linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Ritesh Harjani (IBM)" <ritesh.list@gmail.com>
To: Theodore Ts'o <tytso@mit.edu>
Cc: linux-ext4@vger.kernel.org,
	Harshad Shirwadkar <harshadshirwadkar@gmail.com>,
	Wang Shilong <wshilong@ddn.com>,
	Andreas Dilger <adilger.kernel@dilger.ca>, Li Xi <lixi@ddn.com>,
	Ritesh Harjani <ritesh.list@gmail.com>
Subject: [RFCv1 29/72] e2fsck: add start/end group for thread
Date: Mon,  7 Nov 2022 17:51:17 +0530	[thread overview]
Message-ID: <800f926c9314d97f735fb25c40a3a88efcc05beb.1667822611.git.ritesh.list@gmail.com> (raw)
In-Reply-To: <cover.1667822611.git.ritesh.list@gmail.com>

From: Li Xi <lixi@ddn.com>

When multi-threads are used for check, each thread needs to jump
to different group in pass1 check. This patch adds the group
jumping support. But still, only one thread is used to check.

Signed-off-by: Li Xi <lixi@ddn.com>
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
---
 e2fsck/e2fsck.h           | 23 ++++++++++++++++++--
 e2fsck/logfile.c          |  3 ++-
 e2fsck/pass1.c            | 44 ++++++++++++++++++++++++++++++++++++---
 e2fsck/problem.c          |  5 +++++
 e2fsck/problem.h          |  3 +++
 lib/ext2fs/ext2_err.et.in |  3 +++
 6 files changed, 75 insertions(+), 6 deletions(-)

diff --git a/e2fsck/e2fsck.h b/e2fsck/e2fsck.h
index 5bc24c3f..639f4e80 100644
--- a/e2fsck/e2fsck.h
+++ b/e2fsck/e2fsck.h
@@ -261,6 +261,22 @@ struct e2fsck_fc_replay_state {
 	__u16 fc_super_state;
 };
 
+#ifdef HAVE_PTHREAD
+/*
+ * Fields that used for multi-thread
+ */
+struct e2fsck_thread {
+	/* Thread index */
+	int		et_thread_index;
+	/* The start group number for this thread */
+	dgrp_t		et_group_start;
+	/* The end (not included) group number for this thread*/
+	dgrp_t		et_group_end;
+	/* The next group number to check */
+	dgrp_t		et_group_next;
+};
+#endif
+
 struct e2fsck_struct {
 	/* Global context to get the cancel flag */
 	e2fsck_t global_ctx;
@@ -344,8 +360,11 @@ struct e2fsck_struct {
 	 */
 	ext2_ino_t		stashed_ino;
 	struct ext2_inode	*stashed_inode;
-	/* Thread index, if global_ctx is null, this field is unused */
-	int			thread_index;
+
+	/* if @global_ctx is null, this field is unused */
+#ifdef HAVE_PTHREAD
+	struct e2fsck_thread	 thread_info;
+#endif
 
 	/*
 	 * Location of the lost and found directory
diff --git a/e2fsck/logfile.c b/e2fsck/logfile.c
index 74781f80..cc811d5a 100644
--- a/e2fsck/logfile.c
+++ b/e2fsck/logfile.c
@@ -315,7 +315,8 @@ static FILE *set_up_log_file(e2fsck_t ctx, const char *key, const char *fn)
 	expand_logfn(ctx, log_fn, &s);
 #ifdef HAVE_PTHREAD
 	if (ctx->global_ctx) {
-		sprintf(string_index, "%d", ctx->thread_index);
+		sprintf(string_index, "%d",
+			ctx->thread_info.et_thread_index);
 		append_string(&s, ".", 1);
 		append_string(&s, string_index, 0);
 	}
diff --git a/e2fsck/pass1.c b/e2fsck/pass1.c
index ea432ff2..3bb87669 100644
--- a/e2fsck/pass1.c
+++ b/e2fsck/pass1.c
@@ -1389,6 +1389,23 @@ void e2fsck_pass1_run(e2fsck_t ctx)
 	/* Set up ctx->lost_and_found if possible */
 	(void) e2fsck_get_lost_and_found(ctx, 0);
 
+#ifdef HAVE_PTHREAD
+	if (ctx->global_ctx) {
+		if (ctx->options & E2F_OPT_DEBUG &&
+		    ctx->options & E2F_OPT_MULTITHREAD)
+			fprintf(stderr, "thread %d jumping to group %d\n",
+					ctx->thread_info.et_thread_index,
+					ctx->thread_info.et_group_start);
+		pctx.errcode = ext2fs_inode_scan_goto_blockgroup(scan,
+					ctx->thread_info.et_group_start);
+		if (pctx.errcode) {
+			fix_problem(ctx, PR_1_PASS_HEADER, &pctx);
+			ctx->flags |= E2F_FLAG_ABORT;
+			goto endit;
+		}
+	}
+#endif
+
 	while (1) {
 		if (ino % (fs->super->s_inodes_per_group * 4) == 1) {
 			if (e2fsck_mmp_update(fs))
@@ -1432,6 +1449,8 @@ void e2fsck_pass1_run(e2fsck_t ctx)
 			ext2fs_mark_inode_bitmap2(ctx->inode_used_map, ino);
 			continue;
 		}
+		if (pctx.errcode == EXT2_ET_SCAN_FINISHED)
+			break;
 		if (pctx.errcode &&
 		    pctx.errcode != EXT2_ET_INODE_CSUM_INVALID &&
 		    pctx.errcode != EXT2_ET_INODE_IS_GARBAGE) {
@@ -2211,12 +2230,14 @@ static errcode_t e2fsck_open_channel_fs(ext2_filsys dest, e2fsck_t dest_context,
 	return 0;
 }
 
-static errcode_t e2fsck_pass1_thread_prepare(e2fsck_t global_ctx, e2fsck_t *thread_ctx)
+static errcode_t e2fsck_pass1_thread_prepare(e2fsck_t global_ctx, e2fsck_t *thread_ctx,
+					     int thread_index)
 {
 	errcode_t retval;
 	e2fsck_t thread_context;
 	ext2_filsys thread_fs;
 	ext2_filsys global_fs = global_ctx->fs;
+	struct e2fsck_thread *tinfo;
 
 	assert(global_ctx->inode_used_map == NULL);
 	assert(global_ctx->inode_dir_map == NULL);
@@ -2252,8 +2273,15 @@ static errcode_t e2fsck_pass1_thread_prepare(e2fsck_t global_ctx, e2fsck_t *thre
 		goto out_fs;
 	}
 
-	thread_context->thread_index = 0;
+	thread_context->thread_info.et_thread_index = thread_index;
 	set_up_logging(thread_context);
+
+	assert(thread_index == 0);
+	tinfo = &thread_context->thread_info;
+	tinfo->et_group_start = 0;
+	tinfo->et_group_next = 0;
+	tinfo->et_group_end = thread_fs->group_desc_count;
+
 	*thread_ctx = thread_context;
 	return 0;
 out_fs:
@@ -2495,7 +2523,7 @@ static int e2fsck_pass1_threads_start(struct e2fsck_thread_info **pinfo,
 	for (i = 0; i < num_threads; i++) {
 		tmp_pinfo = &infos[i];
 		tmp_pinfo->eti_thread_index = i;
-		retval = e2fsck_pass1_thread_prepare(global_ctx, &thread_ctx);
+		retval = e2fsck_pass1_thread_prepare(global_ctx, &thread_ctx, i);
 		if (retval) {
 			com_err(global_ctx->program_name, retval,
 				_("while preparing pass1 thread\n"));
@@ -2580,6 +2608,7 @@ static errcode_t scan_callback(ext2_filsys fs,
 {
 	struct scan_callback_struct *scan_struct;
 	e2fsck_t ctx;
+	struct e2fsck_thread *tinfo;
 
 	scan_struct = (struct scan_callback_struct *) priv_data;
 	ctx = scan_struct->ctx;
@@ -2591,6 +2620,15 @@ static errcode_t scan_callback(ext2_filsys fs,
 				    ctx->fs->group_desc_count))
 			return EXT2_ET_CANCEL_REQUESTED;
 
+#ifdef HAVE_PTHREAD
+	if (ctx->global_ctx) {
+		tinfo = &ctx->thread_info;
+		tinfo->et_group_next++;
+		if (tinfo->et_group_next >= tinfo->et_group_end)
+			return EXT2_ET_SCAN_FINISHED;
+	}
+#endif
+
 	return 0;
 }
 
diff --git a/e2fsck/problem.c b/e2fsck/problem.c
index 6ad6fb84..d5452441 100644
--- a/e2fsck/problem.c
+++ b/e2fsck/problem.c
@@ -1309,6 +1309,11 @@ static struct e2fsck_problem problem_table[] = {
 	  N_("Orphan file @i %i is not in use, but contains data.  "),
 	  PROMPT_CLEAR, PR_PREEN_OK },
 
+	/* Failed to goto block group */
+	{ PR_1_SCAN_GOTO,
+	  N_("failed to goto block group"),
+          PROMPT_NONE, PR_FATAL, 0, 0, 0 },
+
 	/* Pass 1b errors */
 
 	/* Pass 1B: Rescan for duplicate/bad blocks */
diff --git a/e2fsck/problem.h b/e2fsck/problem.h
index b47b0c63..a6207428 100644
--- a/e2fsck/problem.h
+++ b/e2fsck/problem.h
@@ -734,6 +734,9 @@ struct problem_context {
 /* Orphan file inode is not in use, but contains data */
 #define PR_1_ORPHAN_FILE_NOT_CLEAR		0x010090
 
+/* Failed to goto block group */
+#define PR_1_SCAN_GOTO				0x0100A0
+
 /*
  * Pass 1b errors
  */
diff --git a/lib/ext2fs/ext2_err.et.in b/lib/ext2fs/ext2_err.et.in
index de140198..b8f45ae2 100644
--- a/lib/ext2fs/ext2_err.et.in
+++ b/lib/ext2fs/ext2_err.et.in
@@ -557,4 +557,7 @@ ec	EXT2_ET_EXTENT_CYCLE,
 ec	EXT2_ET_EXTERNAL_JOURNAL_NOSUPP,
 	"Operation not supported on an external journal"
 
+ec	EXT2_ET_SCAN_FINISHED,
+	"Scanning finished"
+
 	end
-- 
2.37.3


  parent reply	other threads:[~2022-11-07 12:25 UTC|newest]

Thread overview: 104+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-07 12:20 [RFCv1 00/72] e2fsprogs: Parallel fsck support Ritesh Harjani (IBM)
2022-11-07 12:20 ` [RFCv1 01/72] e2fsck: Fix unbalanced mutex unlock for BOUNCE_MTX Ritesh Harjani (IBM)
2022-11-17 16:02   ` Theodore Ts'o
2022-11-17 18:45     ` Ritesh Harjani (IBM)
2022-11-18 10:34   ` Andreas Dilger
2022-11-18 11:37     ` Ritesh Harjani (IBM)
2022-11-18 13:20       ` Andreas Dilger
2022-11-19  3:46         ` Ritesh Harjani (IBM)
2023-01-24 16:40   ` Theodore Ts'o
2022-11-07 12:20 ` [RFCv1 02/72] gen_bitmaps: Fix ext2fs_compare_generic_bmap/bitmap logic Ritesh Harjani (IBM)
2022-11-23  5:04   ` Andreas Dilger
2023-01-24 16:59     ` Theodore Ts'o
2022-11-07 12:20 ` [RFCv1 03/72] blkmap64_ba: Add common helper for bits size calculation Ritesh Harjani (IBM)
2022-11-18 10:40   ` Andreas Dilger
2022-11-07 12:20 ` [RFCv1 04/72] badblocks: Remove unused badblocks_flags Ritesh Harjani (IBM)
2022-11-18 13:26   ` Andreas Dilger
2022-11-07 12:20 ` [RFCv1 05/72] badblocks: Add badblocks merge logic Ritesh Harjani (IBM)
2022-11-18 13:31   ` Andreas Dilger
2022-11-07 12:20 ` [RFCv1 06/72] dblist: add dblist " Ritesh Harjani (IBM)
2022-11-18 13:34   ` Andreas Dilger
2022-11-07 12:20 ` [RFCv1 07/72] libext2fs: Add rbtree bitmap " Ritesh Harjani (IBM)
2022-11-07 12:20 ` [RFCv1 08/72] libext2fs: Add bitmaps merge ops Ritesh Harjani (IBM)
2022-11-18 13:36   ` Andreas Dilger
2022-11-07 12:20 ` [RFCv1 09/72] libext2fs: Add flush cleanup API Ritesh Harjani (IBM)
2022-11-18 13:39   ` Andreas Dilger
2022-11-07 12:20 ` [RFCv1 10/72] libext2fs: merge icounts after thread finishes Ritesh Harjani (IBM)
2022-11-18 13:40   ` Andreas Dilger
2022-11-07 12:20 ` [RFCv1 11/72] libext2fs: merge quota context after threads finish Ritesh Harjani (IBM)
2022-11-18 13:42   ` Andreas Dilger
2022-11-07 12:21 ` [RFCv1 12/72] libext2fs: dupfs: Add fs clone & merge api Ritesh Harjani (IBM)
2022-11-18 19:46   ` Andreas Dilger
2022-11-19  5:02     ` Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 13/72] tst_badblocks: Add unit test to verify badblocks list " Ritesh Harjani (IBM)
2022-12-12 20:35   ` Andreas Dilger
2022-11-07 12:21 ` [RFCv1 14/72] tst_bitmaps_standalone: Add copy and merge bitmaps test Ritesh Harjani (IBM)
2022-12-12 20:40   ` Andreas Dilger
2022-12-14  5:12     ` Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 15/72] tst_bitmaps_pthread: Add merge bitmaps test using pthreads Ritesh Harjani (IBM)
2022-12-14 21:15   ` Andreas Dilger
2022-11-07 12:21 ` [RFCv1 16/72] tst_libext2fs_pthread: Add libext2fs merge/clone unit tests Ritesh Harjani (IBM)
2022-12-14 21:17   ` Andreas Dilger
2022-11-07 12:21 ` [RFCv1 17/72] libext2fs: Add support for ext2fs_test_block_bitmap_range2_valid() Ritesh Harjani (IBM)
2022-12-14 21:21   ` Andreas Dilger
2022-11-07 12:21 ` [RFCv1 18/72] libext2fs: Add support to get average group count Ritesh Harjani (IBM)
2022-12-14 21:24   ` Andreas Dilger
2022-11-07 12:21 ` [RFCv1 19/72] libext2fs: Misc fixes for struct_ext2_filsys Ritesh Harjani (IBM)
2022-12-14 21:22   ` Andreas Dilger
2022-11-07 12:21 ` [RFCv1 20/72] libext2fs: avoid too much memory allocation in case fs_num_threads Ritesh Harjani (IBM)
2022-11-18 13:37   ` Andreas Dilger
2022-11-07 12:21 ` [RFCv1 21/72] e2fsck: add -m option for multithread Ritesh Harjani (IBM)
2022-12-14 21:32   ` Andreas Dilger
2022-11-07 12:21 ` [RFCv1 22/72] e2fsck: copy context when using multi-thread fsck Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 23/72] e2fsck: create logs for multi-threads Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 24/72] e2fsck: configure one pfsck thread Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 25/72] e2fsck: Add e2fsck_pass1_thread_join return value Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 26/72] e2fsck: Use merge/clone apis of libext2fs Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 27/72] e2fsck: Add e2fsck_pass1_merge_bitmap() api Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 28/72] e2fsck: Add asserts in open_channel_fs Ritesh Harjani (IBM)
2022-11-07 12:21 ` Ritesh Harjani (IBM) [this message]
2022-11-07 12:21 ` [RFCv1 30/72] e2fsck: split groups to different threads Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 31/72] e2fsck: print thread log properly Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 32/72] e2fsck: do not change global variables Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 33/72] e2fsck: optimize the inserting of dir_info_db Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 34/72] e2fsck: merge dir_info after thread finishes Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 35/72] e2fsck: rbtree bitmap for dir Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 36/72] e2fsck: merge icounts after thread finishes Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 37/72] e2fsck: add debug codes for multiple threads Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 38/72] e2fsck: merge counts after threads finish Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 39/72] e2fsck: merge dx_dir_info " Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 40/72] e2fsck: merge dirs_to_hash when " Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 41/72] e2fsck: merge context flags properly Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 42/72] e2fsck: merge quota context after threads finish Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 43/72] e2fsck: serialize fix operations Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 44/72] e2fsck: move some fixes out of parallel pthreads Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 45/72] e2fsck: split and merge invalid bitmaps Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 46/72] e2fsck: merge EA blocks properly Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 47/72] e2fsck: kickoff mutex lock for block found map Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 48/72] e2fsck: allow admin specify number of threads Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 49/72] e2fsck: adjust " Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 50/72] e2fsck: fix readahead for pfsck of pass1 Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 51/72] e2fsck: merge options after threads finish Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 52/72] e2fsck: reset lost_and_found " Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 53/72] e2fsck: merge extent depth count " Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 54/72] e2fsck: simplify e2fsck context merging codes Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 55/72] e2fsck: set E2F_FLAG_ALLOC_OK after threads Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 56/72] e2fsck: wait fix thread finish before checking Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 57/72] e2fsck: cleanup e2fsck_pass1_thread_join() Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 58/72] e2fsck: make default smallest RA size to 1M Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 59/72] e2fsck: update mmp block in one thread Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 60/72] e2fsck: reset @inodes_to_rebuild if restart Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 61/72] tests: add pfsck test Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 62/72] e2fsck: fix memory leaks with pfsck enabled Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 63/72] e2fsck: misc cleanups for pfsck Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 64/72] e2fsck: propagate number of threads Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 65/72] e2fsck: Annotating fields in e2fsck_struct Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 66/72] e2fsck: merge casefolded dir lists after thread finish Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 67/72] sec: support encrypted files handling in pfsck mode Ritesh Harjani (IBM)
2022-11-07 19:22   ` Eric Biggers
2022-11-07 12:21 ` [RFCv1 68/72] e2fsck: Fix io->align assert check Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 69/72] e2fsck: Fix double free of inodes_to_process Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 70/72] e2fsck: Fix and simplify update_mmp in case of pfsck Ritesh Harjani (IBM)
2022-11-07 12:21 ` [RFCv1 71/72] e2fsck: Make threads call log_out after pthread_join Ritesh Harjani (IBM)
2022-11-07 12:22 ` [RFCv1 72/72] tests/f_multithread: Fix f_multithread related tests Ritesh Harjani (IBM)
     [not found] ` <B4ED1C86-D3EC-4A0A-97B3-CFCB46617E1A@dilger.ca>
2022-11-19  5:39   ` [RFCv1 00/72] e2fsprogs: Parallel fsck support Ritesh Harjani

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=800f926c9314d97f735fb25c40a3a88efcc05beb.1667822611.git.ritesh.list@gmail.com \
    --to=ritesh.list@gmail.com \
    --cc=adilger.kernel@dilger.ca \
    --cc=harshadshirwadkar@gmail.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=lixi@ddn.com \
    --cc=tytso@mit.edu \
    --cc=wshilong@ddn.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).