.localdomain.localdomain (10.175.113.25) by dggpemm100001.china.huawei.com (7.185.36.93) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.1.2507.35; Fri, 12 Apr 2024 14:20:42 +0800 From: Kefeng Wang To: Andrew Morton CC: "Matthew Wilcox (Oracle)" , , , Kefeng Wang Subject: [PATCH v3 2/2] mm: filemap: batch mm counter updating in filemap_map_pages() Date: Fri, 12 Apr 2024 14:47:51 +0800 Message-ID: <20240412064751.119015-3-wangkefeng.wang@huawei.com> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20240412064751.119015-1-wangkefeng.wang@huawei.com> References: <20240412064751.119015-1-wangkefeng.wang@huawei.com> Precedence: bulk X-Mailing-List: linux-fsdevel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-ClientProxiedBy: dggems701-chm.china.huawei.com (10.3.19.178) To dggpemm100001.china.huawei.com (7.185.36.93) Like copy_pte_range()/zap_pte_range(), make mm counter batch updating in filemap_map_pages(), since folios type are same(MM_SHMEMPAGES or MM_FILEPAGES) in filemap_map_pages(), only check the first folio type is enough, the 'lat_pagefault -P 1 file' test from lmbench shows 12% improvement, and the percpu_counter_add_batch() is gone from perf flame graph. Signed-off-by: Kefeng Wang --- mm/filemap.c | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index 04b813f0146c..531af4acc667 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -3506,7 +3506,7 @@ static struct folio *next_uptodate_folio(struct xa_state *xas, static vm_fault_t filemap_map_folio_range(struct vm_fault *vmf, struct folio *folio, unsigned long start, unsigned long addr, unsigned int nr_pages, - unsigned int *mmap_miss) + unsigned long *rss, unsigned int *mmap_miss) { vm_fault_t ret = 0; struct page *page = folio_page(folio, start); @@ -3540,8 +3540,7 @@ static vm_fault_t filemap_map_folio_range(struct vm_fault *vmf, skip: if (count) { set_pte_range(vmf, folio, page, count, addr); - add_mm_counter(vmf->vma->vm_mm, mm_counter_file(folio), - count); + *rss += count; folio_ref_add(folio, count); if (in_range(vmf->address, addr, count * PAGE_SIZE)) ret = VM_FAULT_NOPAGE; @@ -3556,7 +3555,7 @@ static vm_fault_t filemap_map_folio_range(struct vm_fault *vmf, if (count) { set_pte_range(vmf, folio, page, count, addr); - add_mm_counter(vmf->vma->vm_mm, mm_counter_file(folio), count); + *rss += count; folio_ref_add(folio, count); if (in_range(vmf->address, addr, count * PAGE_SIZE)) ret = VM_FAULT_NOPAGE; @@ -3569,7 +3568,7 @@ static vm_fault_t filemap_map_folio_range(struct vm_fault *vmf, static vm_fault_t filemap_map_order0_folio(struct vm_fault *vmf, struct folio *folio, unsigned long addr, - unsigned int *mmap_miss) + unsigned long *rss, unsigned int *mmap_miss) { vm_fault_t ret = 0; struct page *page = &folio->page; @@ -3593,7 +3592,7 @@ static vm_fault_t filemap_map_order0_folio(struct vm_fault *vmf, ret = VM_FAULT_NOPAGE; set_pte_range(vmf, folio, page, 1, addr); - add_mm_counter(vmf->vma->vm_mm, mm_counter_file(folio), 1); + (*rss)++; folio_ref_inc(folio); return ret; @@ -3610,7 +3609,8 @@ vm_fault_t filemap_map_pages(struct vm_fault *vmf, XA_STATE(xas, &mapping->i_pages, start_pgoff); struct folio *folio; vm_fault_t ret = 0; - unsigned int nr_pages = 0, mmap_miss = 0, mmap_miss_saved; + unsigned long rss = 0; + unsigned int nr_pages = 0, mmap_miss = 0, mmap_miss_saved, folio_type; rcu_read_lock(); folio = next_uptodate_folio(&xas, mapping, end_pgoff); @@ -3629,6 +3629,8 @@ vm_fault_t filemap_map_pages(struct vm_fault *vmf, folio_put(folio); goto out; } + + folio_type = mm_counter_file(folio); do { unsigned long end; @@ -3640,15 +3642,16 @@ vm_fault_t filemap_map_pages(struct vm_fault *vmf, if (!folio_test_large(folio)) ret |= filemap_map_order0_folio(vmf, - folio, addr, &mmap_miss); + folio, addr, &rss, &mmap_miss); else ret |= filemap_map_folio_range(vmf, folio, xas.xa_index - folio->index, addr, - nr_pages, &mmap_miss); + nr_pages, &rss, &mmap_miss); folio_unlock(folio); folio_put(folio); } while ((folio = next_uptodate_folio(&xas, mapping, end_pgoff)) != NULL); + add_mm_counter(vma->vm_mm, folio_type, rss); pte_unmap_unlock(vmf->pte, vmf->ptl); out: rcu_read_unlock();