From nobody Sat Nov 30 05:39:24 2024 Received: from mx0b-001b2d01.pphosted.com (mx0b-001b2d01.pphosted.com [148.163.158.5]) (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 4B6F417CA0A; Wed, 11 Sep 2024 09:01:29 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=148.163.158.5 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1726045292; cv=none; b=FlrEnzJ7JYHbx8Eo0wR2CFk4P4UFiXujvoxQbMokwNv7t5VM0wddjd2/5F40F71yOd0HN0PUfcicTO6s1KVG6llesUtd9jQagnq3xUIB3w1dny77VTpbU3kYVyx619mVTk6mX9ApfBRAmyyu58otSDMT5LA+WZqGVfI9odm3eKw= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1726045292; c=relaxed/simple; bh=jd8FSTGR4XQgbUPpzzHFefpwftEHIkgBoh1yYwPdIXE=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=lpW3KgiHbi9JpUjpjU47OtDWxOok2vhAm4BRc8kSqs40Xev5jKRxUbo8av5jtz6scf/aifQNONh0IoNhWJkv2jgWHqM0f1QsNv+q326gj6KRBCHtbQ5+gRkCCDW8e/7dMvgM02AdKo5XoT1+dbiQF9863lKmpzMev+0wDIRr9ks= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.ibm.com; spf=pass smtp.mailfrom=linux.ibm.com; dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com header.b=ldbJfn3K; arc=none smtp.client-ip=148.163.158.5 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.ibm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.ibm.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com header.b="ldbJfn3K" Received: from pps.filterd (m0353725.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 48B7L9ZF008336; Wed, 11 Sep 2024 09:01:18 GMT DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ibm.com; h=from :to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; s=pp1; bh=MRdPjSH7Wl1rI WISQKQo89B2JKiM+2R/9igZckD2ekw=; b=ldbJfn3KvJZJsqSBL+YY6eEJe08wZ qfE1W5LwFgjDelqdkMOArFZEkbLRitj//oe8s3paw+47gMXRZ7KaNS83xcBkbH6H osCpF+8oPqJWLN1W/wLYCfaqSR2UfwasXAj8q6MW3f4KxDwapRJX8J5euJUVjePD nJ8gvlMCWotPTdl2HTBK/y2Z0X3ojMjMt7x3yrpyXgLdPTI2/J+gWKUS+IM2CnO/ ERFr9xXtvhBOdsiY4r9zBmTf8x3C38orNr8/K58Gmr3bUcGdYLP6If54g6GEvxyT Kky3u/edr7ojfUABxWPY8wgVhgmiW0V3i13f8kdtgM1JiNWSOl2m2GGWQ== Received: from pps.reinject (localhost [127.0.0.1]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 41gd8kmee2-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:17 +0000 (GMT) Received: from m0353725.ppops.net (m0353725.ppops.net [127.0.0.1]) by pps.reinject (8.18.0.8/8.18.0.8) with ESMTP id 48B8vNEH018725; Wed, 11 Sep 2024 09:01:17 GMT Received: from ppma21.wdc07v.mail.ibm.com (5b.69.3da9.ip4.static.sl-reverse.com [169.61.105.91]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 41gd8kmedv-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:16 +0000 (GMT) Received: from pps.filterd (ppma21.wdc07v.mail.ibm.com [127.0.0.1]) by ppma21.wdc07v.mail.ibm.com (8.18.1.2/8.18.1.2) with ESMTP id 48B7To03019847; Wed, 11 Sep 2024 09:01:16 GMT Received: from smtprelay05.fra02v.mail.ibm.com ([9.218.2.225]) by ppma21.wdc07v.mail.ibm.com (PPS) with ESMTPS id 41h25q0bxf-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:16 +0000 Received: from smtpav06.fra02v.mail.ibm.com (smtpav06.fra02v.mail.ibm.com [10.20.54.105]) by smtprelay05.fra02v.mail.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id 48B91EK345810140 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Wed, 11 Sep 2024 09:01:14 GMT Received: from smtpav06.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 37E452004D; Wed, 11 Sep 2024 09:01:14 +0000 (GMT) Received: from smtpav06.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 85A5120049; Wed, 11 Sep 2024 09:01:12 +0000 (GMT) Received: from li-bb2b2a4c-3307-11b2-a85c-8fa5c3a69313.in.ibm.com (unknown [9.109.253.82]) by smtpav06.fra02v.mail.ibm.com (Postfix) with ESMTP; Wed, 11 Sep 2024 09:01:12 +0000 (GMT) From: Ojaswin Mujoo To: linux-ext4@vger.kernel.org, "Theodore Ts'o" Cc: Ritesh Harjani , linux-kernel@vger.kernel.org, "Darrick J . Wong" , linux-fsdevel@vger.kernel.org, John Garry , dchinner@redhat.com Subject: [RFC 1/5] ext4: add aligned allocation hint in mballoc Date: Wed, 11 Sep 2024 14:31:05 +0530 Message-ID: <3a17d8d55c88799a3af9a7d337f183863912e28e.1726034272.git.ojaswin@linux.ibm.com> X-Mailer: git-send-email 2.43.5 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-TM-AS-GCONF: 00 X-Proofpoint-GUID: th1Mj74_lChmgM2Y_8I6D0XwYtNuPs7g X-Proofpoint-ORIG-GUID: zQV5Zz-WHZ97A6519km2HUeqhJMxYnl7 X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-10_12,2024-09-09_02,2024-09-02_01 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 priorityscore=1501 impostorscore=0 suspectscore=0 clxscore=1015 phishscore=0 spamscore=0 lowpriorityscore=0 adultscore=0 mlxscore=0 mlxlogscore=999 bulkscore=0 malwarescore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.19.0-2408220000 definitions=main-2409110064 Content-Type: text/plain; charset="utf-8" Add support in mballoc for allocating blocks that are aligned to a certain power-of-2 offset. 1. We define a new flag EXT4_MB_ALIGNED_HINT to indicate that we want an aligned allocation. This is just a hint, mballoc tries its best to provide aligned blocks but if it can't then it'll fallback to normal allocation 2. The alignment is determined by the length of the allocation, for example if we ask for 8192 bytes, then the alignment of physical blocks will also be 8192 bytes aligned (ie 2 blocks aligned on 4k blocksize). 3. We dont yet support arbitrary alignment. For aligned writes, the length/alignment must be power of 2 in blocks, ie for 4k blocksize we can get 4k byte aligned, 8k byte aligned, 16k byte aligned ... allocation but not 12k byte aligned. 4. We use CR_POWER2_ALIGNED criteria for aligned allocation which by design allocates in an aligned manner. Since CR_POWER2_ALIGNED needs the ac->ac_g_ex.fe_len to be power of 2, thats where the restriction in point 3 above comes from. Since right now aligned allocation support is added mainly for atomic writes use case, this restriction should be fine since atomic write capable devices usually support only power of 2 alignments 5. For ease of review enabling inode preallocation support is done in upcoming patches and is disabled in this patch. Signed-off-by: Ojaswin Mujoo --- fs/ext4/ext4.h | 2 ++ fs/ext4/mballoc.c | 60 +++++++++++++++++++++++++++++++++---- include/trace/events/ext4.h | 1 + 3 files changed, 58 insertions(+), 5 deletions(-) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 8cc15d00e5c8..17964994a049 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -222,6 +222,8 @@ enum criteria { /* Avg fragment size rb tree lookup succeeded at least once for * CR_BEST_AVAIL_LEN */ #define EXT4_MB_CR_BEST_AVAIL_LEN_OPTIMIZED 0x00020000 +/* The allocation must respect alignment requirements for physical blocks = */ +#define EXT4_MB_HINT_ALIGNED 0x40000 =20 struct ext4_allocation_request { /* target inode for block we're allocating */ diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index d73e38323879..724905552f3b 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2177,8 +2177,11 @@ static void ext4_mb_use_best_found(struct ext4_alloc= ation_context *ac, * user requested originally, we store allocated * space in a special descriptor. */ - if (ac->ac_o_ex.fe_len < ac->ac_b_ex.fe_len) + if (ac->ac_o_ex.fe_len < ac->ac_b_ex.fe_len) { + /* Aligned allocation doesn't have preallocation support */ + WARN_ON(ac->ac_flags & EXT4_MB_HINT_ALIGNED); ext4_mb_new_preallocation(ac); + } =20 } =20 @@ -2814,10 +2817,15 @@ ext4_mb_regular_allocator(struct ext4_allocation_co= ntext *ac) =20 BUG_ON(ac->ac_status =3D=3D AC_STATUS_FOUND); =20 - /* first, try the goal */ - err =3D ext4_mb_find_by_goal(ac, &e4b); - if (err || ac->ac_status =3D=3D AC_STATUS_FOUND) - goto out; + /* + * first, try the goal. Skip trying goal for aligned allocations since + * goal determination logic is not alignment aware (yet) + */ + if (!(ac->ac_flags & EXT4_MB_HINT_ALIGNED)) { + err =3D ext4_mb_find_by_goal(ac, &e4b); + if (err || ac->ac_status =3D=3D AC_STATUS_FOUND) + goto out; + } =20 if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY)) goto out; @@ -2858,9 +2866,22 @@ ext4_mb_regular_allocator(struct ext4_allocation_con= text *ac) */ if (ac->ac_2order) cr =3D CR_POWER2_ALIGNED; + else + WARN_ON_ONCE(ac->ac_g_ex.fe_len > 1 && + ac->ac_flags & EXT4_MB_HINT_ALIGNED); repeat: for (; cr < EXT4_MB_NUM_CRS && ac->ac_status =3D=3D AC_STATUS_CONTINUE; c= r++) { ac->ac_criteria =3D cr; + + if (ac->ac_criteria > CR_POWER2_ALIGNED && + ac->ac_flags & EXT4_MB_HINT_ALIGNED && + ac->ac_g_ex.fe_len > 1) { + ext4_warning_inode( + ac->ac_inode, + "Aligned allocation not possible, using unaligned allocation"); + ac->ac_flags &=3D ~EXT4_MB_HINT_ALIGNED; + } + /* * searching for the right group start * from the goal value specified @@ -2993,6 +3014,24 @@ ext4_mb_regular_allocator(struct ext4_allocation_con= text *ac) if (!err && ac->ac_status !=3D AC_STATUS_FOUND && first_err) err =3D first_err; =20 + if (ac->ac_flags & EXT4_MB_HINT_ALIGNED && ac->ac_status =3D=3D AC_STATUS= _FOUND) { + ext4_fsblk_t start =3D ext4_grp_offs_to_block(sb, &ac->ac_b_ex); + ext4_grpblk_t len =3D EXT4_C2B(sbi, ac->ac_b_ex.fe_len); + + if (!len) { + ext4_warning_inode(ac->ac_inode, + "Expected a non zero len extent"); + ac->ac_status =3D AC_STATUS_BREAK; + goto exit; + } + + WARN_ON_ONCE(!is_power_of_2(len)); + WARN_ON_ONCE(start % len); + /* We don't support preallocation yet */ + WARN_ON_ONCE(ac->ac_b_ex.fe_len !=3D ac->ac_o_ex.fe_len); + } + + exit: mb_debug(sb, "Best len %d, origin len %d, ac_status %u, ac_flags 0x%x, cr= %d ret %d\n", ac->ac_b_ex.fe_len, ac->ac_o_ex.fe_len, ac->ac_status, ac->ac_flags, cr, err); @@ -4440,6 +4479,13 @@ ext4_mb_normalize_request(struct ext4_allocation_con= text *ac, if (ac->ac_flags & EXT4_MB_HINT_NOPREALLOC) return; =20 + /* + * caller may have strict alignment requirements. In this case, avoid + * normalization since it is not alignment aware. + */ + if (ac->ac_flags & EXT4_MB_HINT_ALIGNED) + return; + if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) { ext4_mb_normalize_group_request(ac); return ; @@ -4794,6 +4840,10 @@ ext4_mb_use_preallocated(struct ext4_allocation_cont= ext *ac) if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) return false; =20 + /* using preallocated blocks is not alignment aware. */ + if (ac->ac_flags & EXT4_MB_HINT_ALIGNED) + return false; + /* * first, try per-file preallocation by searching the inode pa rbtree. * diff --git a/include/trace/events/ext4.h b/include/trace/events/ext4.h index cc5e9b7b2b44..05441f87c5d2 100644 --- a/include/trace/events/ext4.h +++ b/include/trace/events/ext4.h @@ -36,6 +36,7 @@ struct partial_cluster; { EXT4_MB_STREAM_ALLOC, "STREAM_ALLOC" }, \ { EXT4_MB_USE_ROOT_BLOCKS, "USE_ROOT_BLKS" }, \ { EXT4_MB_USE_RESERVED, "USE_RESV" }, \ + { EXT4_MB_HINT_ALIGNED, "HINT_ALIGNED" }, \ { EXT4_MB_STRICT_CHECK, "STRICT_CHECK" }) =20 #define show_map_flags(flags) __print_flags(flags, "|", \ --=20 2.43.5 From nobody Sat Nov 30 05:39:24 2024 Received: from mx0a-001b2d01.pphosted.com (mx0a-001b2d01.pphosted.com [148.163.156.1]) (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 C1FDD16DEA7; Wed, 11 Sep 2024 09:01:25 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=148.163.156.1 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1726045287; cv=none; b=B6EFa0qSZO6DxHmyNYaFsx1ppUwTlLB9eXucIQ5pBp3/c3VkjYRsWstovYW3u5nErp08ONeDKA0E8C9PfcluWrG1f5i45oQOL3ZjknfYt5lZUNPiElkqpe34AnsU0frpn7jqqdJnP55kug+S92AGjHYbHYhWtZmAi9ssqVKX7TA= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1726045287; c=relaxed/simple; bh=PzNZw6NXBeSBlaIFjYqQOwtAce5PHUf4j/0cLhdP9J4=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=X/K8Pdi5cKF1wI8+IQ8W7ASrepee5G+SGOQMVG3+Jcch8EwN+yH3U+/aqVeLX92IvEJz3Ia2a0WNzgla7SyAtmMS9Eyuu6X43SL48g1qmFvNt5MowrTE7IsEi3580dKsv0DYYg+ELdS+pr4afIGiV0gO7Ha+sOWK4PI4E7/U0Gw= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.ibm.com; spf=pass smtp.mailfrom=linux.ibm.com; dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com header.b=fHkK79Vq; arc=none smtp.client-ip=148.163.156.1 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.ibm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.ibm.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com header.b="fHkK79Vq" Received: from pps.filterd (m0356517.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 48ALVVKu022933; Wed, 11 Sep 2024 09:01:20 GMT DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ibm.com; h=from :to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; s=pp1; bh=QsWsSdrldUDth aoscoTTEvWVk10Hss1tSzmbjQcYOBo=; b=fHkK79VqtH8LLM27nXq0WqudvpFK9 mRPFOM+yOmMReUJh/HeYSRgplmoTFdrbDdfw8E4Z0kr6upvR28H/3b1H6WCJljDN 28AymYheyhFhIQ1ug2OuhIhgWC42NHrOXXlz134nif8rrlkHLBJfkqc7exgVJm0+ W7KUo1fKUqE45YrW2dXOGOZ2wGRSinn6F2pzB581fedK0Q1JKnbLP5lzBLA2yZl4 jyf79G4w2d4V/6A9q0KxkgXaVm0C3UFzC4PVFIvKAltE5MqZ2XeOquqZDVPc7NLv pu+x7+sjYrQYKt6kTvg3W1NjRhTpOt0SIKUyfq+JFqnWOs5nPeXNaW0XA== Received: from pps.reinject (localhost [127.0.0.1]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 41gejamm42-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:20 +0000 (GMT) Received: from m0356517.ppops.net (m0356517.ppops.net [127.0.0.1]) by pps.reinject (8.18.0.8/8.18.0.8) with ESMTP id 48B8rb2S013815; Wed, 11 Sep 2024 09:01:19 GMT Received: from ppma22.wdc07v.mail.ibm.com (5c.69.3da9.ip4.static.sl-reverse.com [169.61.105.92]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 41gejamm3t-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:19 +0000 (GMT) Received: from pps.filterd (ppma22.wdc07v.mail.ibm.com [127.0.0.1]) by ppma22.wdc07v.mail.ibm.com (8.18.1.2/8.18.1.2) with ESMTP id 48B6s7xf032399; Wed, 11 Sep 2024 09:01:18 GMT Received: from smtprelay07.fra02v.mail.ibm.com ([9.218.2.229]) by ppma22.wdc07v.mail.ibm.com (PPS) with ESMTPS id 41h1j0rg2k-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:18 +0000 Received: from smtpav06.fra02v.mail.ibm.com (smtpav06.fra02v.mail.ibm.com [10.20.54.105]) by smtprelay07.fra02v.mail.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id 48B91Grv49807752 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Wed, 11 Sep 2024 09:01:16 GMT Received: from smtpav06.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 413D820049; Wed, 11 Sep 2024 09:01:16 +0000 (GMT) Received: from smtpav06.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 8765420040; Wed, 11 Sep 2024 09:01:14 +0000 (GMT) Received: from li-bb2b2a4c-3307-11b2-a85c-8fa5c3a69313.in.ibm.com (unknown [9.109.253.82]) by smtpav06.fra02v.mail.ibm.com (Postfix) with ESMTP; Wed, 11 Sep 2024 09:01:14 +0000 (GMT) From: Ojaswin Mujoo To: linux-ext4@vger.kernel.org, "Theodore Ts'o" Cc: Ritesh Harjani , linux-kernel@vger.kernel.org, "Darrick J . Wong" , linux-fsdevel@vger.kernel.org, John Garry , dchinner@redhat.com Subject: [RFC 2/5] ext4: allow inode preallocation for aligned alloc Date: Wed, 11 Sep 2024 14:31:06 +0530 Message-ID: <2ab5226e6e5fe27a5cd0a1a756527c22bbd8d7c2.1726034272.git.ojaswin@linux.ibm.com> X-Mailer: git-send-email 2.43.5 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-TM-AS-GCONF: 00 X-Proofpoint-ORIG-GUID: lsKP52efjPTsO4O4cxigR_FzY7g-Su3D X-Proofpoint-GUID: IqzTRvFhoup8hvGGAiKaM3-d0Ou3exuw X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-10_12,2024-09-09_02,2024-09-02_01 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 priorityscore=1501 mlxscore=0 phishscore=0 mlxlogscore=999 spamscore=0 impostorscore=0 adultscore=0 bulkscore=0 suspectscore=0 malwarescore=0 lowpriorityscore=0 clxscore=1015 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.19.0-2408220000 definitions=main-2409110064 Content-Type: text/plain; charset="utf-8" Enable inode preallocation support for aligned allocations. Inode preallocation will only be used if the preallocated blocks are able to satisfy the length and alignment requirements of the allocations, else we disable preallocation for this particular allocation and proceed as usual. Disabling inode preallocation is required otherwise we might end up with overlapping preallocated ranges which can trigger a BUG() later. Further, during normalizing, we usually try to round it up to a power of 2 which can still give us aligned allocation. We also make sure not change the goal start so aligned allocation is more straightforward. If for whatever reason the goal is not power of 2 or doesn't contain the original request, then we throw a warning and proceed as normal. For now, group preallocation is disabled for aligned allocations. Signed-off-by: Ojaswin Mujoo --- fs/ext4/mballoc.c | 96 +++++++++++++++++++++++++++++++---------------- 1 file changed, 63 insertions(+), 33 deletions(-) diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 724905552f3b..23a553ad02fa 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2178,8 +2178,6 @@ static void ext4_mb_use_best_found(struct ext4_alloca= tion_context *ac, * space in a special descriptor. */ if (ac->ac_o_ex.fe_len < ac->ac_b_ex.fe_len) { - /* Aligned allocation doesn't have preallocation support */ - WARN_ON(ac->ac_flags & EXT4_MB_HINT_ALIGNED); ext4_mb_new_preallocation(ac); } =20 @@ -3027,8 +3025,7 @@ ext4_mb_regular_allocator(struct ext4_allocation_cont= ext *ac) =20 WARN_ON_ONCE(!is_power_of_2(len)); WARN_ON_ONCE(start % len); - /* We don't support preallocation yet */ - WARN_ON_ONCE(ac->ac_b_ex.fe_len !=3D ac->ac_o_ex.fe_len); + WARN_ON_ONCE(ac->ac_b_ex.fe_len < ac->ac_o_ex.fe_len); } =20 exit: @@ -4479,13 +4476,6 @@ ext4_mb_normalize_request(struct ext4_allocation_con= text *ac, if (ac->ac_flags & EXT4_MB_HINT_NOPREALLOC) return; =20 - /* - * caller may have strict alignment requirements. In this case, avoid - * normalization since it is not alignment aware. - */ - if (ac->ac_flags & EXT4_MB_HINT_ALIGNED) - return; - if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) { ext4_mb_normalize_group_request(ac); return ; @@ -4542,6 +4532,21 @@ ext4_mb_normalize_request(struct ext4_allocation_con= text *ac, size =3D (loff_t) EXT4_C2B(sbi, ac->ac_o_ex.fe_len) << bsbits; } + + /* + * For aligned allocations, we need to ensure 2 things: + * + * 1. The start should remain same as original start so that finding + * aligned physical blocks for it is straight forward. + * + * 2. The new_size should not be less than the original len. This + * can sometimes happen due to the way we predict size above. + */ + if (ac->ac_flags & EXT4_MB_HINT_ALIGNED) { + start_off =3D ac->ac_o_ex.fe_logical << bsbits; + size =3D max_t(loff_t, size, + EXT4_C2B(sbi, ac->ac_o_ex.fe_len) << bsbits); + } size =3D size >> bsbits; start =3D start_off >> bsbits; =20 @@ -4792,32 +4797,46 @@ ext4_mb_check_group_pa(ext4_fsblk_t goal_block, } =20 /* - * check if found pa meets EXT4_MB_HINT_GOAL_ONLY + * check if found pa meets EXT4_MB_HINT_GOAL_ONLY or EXT4_MB_HINT_ALIGNED */ static bool -ext4_mb_pa_goal_check(struct ext4_allocation_context *ac, +ext4_mb_pa_check(struct ext4_allocation_context *ac, struct ext4_prealloc_space *pa) { struct ext4_sb_info *sbi =3D EXT4_SB(ac->ac_sb); ext4_fsblk_t start; =20 - if (likely(!(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY))) + if (likely(!(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY || + ac->ac_flags & EXT4_MB_HINT_ALIGNED))) return true; =20 - /* - * If EXT4_MB_HINT_GOAL_ONLY is set, ac_g_ex will not be adjusted - * in ext4_mb_normalize_request and will keep same with ac_o_ex - * from ext4_mb_initialize_context. Choose ac_g_ex here to keep - * consistent with ext4_mb_find_by_goal. - */ - start =3D pa->pa_pstart + - (ac->ac_g_ex.fe_logical - pa->pa_lstart); - if (ext4_grp_offs_to_block(ac->ac_sb, &ac->ac_g_ex) !=3D start) - return false; + if (ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY) { + /* + * If EXT4_MB_HINT_GOAL_ONLY is set, ac_g_ex will not be adjusted + * in ext4_mb_normalize_request and will keep same with ac_o_ex + * from ext4_mb_initialize_context. Choose ac_g_ex here to keep + * consistent with ext4_mb_find_by_goal. + */ + start =3D pa->pa_pstart + + (ac->ac_g_ex.fe_logical - pa->pa_lstart); + if (ext4_grp_offs_to_block(ac->ac_sb, &ac->ac_g_ex) !=3D start) + return false; =20 - if (ac->ac_g_ex.fe_len > pa->pa_len - - EXT4_B2C(sbi, ac->ac_g_ex.fe_logical - pa->pa_lstart)) - return false; + if (ac->ac_g_ex.fe_len > + pa->pa_len - EXT4_B2C(sbi, ac->ac_g_ex.fe_logical - + pa->pa_lstart)) + return false; + } else if (ac->ac_flags & EXT4_MB_HINT_ALIGNED) { + start =3D pa->pa_pstart + + (ac->ac_g_ex.fe_logical - pa->pa_lstart); + if (start % EXT4_C2B(sbi, ac->ac_g_ex.fe_len)) + return false; + + if (EXT4_C2B(sbi, ac->ac_g_ex.fe_len) > + (EXT4_C2B(sbi, pa->pa_len) - + (ac->ac_g_ex.fe_logical - pa->pa_lstart))) + return false; + } =20 return true; } @@ -4840,10 +4859,6 @@ ext4_mb_use_preallocated(struct ext4_allocation_cont= ext *ac) if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) return false; =20 - /* using preallocated blocks is not alignment aware. */ - if (ac->ac_flags & EXT4_MB_HINT_ALIGNED) - return false; - /* * first, try per-file preallocation by searching the inode pa rbtree. * @@ -4949,7 +4964,7 @@ ext4_mb_use_preallocated(struct ext4_allocation_conte= xt *ac) goto try_group_pa; } =20 - if (tmp_pa->pa_free && likely(ext4_mb_pa_goal_check(ac, tmp_pa))) { + if (tmp_pa->pa_free && likely(ext4_mb_pa_check(ac, tmp_pa))) { atomic_inc(&tmp_pa->pa_count); ext4_mb_use_inode_pa(ac, tmp_pa); spin_unlock(&tmp_pa->pa_lock); @@ -4984,6 +4999,19 @@ ext4_mb_use_preallocated(struct ext4_allocation_cont= ext *ac) * pa_free =3D=3D 0. */ WARN_ON_ONCE(tmp_pa->pa_free =3D=3D 0); + + /* + * If, for any reason, we reach here then we need to disable PA + * because otherwise ext4_mb_normalize_request() will try to + * allocate a new PA for this logical range where another PA + * already exists. This is not allowed and will trigger BUG_ONs. + * Hence, as a workaround we disable PA. + * + * NOTE: ideally we would want to have some logic to take care + * of the unusable PA. Maybe a more fine grained discard logic + * that could allow us to discard only specific PAs. + */ + ac->ac_flags |=3D EXT4_MB_HINT_NOPREALLOC; } spin_unlock(&tmp_pa->pa_lock); try_group_pa: @@ -5790,6 +5818,7 @@ static void ext4_mb_group_or_file(struct ext4_allocat= ion_context *ac) int bsbits =3D ac->ac_sb->s_blocksize_bits; loff_t size, isize; bool inode_pa_eligible, group_pa_eligible; + bool is_aligned =3D (ac->ac_flags & EXT4_MB_HINT_ALIGNED); =20 if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) return; @@ -5797,7 +5826,8 @@ static void ext4_mb_group_or_file(struct ext4_allocat= ion_context *ac) if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY)) return; =20 - group_pa_eligible =3D sbi->s_mb_group_prealloc > 0; + /* Aligned allocation does not support group pa */ + group_pa_eligible =3D (!is_aligned && sbi->s_mb_group_prealloc > 0); inode_pa_eligible =3D true; size =3D extent_logical_end(sbi, &ac->ac_o_ex); isize =3D (i_size_read(ac->ac_inode) + ac->ac_sb->s_blocksize - 1) --=20 2.43.5 From nobody Sat Nov 30 05:39:24 2024 Received: from mx0b-001b2d01.pphosted.com (mx0b-001b2d01.pphosted.com [148.163.158.5]) (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 0EB96183CCA; Wed, 11 Sep 2024 09:01:33 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=148.163.158.5 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1726045295; cv=none; b=NwiobSOj+kEk+1mlYxaCHR6Izjr6oFROoxiBHOA1k+GyOCKe2NpjUQKEV3H4q8MQRBBp8C2mSfdDFUPDPI6AnMzj9edOBqAUvi8xW16AiV4qc1h/xQLSzYPiYZn3tBvOjqheWrbEahuEHw2duno2jAPdXTMyfo7vvxtPQkdW1wE= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1726045295; c=relaxed/simple; bh=EfnBcbb247aIMImII5MMzigKz7nE+J8+qXUtbzDTD9g=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=pK0Cr2a+3ezzmiYPo9Km/H0uMr2BlAWLUtOaxXN9wt5ao8qFIDeWwrnG7m2X2bUAHSJgVLg8Sd+72DBrrMUArFStbR9/DReW06G0d9/AE1WGKRsFXcxMWHjmDveoFyRZ1JsuGr+woKHTzrBvGPlzmxVWpl+YTP574AEsF3f6GBM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.ibm.com; spf=pass smtp.mailfrom=linux.ibm.com; dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com header.b=qE6nIyzW; arc=none smtp.client-ip=148.163.158.5 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.ibm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.ibm.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com header.b="qE6nIyzW" Received: from pps.filterd (m0356516.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 48B11xJB022813; Wed, 11 Sep 2024 09:01:21 GMT DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ibm.com; h=from :to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; s=pp1; bh=ZlyxyKXS/TEvE Do1qWJs/qa3Pj3ZYUBvlNtlwO/zUAE=; b=qE6nIyzWQZrHcuEl+72X2dT6dflz8 5Used+qYiu9U5B3u7AE6+6c5U8ChRvlPgNkd65gshkTfUy+dnm6bNhADmSVdbXc8 FQl6h5XDlT12Ef6B6+dlBr4uHa4g80ORPhgCE+Fh40zSDtZ16pZmd3CQvFBU8sXn 6gOBK3wBxAe4KgMChKtkyKj2EdNhEeApexoltR8/gskB3VVd+lDTNPMEHkfaZ0to s1yqAI72kwH1JJKKURg1/wob0T6yoB32XG/xMKfqRZju/resDfTKLhETnGjgBa6g t0q1WMGPFdfJCaPmu/2EJTtdGjBIB7JzjunyT3UVY1n+vg1LjjovVdyBQ== Received: from pps.reinject (localhost [127.0.0.1]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 41gc8qcrxm-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:21 +0000 (GMT) Received: from m0356516.ppops.net (m0356516.ppops.net [127.0.0.1]) by pps.reinject (8.18.0.8/8.18.0.8) with ESMTP id 48B91L0q013404; Wed, 11 Sep 2024 09:01:21 GMT Received: from ppma13.dal12v.mail.ibm.com (dd.9e.1632.ip4.static.sl-reverse.com [50.22.158.221]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 41gc8qcrxd-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:21 +0000 (GMT) Received: from pps.filterd (ppma13.dal12v.mail.ibm.com [127.0.0.1]) by ppma13.dal12v.mail.ibm.com (8.18.1.2/8.18.1.2) with ESMTP id 48B804te014576; Wed, 11 Sep 2024 09:01:20 GMT Received: from smtprelay04.fra02v.mail.ibm.com ([9.218.2.228]) by ppma13.dal12v.mail.ibm.com (PPS) with ESMTPS id 41h3cm83ew-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:20 +0000 Received: from smtpav06.fra02v.mail.ibm.com (smtpav06.fra02v.mail.ibm.com [10.20.54.105]) by smtprelay04.fra02v.mail.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id 48B91I1X34538118 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Wed, 11 Sep 2024 09:01:18 GMT Received: from smtpav06.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 441E52004B; Wed, 11 Sep 2024 09:01:18 +0000 (GMT) Received: from smtpav06.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 95FE220040; Wed, 11 Sep 2024 09:01:16 +0000 (GMT) Received: from li-bb2b2a4c-3307-11b2-a85c-8fa5c3a69313.in.ibm.com (unknown [9.109.253.82]) by smtpav06.fra02v.mail.ibm.com (Postfix) with ESMTP; Wed, 11 Sep 2024 09:01:16 +0000 (GMT) From: Ojaswin Mujoo To: linux-ext4@vger.kernel.org, "Theodore Ts'o" Cc: Ritesh Harjani , linux-kernel@vger.kernel.org, "Darrick J . Wong" , linux-fsdevel@vger.kernel.org, John Garry , dchinner@redhat.com Subject: [RFC 3/5] ext4: Support for extsize hint using FS_IOC_FS(GET/SET)XATTR Date: Wed, 11 Sep 2024 14:31:07 +0530 Message-ID: <0fddb518b268164fe565c60f6f52f27c58496aae.1726034272.git.ojaswin@linux.ibm.com> X-Mailer: git-send-email 2.43.5 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-TM-AS-GCONF: 00 X-Proofpoint-GUID: 8A45KZ_unI155SNhpj17fnIs6N5fG8J_ X-Proofpoint-ORIG-GUID: sGKhjU9biDXh6efrzCKfmXdkOC3UuBYi X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-10_12,2024-09-09_02,2024-09-02_01 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 bulkscore=0 spamscore=0 mlxscore=0 lowpriorityscore=0 impostorscore=0 priorityscore=1501 malwarescore=0 adultscore=0 clxscore=1015 mlxlogscore=999 suspectscore=0 phishscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.19.0-2408220000 definitions=main-2409110064 Content-Type: text/plain; charset="utf-8" This patch adds support for getting and setting extsize hint using FS_IOC_GETXATTR and FS_IOC_SETXATTR interface. The extsize is stored in xattr of type EXT4_XATTR_INDEX_SYSTEM. Restrictions on setting extsize: 1. extsize can't be set on files with data 2. extsize can't be set on non regular files 3. extsize hint can't be used with bigalloc (yet) 4. extsize (in blocks) should be power-of-2 for simplicity. 5. extsize must be a multiple of block size The ioctl behavior has been kept as close to the XFS equivalent as possible. Signed-off-by: Ojaswin Mujoo ext4: Some modifications to extsize ioctl (To be Squashed) Signed-off-by: Ojaswin Mujoo --- fs/ext4/ext4.h | 6 +++ fs/ext4/inode.c | 89 ++++++++++++++++++++++++++++++++++++ fs/ext4/ioctl.c | 119 ++++++++++++++++++++++++++++++++++++++++++++++++ fs/ext4/super.c | 1 + 4 files changed, 215 insertions(+) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 17964994a049..d34e60cf6458 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1170,6 +1170,8 @@ struct ext4_inode_info { __u32 i_csum_seed; =20 kprojid_t i_projid; + /* The extentsize hint for the inode in blocks */ + ext4_grpblk_t i_extsize; }; =20 /* @@ -3037,6 +3039,10 @@ extern void ext4_da_update_reserve_space(struct inod= e *inode, int used, int quota_claim); extern int ext4_issue_zeroout(struct inode *inode, ext4_lblk_t lblk, ext4_fsblk_t pblk, ext4_lblk_t len); +int ext4_inode_xattr_get_extsize(struct inode *inode); +int ext4_inode_xattr_set_extsize(struct inode *inode, ext4_grpblk_t extsiz= e); +ext4_grpblk_t ext4_inode_get_extsize(struct ext4_inode_info *ei); +void ext4_inode_set_extsize(struct ext4_inode_info *ei, ext4_grpblk_t exts= ize); =20 /* indirect.c */ extern int ext4_ind_map_blocks(handle_t *handle, struct inode *inode, diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 7475deef9793..898b41751cf4 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -4881,6 +4881,20 @@ struct inode *__ext4_iget(struct super_block *sb, un= signed long ino, INIT_LIST_HEAD(&ei->i_orphan); ext4_fc_init_inode(&ei->vfs_inode); =20 + ret =3D ext4_inode_xattr_get_extsize(&ei->vfs_inode); + if (ret >=3D 0) { + ei->i_extsize =3D ret; + } else if (ret =3D=3D -ENODATA) { + /* extsize is not set */ + ei->i_extsize =3D 0; + } else { + ext4_error_inode( + inode, function, line, 0, + "iget: error while retrieving extsize from xattr: %ld", ret); + ret =3D -EFSCORRUPTED; + goto bad_inode; + } + /* * Set transaction id's of transactions that have to be committed * to finish f[data]sync. We set them to currently running transaction @@ -6216,3 +6230,78 @@ vm_fault_t ext4_page_mkwrite(struct vm_fault *vmf) ext4_journal_stop(handle); goto out; } + +/* + * Returns positive extsize if set, 0 if not set else error + */ +ext4_grpblk_t ext4_inode_xattr_get_extsize(struct inode *inode) +{ + char *buf; + int size, ret =3D 0; + ext4_grpblk_t extsize =3D 0; + + size =3D ext4_xattr_get(inode, EXT4_XATTR_INDEX_SYSTEM, "extsize", NULL, = 0); + + if (size =3D=3D -ENODATA || size =3D=3D 0) { + return 0; + } else if (size < 0) { + ret =3D size; + goto exit; + } + + buf =3D (char *)kmalloc(size + 1, GFP_KERNEL); + if (!buf) { + ret =3D -ENOMEM; + goto exit; + } + + size =3D ext4_xattr_get(inode, EXT4_XATTR_INDEX_SYSTEM, "extsize", buf, + size); + if (size =3D=3D -ENODATA) + /* No extsize is set */ + extsize =3D 0; + else if (size < 0) + ret =3D size; + else { + buf[size] =3D '\0'; + ret =3D kstrtoint(buf, 10, &extsize); + } + + kfree(buf); +exit: + if (ret) + return ret; + return extsize; +} + +int ext4_inode_xattr_set_extsize(struct inode *inode, ext4_grpblk_t extsiz= e) +{ + int err =3D 0; + /* max value of extsize should fit within 11 chars */ + char extsize_str[11]; + + if ((err =3D snprintf(extsize_str, 10, "%u", extsize)) < 0) { + return err; + } + + /* Try to replace the xattr if it exists, else try to create it */ + err =3D ext4_xattr_set(inode, EXT4_XATTR_INDEX_SYSTEM, "extsize", + extsize_str, strlen(extsize_str), XATTR_REPLACE); + + if (err =3D=3D -ENODATA) + err =3D ext4_xattr_set(inode, EXT4_XATTR_INDEX_SYSTEM, "extsize", + extsize_str, strlen(extsize_str), + XATTR_CREATE); + + return err; +} + +ext4_grpblk_t ext4_inode_get_extsize(struct ext4_inode_info *ei) +{ + return ei->i_extsize; +} + +void ext4_inode_set_extsize(struct ext4_inode_info *ei, ext4_grpblk_t exts= ize) +{ + ei->i_extsize =3D extsize; +} diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index e8bf5972dd47..e456e71e6187 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -708,6 +708,90 @@ static int ext4_ioctl_setflags(struct inode *inode, return err; } =20 +static u32 ext4_ioctl_getextsize(struct inode *inode) +{ + ext4_grpblk_t extsize; + + extsize =3D ext4_inode_get_extsize(EXT4_I(inode)); + + return (u32) extsize << inode->i_blkbits; +} + + +static int ext4_ioctl_setextsize(struct inode *inode, u32 extsize, u32 xfl= ags) +{ + int err; + ext4_grpblk_t extsize_blks =3D extsize >> inode->i_blkbits; + struct ext4_sb_info *sbi =3D EXT4_SB(inode->i_sb); + int blksize =3D 1 << inode->i_blkbits; + char *msg =3D NULL; + + if (!S_ISREG(inode->i_mode)) { + msg =3D "Cannot set extsize on non regular file"; + err =3D -EOPNOTSUPP; + goto error; + } + + /* TODO: Can we just use i_size here? */ + if (ext4_has_inline_data(inode) || + READ_ONCE(EXT4_I(inode)->i_disksize) || + EXT4_I(inode)->i_reserved_data_blocks) { + msg =3D "Cannot set extsize on file with data"; + err =3D -EOPNOTSUPP; + goto error; + } + + if (extsize % blksize) { + msg =3D "extsize must be multiple of blocksize"; + err =3D -EINVAL; + goto error; + } + + if (sbi->s_cluster_ratio > 1) { + msg =3D "Can't use extsize hint with bigalloc"; + err =3D -EINVAL; + goto error; + } + + if ((xflags & FS_XFLAG_EXTSIZE) && extsize =3D=3D 0) { + msg =3D "fsx_extsize can't be 0 if FS_XFLAG_EXTSIZE is passed"; + err =3D -EINVAL; + goto error; + } + + if (extsize_blks > sbi->s_blocks_per_group) { + msg =3D "extsize cannot exceed number of bytes in block group"; + err =3D -EINVAL; + goto error; + } + + if (extsize && !is_power_of_2(extsize_blks)) { + msg =3D "extsize must be either power-of-2 in fs blocks or 0"; + err =3D -EINVAL; + goto error; + } + + if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) { + msg =3D "extsize can't be set on non-extent based files"; + err =3D -EINVAL; + goto error; + } + + /* update the extsize in inode xattr */ + if ((err =3D ext4_inode_xattr_set_extsize(inode, extsize_blks)) < 0) + return err; + + /* Update the new extsize in the in-core inode */ + ext4_inode_set_extsize(EXT4_I(inode), extsize_blks); + return 0; + +error: + if (msg) + ext4_warning_inode(inode, "%s\n", msg); + + return err; +} + #ifdef CONFIG_QUOTA static int ext4_ioctl_setproject(struct inode *inode, __u32 projid) { @@ -985,6 +1069,7 @@ int ext4_fileattr_get(struct dentry *dentry, struct fi= leattr *fa) struct inode *inode =3D d_inode(dentry); struct ext4_inode_info *ei =3D EXT4_I(inode); u32 flags =3D ei->i_flags & EXT4_FL_USER_VISIBLE; + u32 extsize=3D0; =20 if (S_ISREG(inode->i_mode)) flags &=3D ~FS_PROJINHERIT_FL; @@ -993,6 +1078,13 @@ int ext4_fileattr_get(struct dentry *dentry, struct f= ileattr *fa) if (ext4_has_feature_project(inode->i_sb)) fa->fsx_projid =3D from_kprojid(&init_user_ns, ei->i_projid); =20 + extsize =3D ext4_ioctl_getextsize(inode); + /* Flag is only set if extsize is non zero */ + if (extsize > 0) { + fa->fsx_extsize =3D extsize; + fa->fsx_xflags |=3D FS_XFLAG_EXTSIZE; + } + return 0; } =20 @@ -1022,6 +1114,33 @@ int ext4_fileattr_set(struct mnt_idmap *idmap, if (err) goto out; err =3D ext4_ioctl_setproject(inode, fa->fsx_projid); + if (err) + goto out; + + if (fa->fsx_xflags & FS_XFLAG_EXTSIZE) { + err =3D ext4_ioctl_setextsize(inode, fa->fsx_extsize, + fa->fsx_xflags); + if (err) + goto out; + } else if (fa->fsx_extsize =3D=3D 0) { + /* + * Even when user explicitly passes extsize=3D0 the flag is cleared in + * fileattr_set_prepare(). + */ + if (ext4_inode_get_extsize(EXT4_I(inode)) !=3D 0) { + err =3D ext4_ioctl_setextsize(inode, fa->fsx_extsize, + fa->fsx_xflags); + if (err) + goto out; + } + + } else { + /* Unexpected usage, reset extsize to 0 */ + err =3D ext4_ioctl_setextsize(inode, 0, fa->fsx_xflags); + if (err) + goto out; + fa->fsx_xflags =3D 0; + } out: return err; } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 16a4ce704460..4e293a2bccd3 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1421,6 +1421,7 @@ static struct inode *ext4_alloc_inode(struct super_bl= ock *sb) spin_lock_init(&ei->i_completed_io_lock); ei->i_sync_tid =3D 0; ei->i_datasync_tid =3D 0; + ei->i_extsize =3D 0; atomic_set(&ei->i_unwritten, 0); INIT_WORK(&ei->i_rsv_conversion_work, ext4_end_io_rsv_work); ext4_fc_init_inode(&ei->vfs_inode); --=20 2.43.5 From nobody Sat Nov 30 05:39:24 2024 Received: from mx0b-001b2d01.pphosted.com (mx0b-001b2d01.pphosted.com [148.163.158.5]) (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 9285815381A; Wed, 11 Sep 2024 09:01:30 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=148.163.158.5 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1726045292; cv=none; b=HQgFh8QdKYjcj3J2LtxiG3uFgzswi9JgUlZf+Eee2nB/K8Vx1xPHyD+iDYjehlzsneSIbkfyQbmz40rG3chYCMZtnKRhkRtetH6UjEUW8vGm485Y70IKSyQNZoYWUJX29WI+7FuROhsjgCMOTLM+sYj9mUB4gGT75QcXnuMBBaU= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1726045292; c=relaxed/simple; bh=yrID4KohIk6xncNm/NMKEFC1LATH0YC8oNkPsYVr6wo=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=eCi+H0IzHJGHd0mmpcFtiHZV3egMvAbqi4UKtz3UhYzOCwDJv4ypwKFQIWEDX4jnmpFYn0vbPE5ghH1GlHyaw9LFhZeOhZ5PUEiAA07tA3RLkgX3sIAxoVH2/UPAjftS7GVYyb2pjvPL24DzzSYriLkIh4r6+TuMtt5+JkTBNzg= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.ibm.com; spf=pass smtp.mailfrom=linux.ibm.com; dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com header.b=NdzsgZSN; arc=none smtp.client-ip=148.163.158.5 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.ibm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.ibm.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com header.b="NdzsgZSN" Received: from pps.filterd (m0353725.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 48AJxktr005571; Wed, 11 Sep 2024 09:01:24 GMT DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ibm.com; h=from :to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; s=pp1; bh=Zlmur8TQDlUpV bfQd7mPaUXRtaFk/VFFlbEreuhKxZM=; b=NdzsgZSNZsKaABIJqs5/dharDIzGu CA+1gF+igkePNhhuUk1WE4G7eYH5FS7RH7GnDQlhpVJBb7GBrplkAfLcuLH37yd6 EarG/GEjuPSOnoXQTtUm8Y0OAowmkd9Z0Gfcg/jhkvFldd43QEP2Infv1ZH00Pdb BYSH5g+FV/RCRK3koi0PqjZtRH231W2sRemFxy0v9GnJ412siTS/AlYubb1GurF/ EpMhDot/qXGqQPwVt5V+lc9f6xksB2ht/A2SRaHpn20HF6yYPf9cg7CPRHJ5Ra8X Zs2X5yIg9YUyqRaLBwyUOvOTFDhLXRof8sVVsLTRniElillBquPaOw2ZQ== Received: from pps.reinject (localhost [127.0.0.1]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 41gd8kmef2-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:23 +0000 (GMT) Received: from m0353725.ppops.net (m0353725.ppops.net [127.0.0.1]) by pps.reinject (8.18.0.8/8.18.0.8) with ESMTP id 48B91NlZ029094; Wed, 11 Sep 2024 09:01:23 GMT Received: from ppma21.wdc07v.mail.ibm.com (5b.69.3da9.ip4.static.sl-reverse.com [169.61.105.91]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 41gd8kmeeu-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:23 +0000 (GMT) Received: from pps.filterd (ppma21.wdc07v.mail.ibm.com [127.0.0.1]) by ppma21.wdc07v.mail.ibm.com (8.18.1.2/8.18.1.2) with ESMTP id 48B7CZml019843; Wed, 11 Sep 2024 09:01:22 GMT Received: from smtprelay01.fra02v.mail.ibm.com ([9.218.2.227]) by ppma21.wdc07v.mail.ibm.com (PPS) with ESMTPS id 41h25q0by2-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:22 +0000 Received: from smtpav06.fra02v.mail.ibm.com (smtpav06.fra02v.mail.ibm.com [10.20.54.105]) by smtprelay01.fra02v.mail.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id 48B91KXN52363670 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Wed, 11 Sep 2024 09:01:20 GMT Received: from smtpav06.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 4519120040; Wed, 11 Sep 2024 09:01:20 +0000 (GMT) Received: from smtpav06.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 941312004D; Wed, 11 Sep 2024 09:01:18 +0000 (GMT) Received: from li-bb2b2a4c-3307-11b2-a85c-8fa5c3a69313.in.ibm.com (unknown [9.109.253.82]) by smtpav06.fra02v.mail.ibm.com (Postfix) with ESMTP; Wed, 11 Sep 2024 09:01:18 +0000 (GMT) From: Ojaswin Mujoo To: linux-ext4@vger.kernel.org, "Theodore Ts'o" Cc: Ritesh Harjani , linux-kernel@vger.kernel.org, "Darrick J . Wong" , linux-fsdevel@vger.kernel.org, John Garry , dchinner@redhat.com Subject: [RFC 4/5] ext4: pass lblk and len explicitly to ext4_split_extent*() Date: Wed, 11 Sep 2024 14:31:08 +0530 Message-ID: <35d703d739deafcdd8e57328de0830aecd0bf2e9.1726034272.git.ojaswin@linux.ibm.com> X-Mailer: git-send-email 2.43.5 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-TM-AS-GCONF: 00 X-Proofpoint-GUID: 5S70imQ3r5t2bIH10SBo5CDtBZWqV4fn X-Proofpoint-ORIG-GUID: 4LQc49d_ts3Tozl_qf6CHCdas8lssgye X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-10_12,2024-09-09_02,2024-09-02_01 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 priorityscore=1501 impostorscore=0 suspectscore=0 clxscore=1015 phishscore=0 spamscore=0 lowpriorityscore=0 adultscore=0 mlxscore=0 mlxlogscore=638 bulkscore=0 malwarescore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.19.0-2408220000 definitions=main-2409110064 Content-Type: text/plain; charset="utf-8" Since these functions only use the map to determine lblk and len of the split, pass them explicitly. This is in preparation for making them work with extent size hints cleanly. No functional change in this patch. Signed-off-by: Ojaswin Mujoo --- fs/ext4/extents.c | 57 +++++++++++++++++++++++++---------------------- 1 file changed, 30 insertions(+), 27 deletions(-) diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 34e25eee6521..94aeb5b47971 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3347,7 +3347,8 @@ static struct ext4_ext_path *ext4_split_extent_at(han= dle_t *handle, static struct ext4_ext_path *ext4_split_extent(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, - struct ext4_map_blocks *map, + ext4_lblk_t lblk, + unsigned int len, int split_flag, int flags, unsigned int *allocated) { @@ -3363,7 +3364,7 @@ static struct ext4_ext_path *ext4_split_extent(handle= _t *handle, ee_len =3D ext4_ext_get_actual_len(ex); unwritten =3D ext4_ext_is_unwritten(ex); =20 - if (map->m_lblk + map->m_len < ee_block + ee_len) { + if (lblk + len < ee_block + ee_len) { split_flag1 =3D split_flag & EXT4_EXT_MAY_ZEROOUT; flags1 =3D flags | EXT4_GET_BLOCKS_PRE_IO; if (unwritten) @@ -3372,28 +3373,28 @@ static struct ext4_ext_path *ext4_split_extent(hand= le_t *handle, if (split_flag & EXT4_EXT_DATA_VALID2) split_flag1 |=3D EXT4_EXT_DATA_VALID1; path =3D ext4_split_extent_at(handle, inode, path, - map->m_lblk + map->m_len, split_flag1, flags1); + lblk + len, split_flag1, flags1); if (IS_ERR(path)) return path; /* * Update path is required because previous ext4_split_extent_at * may result in split of original leaf or extent zeroout. */ - path =3D ext4_find_extent(inode, map->m_lblk, path, flags); + path =3D ext4_find_extent(inode, lblk, path, flags); if (IS_ERR(path)) return path; depth =3D ext_depth(inode); ex =3D path[depth].p_ext; if (!ex) { EXT4_ERROR_INODE(inode, "unexpected hole at %lu", - (unsigned long) map->m_lblk); + (unsigned long) lblk); ext4_free_ext_path(path); return ERR_PTR(-EFSCORRUPTED); } unwritten =3D ext4_ext_is_unwritten(ex); } =20 - if (map->m_lblk >=3D ee_block) { + if (lblk >=3D ee_block) { split_flag1 =3D split_flag & EXT4_EXT_DATA_VALID2; if (unwritten) { split_flag1 |=3D EXT4_EXT_MARK_UNWRIT1; @@ -3401,16 +3402,16 @@ static struct ext4_ext_path *ext4_split_extent(hand= le_t *handle, EXT4_EXT_MARK_UNWRIT2); } path =3D ext4_split_extent_at(handle, inode, path, - map->m_lblk, split_flag1, flags); + lblk, split_flag1, flags); if (IS_ERR(path)) return path; } =20 if (allocated) { - if (map->m_lblk + map->m_len > ee_block + ee_len) - *allocated =3D ee_len - (map->m_lblk - ee_block); + if (lblk + len > ee_block + ee_len) + *allocated =3D ee_len - (lblk - ee_block); else - *allocated =3D map->m_len; + *allocated =3D len; } ext4_ext_show_leaf(inode, path); return path; @@ -3658,8 +3659,8 @@ ext4_ext_convert_to_initialized(handle_t *handle, str= uct inode *inode, } =20 fallback: - path =3D ext4_split_extent(handle, inode, path, &split_map, split_flag, - flags, NULL); + path =3D ext4_split_extent(handle, inode, path, split_map.m_lblk, + split_map.m_len, split_flag, flags, NULL); if (IS_ERR(path)) return path; out: @@ -3699,11 +3700,11 @@ ext4_ext_convert_to_initialized(handle_t *handle, s= truct inode *inode, * allocated pointer. Return an extent path pointer on success, or an error * pointer on failure. */ -static struct ext4_ext_path *ext4_split_convert_extents(handle_t *handle, - struct inode *inode, - struct ext4_map_blocks *map, - struct ext4_ext_path *path, - int flags, unsigned int *allocated) +static struct ext4_ext_path * +ext4_split_convert_extents(handle_t *handle, struct inode *inode, + ext4_lblk_t lblk, unsigned int len, + struct ext4_ext_path *path, int flags, + unsigned int *allocated) { ext4_lblk_t eof_block; ext4_lblk_t ee_block; @@ -3712,12 +3713,12 @@ static struct ext4_ext_path *ext4_split_convert_ext= ents(handle_t *handle, int split_flag =3D 0, depth; =20 ext_debug(inode, "logical block %llu, max_blocks %u\n", - (unsigned long long)map->m_lblk, map->m_len); + (unsigned long long)lblk, len); =20 eof_block =3D (EXT4_I(inode)->i_disksize + inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits; - if (eof_block < map->m_lblk + map->m_len) - eof_block =3D map->m_lblk + map->m_len; + if (eof_block < lblk + len) + eof_block =3D lblk + len; /* * It is safe to convert extent to initialized via explicit * zeroout only if extent is fully inside i_size or new_size. @@ -3737,8 +3738,8 @@ static struct ext4_ext_path *ext4_split_convert_exten= ts(handle_t *handle, split_flag |=3D (EXT4_EXT_MARK_UNWRIT2 | EXT4_EXT_DATA_VALID2); } flags |=3D EXT4_GET_BLOCKS_PRE_IO; - return ext4_split_extent(handle, inode, path, map, split_flag, flags, - allocated); + return ext4_split_extent(handle, inode, path, lblk, len, split_flag, + flags, allocated); } =20 static struct ext4_ext_path * @@ -3773,7 +3774,7 @@ ext4_convert_unwritten_extents_endio(handle_t *handle= , struct inode *inode, inode->i_ino, (unsigned long long)ee_block, ee_len, (unsigned long long)map->m_lblk, map->m_len); #endif - path =3D ext4_split_convert_extents(handle, inode, map, path, + path =3D ext4_split_convert_extents(handle, inode, map->m_lblk, map->m_l= en, path, EXT4_GET_BLOCKS_CONVERT, NULL); if (IS_ERR(path)) return path; @@ -3837,8 +3838,9 @@ convert_initialized_extent(handle_t *handle, struct i= node *inode, (unsigned long long)ee_block, ee_len); =20 if (ee_block !=3D map->m_lblk || ee_len > map->m_len) { - path =3D ext4_split_convert_extents(handle, inode, map, path, - EXT4_GET_BLOCKS_CONVERT_UNWRITTEN, NULL); + path =3D ext4_split_convert_extents( + handle, inode, map->m_lblk, map->m_len, path, + EXT4_GET_BLOCKS_CONVERT_UNWRITTEN, NULL); if (IS_ERR(path)) return path; =20 @@ -3909,8 +3911,9 @@ ext4_ext_handle_unwritten_extents(handle_t *handle, s= truct inode *inode, =20 /* get_block() before submitting IO, split the extent */ if (flags & EXT4_GET_BLOCKS_PRE_IO) { - path =3D ext4_split_convert_extents(handle, inode, map, path, - flags | EXT4_GET_BLOCKS_CONVERT, allocated); + path =3D ext4_split_convert_extents( + handle, inode, map->m_lblk, map->m_len, path, + flags | EXT4_GET_BLOCKS_CONVERT, allocated); if (IS_ERR(path)) return path; /* --=20 2.43.5 From nobody Sat Nov 30 05:39:24 2024 Received: from mx0b-001b2d01.pphosted.com (mx0b-001b2d01.pphosted.com [148.163.158.5]) (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 4E277184527; Wed, 11 Sep 2024 09:01:34 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=148.163.158.5 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1726045297; cv=none; b=T8gbrHFQ6H33KKfMjENpicy2O+sNDB4nAcZAD3p4/5qZHM20knCeK4isWHjkcWvqhMrz/qfB4nb1SZPNRE3A+/zRBHiu3jtkQsIII8KUzLd1KAgI1F23KOeAvvY+DXRfqMVfWaTQ4SD2pE1Aj4a2DGprjZqsZQ9vPhRC3FW3jvQ= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1726045297; c=relaxed/simple; bh=SEg/KNLitmbBup+3QKSOFv/JByCjAbCK3j4RxZ4VLV0=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=rQK1ROQEsk77p4PkdVdPRytQa6xL97AXvc2FeO41BFT0OU72TjqhB5eJLtnRTCt8h9irXmW9JIL1io64QkRC6Jg+wqUYusvYVFJohYBpBhn/82x9vQoi7Gsr+pJRXfsjLlwEgvCxq9RJ4eZO3lgAAmguCDEXRBYHMq6+w7t8a10= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.ibm.com; spf=pass smtp.mailfrom=linux.ibm.com; dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com header.b=gv8jR020; arc=none smtp.client-ip=148.163.158.5 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.ibm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.ibm.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com header.b="gv8jR020" Received: from pps.filterd (m0353725.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 48B0eHfX023695; Wed, 11 Sep 2024 09:01:26 GMT DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ibm.com; h=from :to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; s=pp1; bh=2k+gGoLO4kgdd XknagWGoo8nUxN26+gEWRmWqW88wXc=; b=gv8jR020JIERW00lk8XxT34RSVzef ZicxucFfi7SfuOBvTH9yy4rTRLPAcPMcpphPPVxkGKeFRAhfdehs3p8fQ3bVW41T V/5ZmDN5cdsaDLujEN3jLvAY08KziV4/eNSGvp6SQNb4dBOIIzLQ+wVokZSKSWnJ nbDyRvh18uAPcxgHZm+mHvLe4B9w4MsYA4d3iOFEq7WDlopwrptCP3l4Dt1VjEDG oqas/E/YbulJDLf/0UQr/foAAeJY2L+jm8U3f0KthEwdAANskaxlLMBMc+J8M2u9 tbccmA6IenhE17VMaJd1Pnh2IDxxfgGsKg2sIHV+S40AmJyR6KKfMtgYA== Received: from pps.reinject (localhost [127.0.0.1]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 41gd8kmefb-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:25 +0000 (GMT) Received: from m0353725.ppops.net (m0353725.ppops.net [127.0.0.1]) by pps.reinject (8.18.0.8/8.18.0.8) with ESMTP id 48B8vNEK018725; Wed, 11 Sep 2024 09:01:25 GMT Received: from ppma11.dal12v.mail.ibm.com (db.9e.1632.ip4.static.sl-reverse.com [50.22.158.219]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 41gd8kmef7-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:25 +0000 (GMT) Received: from pps.filterd (ppma11.dal12v.mail.ibm.com [127.0.0.1]) by ppma11.dal12v.mail.ibm.com (8.18.1.2/8.18.1.2) with ESMTP id 48B8IF22027389; Wed, 11 Sep 2024 09:01:24 GMT Received: from smtprelay06.fra02v.mail.ibm.com ([9.218.2.230]) by ppma11.dal12v.mail.ibm.com (PPS) with ESMTPS id 41h3v381jw-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 11 Sep 2024 09:01:24 +0000 Received: from smtpav06.fra02v.mail.ibm.com (smtpav06.fra02v.mail.ibm.com [10.20.54.105]) by smtprelay06.fra02v.mail.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id 48B91M7334669082 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Wed, 11 Sep 2024 09:01:22 GMT Received: from smtpav06.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 6C68E2004B; Wed, 11 Sep 2024 09:01:22 +0000 (GMT) Received: from smtpav06.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 950FC20040; Wed, 11 Sep 2024 09:01:20 +0000 (GMT) Received: from li-bb2b2a4c-3307-11b2-a85c-8fa5c3a69313.in.ibm.com (unknown [9.109.253.82]) by smtpav06.fra02v.mail.ibm.com (Postfix) with ESMTP; Wed, 11 Sep 2024 09:01:20 +0000 (GMT) From: Ojaswin Mujoo To: linux-ext4@vger.kernel.org, "Theodore Ts'o" Cc: Ritesh Harjani , linux-kernel@vger.kernel.org, "Darrick J . Wong" , linux-fsdevel@vger.kernel.org, John Garry , dchinner@redhat.com Subject: [RFC 5/5] ext4: Add extsize hint support Date: Wed, 11 Sep 2024 14:31:09 +0530 Message-ID: X-Mailer: git-send-email 2.43.5 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-TM-AS-GCONF: 00 X-Proofpoint-GUID: VOUTpTc5yP-BHojfpmxz05lRnEL_Vkaf X-Proofpoint-ORIG-GUID: Zkd1BamM66ljODppdIGqa4BU4-afKX6_ X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-10_12,2024-09-09_02,2024-09-02_01 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 priorityscore=1501 impostorscore=0 suspectscore=0 clxscore=1015 phishscore=0 spamscore=0 lowpriorityscore=0 adultscore=0 mlxscore=0 mlxlogscore=999 bulkscore=0 malwarescore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.19.0-2408220000 definitions=main-2409110064 Content-Type: text/plain; charset="utf-8" Now that the ioctl is in place, add the underlying infrastructure to support extent size hints. ** MOTIVATION ** 1. This feature allows us to ask the allocator for blocks that are logically AS WELL AS physically aligned to an extent size hint (aka extsize), that is generally a power of 2. 2. This means both start and the length of the physical and logical range should be aligned to the extsize. 3. This sets up the infra we'll eventually need for supporting non-torn/atomic writes that need to follow a certain alignment as required by hardware. 4. This can also be extent to other use cases like stripe alignment ** DESIGN NOTES ** * Physical Alignment * 1. Since the extsize is always a power-of-2 (for now) in fs blocks, we leverage CR_POWER2_ALIGNED allocation to get the blocks. This ensures the b= locks are physically aligned 2. Since this is just a hint, incase we are not able to get any aligned blocks we simply drop back to non aligned allocation. * Logical Alignment * The flow of extsize aligned allocation with bufferred and direct IO: +--------------------------------------------------------+ | Buffered IO | +--------------------------------------------------------+ | ext4_map_blocks() call with extsize allocation | +--------------------------------------------------------+ | +--------------------------------------------+ | Adjust lblk and len to align to extsize | +--------------------------------------------+ | +--------------------------------------------------------+ |Pre-existing written/unwritten blocks in extsize range? | +--------------------------+-----------------------------+ YES NO | | +---------------v---------------+ +-----------------v-------------= ------+ | Covers orig range? | | Allocate extsize range = | +---------------+---------------+ +---------------------+---------= -----+ | | | YES NO | | | +---------------v-------------= -----+ +--------v-------+ +-------v---------+ | Mark allocated extent as = | | Return blocks | | Fallback to | | unwritten = | +----------------+ | non-extsize | +----------------+------------= -----+ | allocation | | +-----------------+ +----------------v------------= -----+ | Insert extsize extent = | | into tree = | +----------------+------------= -----+ | +----------------v------------= -----+ | Return allocated blocks = | +-----------------------------= -----+ +--------------------------------------------+ | During writeback: | +--------------------------------------------+ | Use PRE_IO to split only the dirty extent | +--------------------------------------------+ +--------------------------------------------+ | After IO: | +--------------------------------------------+ | Convert the extent under IO to written | +--------------------------------------------+ Same flow for direct IO: +----------------------------------------------------------------------+ | Direct IO | +----------------------------------------------------------------------+ | ext4_map_blocks() called with extsize allocation and PRE-IO | +----------------------------------------------------------------------+ | +----------------------------------------------------------------------+ | Adjust lblk and len to align to extsize | +----------------------------------------------------------------------+ | +----------------------------------------------------------------------+ | Pre-existing written blocks in extsize range? | +----------------------------------+-----------------------------------+ YES NO | | +---------v----------+ +------------------v-----------------+ | Covers orig range? | | Unwritten blocks in extsize range? | +---------+----------+ +------------------+-----------------+ | | | | YES NO YES NO | | | | +-------v----+ +-----v--------+ +----------v----------+ +-------v-------= ---+ | Return | | Fallback to | | Call ext4_ext_map_ | | Allocate extsi= ze | | blocks | | non-extsize | | blocks() ->ext4_ext | | range = | +------------+ | allocation | | _handle_unwritten_ | +-------+-------= ---+ +--------------+ | extents() | | +----------+----------+ +-------v-------= ---+ | | Mark complete= | +----------v----------+ | range unwritte= n | | Split orig range | | & insert in = | | from bigger | | tree = | | unwritten extent | +-------+-------= ---+ +----------+----------+ | | +-------v-------= ---+ +----------v----------+ | Split orig ran= ge | | Mark split extent | | from bigger = | | as unwritten | | allocated exte= nt | +----------+----------+ +-------+-------= ---+ | | +----------v----------+ +-------v-------= ---+ | Return split extent | | Mark split ext= ent| | to user | | as unwritten= | +---------------------+ +-------+-------= ---+ | +-------v-------= ---+ | Return split = | | extent to user= | +---------------= ---+ +--------------------------------------------+ | After IO: | +--------------------------------------------+ | Convert the extent under IO to written | +--------------------------------------------+ ** IMPLEMENTATION NOTES ** * Callers of ext4_map_blocks work under the assumption that ext4_map_blocks will always only return as much as requested or less but now we might end up allocating more so make changes to ext4_map_blocks to make sure we adjust the allocated map to only return as much as user requested. * Further, we now maintain 2 maps in ext4_map_blocks - the original map and the extsize map that is used when extsize hint allocation is taking place. We also pass these 2 maps down because some functions might now need information of the original map as well as the extsize map. * For example, when we go for direct IO and there's a hole in the orig range requested, we allocate based on extsize range and then split the bigger unwritten extent onto smaller unwritten extents based on orig range. (Needed so we dont have to split after IO). For this, we need the information of extsize range as well as orig range hence 2 maps. * Since now we allocate more than the user requested, to avoid stale data exposure, we mark the bigger extsize extent as unwritten and then use the similar flow of dioread_nolock to only mark the extent under write as written. * We disable extsize hints when writes are beyond EOF. * When extsize is set on an inode, we drop to no delalloc allocations similar to XFS. Signed-off-by: Ojaswin Mujoo --- fs/ext4/ext4.h | 4 +- fs/ext4/ext4_jbd2.h | 15 ++ fs/ext4/extents.c | 169 +++++++++++++++-- fs/ext4/inode.c | 353 ++++++++++++++++++++++++++++++++---- include/trace/events/ext4.h | 1 + 5 files changed, 491 insertions(+), 51 deletions(-) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index d34e60cf6458..0782f4268c86 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -722,6 +722,7 @@ enum { #define EXT4_GET_BLOCKS_IO_SUBMIT 0x0400 /* Caller is in the atomic contex, find extent if it has been cached */ #define EXT4_GET_BLOCKS_CACHED_NOWAIT 0x0800 +#define EXT4_GET_BLOCKS_EXTSIZE 0x1000 =20 /* * The bit position of these flags must not overlap with any of the @@ -3702,7 +3703,8 @@ struct ext4_extent; extern void ext4_ext_tree_init(handle_t *handle, struct inode *inode); extern int ext4_ext_index_trans_blocks(struct inode *inode, int extents); extern int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, - struct ext4_map_blocks *map, int flags); + struct ext4_map_blocks *orig_map, + struct ext4_map_blocks *extsize_map, int flags); extern int ext4_ext_truncate(handle_t *, struct inode *); extern int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start, ext4_lblk_t end); diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h index 0c77697d5e90..20bbf76c0556 100644 --- a/fs/ext4/ext4_jbd2.h +++ b/fs/ext4/ext4_jbd2.h @@ -513,4 +513,19 @@ static inline int ext4_should_dioread_nolock(struct in= ode *inode) return 1; } =20 +static inline int ext4_should_use_extsize(struct inode *inode) +{ + if (!S_ISREG(inode->i_mode)) + return 0; + if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) + return 0; + return (ext4_inode_get_extsize(EXT4_I(inode)) > 0); +} + +static inline int ext4_should_use_unwrit_extents(struct inode *inode) +{ + return (ext4_should_dioread_nolock(inode) || + ext4_should_use_extsize(inode)); +} + #endif /* _EXT4_JBD2_H */ diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 94aeb5b47971..8a37a1ddfc23 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3889,15 +3889,24 @@ convert_initialized_extent(handle_t *handle, struct= inode *inode, =20 static struct ext4_ext_path * ext4_ext_handle_unwritten_extents(handle_t *handle, struct inode *inode, - struct ext4_map_blocks *map, + struct ext4_map_blocks *orig_map, + struct ext4_map_blocks *extsize_map, struct ext4_ext_path *path, int flags, unsigned int *allocated, ext4_fsblk_t newblock) { + struct ext4_map_blocks *map; int err =3D 0; =20 - ext_debug(inode, "logical block %llu, max_blocks %u, flags 0x%x, allocate= d %u\n", - (unsigned long long)map->m_lblk, map->m_len, flags, - *allocated); + if (flags & EXT4_GET_BLOCKS_EXTSIZE) { + BUG_ON(extsize_map =3D=3D NULL); + map =3D extsize_map; + } else + map =3D orig_map; + + ext_debug( + inode, + "logical block %llu, max_blocks %u, flags 0x%x, allocated %u\n", + (unsigned long long)map->m_lblk, map->m_len, flags, *allocated); ext4_ext_show_leaf(inode, path); =20 /* @@ -3906,13 +3915,14 @@ ext4_ext_handle_unwritten_extents(handle_t *handle,= struct inode *inode, */ flags |=3D EXT4_GET_BLOCKS_METADATA_NOFAIL; =20 - trace_ext4_ext_handle_unwritten_extents(inode, map, flags, - *allocated, newblock); + trace_ext4_ext_handle_unwritten_extents(inode, map, flags, *allocated, + newblock); =20 /* get_block() before submitting IO, split the extent */ if (flags & EXT4_GET_BLOCKS_PRE_IO) { + /* Split should always happen based on original mapping */ path =3D ext4_split_convert_extents( - handle, inode, map->m_lblk, map->m_len, path, + handle, inode, orig_map->m_lblk, orig_map->m_len, path, flags | EXT4_GET_BLOCKS_CONVERT, allocated); if (IS_ERR(path)) return path; @@ -3927,11 +3937,19 @@ ext4_ext_handle_unwritten_extents(handle_t *handle,= struct inode *inode, err =3D -EFSCORRUPTED; goto errout; } + + /* + * For extsize case we need to adjust lblk to start of split + * extent because the m_len will be set to len of split extent. + * No change for non extsize case + */ + map->m_lblk =3D orig_map->m_lblk; map->m_flags |=3D EXT4_MAP_UNWRITTEN; goto out; } /* IO end_io complete, convert the filled extent to written */ if (flags & EXT4_GET_BLOCKS_CONVERT) { + BUG_ON(map =3D=3D extsize_map); path =3D ext4_convert_unwritten_extents_endio(handle, inode, map, path); if (IS_ERR(path)) @@ -4189,7 +4207,8 @@ static ext4_lblk_t ext4_ext_determine_insert_hole(str= uct inode *inode, * return < 0, error case. */ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, - struct ext4_map_blocks *map, int flags) + struct ext4_map_blocks *orig_map, + struct ext4_map_blocks *extsize_map, int flags) { struct ext4_ext_path *path =3D NULL; struct ext4_extent newex, *ex, ex2; @@ -4200,6 +4219,17 @@ int ext4_ext_map_blocks(handle_t *handle, struct ino= de *inode, unsigned int allocated_clusters =3D 0; struct ext4_allocation_request ar; ext4_lblk_t cluster_offset; + struct ext4_map_blocks *map; +#ifdef CONFIG_EXT4_DEBUG + struct ext4_ext_path *test_path =3D NULL; +#endif + + if (flags & EXT4_GET_BLOCKS_EXTSIZE) { + BUG_ON(extsize_map =3D=3D NULL); + map =3D extsize_map; + } else + map =3D orig_map; + =20 ext_debug(inode, "blocks %u/%u requested\n", map->m_lblk, map->m_len); trace_ext4_ext_map_blocks_enter(inode, map->m_lblk, map->m_len, flags); @@ -4256,6 +4286,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inod= e *inode, */ if ((!ext4_ext_is_unwritten(ex)) && (flags & EXT4_GET_BLOCKS_CONVERT_UNWRITTEN)) { + BUG_ON(map =3D=3D extsize_map); path =3D convert_initialized_extent(handle, inode, map, path, &allocated); if (IS_ERR(path)) @@ -4272,8 +4303,8 @@ int ext4_ext_map_blocks(handle_t *handle, struct inod= e *inode, } =20 path =3D ext4_ext_handle_unwritten_extents( - handle, inode, map, path, flags, - &allocated, newblock); + handle, inode, orig_map, extsize_map, path, + flags, &allocated, newblock); if (IS_ERR(path)) err =3D PTR_ERR(path); goto out; @@ -4306,6 +4337,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inod= e *inode, */ if (cluster_offset && ex && get_implied_cluster_alloc(inode->i_sb, map, ex, path)) { + BUG_ON(map =3D=3D extsize_map); ar.len =3D allocated =3D map->m_len; newblock =3D map->m_pblk; goto got_allocated_blocks; @@ -4325,6 +4357,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inod= e *inode, * cluster we can use. */ if ((sbi->s_cluster_ratio > 1) && err && get_implied_cluster_alloc(inode->i_sb, map, &ex2, path)) { + BUG_ON(map =3D=3D extsize_map); ar.len =3D allocated =3D map->m_len; newblock =3D map->m_pblk; err =3D 0; @@ -4379,6 +4412,8 @@ int ext4_ext_map_blocks(handle_t *handle, struct inod= e *inode, ar.flags |=3D EXT4_MB_DELALLOC_RESERVED; if (flags & EXT4_GET_BLOCKS_METADATA_NOFAIL) ar.flags |=3D EXT4_MB_USE_RESERVED; + if (flags & EXT4_GET_BLOCKS_EXTSIZE) + ar.flags |=3D EXT4_MB_HINT_ALIGNED; newblock =3D ext4_mb_new_blocks(handle, &ar, &err); if (!newblock) goto out; @@ -4400,9 +4435,114 @@ int ext4_ext_map_blocks(handle_t *handle, struct in= ode *inode, map->m_flags |=3D EXT4_MAP_UNWRITTEN; } =20 - path =3D ext4_ext_insert_extent(handle, inode, path, &newex, flags); - if (IS_ERR(path)) { - err =3D PTR_ERR(path); + if ((flags & EXT4_GET_BLOCKS_EXTSIZE) && + (flags & EXT4_GET_BLOCKS_PRE_IO)) { + /* + * With EXTSIZE and PRE-IO (direct io case) we have to be careful + * because we want to insert the complete extent but split only the + * originally requested range. + * + * Below are the different (S)cenarios and the (A)ction we take: + * + * S1: New extent covers the original range completely/partially. + * A1: Insert new extent, allow merges. Then split the original + * range from this. Adjust the length of split if new extent only + * partially covers original. + * + * S2: New extent doesn't cover original range at all + * A2: Just insert this range and return. Rest is handled in + * ext4_map_blocks() + * NOTE: We can handle this as an error with EAGAIN in future. + */ + ext4_lblk_t newex_lblk =3D le32_to_cpu(newex.ee_block); + loff_t newex_len =3D ext4_ext_get_actual_len(&newex); + + if (in_range(orig_map->m_lblk, newex_lblk, newex_len)) { + /* S1 */ + loff_t split_len =3D 0; + + BUG_ON(!ext4_ext_is_unwritten(&newex)); + + if (newex_lblk + newex_len >=3D + orig_map->m_lblk + (loff_t)orig_map->m_len) + split_len =3D orig_map->m_len; + else + split_len =3D newex_len - + (orig_map->m_lblk - newex_lblk); + + path =3D ext4_ext_insert_extent( + handle, inode, path, &newex, + (flags & ~EXT4_GET_BLOCKS_PRE_IO)); + if (IS_ERR(path)) { + err =3D PTR_ERR(path); + goto insert_error; + } + + /* + * Update path before split + * NOTE: This might no longer be needed with recent + * changes in ext4_ext_insert_extent() + */ + path =3D ext4_find_extent(inode, orig_map->m_lblk, path, 0); + if (IS_ERR(path)) { + err =3D PTR_ERR(path); + goto insert_error; + } + + /* + * GET_BLOCKS_CONVERT is needed to make sure split + * extent is marked unwritten although the flags itself + * means that the extent should be converted to written. + * + * TODO: This is because ext4_split_convert_extents() + * doesn't respect the flags at all but fixing this + * needs more involved design changes. + */ + path =3D ext4_split_convert_extents( + handle, inode, orig_map->m_lblk, split_len, + path, flags | EXT4_GET_BLOCKS_CONVERT, NULL); + if (IS_ERR(path)) { + err =3D PTR_ERR(path); + goto insert_error; + } + +#ifdef CONFIG_EXT4_DEBUG + test_path =3D ext4_find_extent(inode, orig_map->m_lblk, + NULL, 0); + if (!IS_ERR(test_path)) { + /* Confirm we've correctly split and marked the extent unwritten */ + struct ext4_extent *test_ex =3D + test_path[ext_depth(inode)].p_ext; + WARN_ON(!ext4_ext_is_unwritten(test_ex)); + WARN_ON(test_ex->ee_block !=3D orig_map->m_lblk); + WARN_ON(ext4_ext_get_actual_len(test_ex) !=3D + orig_map->m_len); + kfree(test_path); + } +#endif + } else { + /* S2 */ + BUG_ON(orig_map->m_lblk < newex_lblk + newex_len); + + path =3D ext4_ext_insert_extent( + handle, inode, path, &newex, + (flags & ~EXT4_GET_BLOCKS_PRE_IO)); + if (IS_ERR(path)) { + err =3D PTR_ERR(path); + goto insert_error; + } + } + } else { + path =3D ext4_ext_insert_extent(handle, inode, path, &newex, + flags); + if (IS_ERR(path)) { + err =3D PTR_ERR(path); + goto insert_error; + } + } + +insert_error: + if (err) { if (allocated_clusters) { int fb_flags =3D 0; =20 @@ -4793,6 +4933,9 @@ long ext4_fallocate(struct file *file, int mode, loff= _t offset, loff_t len) =20 inode_lock(inode); =20 + if (ext4_should_use_extsize(inode)) + flags |=3D EXT4_GET_BLOCKS_EXTSIZE; + /* * We only support preallocation for extent-based files only */ diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 898b41751cf4..675f24ba009a 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -434,7 +434,7 @@ static void ext4_map_blocks_es_recheck(handle_t *handle, */ down_read(&EXT4_I(inode)->i_data_sem); if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) { - retval =3D ext4_ext_map_blocks(handle, inode, map, 0); + retval =3D ext4_ext_map_blocks(handle, inode, map, NULL, 0); } else { retval =3D ext4_ind_map_blocks(handle, inode, map, 0); } @@ -459,15 +459,33 @@ static void ext4_map_blocks_es_recheck(handle_t *hand= le, #endif /* ES_AGGRESSIVE_TEST */ =20 static int ext4_map_query_blocks(handle_t *handle, struct inode *inode, - struct ext4_map_blocks *map) + struct ext4_map_blocks *orig_map, + struct ext4_map_blocks *extsize_map, + bool should_extsize) { unsigned int status; int retval; + struct ext4_map_blocks *map; + + if (should_extsize) { + BUG_ON(extsize_map =3D=3D NULL); + map =3D extsize_map; + } else + map =3D orig_map; =20 if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) - retval =3D ext4_ext_map_blocks(handle, inode, map, 0); - else + if (should_extsize) { + retval =3D ext4_ext_map_blocks(handle, inode, orig_map, + map, + EXT4_GET_BLOCKS_EXTSIZE); + } else { + retval =3D ext4_ext_map_blocks(handle, inode, map, NULL, + 0); + } + else { + BUG_ON(should_extsize); retval =3D ext4_ind_map_blocks(handle, inode, map, 0); + } =20 if (retval <=3D 0) return retval; @@ -488,11 +506,20 @@ static int ext4_map_query_blocks(handle_t *handle, st= ruct inode *inode, } =20 static int ext4_map_create_blocks(handle_t *handle, struct inode *inode, - struct ext4_map_blocks *map, int flags) + struct ext4_map_blocks *orig_map, + struct ext4_map_blocks *extsize_map, int flags, + bool should_extsize) { struct extent_status es; unsigned int status; int err, retval =3D 0; + struct ext4_map_blocks *map; + + if (should_extsize) { + BUG_ON(extsize_map =3D=3D NULL); + map =3D extsize_map; + } else + map =3D orig_map; =20 /* * We pass in the magic EXT4_GET_BLOCKS_DELALLOC_RESERVE @@ -513,8 +540,15 @@ static int ext4_map_create_blocks(handle_t *handle, st= ruct inode *inode, * changed the inode type in between. */ if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) { - retval =3D ext4_ext_map_blocks(handle, inode, map, flags); + if (should_extsize) { + retval =3D ext4_ext_map_blocks(handle, inode, orig_map, + map, flags); + } else { + retval =3D ext4_ext_map_blocks(handle, inode, map, NULL, + flags); + } } else { + BUG_ON(should_extsize); retval =3D ext4_ind_map_blocks(handle, inode, map, flags); =20 /* @@ -569,6 +603,80 @@ static int ext4_map_create_blocks(handle_t *handle, st= ruct inode *inode, return retval; } =20 +/** + * Extsize hint will change the mapped range and hence we'll end up mappin= g more. + * To not confuse the caller, adjust the struct ext4_map_blocks to reflect= the + * original mapping requested by them. + * + * @cur_map: The block mapping we are working with (for sanity check) + * @orig_map: The originally requested mapping + * @extsize_map: The mapping after adjusting for extsize hint + * @flags Get block flags (for sanity check) + * + * This function assumes that the orig_mlblk is contained within the mappi= ng + * held in extsize_map. Caller must make sure this is true. + */ +static inline unsigned int ext4_extsize_adjust_map(struct ext4_map_blocks = *cur_map, + struct ext4_map_blocks *orig_map, + struct ext4_map_blocks *extsize_map, + int flags) +{ + __u64 map_end =3D (__u64)extsize_map->m_lblk + extsize_map->m_len; + + BUG_ON(cur_map !=3D extsize_map || !(flags & EXT4_GET_BLOCKS_EXTSIZE)); + + orig_map->m_len =3D min(orig_map->m_len, map_end - orig_map->m_lblk); + orig_map->m_pblk =3D + extsize_map->m_pblk + (orig_map->m_lblk - extsize_map->m_lblk); + orig_map->m_flags =3D extsize_map->m_flags; + + return orig_map->m_len; +} + +/** + * ext4_error_adjust_map - Adjust map returned upon error in ext4_map_bloc= ks() + * + * @cur_map: current map we are working with + * @orig_map: original map that would be returned to the user. + * + * Most of the callers of ext4_map_blocks() ignore the map on error, howev= er + * some use it for debug logging. In this case, they log state of the map = just + * before the error, hence this function ensures that map returned to call= er is + * the one we were working with when error happened. Mostly useful when ex= tsize + * hints are enabled. + */ +static inline void ext4_error_adjust_map(struct ext4_map_blocks *cur_map, + struct ext4_map_blocks *orig_map) +{ + if (cur_map !=3D orig_map) + memcpy(orig_map, cur_map, sizeof(*cur_map)); +} + +/* + * This functions resets the mapping to it's original state after it has b= een + * modified due to extent size hint and drops the extsize hint. To be used + * incase we want to fallback from extsize based aligned allocation to nor= mal + * allocation + * + * @map: The block mapping where lblk and len have been modified + * because of extsize hint + * @flags: The get_block flags + * @orig_mlblk: The originally requested logical block to map + * @orig_mlen: The originally requested len to map + * @orig_flags: The originally requested get_block flags + */ +static inline void ext4_extsize_reset_map(struct ext4_map_blocks *map, + int *flags, ext4_lblk_t orig_mlblk, + unsigned int orig_mlen, + int orig_flags) +{ + /* Drop the extsize hint from original flags */ + *flags =3D orig_flags & ~EXT4_GET_BLOCKS_EXTSIZE; + map->m_lblk =3D orig_mlblk; + map->m_len =3D orig_mlen; + map->m_flags =3D 0; +} + /* * The ext4_map_blocks() function tries to look up the requested blocks, * and returns if the blocks are already mapped. @@ -593,31 +701,110 @@ static int ext4_map_create_blocks(handle_t *handle, = struct inode *inode, * It returns the error in case of allocation failure. */ int ext4_map_blocks(handle_t *handle, struct inode *inode, - struct ext4_map_blocks *map, int flags) + struct ext4_map_blocks *orig_map, int flags) { struct extent_status es; int retval; int ret =3D 0; + + ext4_lblk_t orig_mlblk, extsize_mlblk; + unsigned int orig_mlen, extsize_mlen; + int orig_flags; + + struct ext4_map_blocks *map =3D NULL; + struct ext4_map_blocks extsize_map =3D {0}; + + __u32 extsize =3D ext4_inode_get_extsize(EXT4_I(inode)); + bool should_extsize =3D false; + #ifdef ES_AGGRESSIVE_TEST - struct ext4_map_blocks orig_map; + struct ext4_map_blocks test_map; =20 - memcpy(&orig_map, map, sizeof(*map)); + memcpy(&test_map, map, sizeof(*map)); #endif =20 - map->m_flags =3D 0; - ext_debug(inode, "flag 0x%x, max_blocks %u, logical block %lu\n", - flags, map->m_len, (unsigned long) map->m_lblk); + orig_map->m_flags =3D 0; + ext_debug(inode, "flag 0x%x, max_blocks %u, logical block %lu\n", flags, + orig_map->m_len, (unsigned long)orig_map->m_lblk); =20 /* * ext4_map_blocks returns an int, and m_len is an unsigned int */ - if (unlikely(map->m_len > INT_MAX)) - map->m_len =3D INT_MAX; + if (unlikely(orig_map->m_len > INT_MAX)) + orig_map->m_len =3D INT_MAX; =20 /* We can handle the block number less than EXT_MAX_BLOCKS */ - if (unlikely(map->m_lblk >=3D EXT_MAX_BLOCKS)) + if (unlikely(orig_map->m_lblk >=3D EXT_MAX_BLOCKS)) return -EFSCORRUPTED; =20 + orig_mlblk =3D orig_map->m_lblk; + orig_mlen =3D orig_map->m_len; + orig_flags =3D flags; + +set_map: + should_extsize =3D (extsize && (flags & EXT4_GET_BLOCKS_CREATE) && + (flags & EXT4_GET_BLOCKS_EXTSIZE)); + if (should_extsize) { + /* + * We adjust the extent size here but we still return the original lblk = and len while + * returning to keep the behavior compatible. + */ + int len, align; + /* + * NOTE: Should we import EXT_UNWRITTEN_MAX_LEN from + * ext4_extents.h here? + */ + int max_unwrit_len =3D ((1UL << 15) - 1); + loff_t end; + + align =3D orig_map->m_lblk % extsize; + len =3D orig_map->m_len + align; + + extsize_map.m_lblk =3D orig_map->m_lblk - align; + extsize_map.m_len =3D + max_t(unsigned int, roundup_pow_of_two(len), extsize); + + /* + * For now allocations beyond EOF don't use extsize hints so + * that we can avoid dealing with extra blocks allocated past + * EOF. We have inode lock since extsize allocations are + * non-delalloc so i_size can be accessed safely + */ + end =3D (extsize_map.m_lblk + (loff_t)extsize_map.m_len) << inode->i_blk= bits; + if (end > inode->i_size) { + flags =3D orig_flags & ~EXT4_GET_BLOCKS_EXTSIZE; + goto set_map; + } + + /* Fallback to normal allocation if we go beyond max len */ + if (extsize_map.m_len >=3D max_unwrit_len) { + flags =3D orig_flags & ~EXT4_GET_BLOCKS_EXTSIZE; + goto set_map; + } + + /* + * We are allocating more than requested. We'll have to convert + * the extent to unwritten and then convert only the part + * requested to written. For now we are using the same flow as + * dioread nolock to achieve this. Hence the caller has to pass + * CREATE_UNWRIT with EXTSIZE + */ + if (!(flags | EXT4_GET_BLOCKS_CREATE_UNWRIT_EXT)) { + WARN_ON(true); + + /* Fallback to non extsize allocation */ + flags =3D orig_flags & ~EXT4_GET_BLOCKS_EXTSIZE; + goto set_map; + } + + extsize_mlblk =3D extsize_map.m_lblk; + extsize_mlen =3D extsize_map.m_len; + + extsize_map.m_flags =3D orig_map->m_flags; + map =3D &extsize_map; + } else + map =3D orig_map; + /* Lookup extent status tree firstly */ if (!(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY) && ext4_es_lookup_extent(inode, map->m_lblk, NULL, &es)) { @@ -647,7 +834,7 @@ int ext4_map_blocks(handle_t *handle, struct inode *ino= de, return retval; #ifdef ES_AGGRESSIVE_TEST ext4_map_blocks_es_recheck(handle, inode, map, - &orig_map, flags); + &test_map, flags); #endif goto found; } @@ -663,19 +850,60 @@ int ext4_map_blocks(handle_t *handle, struct inode *i= node, * file system block. */ down_read(&EXT4_I(inode)->i_data_sem); - retval =3D ext4_map_query_blocks(handle, inode, map); + if (should_extsize) { + BUG_ON(map !=3D &extsize_map); + retval =3D ext4_map_query_blocks(handle, inode, orig_map, map, + should_extsize); + } else { + BUG_ON(map !=3D orig_map); + retval =3D ext4_map_query_blocks(handle, inode, map, NULL, + should_extsize); + } up_read((&EXT4_I(inode)->i_data_sem)); =20 found: if (retval > 0 && map->m_flags & EXT4_MAP_MAPPED) { ret =3D check_block_validity(inode, map); - if (ret !=3D 0) + if (ret !=3D 0) { + ext4_error_adjust_map(map, orig_map); return ret; + } } =20 /* If it is only a block(s) look up */ - if ((flags & EXT4_GET_BLOCKS_CREATE) =3D=3D 0) + if ((flags & EXT4_GET_BLOCKS_CREATE) =3D=3D 0) { + BUG_ON(flags & EXT4_GET_BLOCKS_EXTSIZE); return retval; + } + + /* Handle some special cases when extsize based allocation is needed */ + if (retval >=3D 0 && flags & EXT4_GET_BLOCKS_EXTSIZE) { + bool orig_in_range =3D + in_range(orig_mlblk, (__u64)map->m_lblk, map->m_len); + /* + * Special case: if the extsize range is mapped already and + * covers the original start, we return it. + */ + if (map->m_flags & EXT4_MAP_MAPPED && orig_in_range) { + /* + * We don't use EXTSIZE with CONVERT_UNWRITTEN so + * we can directly return the written extent + */ + return ext4_extsize_adjust_map(map, orig_map, &extsize_map, flags); + } + + /* + * Fallback case: if the found mapping (or hole) doesn't cover + * the extsize required, then just fall back to normal + * allocation to keep things simple. + */ + + if (map->m_lblk !=3D extsize_mlblk || + map->m_len !=3D extsize_mlen) { + flags =3D orig_flags & ~EXT4_GET_BLOCKS_EXTSIZE; + goto set_map; + } + } =20 /* * Returns if the blocks have already allocated @@ -699,12 +927,22 @@ int ext4_map_blocks(handle_t *handle, struct inode *i= node, * with create =3D=3D 1 flag. */ down_write(&EXT4_I(inode)->i_data_sem); - retval =3D ext4_map_create_blocks(handle, inode, map, flags); + if (should_extsize) { + BUG_ON(map !=3D &extsize_map); + retval =3D ext4_map_create_blocks(handle, inode, orig_map, map, flags, + should_extsize); + } else { + BUG_ON(map !=3D orig_map); + retval =3D ext4_map_create_blocks(handle, inode, map, NULL, flags, + should_extsize); + } up_write((&EXT4_I(inode)->i_data_sem)); if (retval > 0 && map->m_flags & EXT4_MAP_MAPPED) { ret =3D check_block_validity(inode, map); - if (ret !=3D 0) + if (ret !=3D 0) { + ext4_error_adjust_map(map, orig_map); return ret; + } =20 /* * Inodes with freshly allocated blocks where contents will be @@ -726,16 +964,38 @@ int ext4_map_blocks(handle_t *handle, struct inode *i= node, else ret =3D ext4_jbd2_inode_add_write(handle, inode, start_byte, length); - if (ret) + if (ret) { + ext4_error_adjust_map(map, orig_map); return ret; + } } } if (retval > 0 && (map->m_flags & EXT4_MAP_UNWRITTEN || map->m_flags & EXT4_MAP_MAPPED)) ext4_fc_track_range(handle, inode, map->m_lblk, map->m_lblk + map->m_len - 1); - if (retval < 0) + + if (retval > 0 && flags & EXT4_GET_BLOCKS_EXTSIZE) { + /* + * In the rare case that we have a short allocation and orig + * lblk doesn't lie in mapped range just try to retry with + * actual allocation. This is not ideal but this should be an + * edge case near ENOSPC. + * + * NOTE: This has a side effect that blocks are allocated but + * not used. Can we avoid that? + */ + if (!in_range(orig_mlblk, (__u64)map->m_lblk, map->m_len)) { + flags =3D orig_flags & ~EXT4_GET_BLOCKS_EXTSIZE; + goto set_map; + } + return ext4_extsize_adjust_map(map, orig_map, &extsize_map, flags); + } + + if (retval < 0) { + ext4_error_adjust_map(map, orig_map); ext_debug(inode, "failed with err %d\n", retval); + } return retval; } =20 @@ -771,18 +1031,20 @@ static int _ext4_get_block(struct inode *inode, sect= or_t iblock, { struct ext4_map_blocks map; int ret =3D 0; + unsigned int orig_mlen =3D bh->b_size >> inode->i_blkbits; =20 if (ext4_has_inline_data(inode)) return -ERANGE; =20 map.m_lblk =3D iblock; - map.m_len =3D bh->b_size >> inode->i_blkbits; + map.m_len =3D orig_mlen; =20 ret =3D ext4_map_blocks(ext4_journal_current_handle(), inode, &map, flags); if (ret > 0) { map_bh(bh, inode->i_sb, map.m_pblk); ext4_update_bh_state(bh, map.m_flags); + WARN_ON(map.m_len !=3D orig_mlen); bh->b_size =3D inode->i_sb->s_blocksize * map.m_len; ret =3D 0; } else if (ret =3D=3D 0) { @@ -808,11 +1070,14 @@ int ext4_get_block_unwritten(struct inode *inode, se= ctor_t iblock, struct buffer_head *bh_result, int create) { int ret =3D 0; + int flags =3D EXT4_GET_BLOCKS_CREATE_UNWRIT_EXT; + + if (ext4_should_use_extsize(inode)) + flags |=3D EXT4_GET_BLOCKS_EXTSIZE; =20 ext4_debug("ext4_get_block_unwritten: inode %lu, create flag %d\n", inode->i_ino, create); - ret =3D _ext4_get_block(inode, iblock, bh_result, - EXT4_GET_BLOCKS_CREATE_UNWRIT_EXT); + ret =3D _ext4_get_block(inode, iblock, bh_result, flags); =20 /* * If the buffer is marked unwritten, mark it as new to make sure it is @@ -1155,7 +1420,8 @@ static int ext4_write_begin(struct file *file, struct= address_space *mapping, from =3D pos & (PAGE_SIZE - 1); to =3D from + len; =20 - if (ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) { + if (!ext4_should_use_extsize(inode) && + ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) { ret =3D ext4_try_to_write_inline_data(mapping, inode, pos, len, pagep); if (ret < 0) @@ -1203,7 +1469,7 @@ static int ext4_write_begin(struct file *file, struct= address_space *mapping, /* In case writeback began while the folio was unlocked */ folio_wait_stable(folio); =20 - if (ext4_should_dioread_nolock(inode)) + if (ext4_should_use_unwrit_extents(inode)) ret =3D ext4_block_write_begin(handle, folio, pos, len, ext4_get_block_unwritten); else @@ -1791,7 +2057,7 @@ static int ext4_da_map_blocks(struct inode *inode, st= ruct ext4_map_blocks *map) if (ext4_has_inline_data(inode)) retval =3D 0; else - retval =3D ext4_map_query_blocks(NULL, inode, map); + retval =3D ext4_map_query_blocks(NULL, inode, map, NULL, false); up_read(&EXT4_I(inode)->i_data_sem); if (retval) return retval < 0 ? retval : 0; @@ -1814,7 +2080,7 @@ static int ext4_da_map_blocks(struct inode *inode, st= ruct ext4_map_blocks *map) goto found; } } else if (!ext4_has_inline_data(inode)) { - retval =3D ext4_map_query_blocks(NULL, inode, map); + retval =3D ext4_map_query_blocks(NULL, inode, map, NULL, false); if (retval) { up_write(&EXT4_I(inode)->i_data_sem); return retval < 0 ? retval : 0; @@ -2188,6 +2454,7 @@ static int mpage_map_one_extent(handle_t *handle, str= uct mpage_da_data *mpd) struct ext4_map_blocks *map =3D &mpd->map; int get_blocks_flags; int err, dioread_nolock; + int extsize =3D ext4_should_use_extsize(inode); =20 trace_ext4_da_write_pages_extent(inode, map); /* @@ -2206,11 +2473,14 @@ static int mpage_map_one_extent(handle_t *handle, s= truct mpage_da_data *mpd) dioread_nolock =3D ext4_should_dioread_nolock(inode); if (dioread_nolock) get_blocks_flags |=3D EXT4_GET_BLOCKS_IO_CREATE_EXT; + if (extsize) + get_blocks_flags |=3D EXT4_GET_BLOCKS_PRE_IO; =20 err =3D ext4_map_blocks(handle, inode, map, get_blocks_flags); if (err < 0) return err; - if (dioread_nolock && (map->m_flags & EXT4_MAP_UNWRITTEN)) { + if ((extsize || dioread_nolock) && + (map->m_flags & EXT4_MAP_UNWRITTEN)) { if (!mpd->io_submit.io_end->handle && ext4_handle_valid(handle)) { mpd->io_submit.io_end->handle =3D handle->h_rsv_handle; @@ -2633,10 +2903,11 @@ static int ext4_do_writepages(struct mpage_da_data = *mpd) } mpd->journalled_more_data =3D 0; =20 - if (ext4_should_dioread_nolock(inode)) { + if (ext4_should_use_unwrit_extents(inode)) { /* - * We may need to convert up to one extent per block in - * the page and we may dirty the inode. + * For extsize allocation or dioread_nolock, we may need to + * convert up to one extent per block in the page and we may + * dirty the inode. */ rsv_blocks =3D 1 + ext4_chunk_trans_blocks(inode, PAGE_SIZE >> inode->i_blkbits); @@ -2911,7 +3182,8 @@ static int ext4_da_write_begin(struct file *file, str= uct address_space *mapping, =20 index =3D pos >> PAGE_SHIFT; =20 - if (ext4_nonda_switch(inode->i_sb) || ext4_verity_in_progress(inode)) { + if (ext4_nonda_switch(inode->i_sb) || ext4_verity_in_progress(inode) || + ext4_should_use_extsize(inode)) { *fsdata =3D (void *)FALL_BACK_TO_NONDELALLOC; return ext4_write_begin(file, mapping, pos, len, pagep, fsdata); @@ -3355,12 +3627,19 @@ static int ext4_iomap_alloc(struct inode *inode, st= ruct ext4_map_blocks *map, * can complete at any point during the I/O and subsequently push the * i_disksize out to i_size. This could be beyond where direct I/O is * happening and thus expose allocated blocks to direct I/O reads. + * + * NOTE for extsize hints: We only support it for writes inside + * EOF (for now) to not have to deal with blocks past EOF */ else if (((loff_t)map->m_lblk << blkbits) >=3D i_size_read(inode)) m_flags =3D EXT4_GET_BLOCKS_CREATE; - else if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) + else if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) { m_flags =3D EXT4_GET_BLOCKS_IO_CREATE_EXT; =20 + if (ext4_should_use_extsize(inode) && retries =3D=3D 0) + m_flags |=3D EXT4_GET_BLOCKS_EXTSIZE; + } + ret =3D ext4_map_blocks(handle, inode, map, m_flags); =20 /* @@ -6175,7 +6454,7 @@ vm_fault_t ext4_page_mkwrite(struct vm_fault *vmf) } folio_unlock(folio); /* OK, we need to fill the hole... */ - if (ext4_should_dioread_nolock(inode)) + if (ext4_should_use_unwrit_extents(inode)) get_block =3D ext4_get_block_unwritten; else get_block =3D ext4_get_block; diff --git a/include/trace/events/ext4.h b/include/trace/events/ext4.h index 05441f87c5d2..0942578a97bc 100644 --- a/include/trace/events/ext4.h +++ b/include/trace/events/ext4.h @@ -50,6 +50,7 @@ struct partial_cluster; { EXT4_GET_BLOCKS_CONVERT_UNWRITTEN, "CONVERT_UNWRITTEN" }, \ { EXT4_GET_BLOCKS_ZERO, "ZERO" }, \ { EXT4_GET_BLOCKS_IO_SUBMIT, "IO_SUBMIT" }, \ + { EXT4_GET_BLOCKS_EXTSIZE, "EXTSIZE" }, \ { EXT4_EX_NOCACHE, "EX_NOCACHE" }) =20 /* --=20 2.43.5