From patchwork Mon Jun 14 21:19:01 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Johannes Weiner X-Patchwork-Id: 12320037 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=-16.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED,USER_AGENT_GIT 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 4B909C49361 for ; Mon, 14 Jun 2021 21:19:13 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 262606128B for ; Mon, 14 Jun 2021 21:19:13 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234855AbhFNVVO (ORCPT ); Mon, 14 Jun 2021 17:21:14 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:33648 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234706AbhFNVVN (ORCPT ); Mon, 14 Jun 2021 17:21:13 -0400 Received: from mail-qk1-x72e.google.com (mail-qk1-x72e.google.com [IPv6:2607:f8b0:4864:20::72e]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 273C5C061574 for ; Mon, 14 Jun 2021 14:19:10 -0700 (PDT) Received: by mail-qk1-x72e.google.com with SMTP id c18so24822939qkc.11 for ; Mon, 14 Jun 2021 14:19:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cmpxchg-org.20150623.gappssmtp.com; s=20150623; h=from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding; bh=Sgb+fuaFTgoZ/okWwz0fCGk2erI6Ba+62W428kydMsA=; b=cVWUH0f1+aGvFTLqzf7RJTBR7K+/orq4vCRKi2GARBla0mItJiAjqyUPfz33yCFaI+ d0JyTzYOR/pT75cxDCwyBcC0FWBIWv05whp2rKIiIhZOQO2mzw6aQcsAwD6UhGJqu6gR 1ygYrnw9gO6hV74fvHExc+Ffz2U3L6lPZXvn3btl4L55IRHmMKjR8ZNQRvkNQssayzvg TWMG+irrP+wXze7QsjXXVzJOzyK0xtmZ51LYuG0lldGynOKFf74cCyTR+OQjf4RVD1sq isP6Zq5/uhz2bHgliaTpKb8krNnXDulKBT0ULL4UM2pJrlJ/q50J0vCzTfltNPAVV+pG RKNQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding; bh=Sgb+fuaFTgoZ/okWwz0fCGk2erI6Ba+62W428kydMsA=; b=UJyFh7wxS4kslTMOaDCGHdPL/Z13b3rNhixUvf2ETyJwFuqBpMc2g3kDUazGTns0I8 IPYueqCN3x1uWGOC866vjEMfWvC1oUjfqobvCS0UniQzIOjzj5PPxOaNnU+q2DOlRuMW SpThMqBZSQ7QUJXq7WxxK+1R/uk5nYXpbvhYuMhIGfB5/bDx8VljsmPTMBlF6EYj08y5 aZImsCA2fuDcs26rEkb/qCXCNnUTovoyAZC4cGzO2AWa8skAbO0lmr1nSmhsfNsb9+Tb FYQgAF9Xd68cQTltV3t8dkVfq5X70NnJrNnKlTzfPQR7OxTR/tUxph9jXqFP+Nuo+LvP /kNg== X-Gm-Message-State: AOAM533kHz5FlN2G3rg5BqnTOBWRZKiH3Z2oEDLEnKxuju3NtG94zP1y VHd/vcNpSQfTd4jN2tLjRIaw1w== X-Google-Smtp-Source: ABdhPJzSF4akbF9UeyuNlv52c3S5EjiIMmK/poM1yMN0hqxBsN6pWFJcNTaVJKemRIyhzASAXKFzzQ== X-Received: by 2002:ae9:f205:: with SMTP id m5mr12202354qkg.412.1623705545942; Mon, 14 Jun 2021 14:19:05 -0700 (PDT) Received: from localhost (70.44.39.90.res-cmts.bus.ptd.net. [70.44.39.90]) by smtp.gmail.com with ESMTPSA id q16sm10895592qkq.121.2021.06.14.14.19.04 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 14 Jun 2021 14:19:05 -0700 (PDT) From: Johannes Weiner To: Andrew Morton Cc: Roman Gushchin , Tejun Heo , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, kernel-team@fb.com Subject: [PATCH 1/4] mm: remove irqsave/restore locking from contexts with irqs enabled Date: Mon, 14 Jun 2021 17:19:01 -0400 Message-Id: <20210614211904.14420-1-hannes@cmpxchg.org> X-Mailer: git-send-email 2.32.0 MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org The page cache deletion paths all have interrupts enabled, so no need to use irqsafe/irqrestore locking variants. They used to have irqs disabled by the memcg lock added in commit c4843a7593a9 ("memcg: add per cgroup dirty page accounting"), but that has since been replaced by memcg taking the page lock instead, commit 0a31bc97c80c ("mm: memcontrol: rewrite uncharge AP"). Signed-off-by: Johannes Weiner --- mm/filemap.c | 15 ++++++--------- mm/truncate.c | 8 +++----- mm/vmscan.c | 9 ++++----- 3 files changed, 13 insertions(+), 19 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index ba1068a1837f..819d2589abef 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -258,12 +258,11 @@ static void page_cache_free_page(struct address_space *mapping, void delete_from_page_cache(struct page *page) { struct address_space *mapping = page_mapping(page); - unsigned long flags; BUG_ON(!PageLocked(page)); - xa_lock_irqsave(&mapping->i_pages, flags); + xa_lock_irq(&mapping->i_pages); __delete_from_page_cache(page, NULL); - xa_unlock_irqrestore(&mapping->i_pages, flags); + xa_unlock_irq(&mapping->i_pages); page_cache_free_page(mapping, page); } @@ -335,19 +334,18 @@ void delete_from_page_cache_batch(struct address_space *mapping, struct pagevec *pvec) { int i; - unsigned long flags; if (!pagevec_count(pvec)) return; - xa_lock_irqsave(&mapping->i_pages, flags); + xa_lock_irq(&mapping->i_pages); for (i = 0; i < pagevec_count(pvec); i++) { trace_mm_filemap_delete_from_page_cache(pvec->pages[i]); unaccount_page_cache_page(mapping, pvec->pages[i]); } page_cache_delete_batch(mapping, pvec); - xa_unlock_irqrestore(&mapping->i_pages, flags); + xa_unlock_irq(&mapping->i_pages); for (i = 0; i < pagevec_count(pvec); i++) page_cache_free_page(mapping, pvec->pages[i]); @@ -821,7 +819,6 @@ void replace_page_cache_page(struct page *old, struct page *new) void (*freepage)(struct page *) = mapping->a_ops->freepage; pgoff_t offset = old->index; XA_STATE(xas, &mapping->i_pages, offset); - unsigned long flags; VM_BUG_ON_PAGE(!PageLocked(old), old); VM_BUG_ON_PAGE(!PageLocked(new), new); @@ -833,7 +830,7 @@ void replace_page_cache_page(struct page *old, struct page *new) mem_cgroup_migrate(old, new); - xas_lock_irqsave(&xas, flags); + xas_lock_irq(&xas); xas_store(&xas, new); old->mapping = NULL; @@ -846,7 +843,7 @@ void replace_page_cache_page(struct page *old, struct page *new) __dec_lruvec_page_state(old, NR_SHMEM); if (PageSwapBacked(new)) __inc_lruvec_page_state(new, NR_SHMEM); - xas_unlock_irqrestore(&xas, flags); + xas_unlock_irq(&xas); if (freepage) freepage(old); put_page(old); diff --git a/mm/truncate.c b/mm/truncate.c index 57a618c4a0d6..b92b86222625 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -563,21 +563,19 @@ void invalidate_mapping_pagevec(struct address_space *mapping, static int invalidate_complete_page2(struct address_space *mapping, struct page *page) { - unsigned long flags; - if (page->mapping != mapping) return 0; if (page_has_private(page) && !try_to_release_page(page, GFP_KERNEL)) return 0; - xa_lock_irqsave(&mapping->i_pages, flags); + xa_lock_irq(&mapping->i_pages); if (PageDirty(page)) goto failed; BUG_ON(page_has_private(page)); __delete_from_page_cache(page, NULL); - xa_unlock_irqrestore(&mapping->i_pages, flags); + xa_unlock_irq(&mapping->i_pages); if (mapping->a_ops->freepage) mapping->a_ops->freepage(page); @@ -585,7 +583,7 @@ invalidate_complete_page2(struct address_space *mapping, struct page *page) put_page(page); /* pagecache ref */ return 1; failed: - xa_unlock_irqrestore(&mapping->i_pages, flags); + xa_unlock_irq(&mapping->i_pages); return 0; } diff --git a/mm/vmscan.c b/mm/vmscan.c index eb314525c889..cc5d7cd75935 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1049,14 +1049,13 @@ static pageout_t pageout(struct page *page, struct address_space *mapping) static int __remove_mapping(struct address_space *mapping, struct page *page, bool reclaimed, struct mem_cgroup *target_memcg) { - unsigned long flags; int refcount; void *shadow = NULL; BUG_ON(!PageLocked(page)); BUG_ON(mapping != page_mapping(page)); - xa_lock_irqsave(&mapping->i_pages, flags); + xa_lock_irq(&mapping->i_pages); /* * The non racy check for a busy page. * @@ -1097,7 +1096,7 @@ static int __remove_mapping(struct address_space *mapping, struct page *page, if (reclaimed && !mapping_exiting(mapping)) shadow = workingset_eviction(page, target_memcg); __delete_from_swap_cache(page, swap, shadow); - xa_unlock_irqrestore(&mapping->i_pages, flags); + xa_unlock_irq(&mapping->i_pages); put_swap_page(page, swap); } else { void (*freepage)(struct page *); @@ -1123,7 +1122,7 @@ static int __remove_mapping(struct address_space *mapping, struct page *page, !mapping_exiting(mapping) && !dax_mapping(mapping)) shadow = workingset_eviction(page, target_memcg); __delete_from_page_cache(page, shadow); - xa_unlock_irqrestore(&mapping->i_pages, flags); + xa_unlock_irq(&mapping->i_pages); if (freepage != NULL) freepage(page); @@ -1132,7 +1131,7 @@ static int __remove_mapping(struct address_space *mapping, struct page *page, return 1; cannot_free: - xa_unlock_irqrestore(&mapping->i_pages, flags); + xa_unlock_irq(&mapping->i_pages); return 0; } From patchwork Mon Jun 14 21:19:02 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Johannes Weiner X-Patchwork-Id: 12320035 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=-16.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED,USER_AGENT_GIT 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 BAAF9C2B9F4 for ; Mon, 14 Jun 2021 21:19:09 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 980C8611CE for ; Mon, 14 Jun 2021 21:19:09 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234157AbhFNVVM (ORCPT ); Mon, 14 Jun 2021 17:21:12 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:33634 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230081AbhFNVVL (ORCPT ); Mon, 14 Jun 2021 17:21:11 -0400 Received: from mail-qk1-x733.google.com (mail-qk1-x733.google.com [IPv6:2607:f8b0:4864:20::733]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 411D9C061574 for ; Mon, 14 Jun 2021 14:19:08 -0700 (PDT) Received: by mail-qk1-x733.google.com with SMTP id i68so36351075qke.3 for ; Mon, 14 Jun 2021 14:19:07 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cmpxchg-org.20150623.gappssmtp.com; s=20150623; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=kPFT6nLLi1vvs4YFnK7DHOeGhzIW1Fs6DFQ1T0gCrck=; b=biiUnj1ExqttolBbEAb89QL8bsuWSOP8ozPOvpGyIPVXaX48+dqGxyTo+1pSUVs7xJ DOM157n2vskD+MQ7M6Wa2ZyYhIffBRELYKQfSKi6vCy6qJb5yq2/FsmsOx3G9MErNhE5 AckuooFjPNHwzfRp7zu4otoudRs+NTd3qiCWoDG9CwIpCJxBaNR4gt/3xYyuQVyhD6OG 1UVooRbjWx3DLZnqMCgS4WVnS93O2FIHISxYKfYVa51c5PITxwNolcDZfLaMdSaftBNp RqHBG2wZ8AvQf0JWzVViLZjJbtEaow1MSGLmMKdErHotkPeWlY2SbSgzZPJdDRltjrUm Z4ZQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=kPFT6nLLi1vvs4YFnK7DHOeGhzIW1Fs6DFQ1T0gCrck=; b=rmcnMq2JDCgYshh5zTanyTbT7XAOhxkqaB2VBrh+gdxNXzpPjSy/pn3GhxVnwHUP6S Uk9EXpkxEh+7dXsEW/JYaq7VNfmVj+tbzFrmw1MpC0iDxczwx4ypO7TOsL1kTvOdAvlZ ON91+i3AFdSrwb6hHSmNKX438Gg3JqQnuq05vTH52B2Dc3AH9h3SIn8dnLlK89A1O4IP jjCDGZp8yVIsdm+dJ7dc4//Ywlsd57BLSK1YzLKwYreVhdOKchL5qaWrlV6ICV8KLcPy EtFExx7jqp3GbxO8gje0z0BILnfvECb3c/pk6F4ev7iCYay8Tkx85tS9YFdwfLh8BQsQ SVUA== X-Gm-Message-State: AOAM533aC3vzW5LCNKuC9PyhfJuVzjVc2sZDJrhCwLRieAwMCcH45Tp8 bR315cRnINOhsUtpJdBk2vXNNg== X-Google-Smtp-Source: ABdhPJzvbqbfUECNW/9wGSUb5hxtP8w89hrKHvoC2k0yCleVRDRQ6BXu2QO6aCEMxPYVsmeHcsi6pg== X-Received: by 2002:a37:814:: with SMTP id 20mr18822511qki.243.1623705547189; Mon, 14 Jun 2021 14:19:07 -0700 (PDT) Received: from localhost (70.44.39.90.res-cmts.bus.ptd.net. [70.44.39.90]) by smtp.gmail.com with ESMTPSA id e4sm10146101qtw.62.2021.06.14.14.19.06 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 14 Jun 2021 14:19:06 -0700 (PDT) From: Johannes Weiner To: Andrew Morton Cc: Roman Gushchin , Tejun Heo , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, kernel-team@fb.com Subject: [PATCH 2/4] fs: drop_caches: fix skipping over shadow cache inodes Date: Mon, 14 Jun 2021 17:19:02 -0400 Message-Id: <20210614211904.14420-2-hannes@cmpxchg.org> X-Mailer: git-send-email 2.32.0 In-Reply-To: <20210614211904.14420-1-hannes@cmpxchg.org> References: <20210614211904.14420-1-hannes@cmpxchg.org> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org When drop_caches truncates the page cache in an inode it also includes any shadow entries for evicted pages. However, there is a preliminary check on whether the inode has pages: if it has *only* shadow entries, it will skip running truncation on the inode and leave it behind. Fix the check to mapping_empty(), such that it runs truncation on any inode that has cache entries at all. Reported-by: Roman Gushchin Signed-off-by: Johannes Weiner Acked-by: Roman Gushchin --- fs/drop_caches.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/drop_caches.c b/fs/drop_caches.c index f00fcc4a4f72..e619c31b6bd9 100644 --- a/fs/drop_caches.c +++ b/fs/drop_caches.c @@ -3,6 +3,7 @@ * Implement the manual drop-all-pagecache function */ +#include #include #include #include @@ -27,7 +28,7 @@ static void drop_pagecache_sb(struct super_block *sb, void *unused) * we need to reschedule to avoid softlockups. */ if ((inode->i_state & (I_FREEING|I_WILL_FREE|I_NEW)) || - (inode->i_mapping->nrpages == 0 && !need_resched())) { + (mapping_empty(inode->i_mapping) && !need_resched())) { spin_unlock(&inode->i_lock); continue; } From patchwork Mon Jun 14 21:19:03 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Johannes Weiner X-Patchwork-Id: 12320041 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=-16.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED,USER_AGENT_GIT 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 011CAC2B9F4 for ; Mon, 14 Jun 2021 21:20:10 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id D599B6128C for ; Mon, 14 Jun 2021 21:20:09 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235178AbhFNVWM (ORCPT ); Mon, 14 Jun 2021 17:22:12 -0400 Received: from mail-qk1-f175.google.com ([209.85.222.175]:33516 "EHLO mail-qk1-f175.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235156AbhFNVWM (ORCPT ); Mon, 14 Jun 2021 17:22:12 -0400 Received: by mail-qk1-f175.google.com with SMTP id c9so13608645qkm.0 for ; Mon, 14 Jun 2021 14:20:08 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cmpxchg-org.20150623.gappssmtp.com; s=20150623; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=dPrRC5GgzYjCmG/sOcyalcGlRFIg9CpzRtffxeyBsJs=; b=Cq8Kh47l8CTW8cX2gegKOmPYyRubXDBLX8Wra8IzL6m6emUBcjZl3GKYuVRjuPm0v9 c8de88BP5fIqRjTKx231bOte4GqGnXQa5hRxJAXZ+iTiOAm2AKnIdPPGwFBl7rm+BupS rtTtKLbNKohQeqdYJVJSHO92kBetmnTG+rxFCT9JPqkF3aeC6dM+TflTPVIG9wTwdhjt aprvCNb3nTKFMGz12ZOCSNZuyeBhL6XL5r4uMlViOup2ZnYSFgNBSarroZlxVKO+JybG syK0UQnL89xa3dQyaH9afsouyOR1DSfr+T31Aw3k56HwDy//BIeRhq00BKxnJLzo1dvn xHPg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=dPrRC5GgzYjCmG/sOcyalcGlRFIg9CpzRtffxeyBsJs=; b=oV0uEueZVGF/olrJJWuHEA0rqTktebKljCt7+ecLV1HYKWQ5e9pgKNHRT3XSVAbr0m p0DaPFe3DKYM1AaAltoKgWbQMofEyog/loVCWEhWs0byxl9hcEKwH8PK4AAed5HwqszL Ch3JxvkTeMyPgjf8lkn3vD2QyXW55I1FSFw430WniXqV9uqy1NW0S3zirPeN9zp2S/vb blRLwEOyjGOr/uj5Kyw5nRx8gVmiPUyQzujo4gOOQhBm5F4Nvj3TRvs/CLZ5HAhMY0iM tgKXv+gUnD54cNPn0tII13OqN/LrCV6SE+GlslU106IUeyufdbwAvGfTFQ2SU1Z/2iny BCHA== X-Gm-Message-State: AOAM531RFBHJBV7oPCkSBEI8KWODlhQqyS9/LhqhR5zwBovawx6s+wkY NTiSFcMeCQjpf2Wox3ysI7/t5w== X-Google-Smtp-Source: ABdhPJzhFoyCqJfWV/336OnsiH+zuoKpzlTCfmft41UKsBXIgezPkNBEAf1ehK2PH+sWsE9isbmljQ== X-Received: by 2002:a37:5943:: with SMTP id n64mr18642452qkb.122.1623705548426; Mon, 14 Jun 2021 14:19:08 -0700 (PDT) Received: from localhost (70.44.39.90.res-cmts.bus.ptd.net. [70.44.39.90]) by smtp.gmail.com with ESMTPSA id 4sm10968318qkv.134.2021.06.14.14.19.07 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 14 Jun 2021 14:19:07 -0700 (PDT) From: Johannes Weiner To: Andrew Morton Cc: Roman Gushchin , Tejun Heo , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, kernel-team@fb.com Subject: [PATCH 3/4] fs: inode: count invalidated shadow pages in pginodesteal Date: Mon, 14 Jun 2021 17:19:03 -0400 Message-Id: <20210614211904.14420-3-hannes@cmpxchg.org> X-Mailer: git-send-email 2.32.0 In-Reply-To: <20210614211904.14420-1-hannes@cmpxchg.org> References: <20210614211904.14420-1-hannes@cmpxchg.org> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org pginodesteal is supposed to capture the impact that inode reclaim has on the page cache state. Currently, it doesn't consider shadow pages that get dropped this way, even though this can have a significant impact on paging behavior, memory pressure calculations etc. To improve visibility into these effects, make sure shadow pages get counted when they get dropped through inode reclaim. This changes the return value semantics of invalidate_mapping_pages() semantics slightly, but the only two users are the inode shrinker itsel and a usb driver that logs it for debugging purposes. Signed-off-by: Johannes Weiner --- fs/inode.c | 2 +- mm/truncate.c | 20 ++++++++++---------- 2 files changed, 11 insertions(+), 11 deletions(-) diff --git a/fs/inode.c b/fs/inode.c index c93500d84264..8830a727b0af 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -768,7 +768,7 @@ static enum lru_status inode_lru_isolate(struct list_head *item, return LRU_ROTATE; } - if (inode_has_buffers(inode) || inode->i_data.nrpages) { + if (inode_has_buffers(inode) || !mapping_empty(&inode->i_data)) { __iget(inode); spin_unlock(&inode->i_lock); spin_unlock(lru_lock); diff --git a/mm/truncate.c b/mm/truncate.c index b92b86222625..95934c98259a 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -486,8 +486,9 @@ static unsigned long __invalidate_mapping_pages(struct address_space *mapping, index = indices[i]; if (xa_is_value(page)) { - invalidate_exceptional_entry(mapping, index, - page); + count += invalidate_exceptional_entry(mapping, + index, + page); continue; } index += thp_nr_pages(page) - 1; @@ -515,19 +516,18 @@ static unsigned long __invalidate_mapping_pages(struct address_space *mapping, } /** - * invalidate_mapping_pages - Invalidate all the unlocked pages of one inode - * @mapping: the address_space which holds the pages to invalidate + * invalidate_mapping_pages - Invalidate all clean, unlocked cache of one inode + * @mapping: the address_space which holds the cache to invalidate * @start: the offset 'from' which to invalidate * @end: the offset 'to' which to invalidate (inclusive) * - * This function only removes the unlocked pages, if you want to - * remove all the pages of one inode, you must call truncate_inode_pages. + * This function removes pages that are clean, unmapped and unlocked, + * as well as shadow entries. It will not block on IO activity. * - * invalidate_mapping_pages() will not block on IO activity. It will not - * invalidate pages which are dirty, locked, under writeback or mapped into - * pagetables. + * If you want to remove all the pages of one inode, regardless of + * their use and writeback state, use truncate_inode_pages(). * - * Return: the number of the pages that were invalidated + * Return: the number of the cache entries that were invalidated */ unsigned long invalidate_mapping_pages(struct address_space *mapping, pgoff_t start, pgoff_t end) From patchwork Mon Jun 14 21:19:04 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Johannes Weiner X-Patchwork-Id: 12320039 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=-16.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED,USER_AGENT_GIT 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 21A90C2B9F4 for ; Mon, 14 Jun 2021 21:19:32 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 00B0D6128C for ; Mon, 14 Jun 2021 21:19:31 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235118AbhFNVVd (ORCPT ); Mon, 14 Jun 2021 17:21:33 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:33650 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235056AbhFNVV1 (ORCPT ); Mon, 14 Jun 2021 17:21:27 -0400 Received: from mail-qt1-x82e.google.com (mail-qt1-x82e.google.com [IPv6:2607:f8b0:4864:20::82e]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id AD36DC0617AF for ; Mon, 14 Jun 2021 14:19:10 -0700 (PDT) Received: by mail-qt1-x82e.google.com with SMTP id t9so9704905qtw.7 for ; Mon, 14 Jun 2021 14:19:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cmpxchg-org.20150623.gappssmtp.com; s=20150623; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=QcD57FtNiWWrNimGrimqF/td7GBZ9gYAGkV/5ivnuQo=; b=QKRPAgQhM9SUJn2ILF83S8tR3b5HRzWek6zKuhYZsa42SpQMDIvDh7tiWzw0gnDIb0 on+8jWjW6+zeQFL+9zFsNDlO16NfOLlR8AGqc5P71prto0IvIS6HzMa3lQbddejj9nS6 Knkp1qSmpQd//eJ1TOe36iMX4b4dHJpynKFXV0FN7DWm5jL5iPNWIBSWBaN4GNVe4pzW vz98Ufg1VhJ6PlD/etx37Dc4tAnkj0GCvaqvDrkrf1VpFjqCQMF1tGh+bLw3umyG4+wj mxUPpmcP+9H/1ahep+qwu18iKtRLQIllIeZaABt4Am8MIH5lh+FzrAzqLnDl8lG0aab8 M0IA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=QcD57FtNiWWrNimGrimqF/td7GBZ9gYAGkV/5ivnuQo=; b=Y2x2Y3OkCWMdlzEDsKfiYEoZ+ihnzjWkJ36WQknBrBsBqDR4NPAu+h75qJvqljv/Va CQfQ+mFBI1Vps8HuXmgZgJJl3QQe8BwLrrNZSrxSHJI49/scznOEc45tfwuvWziEq0Ti EIoks7lXzOwSa4i8f357MufTxHyNt/v3XKfUdrd3XAK8ilDr+2CovFW+gZP7jim7YAKU TvOcsAYqya88tKBco8Gim1pBrj8ucU2pgChnkSeMsz7xvQVVqmvoJSvJhJJ0+PVOUVaL jNZsgaAt3WUb+LQ4+Tr43HXQJtUU6OCG/kg+4BYwa8ju3I8WpmGTpKkT7+cC0O+scFPv X0DQ== X-Gm-Message-State: AOAM532lz7H3UO0jhwBfAGQP9/fe6Vmmm2IL8Bj2FRRoKMVGwD25UymW A8jrvUPzhK5lTR1TmA9qfwtlhg== X-Google-Smtp-Source: ABdhPJwzrRrY2GPbYHtXtlQCHq58Du7ysF5zlWkgtjH22uYcEOuDKyojOhKRZpsNm2uJdI9Be4JxQg== X-Received: by 2002:aed:204c:: with SMTP id 70mr18858717qta.260.1623705549819; Mon, 14 Jun 2021 14:19:09 -0700 (PDT) Received: from localhost (70.44.39.90.res-cmts.bus.ptd.net. [70.44.39.90]) by smtp.gmail.com with ESMTPSA id 5sm8326659qkj.99.2021.06.14.14.19.09 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 14 Jun 2021 14:19:09 -0700 (PDT) From: Johannes Weiner To: Andrew Morton Cc: Roman Gushchin , Tejun Heo , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, kernel-team@fb.com Subject: [PATCH 4/4] vfs: keep inodes with page cache off the inode shrinker LRU Date: Mon, 14 Jun 2021 17:19:04 -0400 Message-Id: <20210614211904.14420-4-hannes@cmpxchg.org> X-Mailer: git-send-email 2.32.0 In-Reply-To: <20210614211904.14420-1-hannes@cmpxchg.org> References: <20210614211904.14420-1-hannes@cmpxchg.org> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org Historically (pre-2.5), the inode shrinker used to reclaim only empty inodes and skip over those that still contained page cache. This caused problems on highmem hosts: struct inode could put fill lowmem zones before the cache was getting reclaimed in the highmem zones. To address this, the inode shrinker started to strip page cache to facilitate reclaiming lowmem. However, this comes with its own set of problems: the shrinkers may drop actively used page cache just because the inodes are not currently open or dirty - think working with a large git tree. It further doesn't respect cgroup memory protection settings and can cause priority inversions between containers. Nowadays, the page cache also holds non-resident info for evicted cache pages in order to detect refaults. We've come to rely heavily on this data inside reclaim for protecting the cache workingset and driving swap behavior. We also use it to quantify and report workload health through psi. The latter in turn is used for fleet health monitoring, as well as driving automated memory sizing of workloads and containers, proactive reclaim and memory offloading schemes. The consequences of dropping page cache prematurely is that we're seeing subtle and not-so-subtle failures in all of the above-mentioned scenarios, with the workload generally entering unexpected thrashing states while losing the ability to reliably detect it. To fix this on non-highmem systems at least, going back to rotating inodes on the LRU isn't feasible. We've tried (commit a76cf1a474d7 ("mm: don't reclaim inodes with many attached pages")) and failed (commit 69056ee6a8a3 ("Revert "mm: don't reclaim inodes with many attached pages"")). The issue is mostly that shrinker pools attract pressure based on their size, and when objects get skipped the shrinkers remember this as deferred reclaim work. This accumulates excessive pressure on the remaining inodes, and we can quickly eat into heavily used ones, or dirty ones that require IO to reclaim, when there potentially is plenty of cold, clean cache around still. Instead, this patch keeps populated inodes off the inode LRU in the first place - just like an open file or dirty state would. An otherwise clean and unused inode then gets queued when the last cache entry disappears. This solves the problem without reintroducing the reclaim issues, and generally is a bit more scalable than having to wade through potentially hundreds of thousands of busy inodes. Locking is a bit tricky because the locks protecting the inode state (i_lock) and the inode LRU (lru_list.lock) don't nest inside the irq-safe page cache lock (i_pages.xa_lock). Page cache deletions are serialized through i_lock, taken before the i_pages lock, to make sure depopulated inodes are queued reliably. Additions may race with deletions, but we'll check again in the shrinker. If additions race with the shrinker itself, we're protected by the i_lock: if find_inode() or iput() win, the shrinker will bail on the elevated i_count or I_REFERENCED; if the shrinker wins and goes ahead with the inode, it will set I_FREEING and inhibit further igets(), which will cause the other side to create a new instance of the inode instead. Signed-off-by: Johannes Weiner --- fs/inode.c | 46 +++++++++++++++++++++---------------- fs/internal.h | 1 - include/linux/fs.h | 1 + include/linux/pagemap.h | 50 +++++++++++++++++++++++++++++++++++++++++ mm/filemap.c | 8 +++++++ mm/truncate.c | 19 ++++++++++++++-- mm/vmscan.c | 7 ++++++ mm/workingset.c | 10 +++++++++ 8 files changed, 120 insertions(+), 22 deletions(-) diff --git a/fs/inode.c b/fs/inode.c index 8830a727b0af..6b74701c1954 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -424,11 +424,20 @@ void ihold(struct inode *inode) } EXPORT_SYMBOL(ihold); -static void inode_lru_list_add(struct inode *inode) +static void __inode_add_lru(struct inode *inode, bool rotate) { + if (inode->i_state & (I_DIRTY_ALL | I_SYNC | I_FREEING | I_WILL_FREE)) + return; + if (atomic_read(&inode->i_count)) + return; + if (!(inode->i_sb->s_flags & SB_ACTIVE)) + return; + if (!mapping_shrinkable(&inode->i_data)) + return; + if (list_lru_add(&inode->i_sb->s_inode_lru, &inode->i_lru)) this_cpu_inc(nr_unused); - else + else if (rotate) inode->i_state |= I_REFERENCED; } @@ -439,16 +448,11 @@ static void inode_lru_list_add(struct inode *inode) */ void inode_add_lru(struct inode *inode) { - if (!(inode->i_state & (I_DIRTY_ALL | I_SYNC | - I_FREEING | I_WILL_FREE)) && - !atomic_read(&inode->i_count) && inode->i_sb->s_flags & SB_ACTIVE) - inode_lru_list_add(inode); + __inode_add_lru(inode, false); } - static void inode_lru_list_del(struct inode *inode) { - if (list_lru_del(&inode->i_sb->s_inode_lru, &inode->i_lru)) this_cpu_dec(nr_unused); } @@ -724,10 +728,6 @@ int invalidate_inodes(struct super_block *sb, bool kill_dirty) /* * Isolate the inode from the LRU in preparation for freeing it. * - * Any inodes which are pinned purely because of attached pagecache have their - * pagecache removed. If the inode has metadata buffers attached to - * mapping->private_list then try to remove them. - * * If the inode has the I_REFERENCED flag set, then it means that it has been * used recently - the flag is set in iput_final(). When we encounter such an * inode, clear the flag and move it to the back of the LRU so it gets another @@ -743,31 +743,39 @@ static enum lru_status inode_lru_isolate(struct list_head *item, struct inode *inode = container_of(item, struct inode, i_lru); /* - * we are inverting the lru lock/inode->i_lock here, so use a trylock. - * If we fail to get the lock, just skip it. + * We are inverting the lru lock/inode->i_lock here, so use a + * trylock. If we fail to get the lock, just skip it. */ if (!spin_trylock(&inode->i_lock)) return LRU_SKIP; /* - * Referenced or dirty inodes are still in use. Give them another pass - * through the LRU as we canot reclaim them now. + * Inodes can get referenced, redirtied, or repopulated while + * they're already on the LRU, and this can make them + * unreclaimable for a while. Remove them lazily here; iput, + * sync, or the last page cache deletion will requeue them. */ if (atomic_read(&inode->i_count) || - (inode->i_state & ~I_REFERENCED)) { + (inode->i_state & ~I_REFERENCED) || + !mapping_shrinkable(&inode->i_data)) { list_lru_isolate(lru, &inode->i_lru); spin_unlock(&inode->i_lock); this_cpu_dec(nr_unused); return LRU_REMOVED; } - /* recently referenced inodes get one more pass */ + /* Recently referenced inodes get one more pass */ if (inode->i_state & I_REFERENCED) { inode->i_state &= ~I_REFERENCED; spin_unlock(&inode->i_lock); return LRU_ROTATE; } + /* + * On highmem systems, mapping_shrinkable() permits dropping + * page cache in order to free up struct inodes: lowmem might + * be under pressure before the cache inside the highmem zone. + */ if (inode_has_buffers(inode) || !mapping_empty(&inode->i_data)) { __iget(inode); spin_unlock(&inode->i_lock); @@ -1634,7 +1642,7 @@ static void iput_final(struct inode *inode) if (!drop && !(inode->i_state & I_DONTCACHE) && (sb->s_flags & SB_ACTIVE)) { - inode_add_lru(inode); + __inode_add_lru(inode, true); spin_unlock(&inode->i_lock); return; } diff --git a/fs/internal.h b/fs/internal.h index 6aeae7ef3380..3eb90dde62bd 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -146,7 +146,6 @@ extern int vfs_open(const struct path *, struct file *); * inode.c */ extern long prune_icache_sb(struct super_block *sb, struct shrink_control *sc); -extern void inode_add_lru(struct inode *inode); extern int dentry_needs_remove_privs(struct dentry *dentry); /* diff --git a/include/linux/fs.h b/include/linux/fs.h index 8652ed7cdce8..301cd0195036 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -3214,6 +3214,7 @@ static inline void remove_inode_hash(struct inode *inode) } extern void inode_sb_list_add(struct inode *inode); +extern void inode_add_lru(struct inode *inode); extern int sb_set_blocksize(struct super_block *, int); extern int sb_min_blocksize(struct super_block *, int); diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index e89df447fae3..c9956fac640e 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -23,6 +23,56 @@ static inline bool mapping_empty(struct address_space *mapping) return xa_empty(&mapping->i_pages); } +/* + * mapping_shrinkable - test if page cache state allows inode reclaim + * @mapping: the page cache mapping + * + * This checks the mapping's cache state for the pupose of inode + * reclaim and LRU management. + * + * The caller is expected to hold the i_lock, but is not required to + * hold the i_pages lock, which usually protects cache state. That's + * because the i_lock and the list_lru lock that protect the inode and + * its LRU state don't nest inside the irq-safe i_pages lock. + * + * Cache deletions are performed under the i_lock, which ensures that + * when an inode goes empty, it will reliably get queued on the LRU. + * + * Cache additions do not acquire the i_lock and may race with this + * check, in which case we'll report the inode as shrinkable when it + * has cache pages. This is okay: the shrinker also checks the + * refcount and the referenced bit, which will be elevated or set in + * the process of adding new cache pages to an inode. + */ +static inline bool mapping_shrinkable(struct address_space *mapping) +{ + void *head; + + /* + * On highmem systems, there could be lowmem pressure from the + * inodes before there is highmem pressure from the page + * cache. Make inodes shrinkable regardless of cache state. + */ + if (IS_ENABLED(CONFIG_HIGHMEM)) + return true; + + /* Cache completely empty? Shrink away. */ + head = rcu_access_pointer(mapping->i_pages.xa_head); + if (!head) + return true; + + /* + * The xarray stores single offset-0 entries directly in the + * head pointer, which allows non-resident page cache entries + * to escape the shadow shrinker's list of xarray nodes. The + * inode shrinker needs to pick them up under memory pressure. + */ + if (!xa_is_node(head) && xa_is_value(head)) + return true; + + return false; +} + /* * Bits in mapping->flags. */ diff --git a/mm/filemap.c b/mm/filemap.c index 819d2589abef..0d0d72ced961 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -260,9 +260,13 @@ void delete_from_page_cache(struct page *page) struct address_space *mapping = page_mapping(page); BUG_ON(!PageLocked(page)); + spin_lock(&mapping->host->i_lock); xa_lock_irq(&mapping->i_pages); __delete_from_page_cache(page, NULL); xa_unlock_irq(&mapping->i_pages); + if (mapping_shrinkable(mapping)) + inode_add_lru(mapping->host); + spin_unlock(&mapping->host->i_lock); page_cache_free_page(mapping, page); } @@ -338,6 +342,7 @@ void delete_from_page_cache_batch(struct address_space *mapping, if (!pagevec_count(pvec)) return; + spin_lock(&mapping->host->i_lock); xa_lock_irq(&mapping->i_pages); for (i = 0; i < pagevec_count(pvec); i++) { trace_mm_filemap_delete_from_page_cache(pvec->pages[i]); @@ -346,6 +351,9 @@ void delete_from_page_cache_batch(struct address_space *mapping, } page_cache_delete_batch(mapping, pvec); xa_unlock_irq(&mapping->i_pages); + if (mapping_shrinkable(mapping)) + inode_add_lru(mapping->host); + spin_unlock(&mapping->host->i_lock); for (i = 0; i < pagevec_count(pvec); i++) page_cache_free_page(mapping, pvec->pages[i]); diff --git a/mm/truncate.c b/mm/truncate.c index 95934c98259a..950d73fa995d 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -45,9 +45,13 @@ static inline void __clear_shadow_entry(struct address_space *mapping, static void clear_shadow_entry(struct address_space *mapping, pgoff_t index, void *entry) { + spin_lock(&mapping->host->i_lock); xa_lock_irq(&mapping->i_pages); __clear_shadow_entry(mapping, index, entry); xa_unlock_irq(&mapping->i_pages); + if (mapping_shrinkable(mapping)) + inode_add_lru(mapping->host); + spin_unlock(&mapping->host->i_lock); } /* @@ -73,8 +77,10 @@ static void truncate_exceptional_pvec_entries(struct address_space *mapping, return; dax = dax_mapping(mapping); - if (!dax) + if (!dax) { + spin_lock(&mapping->host->i_lock); xa_lock_irq(&mapping->i_pages); + } for (i = j; i < pagevec_count(pvec); i++) { struct page *page = pvec->pages[i]; @@ -93,8 +99,12 @@ static void truncate_exceptional_pvec_entries(struct address_space *mapping, __clear_shadow_entry(mapping, index, page); } - if (!dax) + if (!dax) { xa_unlock_irq(&mapping->i_pages); + if (mapping_shrinkable(mapping)) + inode_add_lru(mapping->host); + spin_unlock(&mapping->host->i_lock); + } pvec->nr = j; } @@ -569,6 +579,7 @@ invalidate_complete_page2(struct address_space *mapping, struct page *page) if (page_has_private(page) && !try_to_release_page(page, GFP_KERNEL)) return 0; + spin_lock(&mapping->host->i_lock); xa_lock_irq(&mapping->i_pages); if (PageDirty(page)) goto failed; @@ -576,6 +587,9 @@ invalidate_complete_page2(struct address_space *mapping, struct page *page) BUG_ON(page_has_private(page)); __delete_from_page_cache(page, NULL); xa_unlock_irq(&mapping->i_pages); + if (mapping_shrinkable(mapping)) + inode_add_lru(mapping->host); + spin_unlock(&mapping->host->i_lock); if (mapping->a_ops->freepage) mapping->a_ops->freepage(page); @@ -584,6 +598,7 @@ invalidate_complete_page2(struct address_space *mapping, struct page *page) return 1; failed: xa_unlock_irq(&mapping->i_pages); + spin_unlock(&mapping->host->i_lock); return 0; } diff --git a/mm/vmscan.c b/mm/vmscan.c index cc5d7cd75935..6dd5ef8a11bc 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1055,6 +1055,8 @@ static int __remove_mapping(struct address_space *mapping, struct page *page, BUG_ON(!PageLocked(page)); BUG_ON(mapping != page_mapping(page)); + if (!PageSwapCache(page)) + spin_lock(&mapping->host->i_lock); xa_lock_irq(&mapping->i_pages); /* * The non racy check for a busy page. @@ -1123,6 +1125,9 @@ static int __remove_mapping(struct address_space *mapping, struct page *page, shadow = workingset_eviction(page, target_memcg); __delete_from_page_cache(page, shadow); xa_unlock_irq(&mapping->i_pages); + if (mapping_shrinkable(mapping)) + inode_add_lru(mapping->host); + spin_unlock(&mapping->host->i_lock); if (freepage != NULL) freepage(page); @@ -1132,6 +1137,8 @@ static int __remove_mapping(struct address_space *mapping, struct page *page, cannot_free: xa_unlock_irq(&mapping->i_pages); + if (!PageSwapCache(page)) + spin_unlock(&mapping->host->i_lock); return 0; } diff --git a/mm/workingset.c b/mm/workingset.c index 4f7a306ce75a..9d3d2b4ce44d 100644 --- a/mm/workingset.c +++ b/mm/workingset.c @@ -540,6 +540,13 @@ static enum lru_status shadow_lru_isolate(struct list_head *item, goto out; } + if (!spin_trylock(&mapping->host->i_lock)) { + xa_unlock(&mapping->i_pages); + spin_unlock_irq(lru_lock); + ret = LRU_RETRY; + goto out; + } + list_lru_isolate(lru, item); __dec_lruvec_kmem_state(node, WORKINGSET_NODES); @@ -559,6 +566,9 @@ static enum lru_status shadow_lru_isolate(struct list_head *item, out_invalid: xa_unlock_irq(&mapping->i_pages); + if (mapping_shrinkable(mapping)) + inode_add_lru(mapping->host); + spin_unlock(&mapping->host->i_lock); ret = LRU_REMOVED_RETRY; out: cond_resched();