From: David Sterba <dsterba@suse.cz>
To: Su Yue <suy.fnst@cn.fujitsu.com>
Cc: linux-btrfs@vger.kernel.org, dsterba@suse.cz
Subject: Re: [PATCH v2 3/3] btrfs-progs: fi defrag: extend -c to drop nocompress flag on files
Date: Tue, 28 Nov 2017 17:07:00 +0100 [thread overview]
Message-ID: <20171128160659.GV3553@twin.jikos.cz> (raw)
In-Reply-To: <20171128091450.21789-3-suy.fnst@cn.fujitsu.com>
On Tue, Nov 28, 2017 at 05:14:50PM +0800, Su Yue wrote:
> Now, files which have nocompress flag also will be defraged
> with compression. However, nocompress flag is still existed
> and have to be cleared manually.
>
> So add an option '--clear-nocompress' to extend -c to drop
> nocompress flag after defragement.
> Suggested-by: David Sterba <dsterba@suse.com>
> Suggested-by: Anand Jain <anand.jain@oracle.com>
Do you have the pointer to the discussion? The whole idea sounds
familiar and seeing my name here means I must have been involved, but I
have only vague memories.
> Signed-off-by: Su Yue <suy.fnst@cn.fujitsu.com>
> ---
> Changelog:
> v2: Remove change about indentation of defrag_callback().
>
> ---
> cmds-filesystem.c | 65 ++++++++++++++++++++++++++++++++++++++++++++++++++++++-
> 1 file changed, 64 insertions(+), 1 deletion(-)
>
> diff --git a/cmds-filesystem.c b/cmds-filesystem.c
> index 3931333f76c6..84242814798e 100644
> --- a/cmds-filesystem.c
> +++ b/cmds-filesystem.c
> @@ -27,6 +27,7 @@
> #include <mntent.h>
> #include <linux/limits.h>
> #include <getopt.h>
> +#include <linux/fs.h>
>
> #include "kerncompat.h"
> #include "ctree.h"
> @@ -867,6 +868,8 @@ static const char * const cmd_filesystem_defrag_usage[] = {
> "-l len defragment only up to len bytes",
> "-t size target extent size hint (default: 32M)",
> "",
> + "--compress-force clear nocompress flag on files after defragment, only work with option -c",
This is probably --clear-nocompress .
The "force compress" semantics is up to kernel, so I think naming the
options 'clear-nocompress' is the right thing, as it's what it really
does.
> + "",
> "Warning: most Linux kernels will break up the ref-links of COW data",
> "(e.g., files copied with 'cp --reflink', snapshots) which may cause",
> "considerable increase of space usage. See btrfs-filesystem(8) for",
> @@ -874,9 +877,39 @@ static const char * const cmd_filesystem_defrag_usage[] = {
> NULL
> };
>
> +static int clear_nocompress_flag(int fd)
> +{
> + unsigned int flags;
> + int ret = 0;
> +
> + ret = ioctl(fd, FS_IOC_GETFLAGS, &flags);
> + if (ret < 0) {
> + ret = -errno;
> + error("failed to get flags: %s", strerror(-ret));
> + goto out;
> + }
> +
> + if (!(flags & FS_NOCOMP_FL)) {
> + ret = 0;
> + goto out;
> + }
> + flags &= ~FS_NOCOMP_FL;
> + ret = ioctl(fd, FS_IOC_SETFLAGS, &flags);
This is inherently racy, but the inode flags do not change that often so
concurrent change is unlikely.
> + if (ret < 0) {
> + ret = -errno;
> + error("failed to set flags: %s", strerror(-ret));
> + goto out;
> + }
> +
> + ret = 0;
> +out:
> + return ret;
> +}
> +
> static struct btrfs_ioctl_defrag_range_args defrag_global_range;
> static int defrag_global_verbose;
> static int defrag_global_errors;
> +static int defrag_global_clear_nocompress;
> static int defrag_callback(const char *fpath, const struct stat *sb,
> int typeflag, struct FTW *ftwbuf)
> {
> @@ -904,6 +937,14 @@ static int defrag_callback(const char *fpath, const struct stat *sb,
> err = errno;
> goto error;
> }
> +
> + if (defrag_global_clear_nocompress) {
> + ret = clear_nocompress_flag(fd);
> + if (ret) {
> + err = -ret;
> + goto error;
> + }
> + }
> }
> return 0;
>
> @@ -926,6 +967,12 @@ static int cmd_filesystem_defrag(int argc, char **argv)
> int compress_type = BTRFS_COMPRESS_NONE;
> DIR *dirstream;
>
> + enum { GETOPT_VAL_CLEAR_NOCOMPRESS = 257};
> + static const struct option long_options[] = {
> + { "clear-nocompress", no_argument, NULL,
> + GETOPT_VAL_CLEAR_NOCOMPRESS},
> + { NULL, 0, NULL, 0}
> + };
> /*
> * Kernel has a different default (256K) that is supposed to be safe,
> * but it does not defragment very well. The 32M will likely lead to
> @@ -937,8 +984,10 @@ static int cmd_filesystem_defrag(int argc, char **argv)
> defrag_global_errors = 0;
> defrag_global_verbose = 0;
> defrag_global_errors = 0;
> + defrag_global_clear_nocompress = 0;
> while(1) {
> - int c = getopt(argc, argv, "vrc::fs:l:t:");
> + int c = getopt_long(argc, argv, "vrc::fs:l:t:", long_options,
> + NULL);
> if (c < 0)
> break;
>
> @@ -972,6 +1021,9 @@ static int cmd_filesystem_defrag(int argc, char **argv)
> case 'r':
> recursive = 1;
> break;
> + case GETOPT_VAL_CLEAR_NOCOMPRESS:
> + defrag_global_clear_nocompress = 1;
> + break;
> default:
> usage(cmd_filesystem_defrag_usage);
> }
> @@ -987,6 +1039,8 @@ static int cmd_filesystem_defrag(int argc, char **argv)
> if (compress_type) {
> defrag_global_range.flags |= BTRFS_DEFRAG_RANGE_COMPRESS;
> defrag_global_range.compress_type = compress_type;
> + } else if (defrag_global_clear_nocompress) {
> + warning("Option --clear-nocompress only works for -c");
> }
> if (flush)
> defrag_global_range.flags |= BTRFS_DEFRAG_RANGE_START_IO;
> @@ -1071,6 +1125,15 @@ static int cmd_filesystem_defrag(int argc, char **argv)
> strerror(defrag_err));
> goto next;
> }
> +
> + if (defrag_global_clear_nocompress)
> + ret = clear_nocompress_flag(fd);
> + if (ret) {
> + error(
> + "failed to drop nocompress flag on %s: %s",
> + argv[i], strerror(-ret));
> + goto next;
> + }
> }
> next:
> if (ret)
> --
> 2.15.0
>
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2017-11-28 16:08 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-28 9:14 [PATCH v2 1/3] btrfs-progs: fi defrag: clean up duplicate code if find errors Su Yue
2017-11-28 9:14 ` [PATCH v2 2/3] btrfs-progs: fi defrag: do not exit if defrag range ioctl is unsupported Su Yue
2017-11-28 15:57 ` David Sterba
2017-11-28 9:14 ` [PATCH v2 3/3] btrfs-progs: fi defrag: extend -c to drop nocompress flag on files Su Yue
2017-11-28 16:07 ` David Sterba [this message]
2017-11-29 1:54 ` Su Yue
2017-11-30 18:30 ` David Sterba
2017-11-28 16:07 ` [PATCH v2 1/3] btrfs-progs: fi defrag: clean up duplicate code if find errors David Sterba
2017-11-29 2:12 ` [PATCH v3 2/3] btrfs-progs: fi defrag: do not exit if defrag range ioctl is unsupported Su Yue
2017-11-30 6:37 ` Qu Wenruo
2017-11-30 6:49 ` Qu Wenruo
2017-11-30 6:47 ` [PATCH v2 1/3] btrfs-progs: fi defrag: clean up duplicate code if find errors Qu Wenruo
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=20171128160659.GV3553@twin.jikos.cz \
--to=dsterba@suse.cz \
--cc=linux-btrfs@vger.kernel.org \
--cc=suy.fnst@cn.fujitsu.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).