xfsprogs: fixes for 5.10-rc1
v3 cover xfsprogs: fixes for 5.10-rc1
Hi all,
Fix various minor bugs in 5.10-rc1 (hence v3).
If you're going to start using this mess, you probably ought to just pull from my git trees, which are linked below.
This is an extraordinary way to destroy everything. Enjoy! Comments and questions are, as always, welcome.
--D
xfsprogs git tree:
https://git.kernel.org/cgit/linux/kernel/git/djwong/xfsprogs-dev.git/log/?h=xfsprogs-5.10-fixes
--- debian/changelog | 2 +- debian/control | 2 +- include/platform_defs.h.in | 6 +++--- libxfs/trans.c | 14 ++++++++++++++ tools/libxfs-apply | 14 +++++++++++--- 5 files changed, 30 insertions(+), 8 deletions(-)
PATCH v3 1/5 libxfs-apply: don't add duplicate headers
From: Darrick J. Wong <[email protected]>
When we're backporting patches from libxfs, don't add a S-o-b header if there's already one at the end of the headers of the patch being ported.
That way, we avoid things like:
--- tools/libxfs-apply | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-)
tools/libxfs-apply +11 -3
@@ -193,6 +193,14 @@ filter_xfsprogs_patch() rm -f $_libxfs_files } +add_header() +{ + local hdr="$1" + local hdrfile="$2" + + tail -n 1 "$hdrfile" | grep -q "^${hdr}$" || echo "$hdr" >> "$hdrfile" +} + fixup_header_format() { local _source=$1
@@ -280,13 +288,13 @@ fixup_header_format() sed -i '${/^[[:space:]]*$/d;}' $_hdr.new # Add Signed-off-by: header if specified - if [ ! -z ${SIGNED_OFF_BY+x} ]; then - echo "Signed-off-by: $SIGNED_OFF_BY" >> $_hdr.new + if [ ! -z ${SIGNED_OFF_BY+x} ]; then + add_header "Signed-off-by: $SIGNED_OFF_BY" $_hdr.new else # get it from git config if present SOB_NAME=`git config --get user.name` SOB_EMAIL=`git config --get user.email` if [ ! -z ${SOB_NAME+x} ]; then - echo "Signed-off-by: $SOB_NAME <$SOB_EMAIL>" >> $_hdr.new + add_header "Signed-off-by: $SOB_NAME <$SOB_EMAIL>" $_hdr.new fi fi
Re: [PATCH 1/5] libxfs-apply: don't add duplicate headers
Quoted 8 lines · Darrick J. Wong
From: Darrick J. Wong <[email protected]>
When we're backporting patches from libxfs, don't add a S-o-b header if there's already one at the end of the headers of the patch being ported.
That way, we avoid things like:
Signed-off-by Darrick J. Wong <[email protected]>Signed-off-by Darrick J. Wong <[email protected]>
But it will still not add my additional SOB if I merge something across to userspace that starts out with:
And I always felt like the committer should be adding their SOB to the end of the chain when they move code from one place to another, especially if any tweaks got made along the way.
IOWs I see the rationale for removing duplicate /sequential/ SOBs, but not for removing duplicate SOBs in general.
Am I thinking about that wrong?
Quoted 42 lines
--- tools/libxfs-apply | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-)tools/libxfs-apply +11
-3@@ -193,6 +193,14 @@ filter_xfsprogs_patch() rm -f $_libxfs_files } +add_header() +{ + local hdr="$1" + local hdrfile="$2" + + tail -n 1 "$hdrfile" | grep -q "^${hdr}$" || echo "$hdr" >> "$hdrfile" +} + fixup_header_format() { local _source=$1@@ -280,13 +288,13 @@ fixup_header_format() sed -i '${/^[[:space:]]*$/d;}' $_hdr.new # Add Signed-off-by: header if specified - if [ ! -z ${SIGNED_OFF_BY+x} ]; then - echo "Signed-off-by: $SIGNED_OFF_BY" >> $_hdr.new + if [ ! -z ${SIGNED_OFF_BY+x} ]; then + add_header "Signed-off-by: $SIGNED_OFF_BY" $_hdr.new else # get it from git config if present SOB_NAME=`git config --get user.name` SOB_EMAIL=`git config --get user.email` if [ ! -z ${SOB_NAME+x} ]; then - echo "Signed-off-by: $SOB_NAME <$SOB_EMAIL>" >> $_hdr.new + add_header "Signed-off-by: $SOB_NAME <$SOB_EMAIL>" $_hdr.new fi fi
Re: [PATCH 1/5] libxfs-apply: don't add duplicate headers
Quoted 17 lines · Eric Sandeen
Quoted 8 lines · Darrick J. Wong
From: Darrick J. Wong <[email protected]>
When we're backporting patches from libxfs, don't add a S-o-b header if there's already one at the end of the headers of the patch being ported.
That way, we avoid things like:
Signed-off-by Darrick J. Wong <[email protected]>Signed-off-by Darrick J. Wong <[email protected]>But it will still not add my additional SOB if I merge something across to userspace that starts out with:
Signed-off-by Eric Sandeen <[email protected]>Reviewed-by Dave Chinner <[email protected]>Signed-off-by Darrick J. Wong <[email protected]>
Quoted 8 lines
And I always felt like the committer should be adding their SOB to the end of the chain when they move code from one place to another, especially if any tweaks got made along the way.
IOWs I see the rationale for removing duplicate /sequential/ SOBs, but not for removing duplicate SOBs in general.
Am I thinking about that wrong?
add_header is different from last time -- whereas before it would search the entire $hdrfile for $hdr and add $hdr if it didn't find a match, now it only looks at the last line of $hdrfile for a match.
For your example above, it would not add another "SOB: Darrick" because there's already one at the bottom of the headers; but it would add another "SOB: Eric" because this commit acquired other tags after your first signoff.
--D
Quoted 42 lines
Quoted 42 lines
--- tools/libxfs-apply | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-)tools/libxfs-apply +11
-3@@ -193,6 +193,14 @@ filter_xfsprogs_patch() rm -f $_libxfs_files } +add_header() +{ + local hdr="$1" + local hdrfile="$2" + + tail -n 1 "$hdrfile" | grep -q "^${hdr}$" || echo "$hdr" >> "$hdrfile" +} + fixup_header_format() { local _source=$1@@ -280,13 +288,13 @@ fixup_header_format() sed -i '${/^[[:space:]]*$/d;}' $_hdr.new # Add Signed-off-by: header if specified - if [ ! -z ${SIGNED_OFF_BY+x} ]; then - echo "Signed-off-by: $SIGNED_OFF_BY" >> $_hdr.new + if [ ! -z ${SIGNED_OFF_BY+x} ]; then + add_header "Signed-off-by: $SIGNED_OFF_BY" $_hdr.new else # get it from git config if present SOB_NAME=`git config --get user.name` SOB_EMAIL=`git config --get user.email` if [ ! -z ${SOB_NAME+x} ]; then - echo "Signed-off-by: $SOB_NAME <$SOB_EMAIL>" >> $_hdr.new + add_header "Signed-off-by: $SOB_NAME <$SOB_EMAIL>" $_hdr.new fi fi
Re: [PATCH 1/5] libxfs-apply: don't add duplicate headers
Quoted 22 lines · Darrick J. Wong
Quoted 21 lines · Eric Sandeen
Quoted 8 lines · Darrick J. Wong
From: Darrick J. Wong <[email protected]>
When we're backporting patches from libxfs, don't add a S-o-b header if there's already one at the end of the headers of the patch being ported.
That way, we avoid things like:
Signed-off-by Darrick J. Wong <[email protected]>Signed-off-by Darrick J. Wong <[email protected]>But it will still not add my additional SOB if I merge something across to userspace that starts out with:
Signed-off-by Eric Sandeen <[email protected]>Reviewed-by Dave Chinner <[email protected]>Signed-off-by Darrick J. Wong <[email protected]>And I always felt like the committer should be adding their SOB to the end of the chain when they move code from one place to another, especially if any tweaks got made along the way.
Quoted 8 lines
Quoted 4 lines
IOWs I see the rationale for removing duplicate /sequential/ SOBs, but not for removing duplicate SOBs in general.
Am I thinking about that wrong?
add_headeris different from last time -- whereas before it would search the entire $hdrfile for $hdr and add $hdr if it didn't find a match, now it only looks at the last line of $hdrfile for a match.
I'm sorry, I visually checked and thought it was same as last time. Will look again, more closely.
Thanks,
-Eric
Quoted 50 lines
For your example above, it would not add another "SOB: Darrick" because there's already one at the bottom of the headers; but it would add another "SOB: Eric" because this commit acquired other tags after your first signoff.
--D
Quoted 42 lines
Quoted 42 lines
--- tools/libxfs-apply | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-)tools/libxfs-apply +11
-3@@ -193,6 +193,14 @@ filter_xfsprogs_patch() rm -f $_libxfs_files } +add_header() +{ + local hdr="$1" + local hdrfile="$2" + + tail -n 1 "$hdrfile" | grep -q "^${hdr}$" || echo "$hdr" >> "$hdrfile" +} + fixup_header_format() { local _source=$1@@ -280,13 +288,13 @@ fixup_header_format() sed -i '${/^[[:space:]]*$/d;}' $_hdr.new # Add Signed-off-by: header if specified - if [ ! -z ${SIGNED_OFF_BY+x} ]; then - echo "Signed-off-by: $SIGNED_OFF_BY" >> $_hdr.new + if [ ! -z ${SIGNED_OFF_BY+x} ]; then + add_header "Signed-off-by: $SIGNED_OFF_BY" $_hdr.new else # get it from git config if present SOB_NAME=`git config --get user.name` SOB_EMAIL=`git config --get user.email` if [ ! -z ${SOB_NAME+x} ]; then - echo "Signed-off-by: $SOB_NAME <$SOB_EMAIL>" >> $_hdr.new + add_header "Signed-off-by: $SOB_NAME <$SOB_EMAIL>" $_hdr.new fi fi
Re: [PATCH 1/5] libxfs-apply: don't add duplicate headers
In reply to Darrick J. WongQuoted 8 lines · Darrick J. Wong
From: Darrick J. Wong <[email protected]>
When we're backporting patches from libxfs, don't add a S-o-b header if there's already one at the end of the headers of the patch being ported.
That way, we avoid things like:
Signed-off-by Darrick J. Wong <[email protected]>Signed-off-by Darrick J. Wong <[email protected]>
I like how that SOB is part of the changelog and is also part of the changelog ;)
Quoted 42 lines
--- tools/libxfs-apply | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-)tools/libxfs-apply +11
-3@@ -193,6 +193,14 @@ filter_xfsprogs_patch() rm -f $_libxfs_files } +add_header() +{ + local hdr="$1" + local hdrfile="$2" + + tail -n 1 "$hdrfile" | grep -q "^${hdr}$" || echo "$hdr" >> "$hdrfile" +} + fixup_header_format() { local _source=$1@@ -280,13 +288,13 @@ fixup_header_format() sed -i '${/^[[:space:]]*$/d;}' $_hdr.new # Add Signed-off-by: header if specified - if [ ! -z ${SIGNED_OFF_BY+x} ]; then - echo "Signed-off-by: $SIGNED_OFF_BY" >> $_hdr.new + if [ ! -z ${SIGNED_OFF_BY+x} ]; then + add_header "Signed-off-by: $SIGNED_OFF_BY" $_hdr.new else # get it from git config if present SOB_NAME=`git config --get user.name` SOB_EMAIL=`git config --get user.email` if [ ! -z ${SOB_NAME+x} ]; then - echo "Signed-off-by: $SOB_NAME <$SOB_EMAIL>" >> $_hdr.new + add_header "Signed-off-by: $SOB_NAME <$SOB_EMAIL>" $_hdr.new fi fi
PATCH v3 2/5 libxfs: fix weird comment
In reply to Darrick J. WongFrom: Darrick J. Wong <[email protected]>
Not sure what happened with this multiline comment, but clean up all the stars.
--- include/platform_defs.h.in | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
include/platform_defs.h.in +3 -3
@@ -86,9 +86,9 @@ extern int platform_nproc(void); /* Simplified from version in include/linux/overflow.h */ /* - * * Compute a*b+c, returning SIZE_MAX on overflow. Internal helper for - * * struct_size() below. - * */ + * Compute a*b+c, returning SIZE_MAX on overflow. Internal helper for + * struct_size() below. + */ static inline size_t __ab_c_size(size_t a, size_t b, size_t c) { return (a * b) + c;
Re: [PATCH 2/5] libxfs: fix weird comment
Quoted 4 lines · Darrick J. Wong
From: Darrick J. Wong <[email protected]>
Not sure what happened with this multiline comment, but clean up all the stars.
urk, that looks like something I would do, thanks :)
"My God, it's full of stars!"
Quoted 25 lines
Signed-off-by Darrick J. Wong <[email protected]>--- include/platform_defs.h.in | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)include/platform_defs.h.in +3
-3@@ -86,9 +86,9 @@ extern int platform_nproc(void); /* Simplified from version in include/linux/overflow.h */ /* - * * Compute a*b+c, returning SIZE_MAX on overflow. Internal helper for - * * struct_size() below. - * */ + * Compute a*b+c, returning SIZE_MAX on overflow. Internal helper for + * struct_size() below. + */ static inline size_t __ab_c_size(size_t a, size_t b, size_t c) { return (a * b) + c;
Re: [PATCH 2/5] libxfs: fix weird comment
In reply to Darrick J. WongQuoted 4 lines · Darrick J. Wong
From: Darrick J. Wong <[email protected]>
Not sure what happened with this multiline comment, but clean up all the stars.
Looks like a typical copy and paste error when the editor tries to be "smart".
PATCH v3 3/5 libxfs: add realtime extent reservation and usage tracking to transactions
In reply to Darrick J. WongFrom: Darrick J. Wong <[email protected]>
The libxfs resync added to the deferred ops code the ability to capture the unfinished deferred ops and transaction reservation for later replay during log recovery. This nominally requires transactions to have the ability to track rt extent reservations and usage, so port that missing piece from the kernel now to avoid leaving logic bombs in case anyone ever /does/ start messing with realtime.
--- libxfs/trans.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+)
libxfs/trans.c +14 -0
@@ -230,6 +230,7 @@ xfs_trans_reserve( error = -ENOSPC ; goto undo_blocks; } + tp->t_rtx_res += rtextents; }return 0 ;
@@ -765,6 +766,19 @@ _("Transaction block reservation exceeded! %u > %u\n"), tp->t_ifree_delta += delta;break ;case XFS_TRANS_SB_FREXTENTS : +/* +* Track the number of rt extents allocated in the transaction. +* Make sure it does not exceed the number reserved. +*/ +if (delta <0 ) { + tp->t_rtx_res_used += (uint)-delta; +if (tp->t_rtx_res_used > tp->t_rtx_res) { +fprintf (stderr, +_ ("Transaction rt block reservation exceeded! %u > %u\n" ), + tp->t_rtx_res_used, tp->t_rtx_res); +ASSERT (0 ); + } + } tp->t_frextents_delta += delta;break ;default :
Re: [PATCH 3/5] libxfs: add realtime extent reservation and usage tracking to transactions
Looks good,
PATCH v3 4/5 debian: fix version in changelog
In reply to Darrick J. WongFrom: Darrick J. Wong <[email protected]>
We're still at 5.10-rc0, at least according to the tags.
--- debian/changelog | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
debian/changelog +1 -1
@@ -1,4 +1,4 @@ -xfsprogs (5.10.0-1) unstable; urgency=low +xfsprogs (5.10.0-rc0-1) unstable; urgency=low * New upstream prerelease
Re: [PATCH 4/5] debian: fix version in changelog
Quoted 5 lines · Darrick J. Wong
From: Darrick J. Wong <[email protected]>
We're still at 5.10-rc0, at least according to the tags.
Signed-off-by Darrick J. Wong <[email protected]>
Thanks for spotting this
Quoted 16 lines
--- debian/changelog | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)debian/changelog +1
-1@@ -1,4 +1,4 @@ -xfsprogs (5.10.0-1) unstable; urgency=low +xfsprogs (5.10.0-rc0-1) unstable; urgency=low * New upstream prerelease
PATCH v3 5/5 debian: add build dependency on libinih-dev
In reply to Darrick J. WongFrom: Darrick J. Wong <[email protected]>
mkfs now supports configuration files, which are parsed using libinih. Add this dependency to the debian build.
--- debian/control | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
debian/control +1 -1
@@ -3,7 +3,7 @@ Section: admin Priority: optional Maintainer: XFS Development Team <[email protected]> Uploaders: Nathan Scott <[email protected]>, Anibal Monsalve Salazar <[email protected]> -Build-Depends: uuid-dev, dh-autoreconf, debhelper (>= 5), gettext, libtool, libedit-dev, libblkid-dev (>= 2.17), linux-libc-dev, libdevmapper-dev, libattr1-dev, libicu-dev, dh-python, pkg-config +Build-Depends: libinih-dev, uuid-dev, dh-autoreconf, debhelper (>= 5), gettext, libtool, libedit-dev, libblkid-dev (>= 2.17), linux-libc-dev, libdevmapper-dev, libattr1-dev, libicu-dev, dh-python, pkg-config Standards-Version: 4.0.0 Homepage: https://xfs.wiki.kernel.org/
Re: [PATCH 5/5] debian: add build dependency on libinih-dev
Quoted 6 lines · Darrick J. Wong
From: Darrick J. Wong <[email protected]>
mkfs now supports configuration files, which are parsed using libinih. Add this dependency to the debian build.
Signed-off-by Darrick J. Wong <[email protected]>
I don't speak Debian very well but this looks fine.
Quoted 19 lines
--- debian/control | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)debian/control +1
-1@@ -3,7 +3,7 @@ Section: admin Priority: optional Maintainer: XFS Development Team <[email protected]> Uploaders: Nathan Scott <[email protected]>, Anibal Monsalve Salazar <[email protected]> -Build-Depends: uuid-dev, dh-autoreconf, debhelper (>= 5), gettext, libtool, libedit-dev, libblkid-dev (>= 2.17), linux-libc-dev, libdevmapper-dev, libattr1-dev, libicu-dev, dh-python, pkg-config +Build-Depends: libinih-dev, uuid-dev, dh-autoreconf, debhelper (>= 5), gettext, libtool, libedit-dev, libblkid-dev (>= 2.17), linux-libc-dev, libdevmapper-dev, libattr1-dev, libicu-dev, dh-python, pkg-config Standards-Version: 4.0.0 Homepage: https://xfs.wiki.kernel.org/
Re: [PATCH 5/5] debian: add build dependency on libinih-dev
In reply to Darrick J. WongQuoted 16 lines · Darrick J. Wong
From: Darrick J. Wong <[email protected]>
mkfs now supports configuration files, which are parsed using libinih. Add this dependency to the debian build.
Signed-off-by Darrick J. Wong <[email protected]>--- debian/control | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)debian/control +0
-0
debian/control Section: admin · line 6
Quoted 5 lines
debian/control +1
-1@@ -3,7 +3,7 @@ Section: admin Priority: optional Maintainer: XFS Development Team <[email protected]> Uploaders: Nathan Scott <[email protected]>, Anibal Monsalve Salazar <[email protected]> -Build-Depends: uuid-dev, dh-autoreconf, debhelper (>= 5), gettext, libtool, libedit-dev, libblkid-dev (>= 2.17), linux-libc-dev, libdevmapper-dev, libattr1-dev, libicu-dev, dh-python, pkg-config +Build-Depends: libinih-dev, uuid-dev, dh-autoreconf, debhelper (>= 5), gettext, libtool, libedit-dev, libblkid-dev (>= 2.17), linux-libc-dev, libdevmapper-dev, libattr1-dev, libicu-dev, dh-python, pkg-config
Oops, yeah, forgot that because my build machines already had that library installed....
Signature · 2 lines
Dave Chinner [email protected]