Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

[Kirill Smelkov]

[-- Attachment #1: Type: text/plain, Size: 12045 bytes --]

Android people,

Please see inside if Android 4.9-o wants to pick up stream_open
related deadlock fixes. FUSE part of it is already applied to upstream
stable/linux-4.9.y as 9c829b6e3fe2 and cfd8d2e79524, but the Android
tree does not have any of the fixes at all.

All 3 patches should be safe to apply, but they are the
least intresting bits of stream_open conversion as other changes from
commit c5bf68fe0c86 (*: convert stream-like files from nonseekable_open
-> stream_open) deal with several real deadlocks (and event not all of them
are marked as errors in that commit). Also there is bbd84f33 (fuse:
Add FOPEN_STREAM to use stream_open()) which fixes FUSE deadlock
regression visible to userspace.

Kirill


---- 8< ----
From: kbuild test robot <lkp@intel.com>
Subject: [PATCH 1/3] objtool, perf: fix stream_open.cocci warnings

sound/core/rawmidi.c:376:7-23: WARNING: snd_rawmidi_f_ops: .read() and .write() have stream semantic; safe to change nonseekable_open -> stream_open.

Generated by: scripts/coccinelle/api/stream_open.cocci

Signed-off-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Kirill Smelkov <kirr@nexedi.com>
---

tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
head:   58ce54880a03d9c27327434ba15281f5a7139671
commit: 4864b070265301ba88506ee407d51e058dcb0ad6 [42/51] objtool, perf: Fix GCC 8 -Wrestrict error
:::::: branch date: 27 hours ago
:::::: commit date: 9 weeks ago

Please take the patch only if it's a positive warning. Thanks!

 rawmidi.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/sound/core/rawmidi.c
+++ b/sound/core/rawmidi.c
@@ -373,7 +373,7 @@ static int snd_rawmidi_open(struct inode
 	if ((file->f_flags & O_APPEND) && !(file->f_flags & O_NONBLOCK))
 		return -EINVAL;		/* invalid combination */

-	err = nonseekable_open(inode, file);
+	err = stream_open(inode, file);
 	if (err < 0)
 		return err;

---- 8< ----
From: kbuild test robot <lkp@intel.com>
Subject: [PATCH 2/3] objtool, perf: fix stream_open.cocci warnings

sound/core/control.c:55:7-23: WARNING: snd_ctl_f_ops: .read() has stream semantic; safe to change nonseekable_open -> stream_open.

Generated by: scripts/coccinelle/api/stream_open.cocci

Signed-off-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Kirill Smelkov <kirr@nexedi.com>
---

tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
head:   58ce54880a03d9c27327434ba15281f5a7139671
commit: 4864b070265301ba88506ee407d51e058dcb0ad6 [42/51] objtool, perf: Fix GCC 8 -Wrestrict error
:::::: branch date: 27 hours ago
:::::: commit date: 9 weeks ago

Please take the patch only if it's a positive warning. Thanks!

 control.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/sound/core/control.c
+++ b/sound/core/control.c
@@ -52,7 +52,7 @@ static int snd_ctl_open(struct inode *in
 	struct snd_ctl_file *ctl;
 	int i, err;

-	err = nonseekable_open(inode, file);
+	err = stream_open(inode, file);
 	if (err < 0)
 		return err;


---- 8< ----
From: kbuild test robot <lkp@intel.com>
Subject: [PATCH 3/3] objtool, perf: fix stream_open.cocci warnings

sound/core/seq/seq_clientmgr.c:309:7-23: WARNING: snd_seq_f_ops: .read() and .write() have stream semantic; safe to change nonseekable_open -> stream_open.

Generated by: scripts/coccinelle/api/stream_open.cocci

Signed-off-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Kirill Smelkov <kirr@nexedi.com>
---

tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
head:   58ce54880a03d9c27327434ba15281f5a7139671
commit: 4864b070265301ba88506ee407d51e058dcb0ad6 [42/51] objtool, perf: Fix GCC 8 -Wrestrict error
:::::: branch date: 27 hours ago
:::::: commit date: 9 weeks ago

Please take the patch only if it's a positive warning. Thanks!

 seq_clientmgr.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/sound/core/seq/seq_clientmgr.c
+++ b/sound/core/seq/seq_clientmgr.c
@@ -306,7 +306,7 @@ static int snd_seq_open(struct inode *in
 	struct snd_seq_user_client *user;
 	int err;

-	err = nonseekable_open(inode, file);
+	err = stream_open(inode, file);
 	if (err < 0)
 		return err;




On Mon, Jan 13, 2020@05:18:48PM +0100, Julia Lawall wrote:
> On Mon, 13 Jan 2020, Kirill Smelkov wrote:
> 
> > +gregkh
> >
> > ( Greg, please see inside if Android 4.9-o wants to pick up stream_open
> >   related deadlock fixes. You already applied FUSE part to stable/linux-4.9.y
> >   as 9c829b6e3fe2 and cfd8d2e79524, but the Android tree does not have
> >   any of the fixes at all )
> >
> > Hi Julia,
> >
> > On Fri, Jan 10, 2020 at 01:46:59PM +0100, Julia Lawall wrote:
> > > Hi,
> > >
> > > Do you want to check this, and forward it on if it seems ok?
> > >
> > > thanks,
> > > julia
> > >
> > > ---------- Forwarded message ----------
> > > Date: Thu, 9 Jan 2020 07:00:42 +0800
> > > From: kbuild test robot <lkp@intel.com>
> > > To: kbuild(a)lists.01.org
> > > Cc: Julia Lawall <julia.lawall@lip6.fr>,
> > >     Elena Reshetova <elena.reshetova@intel.com>
> > > Subject: [PATCH] objtool, perf: fix stream_open.cocci warnings
> > >
> > > CC: kbuild-all(a)lists.01.org
> > > CC: Guenter Roeck <groeck@google.com>
> > > BCC: philip.li(a)intel.com
> > > TO: cros-kernel-buildreports(a)googlegroups.com
> > >
> > > From: kbuild test robot <lkp@intel.com>
> > >
> > > sound/core/rawmidi.c:376:7-23: WARNING: snd_rawmidi_f_ops: .read() and .write() have stream semantic; safe to change nonseekable_open -> stream_open.
> > >
> > > Generated by: scripts/coccinelle/api/stream_open.cocci
> > >
> > > Fixes: 4864b0702653 ("objtool, perf: Fix GCC 8 -Wrestrict error")
> > > CC: Josh Poimboeuf <jpoimboe@redhat.com>
> > > Signed-off-by: kbuild test robot <lkp@intel.com>
> > > ---
> > >
> > > tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
> > > head:   58ce54880a03d9c27327434ba15281f5a7139671
> > > commit: 4864b070265301ba88506ee407d51e058dcb0ad6 [42/51] objtool, perf: Fix GCC 8 -Wrestrict error
> > > :::::: branch date: 27 hours ago
> > > :::::: commit date: 9 weeks ago
> > >
> > > Please take the patch only if it's a positive warning. Thanks!
> > >
> > >  rawmidi.c |    2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > --- a/sound/core/rawmidi.c
> > > +++ b/sound/core/rawmidi.c
> > > @@ -373,7 +373,7 @@ static int snd_rawmidi_open(struct inode
> > >  	if ((file->f_flags & O_APPEND) && !(file->f_flags & O_NONBLOCK))
> > >  		return -EINVAL;		/* invalid combination */
> > >
> > > -	err = nonseekable_open(inode, file);
> > > +	err = stream_open(inode, file);
> > >  	if (err < 0)
> > >  		return err;
> >
> > Strange, sound/core/rawmidi.c was aleady converted to stream_open by my
> > commmit c5bf68fe0c86 (*: convert stream-like files from nonseekable_open
> > -> stream_open). I also don't have commit 4864b0702653 in my tree (see
> > "Fixes: ..." above) even though I just pulled from linus and next. The
> > same story applies to other 2 patches as well.
> >
> > Ok, I just checked the android tree ("tree: ..." referenced above) and
> > commit 4864b0702653 is there, but it has nothing to do with
> > nonseekable_open at all (it touches just tools/lib/str_error_r.c).
> >
> > In any way, all 3 patches should be safe to apply, but they are the
> > least intresting bits of stream_open conversion as other changes from
> > c5bf68fe0c86 deal with several real deadlocks (and event not all of them
> > are marked as errors in that patch), and also there is bbd84f33 (fuse:
> > Add FOPEN_STREAM to use stream_open()) which fixes FUSE deadlock
> > regression visible to userspace.
> >
> > I'm also not sure to where to forward the patches if I wanted to do so.
> 
> You can forward them to the people listed in the message:
> 
> > > CC: kbuild-all(a)lists.01.org
> > > CC: Guenter Roeck <groeck@google.com>
> > > BCC: philip.li(a)intel.com
> > > TO: cros-kernel-buildreports(a)googlegroups.com
> 
> julia
> 
> 
> >
> > Kirill
> >
> >
> > On Fri, Jan 10, 2020 at 01:47:13PM +0100, Julia Lawall wrote:
> > > This one too...
> > >
> > > ---------- Forwarded message ----------
> > > Date: Thu, 9 Jan 2020 07:00:41 +0800
> > > From: kbuild test robot <lkp@intel.com>
> > > To: kbuild(a)lists.01.org
> > > Cc: Julia Lawall <julia.lawall@lip6.fr>,
> > >     Elena Reshetova <elena.reshetova@intel.com>
> > > Subject: [PATCH] objtool, perf: fix stream_open.cocci warnings
> > >
> > > CC: kbuild-all(a)lists.01.org
> > > CC: Guenter Roeck <groeck@google.com>
> > > BCC: philip.li(a)intel.com
> > > TO: cros-kernel-buildreports(a)googlegroups.com
> > >
> > > From: kbuild test robot <lkp@intel.com>
> > >
> > > sound/core/control.c:55:7-23: WARNING: snd_ctl_f_ops: .read() has stream semantic; safe to change nonseekable_open -> stream_open.
> > >
> > > Generated by: scripts/coccinelle/api/stream_open.cocci
> > >
> > > Fixes: 4864b0702653 ("objtool, perf: Fix GCC 8 -Wrestrict error")
> > > CC: Josh Poimboeuf <jpoimboe@redhat.com>
> > > Signed-off-by: kbuild test robot <lkp@intel.com>
> > > ---
> > >
> > > tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
> > > head:   58ce54880a03d9c27327434ba15281f5a7139671
> > > commit: 4864b070265301ba88506ee407d51e058dcb0ad6 [42/51] objtool, perf: Fix GCC 8 -Wrestrict error
> > > :::::: branch date: 27 hours ago
> > > :::::: commit date: 9 weeks ago
> > >
> > > Please take the patch only if it's a positive warning. Thanks!
> > >
> > >  control.c |    2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > --- a/sound/core/control.c
> > > +++ b/sound/core/control.c
> > > @@ -52,7 +52,7 @@ static int snd_ctl_open(struct inode *in
> > >  	struct snd_ctl_file *ctl;
> > >  	int i, err;
> > >
> > > -	err = nonseekable_open(inode, file);
> > > +	err = stream_open(inode, file);
> > >  	if (err < 0)
> > >  		return err;
> > >
> >
> > On Fri, Jan 10, 2020 at 01:47:25PM +0100, Julia Lawall wrote:
> > > And this one.
> > >
> > > ---------- Forwarded message ----------
> > > Date: Thu, 9 Jan 2020 07:00:42 +0800
> > > From: kbuild test robot <lkp@intel.com>
> > > To: kbuild(a)lists.01.org
> > > Cc: Julia Lawall <julia.lawall@lip6.fr>,
> > >     Elena Reshetova <elena.reshetova@intel.com>
> > > Subject: [PATCH] objtool, perf: fix stream_open.cocci warnings
> > >
> > > CC: kbuild-all(a)lists.01.org
> > > CC: Guenter Roeck <groeck@google.com>
> > > BCC: philip.li(a)intel.com
> > > TO: cros-kernel-buildreports(a)googlegroups.com
> > >
> > > From: kbuild test robot <lkp@intel.com>
> > >
> > > sound/core/seq/seq_clientmgr.c:309:7-23: WARNING: snd_seq_f_ops: .read() and .write() have stream semantic; safe to change nonseekable_open -> stream_open.
> > >
> > > Generated by: scripts/coccinelle/api/stream_open.cocci
> > >
> > > Fixes: 4864b0702653 ("objtool, perf: Fix GCC 8 -Wrestrict error")
> > > CC: Josh Poimboeuf <jpoimboe@redhat.com>
> > > Signed-off-by: kbuild test robot <lkp@intel.com>
> > > ---
> > >
> > > tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
> > > head:   58ce54880a03d9c27327434ba15281f5a7139671
> > > commit: 4864b070265301ba88506ee407d51e058dcb0ad6 [42/51] objtool, perf: Fix GCC 8 -Wrestrict error
> > > :::::: branch date: 27 hours ago
> > > :::::: commit date: 9 weeks ago
> > >
> > > Please take the patch only if it's a positive warning. Thanks!
> > >
> > >  seq_clientmgr.c |    2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > --- a/sound/core/seq/seq_clientmgr.c
> > > +++ b/sound/core/seq/seq_clientmgr.c
> > > @@ -306,7 +306,7 @@ static int snd_seq_open(struct inode *in
> > >  	struct snd_seq_user_client *user;
> > >  	int err;
> > >
> > > -	err = nonseekable_open(inode, file);
> > > +	err = stream_open(inode, file);
> > >  	if (err < 0)
> > >  		return err;

Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

[Greg Kroah-Hartman]

[-- Attachment #1: Type: text/plain, Size: 1354 bytes --]

On Mon, Jan 13, 2020 at 06:04:44PM +0000, Kirill Smelkov wrote:
> Android people,
> 
> Please see inside if Android 4.9-o wants to pick up stream_open
> related deadlock fixes. FUSE part of it is already applied to upstream
> stable/linux-4.9.y as 9c829b6e3fe2 and cfd8d2e79524, but the Android
> tree does not have any of the fixes at all.

I do not understand, the commit 10dce8af3422 ("fs: stream_open - opener
for stream-like files so that read and write can run simultaneously
without deadlock") is commit 9c829b6e3fe2 in the linux stable 4.9 branch
and commit bbd84f33652f ("fuse: Add FOPEN_STREAM to use stream_open()")
is commit cfd8d2e79524 in the linux stable 4.9 branch, right?

If so, that commit is in the 4.9.181 release, which _is_ included in all
of the android common branches already, as all of the LTS/stable kernel
releases are merged into the android common trees.

Right now the tip-of-tree of the android-4.9-o branch is 2641bba251cd
("Merge 4.9.209 into android-4.9-o") so of course 4.9.181 is also in
there :)

So I really do not understand what you are asking about here.  Is there
a specific patch that did not get into the stable releases that you feel
should go only into the android tree?  If so, what is that upstream
commit and why wouldn't it go into the stable releases?

confused,

greg k-h

Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

[Todd Kjos]

[-- Attachment #1: Type: text/plain, Size: 2257 bytes --]

> tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1

Looks like the comment was related to an issue found on the
android-4.9-o-mr1 branch which doesn't get LTS merges and is really
just a patch repository for patches cited in the Android Security
Bulletin. We don't do general bug-fixes or take LTS merges on that
branch.

On Mon, Jan 13, 2020 at 10:21 AM Greg Kroah-Hartman
<gregkh@linuxfoundation.org> wrote:
>
> On Mon, Jan 13, 2020 at 06:04:44PM +0000, Kirill Smelkov wrote:
> > Android people,
> >
> > Please see inside if Android 4.9-o wants to pick up stream_open
> > related deadlock fixes. FUSE part of it is already applied to upstream
> > stable/linux-4.9.y as 9c829b6e3fe2 and cfd8d2e79524, but the Android
> > tree does not have any of the fixes at all.
>
> I do not understand, the commit 10dce8af3422 ("fs: stream_open - opener
> for stream-like files so that read and write can run simultaneously
> without deadlock") is commit 9c829b6e3fe2 in the linux stable 4.9 branch
> and commit bbd84f33652f ("fuse: Add FOPEN_STREAM to use stream_open()")
> is commit cfd8d2e79524 in the linux stable 4.9 branch, right?
>
> If so, that commit is in the 4.9.181 release, which _is_ included in all
> of the android common branches already, as all of the LTS/stable kernel
> releases are merged into the android common trees.
>
> Right now the tip-of-tree of the android-4.9-o branch is 2641bba251cd
> ("Merge 4.9.209 into android-4.9-o") so of course 4.9.181 is also in
> there :)
>
> So I really do not understand what you are asking about here.  Is there
> a specific patch that did not get into the stable releases that you feel
> should go only into the android tree?  If so, what is that upstream
> commit and why wouldn't it go into the stable releases?
>
> confused,
>
> greg k-h
>
> --
> You received this message because you are subscribed to the Google Groups "cros-kernel-buildreports" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to cros-kernel-buildreports+unsubscribe(a)googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/cros-kernel-buildreports/20200113182145.GA411698%40kroah.com.

Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

[Kirill Smelkov]

[-- Attachment #1: Type: text/plain, Size: 2770 bytes --]

On Mon, Jan 13, 2020 at 10:32:53AM -0800, Todd Kjos wrote:
> > tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
> 
> Looks like the comment was related to an issue found on the
> android-4.9-o-mr1 branch which doesn't get LTS merges and is really
> just a patch repository for patches cited in the Android Security
> Bulletin. We don't do general bug-fixes or take LTS merges on that
> branch.

Greg, Todd, thanks for feedback. Indeed, it was about android-4.9-o-mr1,
which I mistakenly attributed to only android-4.9-o . Good to hear that
main Android branches are in a good shape.

Julia, if I understand correctly, it seems it is better to remove -mrX
branches from being tracked by kbuild robot as that is causing
confusion to people.

Kirill

> On Mon, Jan 13, 2020 at 10:21 AM Greg Kroah-Hartman
> <gregkh@linuxfoundation.org> wrote:
> >
> > On Mon, Jan 13, 2020 at 06:04:44PM +0000, Kirill Smelkov wrote:
> > > Android people,
> > >
> > > Please see inside if Android 4.9-o wants to pick up stream_open
> > > related deadlock fixes. FUSE part of it is already applied to upstream
> > > stable/linux-4.9.y as 9c829b6e3fe2 and cfd8d2e79524, but the Android
> > > tree does not have any of the fixes at all.
> >
> > I do not understand, the commit 10dce8af3422 ("fs: stream_open - opener
> > for stream-like files so that read and write can run simultaneously
> > without deadlock") is commit 9c829b6e3fe2 in the linux stable 4.9 branch
> > and commit bbd84f33652f ("fuse: Add FOPEN_STREAM to use stream_open()")
> > is commit cfd8d2e79524 in the linux stable 4.9 branch, right?
> >
> > If so, that commit is in the 4.9.181 release, which _is_ included in all
> > of the android common branches already, as all of the LTS/stable kernel
> > releases are merged into the android common trees.
> >
> > Right now the tip-of-tree of the android-4.9-o branch is 2641bba251cd
> > ("Merge 4.9.209 into android-4.9-o") so of course 4.9.181 is also in
> > there :)
> >
> > So I really do not understand what you are asking about here.  Is there
> > a specific patch that did not get into the stable releases that you feel
> > should go only into the android tree?  If so, what is that upstream
> > commit and why wouldn't it go into the stable releases?
> >
> > confused,
> >
> > greg k-h
> >
> > --
> > You received this message because you are subscribed to the Google Groups "cros-kernel-buildreports" group.
> > To unsubscribe from this group and stop receiving emails from it, send an email to cros-kernel-buildreports+unsubscribe(a)googlegroups.com.
> > To view this discussion on the web visit https://groups.google.com/d/msgid/cros-kernel-buildreports/20200113182145.GA411698%40kroah.com.

Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

[Todd Kjos]

[-- Attachment #1: Type: text/plain, Size: 3005 bytes --]

On Mon, Jan 13, 2020 at 10:52 AM Kirill Smelkov <kirr@nexedi.com> wrote:
>
> On Mon, Jan 13, 2020 at 10:32:53AM -0800, Todd Kjos wrote:
> > > tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
> >
> > Looks like the comment was related to an issue found on the
> > android-4.9-o-mr1 branch which doesn't get LTS merges and is really
> > just a patch repository for patches cited in the Android Security
> > Bulletin. We don't do general bug-fixes or take LTS merges on that
> > branch.
>
> Greg, Todd, thanks for feedback. Indeed, it was about android-4.9-o-mr1,
> which I mistakenly attributed to only android-4.9-o . Good to hear that
> main Android branches are in a good shape.
>
> Julia, if I understand correctly, it seems it is better to remove -mrX
> branches from being tracked by kbuild robot as that is causing
> confusion to people.

Yes, remove "-mr1" and "-release" branches.

>
> Kirill
>
> > On Mon, Jan 13, 2020 at 10:21 AM Greg Kroah-Hartman
> > <gregkh@linuxfoundation.org> wrote:
> > >
> > > On Mon, Jan 13, 2020 at 06:04:44PM +0000, Kirill Smelkov wrote:
> > > > Android people,
> > > >
> > > > Please see inside if Android 4.9-o wants to pick up stream_open
> > > > related deadlock fixes. FUSE part of it is already applied to upstream
> > > > stable/linux-4.9.y as 9c829b6e3fe2 and cfd8d2e79524, but the Android
> > > > tree does not have any of the fixes at all.
> > >
> > > I do not understand, the commit 10dce8af3422 ("fs: stream_open - opener
> > > for stream-like files so that read and write can run simultaneously
> > > without deadlock") is commit 9c829b6e3fe2 in the linux stable 4.9 branch
> > > and commit bbd84f33652f ("fuse: Add FOPEN_STREAM to use stream_open()")
> > > is commit cfd8d2e79524 in the linux stable 4.9 branch, right?
> > >
> > > If so, that commit is in the 4.9.181 release, which _is_ included in all
> > > of the android common branches already, as all of the LTS/stable kernel
> > > releases are merged into the android common trees.
> > >
> > > Right now the tip-of-tree of the android-4.9-o branch is 2641bba251cd
> > > ("Merge 4.9.209 into android-4.9-o") so of course 4.9.181 is also in
> > > there :)
> > >
> > > So I really do not understand what you are asking about here.  Is there
> > > a specific patch that did not get into the stable releases that you feel
> > > should go only into the android tree?  If so, what is that upstream
> > > commit and why wouldn't it go into the stable releases?
> > >
> > > confused,
> > >
> > > greg k-h
> > >
> > > --
> > > You received this message because you are subscribed to the Google Groups "cros-kernel-buildreports" group.
> > > To unsubscribe from this group and stop receiving emails from it, send an email to cros-kernel-buildreports+unsubscribe(a)googlegroups.com.
> > > To view this discussion on the web visit https://groups.google.com/d/msgid/cros-kernel-buildreports/20200113182145.GA411698%40kroah.com.

Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

[Julia Lawall]

[-- Attachment #1: Type: text/plain, Size: 2993 bytes --]

On Mon, 13 Jan 2020, Kirill Smelkov wrote:

> On Mon, Jan 13, 2020 at 10:32:53AM -0800, Todd Kjos wrote:
> > > tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
> >
> > Looks like the comment was related to an issue found on the
> > android-4.9-o-mr1 branch which doesn't get LTS merges and is really
> > just a patch repository for patches cited in the Android Security
> > Bulletin. We don't do general bug-fixes or take LTS merges on that
> > branch.
>
> Greg, Todd, thanks for feedback. Indeed, it was about android-4.9-o-mr1,
> which I mistakenly attributed to only android-4.9-o . Good to hear that
> main Android branches are in a good shape.
>
> Julia, if I understand correctly, it seems it is better to remove -mrX
> branches from being tracked by kbuild robot as that is causing
> confusion to people.

I forwarded the request to Philip Li.

thanks,
julia



>
> Kirill
>
> > On Mon, Jan 13, 2020 at 10:21 AM Greg Kroah-Hartman
> > <gregkh@linuxfoundation.org> wrote:
> > >
> > > On Mon, Jan 13, 2020 at 06:04:44PM +0000, Kirill Smelkov wrote:
> > > > Android people,
> > > >
> > > > Please see inside if Android 4.9-o wants to pick up stream_open
> > > > related deadlock fixes. FUSE part of it is already applied to upstream
> > > > stable/linux-4.9.y as 9c829b6e3fe2 and cfd8d2e79524, but the Android
> > > > tree does not have any of the fixes at all.
> > >
> > > I do not understand, the commit 10dce8af3422 ("fs: stream_open - opener
> > > for stream-like files so that read and write can run simultaneously
> > > without deadlock") is commit 9c829b6e3fe2 in the linux stable 4.9 branch
> > > and commit bbd84f33652f ("fuse: Add FOPEN_STREAM to use stream_open()")
> > > is commit cfd8d2e79524 in the linux stable 4.9 branch, right?
> > >
> > > If so, that commit is in the 4.9.181 release, which _is_ included in all
> > > of the android common branches already, as all of the LTS/stable kernel
> > > releases are merged into the android common trees.
> > >
> > > Right now the tip-of-tree of the android-4.9-o branch is 2641bba251cd
> > > ("Merge 4.9.209 into android-4.9-o") so of course 4.9.181 is also in
> > > there :)
> > >
> > > So I really do not understand what you are asking about here.  Is there
> > > a specific patch that did not get into the stable releases that you feel
> > > should go only into the android tree?  If so, what is that upstream
> > > commit and why wouldn't it go into the stable releases?
> > >
> > > confused,
> > >
> > > greg k-h
> > >
> > > --
> > > You received this message because you are subscribed to the Google Groups "cros-kernel-buildreports" group.
> > > To unsubscribe from this group and stop receiving emails from it, send an email to cros-kernel-buildreports+unsubscribe(a)googlegroups.com.
> > > To view this discussion on the web visit https://groups.google.com/d/msgid/cros-kernel-buildreports/20200113182145.GA411698%40kroah.com.
>

Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

[Li, Philip]

[-- Attachment #1: Type: text/plain, Size: 3393 bytes --]

> On Mon, Jan 13, 2020 at 10:52 AM Kirill Smelkov <kirr@nexedi.com> wrote:
> >
> > On Mon, Jan 13, 2020 at 10:32:53AM -0800, Todd Kjos wrote:
> > > > tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
> > >
> > > Looks like the comment was related to an issue found on the
> > > android-4.9-o-mr1 branch which doesn't get LTS merges and is really
> > > just a patch repository for patches cited in the Android Security
> > > Bulletin. We don't do general bug-fixes or take LTS merges on that
> > > branch.
> >
> > Greg, Todd, thanks for feedback. Indeed, it was about android-4.9-o-mr1,
> > which I mistakenly attributed to only android-4.9-o . Good to hear that
> > main Android branches are in a good shape.
> >
> > Julia, if I understand correctly, it seems it is better to remove -mrX
> > branches from being tracked by kbuild robot as that is causing
> > confusion to people.
> 
> Yes, remove "-mr1" and "-release" branches.
Got it, we will blacklist these branches -mrX and -release in the bot.

> 
> >
> > Kirill
> >
> > > On Mon, Jan 13, 2020 at 10:21 AM Greg Kroah-Hartman
> > > <gregkh@linuxfoundation.org> wrote:
> > > >
> > > > On Mon, Jan 13, 2020 at 06:04:44PM +0000, Kirill Smelkov wrote:
> > > > > Android people,
> > > > >
> > > > > Please see inside if Android 4.9-o wants to pick up stream_open
> > > > > related deadlock fixes. FUSE part of it is already applied to upstream
> > > > > stable/linux-4.9.y as 9c829b6e3fe2 and cfd8d2e79524, but the Android
> > > > > tree does not have any of the fixes at all.
> > > >
> > > > I do not understand, the commit 10dce8af3422 ("fs: stream_open - opener
> > > > for stream-like files so that read and write can run simultaneously
> > > > without deadlock") is commit 9c829b6e3fe2 in the linux stable 4.9 branch
> > > > and commit bbd84f33652f ("fuse: Add FOPEN_STREAM to use
> stream_open()")
> > > > is commit cfd8d2e79524 in the linux stable 4.9 branch, right?
> > > >
> > > > If so, that commit is in the 4.9.181 release, which _is_ included in all
> > > > of the android common branches already, as all of the LTS/stable kernel
> > > > releases are merged into the android common trees.
> > > >
> > > > Right now the tip-of-tree of the android-4.9-o branch is 2641bba251cd
> > > > ("Merge 4.9.209 into android-4.9-o") so of course 4.9.181 is also in
> > > > there :)
> > > >
> > > > So I really do not understand what you are asking about here.  Is there
> > > > a specific patch that did not get into the stable releases that you feel
> > > > should go only into the android tree?  If so, what is that upstream
> > > > commit and why wouldn't it go into the stable releases?
> > > >
> > > > confused,
> > > >
> > > > greg k-h
> > > >
> > > > --
> > > > You received this message because you are subscribed to the Google
> Groups "cros-kernel-buildreports" group.
> > > > To unsubscribe from this group and stop receiving emails from it, send an
> email to cros-kernel-buildreports+unsubscribe(a)googlegroups.com.
> > > > To view this discussion on the web visit
> https://groups.google.com/d/msgid/cros-kernel-
> buildreports/20200113182145.GA411698%40kroah.com.
> _______________________________________________
> kbuild-all mailing list -- kbuild-all(a)lists.01.org
> To unsubscribe send an email to kbuild-all-leave(a)lists.01.org

Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

[Guenter Roeck]

[-- Attachment #1: Type: text/plain, Size: 1392 bytes --]

Hi Philip,

On Mon, Jan 13, 2020 at 3:08 PM Li, Philip <philip.li@intel.com> wrote:
>
> > On Mon, Jan 13, 2020 at 10:52 AM Kirill Smelkov <kirr@nexedi.com> wrote:
> > >
> > > On Mon, Jan 13, 2020 at 10:32:53AM -0800, Todd Kjos wrote:
> > > > > tree:   https://android.googlesource.com/kernel/common android-4.9-o-mr1
> > > >
> > > > Looks like the comment was related to an issue found on the
> > > > android-4.9-o-mr1 branch which doesn't get LTS merges and is really
> > > > just a patch repository for patches cited in the Android Security
> > > > Bulletin. We don't do general bug-fixes or take LTS merges on that
> > > > branch.
> > >
> > > Greg, Todd, thanks for feedback. Indeed, it was about android-4.9-o-mr1,
> > > which I mistakenly attributed to only android-4.9-o . Good to hear that
> > > main Android branches are in a good shape.
> > >
> > > Julia, if I understand correctly, it seems it is better to remove -mrX
> > > branches from being tracked by kbuild robot as that is causing
> > > confusion to people.
> >
> > Yes, remove "-mr1" and "-release" branches.
> Got it, we will blacklist these branches -mrX and -release in the bot.
>

Since you are at it, would it be possible to add chromeos-5.4 to the
list of tested branches ? I sent a pull request at github a few of
weeks ago, but it doesn't look like those are handled.

Thanks,
Guenter

Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

[Li, Philip]

[-- Attachment #1: Type: text/plain, Size: 1722 bytes --]

> Subject: Re: [kbuild-all] Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci
> warnings
> 
> Hi Philip,
> 
> On Mon, Jan 13, 2020 at 3:08 PM Li, Philip <philip.li@intel.com> wrote:
> >
> > > On Mon, Jan 13, 2020 at 10:52 AM Kirill Smelkov <kirr@nexedi.com> wrote:
> > > >
> > > > On Mon, Jan 13, 2020 at 10:32:53AM -0800, Todd Kjos wrote:
> > > > > > tree:   https://android.googlesource.com/kernel/common android-4.9-o-
> mr1
> > > > >
> > > > > Looks like the comment was related to an issue found on the
> > > > > android-4.9-o-mr1 branch which doesn't get LTS merges and is really
> > > > > just a patch repository for patches cited in the Android Security
> > > > > Bulletin. We don't do general bug-fixes or take LTS merges on that
> > > > > branch.
> > > >
> > > > Greg, Todd, thanks for feedback. Indeed, it was about android-4.9-o-mr1,
> > > > which I mistakenly attributed to only android-4.9-o . Good to hear that
> > > > main Android branches are in a good shape.
> > > >
> > > > Julia, if I understand correctly, it seems it is better to remove -mrX
> > > > branches from being tracked by kbuild robot as that is causing
> > > > confusion to people.
> > >
> > > Yes, remove "-mr1" and "-release" branches.
> > Got it, we will blacklist these branches -mrX and -release in the bot.
> >
> 
> Since you are at it, would it be possible to add chromeos-5.4 to the
> list of tested branches ? I sent a pull request at github a few of
> weeks ago, but it doesn't look like those are handled.
Sure, I will add it very soon.

BTW: the official repo is at https://github.com/intel/lkp-tests, probably
your request goes to fengguang's original one.

> 
> Thanks,
> Guenter

Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci warnings

[Guenter Roeck]

[-- Attachment #1: Type: text/plain, Size: 1933 bytes --]

On Mon, Jan 13, 2020 at 3:18 PM Li, Philip <philip.li@intel.com> wrote:
>
> > Subject: Re: [kbuild-all] Re: Fwd: [PATCH] objtool, perf: fix stream_open.cocci
> > warnings
> >
> > Hi Philip,
> >
> > On Mon, Jan 13, 2020 at 3:08 PM Li, Philip <philip.li@intel.com> wrote:
> > >
> > > > On Mon, Jan 13, 2020 at 10:52 AM Kirill Smelkov <kirr@nexedi.com> wrote:
> > > > >
> > > > > On Mon, Jan 13, 2020 at 10:32:53AM -0800, Todd Kjos wrote:
> > > > > > > tree:   https://android.googlesource.com/kernel/common android-4.9-o-
> > mr1
> > > > > >
> > > > > > Looks like the comment was related to an issue found on the
> > > > > > android-4.9-o-mr1 branch which doesn't get LTS merges and is really
> > > > > > just a patch repository for patches cited in the Android Security
> > > > > > Bulletin. We don't do general bug-fixes or take LTS merges on that
> > > > > > branch.
> > > > >
> > > > > Greg, Todd, thanks for feedback. Indeed, it was about android-4.9-o-mr1,
> > > > > which I mistakenly attributed to only android-4.9-o . Good to hear that
> > > > > main Android branches are in a good shape.
> > > > >
> > > > > Julia, if I understand correctly, it seems it is better to remove -mrX
> > > > > branches from being tracked by kbuild robot as that is causing
> > > > > confusion to people.
> > > >
> > > > Yes, remove "-mr1" and "-release" branches.
> > > Got it, we will blacklist these branches -mrX and -release in the bot.
> > >
> >
> > Since you are at it, would it be possible to add chromeos-5.4 to the
> > list of tested branches ? I sent a pull request at github a few of
> > weeks ago, but it doesn't look like those are handled.
> Sure, I will add it very soon.
>
> BTW: the official repo is at https://github.com/intel/lkp-tests, probably
> your request goes to fengguang's original one.
>

Ah yes, of course. Thanks a lot for the pointer, and for adding the branch.

Guenter