From nobody Sat Feb 7 19:41:11 2026 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 81B62EB64D7 for ; Wed, 21 Jun 2023 08:32:21 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231325AbjFUIcU (ORCPT ); Wed, 21 Jun 2023 04:32:20 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:59214 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229999AbjFUIcP (ORCPT ); Wed, 21 Jun 2023 04:32:15 -0400 Received: from out30-119.freemail.mail.aliyun.com (out30-119.freemail.mail.aliyun.com [115.124.30.119]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 61BF110E6 for ; Wed, 21 Jun 2023 01:32:14 -0700 (PDT) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R121e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=ay29a033018045170;MF=jefflexu@linux.alibaba.com;NM=1;PH=DS;RN=6;SR=0;TI=SMTPD_---0VlfGJyV_1687336330; Received: from localhost(mailfrom:jefflexu@linux.alibaba.com fp:SMTPD_---0VlfGJyV_1687336330) by smtp.aliyun-inc.com; Wed, 21 Jun 2023 16:32:10 +0800 From: Jingbo Xu To: hsiangkao@linux.alibaba.com, chao@kernel.org, huyue2@coolpad.com, linux-erofs@lists.ozlabs.org Cc: linux-kernel@vger.kernel.org, alexl@redhat.com Subject: [RFC 1/2] erofs: update on-disk format for xattr bloom filter Date: Wed, 21 Jun 2023 16:32:08 +0800 Message-Id: <20230621083209.116024-2-jefflexu@linux.alibaba.com> X-Mailer: git-send-email 2.19.1.6.gb485710b In-Reply-To: <20230621083209.116024-1-jefflexu@linux.alibaba.com> References: <20230621083209.116024-1-jefflexu@linux.alibaba.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" The xattr bloom filter feature is going to be introduced to speed up the negative xattr lookup, e.g. system.posix_acl_[access|default] lookup when running "ls -lR" workload. The number of common used xattr (n) is approximately 8, including system.[posix_acl_access|posix_acl_default], security.[capability|selinux] and security.[SMACK64|SMACK64TRANSMUTE|SMACK64EXEC|SMACK64MMAP]. Given the number of bits of the bloom filter (m) is 32, the optimal value for the number of the hash functions (k) is 2 (ln2 * m/n =3D 2.7). Signed-off-by: Jingbo Xu --- fs/erofs/erofs_fs.h | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/fs/erofs/erofs_fs.h b/fs/erofs/erofs_fs.h index 2c7b16e340fe..9daea86cdb52 100644 --- a/fs/erofs/erofs_fs.h +++ b/fs/erofs/erofs_fs.h @@ -13,6 +13,7 @@ =20 #define EROFS_FEATURE_COMPAT_SB_CHKSUM 0x00000001 #define EROFS_FEATURE_COMPAT_MTIME 0x00000002 +#define EROFS_FEATURE_COMPAT_XATTR_BLOOM 0x00000003 =20 /* * Any bits that aren't in EROFS_ALL_FEATURE_INCOMPAT should @@ -200,7 +201,7 @@ struct erofs_inode_extended { * for read-only fs, no need to introduce h_refcount */ struct erofs_xattr_ibody_header { - __le32 h_reserved; + __le32 h_map; /* bloom filter, bit value 1 indicates not-present */ __u8 h_shared_count; __u8 h_reserved2[7]; __le32 h_shared_xattrs[]; /* shared xattr id array */ @@ -221,6 +222,11 @@ struct erofs_xattr_ibody_header { #define EROFS_XATTR_LONG_PREFIX 0x80 #define EROFS_XATTR_LONG_PREFIX_MASK 0x7f =20 +#define EROFS_XATTR_BLOOM_BITS 32 +#define EROFS_XATTR_BLOOM_MASK (EROFS_XATTR_BLOOM_BITS - 1) +#define EROFS_XATTR_BLOOM_DEFAULT UINT32_MAX +#define EROFS_XATTR_BLOOM_COUNTS 2 + /* xattr entry (for both inline & shared xattrs) */ struct erofs_xattr_entry { __u8 e_name_len; /* length of name */ --=20 2.19.1.6.gb485710b From nobody Sat Feb 7 19:41:11 2026 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id DA59FEB64D8 for ; Wed, 21 Jun 2023 08:32:27 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229675AbjFUIc0 (ORCPT ); Wed, 21 Jun 2023 04:32:26 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:59224 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229759AbjFUIcS (ORCPT ); Wed, 21 Jun 2023 04:32:18 -0400 Received: from out30-133.freemail.mail.aliyun.com (out30-133.freemail.mail.aliyun.com [115.124.30.133]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id C841A19F for ; Wed, 21 Jun 2023 01:32:16 -0700 (PDT) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R411e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=ay29a033018045170;MF=jefflexu@linux.alibaba.com;NM=1;PH=DS;RN=6;SR=0;TI=SMTPD_---0VlfH3v6_1687336331; Received: from localhost(mailfrom:jefflexu@linux.alibaba.com fp:SMTPD_---0VlfH3v6_1687336331) by smtp.aliyun-inc.com; Wed, 21 Jun 2023 16:32:11 +0800 From: Jingbo Xu To: hsiangkao@linux.alibaba.com, chao@kernel.org, huyue2@coolpad.com, linux-erofs@lists.ozlabs.org Cc: linux-kernel@vger.kernel.org, alexl@redhat.com Subject: [RFC 2/2] erofs: optimize getxattr with bloom filter Date: Wed, 21 Jun 2023 16:32:09 +0800 Message-Id: <20230621083209.116024-3-jefflexu@linux.alibaba.com> X-Mailer: git-send-email 2.19.1.6.gb485710b In-Reply-To: <20230621083209.116024-1-jefflexu@linux.alibaba.com> References: <20230621083209.116024-1-jefflexu@linux.alibaba.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Boost the negative xattr lookup with bloom filter. The bit value for the bloom filter map has a reverse semantics for compatibility. That is, the mapped bits will be cleared to 0, while the bit value of 1 indicates the absence of corresponding xattr. Signed-off-by: Jingbo Xu --- fs/erofs/internal.h | 2 ++ fs/erofs/xattr.c | 16 +++++++++++++++- 2 files changed, 17 insertions(+), 1 deletion(-) diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index 1e39c03357d1..49b4b350af8a 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -285,6 +285,7 @@ EROFS_FEATURE_FUNCS(fragments, incompat, INCOMPAT_FRAGM= ENTS) EROFS_FEATURE_FUNCS(dedupe, incompat, INCOMPAT_DEDUPE) EROFS_FEATURE_FUNCS(xattr_prefixes, incompat, INCOMPAT_XATTR_PREFIXES) EROFS_FEATURE_FUNCS(sb_chksum, compat, COMPAT_SB_CHKSUM) +EROFS_FEATURE_FUNCS(xattr_bloom, compat, COMPAT_XATTR_BLOOM) =20 /* atomic flag definitions */ #define EROFS_I_EA_INITED_BIT 0 @@ -304,6 +305,7 @@ struct erofs_inode { unsigned char inode_isize; unsigned int xattr_isize; =20 + unsigned long xattr_bloom_map; unsigned int xattr_shared_count; unsigned int *xattr_shared_xattrs; =20 diff --git a/fs/erofs/xattr.c b/fs/erofs/xattr.c index 4376f654474d..1ab481b46e8d 100644 --- a/fs/erofs/xattr.c +++ b/fs/erofs/xattr.c @@ -5,6 +5,7 @@ * Copyright (C) 2021-2022, Alibaba Cloud */ #include +#include #include "xattr.h" =20 struct erofs_xattr_iter { @@ -87,6 +88,7 @@ static int erofs_init_inode_xattrs(struct inode *inode) } =20 ih =3D it.kaddr + erofs_blkoff(sb, it.pos); + vi->xattr_bloom_map =3D le32_to_cpu(ih->h_map); vi->xattr_shared_count =3D ih->h_shared_count; vi->xattr_shared_xattrs =3D kmalloc_array(vi->xattr_shared_count, sizeof(uint), GFP_KERNEL); @@ -392,8 +394,11 @@ int erofs_getxattr(struct inode *inode, int index, const char *name, void *buffer, size_t buffer_size) { - int ret; + int i, ret; + uint32_t bit; struct erofs_xattr_iter it; + struct erofs_inode *const vi =3D EROFS_I(inode); + struct erofs_sb_info *sbi =3D EROFS_SB(inode->i_sb); =20 if (!name) return -EINVAL; @@ -402,6 +407,15 @@ int erofs_getxattr(struct inode *inode, int index, if (ret) return ret; =20 + if (erofs_sb_has_xattr_bloom(sbi) && vi->xattr_bloom_map) { + for (i =3D 0; i < EROFS_XATTR_BLOOM_COUNTS; i++) { + bit =3D xxh32(name, strlen(name), index + i); + bit &=3D EROFS_XATTR_BLOOM_MASK; + if (test_bit(bit, &vi->xattr_bloom_map)) + return -ENOATTR; + } + } + it.index =3D index; it.name =3D (struct qstr)QSTR_INIT(name, strlen(name)); if (it.name.len > EROFS_NAME_LEN) --=20 2.19.1.6.gb485710b