linux-next: manual merge of the vfs-brauner tree with the vfs-brauner-fixes tree

linux-next: manual merge of the vfs-brauner tree with the vfs-brauner-fixes tree

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 849 bytes --]

Hi all,

Today's linux-next merge of the vfs-brauner tree got a conflict in:

  fs/open.c

between commit:

  7536b2f06724e ("VFS: generate FS_CREATE before FS_OPEN when ->atomic_open used.")

from the vfs-brauner-fixes tree and commit:

  7d1cf5e624ef5 ("vfs: generate FS_CREATE before FS_OPEN when ->atomic_open used.")

from the vfs-brauner tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

diff --cc fs/open.c
index 58110f0bf5ac8,28f2fcbebb1ba..0000000000000
--- a/fs/open.c
+++ b/fs/open.c

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the vfs-brauner tree with the vfs-brauner-fixes tree

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 293 bytes --]

On Thu, Jun 20, 2024 at 12:34:17PM +0100, Mark Brown wrote:
> Hi all,
> 
> Today's linux-next merge of the vfs-brauner tree got a conflict in:
> 
>   fs/open.c
> 
> between commit:

Sorry, the scripting got confused and didn't start the merge properly so
this should just be noise.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

linux-next: manual merge of the vfs-brauner tree with the vfs-brauner-fixes tree

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 3286 bytes --]

Hi all,

Today's linux-next merge of the vfs-brauner tree got a conflict in:

  fs/iomap/bio.c

between commit:

  f621324dfb3d6 ("iomap: fix lockdep complaint when reads fail")

from the vfs-brauner-fixes tree and commit:

  e8f9cf03c9dc9 ("iomap: support ioends for buffered reads")

from the vfs-brauner tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

diff --cc fs/iomap/bio.c
index edd908183058f,f989ffcaac96d..0000000000000
--- a/fs/iomap/bio.c
+++ b/fs/iomap/bio.c
@@@ -8,66 -9,33 +9,78 @@@
  #include "internal.h"
  #include "trace.h"
  
 +static DEFINE_SPINLOCK(failed_read_lock);
 +static struct bio_list failed_read_list = BIO_EMPTY_LIST;
 +
- static void __iomap_read_end_io(struct bio *bio)
+ static u32 __iomap_read_end_io(struct bio *bio, int error)
  {
- 	int error = blk_status_to_errno(bio->bi_status);
  	struct folio_iter fi;
+ 	u32 folio_count = 0;
  
- 	bio_for_each_folio_all(fi, bio)
+ 	bio_for_each_folio_all(fi, bio) {
  		iomap_finish_folio_read(fi.folio, fi.offset, fi.length, error);
+ 		folio_count++;
+ 	}
+ 	if (bio_integrity(bio))
+ 		fs_bio_integrity_free(bio);
  	bio_put(bio);
+ 	return folio_count;
  }
  
 +static void
 +iomap_fail_reads(
 +	struct work_struct	*work)
 +{
 +	struct bio		*bio;
 +	struct bio_list		tmp = BIO_EMPTY_LIST;
 +	unsigned long		flags;
 +
 +	spin_lock_irqsave(&failed_read_lock, flags);
 +	bio_list_merge_init(&tmp, &failed_read_list);
 +	spin_unlock_irqrestore(&failed_read_lock, flags);
 +
 +	while ((bio = bio_list_pop(&tmp)) != NULL) {
- 		__iomap_read_end_io(bio);
++		__iomap_read_end_io(bio, blk_status_to_errno(bio->bi_status));
 +		cond_resched();
 +	}
 +}
 +
 +static DECLARE_WORK(failed_read_work, iomap_fail_reads);
 +
 +static void iomap_fail_buffered_read(struct bio *bio)
 +{
 +	unsigned long flags;
 +
 +	/*
 +	 * Bounce I/O errors to a workqueue to avoid nested i_lock acquisitions
 +	 * in the fserror code.  The caller no longer owns the bio reference
 +	 * after the spinlock drops.
 +	 */
 +	spin_lock_irqsave(&failed_read_lock, flags);
 +	if (bio_list_empty(&failed_read_list))
 +		WARN_ON_ONCE(!schedule_work(&failed_read_work));
 +	bio_list_add(&failed_read_list, bio);
 +	spin_unlock_irqrestore(&failed_read_lock, flags);
 +}
 +
  static void iomap_read_end_io(struct bio *bio)
  {
 -	__iomap_read_end_io(bio, blk_status_to_errno(bio->bi_status));
 +	if (bio->bi_status) {
 +		iomap_fail_buffered_read(bio);
 +		return;
 +	}
 +
- 	__iomap_read_end_io(bio);
++	__iomap_read_end_io(bio, 0);
  }
  
- static void iomap_bio_submit_read(struct iomap_read_folio_ctx *ctx)
++
+ u32 iomap_finish_ioend_buffered_read(struct iomap_ioend *ioend)
+ {
+ 	return __iomap_read_end_io(&ioend->io_bio, ioend->io_error);
+ }
+ 
+ static void iomap_bio_submit_read(const struct iomap_iter *iter,
+ 		struct iomap_read_folio_ctx *ctx)
  {
  	struct bio *bio = ctx->read_ctx;
  

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the vfs-brauner tree with the vfs-brauner-fixes tree

[Darrick J. Wong]

On Wed, Mar 25, 2026 at 01:29:37PM +0000, Mark Brown wrote:
> Hi all,
> 
> Today's linux-next merge of the vfs-brauner tree got a conflict in:
> 
>   fs/iomap/bio.c
> 
> between commit:
> 
>   f621324dfb3d6 ("iomap: fix lockdep complaint when reads fail")
> 
> from the vfs-brauner-fixes tree and commit:
> 
>   e8f9cf03c9dc9 ("iomap: support ioends for buffered reads")
> 
> from the vfs-brauner tree.
> 
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.

That looks correct to me, thanks for pointing out the merge conflict. :)

--D

> diff --cc fs/iomap/bio.c
> index edd908183058f,f989ffcaac96d..0000000000000
> --- a/fs/iomap/bio.c
> +++ b/fs/iomap/bio.c
> @@@ -8,66 -9,33 +9,78 @@@
>   #include "internal.h"
>   #include "trace.h"
>   
>  +static DEFINE_SPINLOCK(failed_read_lock);
>  +static struct bio_list failed_read_list = BIO_EMPTY_LIST;
>  +
> - static void __iomap_read_end_io(struct bio *bio)
> + static u32 __iomap_read_end_io(struct bio *bio, int error)
>   {
> - 	int error = blk_status_to_errno(bio->bi_status);
>   	struct folio_iter fi;
> + 	u32 folio_count = 0;
>   
> - 	bio_for_each_folio_all(fi, bio)
> + 	bio_for_each_folio_all(fi, bio) {
>   		iomap_finish_folio_read(fi.folio, fi.offset, fi.length, error);
> + 		folio_count++;
> + 	}
> + 	if (bio_integrity(bio))
> + 		fs_bio_integrity_free(bio);
>   	bio_put(bio);
> + 	return folio_count;
>   }
>   
>  +static void
>  +iomap_fail_reads(
>  +	struct work_struct	*work)
>  +{
>  +	struct bio		*bio;
>  +	struct bio_list		tmp = BIO_EMPTY_LIST;
>  +	unsigned long		flags;
>  +
>  +	spin_lock_irqsave(&failed_read_lock, flags);
>  +	bio_list_merge_init(&tmp, &failed_read_list);
>  +	spin_unlock_irqrestore(&failed_read_lock, flags);
>  +
>  +	while ((bio = bio_list_pop(&tmp)) != NULL) {
> - 		__iomap_read_end_io(bio);
> ++		__iomap_read_end_io(bio, blk_status_to_errno(bio->bi_status));
>  +		cond_resched();
>  +	}
>  +}
>  +
>  +static DECLARE_WORK(failed_read_work, iomap_fail_reads);
>  +
>  +static void iomap_fail_buffered_read(struct bio *bio)
>  +{
>  +	unsigned long flags;
>  +
>  +	/*
>  +	 * Bounce I/O errors to a workqueue to avoid nested i_lock acquisitions
>  +	 * in the fserror code.  The caller no longer owns the bio reference
>  +	 * after the spinlock drops.
>  +	 */
>  +	spin_lock_irqsave(&failed_read_lock, flags);
>  +	if (bio_list_empty(&failed_read_list))
>  +		WARN_ON_ONCE(!schedule_work(&failed_read_work));
>  +	bio_list_add(&failed_read_list, bio);
>  +	spin_unlock_irqrestore(&failed_read_lock, flags);
>  +}
>  +
>   static void iomap_read_end_io(struct bio *bio)
>   {
>  -	__iomap_read_end_io(bio, blk_status_to_errno(bio->bi_status));
>  +	if (bio->bi_status) {
>  +		iomap_fail_buffered_read(bio);
>  +		return;
>  +	}
>  +
> - 	__iomap_read_end_io(bio);
> ++	__iomap_read_end_io(bio, 0);
>   }
>   
> - static void iomap_bio_submit_read(struct iomap_read_folio_ctx *ctx)
> ++
> + u32 iomap_finish_ioend_buffered_read(struct iomap_ioend *ioend)
> + {
> + 	return __iomap_read_end_io(&ioend->io_bio, ioend->io_error);
> + }
> + 
> + static void iomap_bio_submit_read(const struct iomap_iter *iter,
> + 		struct iomap_read_folio_ctx *ctx)
>   {
>   	struct bio *bio = ctx->read_ctx;
>   

Re: linux-next: manual merge of the vfs-brauner tree with the vfs-brauner-fixes tree

[Christoph Hellwig]

The merge looks good, thanks!

Re: linux-next: manual merge of the vfs-brauner tree with the vfs-brauner-fixes tree

[Christian Brauner]

On Wed, Mar 25, 2026 at 08:13:56AM -0700, Darrick J. Wong wrote:
> On Wed, Mar 25, 2026 at 01:29:37PM +0000, Mark Brown wrote:
> > Hi all,
> > 
> > Today's linux-next merge of the vfs-brauner tree got a conflict in:
> > 
> >   fs/iomap/bio.c
> > 
> > between commit:
> > 
> >   f621324dfb3d6 ("iomap: fix lockdep complaint when reads fail")
> > 
> > from the vfs-brauner-fixes tree and commit:
> > 
> >   e8f9cf03c9dc9 ("iomap: support ioends for buffered reads")
> > 
> > from the vfs-brauner tree.
> > 
> > I fixed it up (see below) and can carry the fix as necessary. This
> > is now fixed as far as linux-next is concerned, but any non trivial
> > conflicts should be mentioned to your upstream maintainer when your tree
> > is submitted for merging.  You may also want to consider cooperating
> > with the maintainer of the conflicting tree to minimise any particularly
> > complex conflicts.
> 
> That looks correct to me, thanks for pointing out the merge conflict. :)

Wait wait, the vfs.fixes tree is merged into the vfs.all tree which
should have the merge conflict resolution:

commit 45933bb2dca2b8a5c81cd50947b7d0f9c381c867
Merge: a61ed0d2b28c 1b63f91d1c90
Author:     Christian Brauner <brauner@kernel.org>
AuthorDate: Tue Mar 24 23:38:16 2026 +0100
Commit:     Christian Brauner <brauner@kernel.org>
CommitDate: Tue Mar 24 23:38:16 2026 +0100

    Merge branch 'vfs-7.1.integrity' into vfs.all
    
    Signed-off-by: Christian Brauner <brauner@kernel.org>
    
    # Conflicts:
    #       fs/iomap/bio.c

diff --cc fs/iomap/bio.c
index edd908183058,f989ffcaac96..4504f4633f17
--- a/fs/iomap/bio.c
+++ b/fs/iomap/bio.c
@@@ -8,66 -9,33 +9,77 @@@
  #include "internal.h"
  #include "trace.h"
  
 +static DEFINE_SPINLOCK(failed_read_lock);
 +static struct bio_list failed_read_list = BIO_EMPTY_LIST;
 +
- static void __iomap_read_end_io(struct bio *bio)
+ static u32 __iomap_read_end_io(struct bio *bio, int error)
  {
- 	int error = blk_status_to_errno(bio->bi_status);
  	struct folio_iter fi;
+ 	u32 folio_count = 0;
  
- 	bio_for_each_folio_all(fi, bio)
+ 	bio_for_each_folio_all(fi, bio) {
  		iomap_finish_folio_read(fi.folio, fi.offset, fi.length, error);
+ 		folio_count++;
+ 	}
+ 	if (bio_integrity(bio))
+ 		fs_bio_integrity_free(bio);
  	bio_put(bio);
+ 	return folio_count;
  }
  
 +static void
 +iomap_fail_reads(
 +	struct work_struct	*work)
 +{
 +	struct bio		*bio;
 +	struct bio_list		tmp = BIO_EMPTY_LIST;
 +	unsigned long		flags;
 +
 +	spin_lock_irqsave(&failed_read_lock, flags);
 +	bio_list_merge_init(&tmp, &failed_read_list);
 +	spin_unlock_irqrestore(&failed_read_lock, flags);
 +
 +	while ((bio = bio_list_pop(&tmp)) != NULL) {
- 		__iomap_read_end_io(bio);
++		__iomap_read_end_io(bio, blk_status_to_errno(bio->bi_status));
 +		cond_resched();
 +	}
 +}
 +
 +static DECLARE_WORK(failed_read_work, iomap_fail_reads);
 +
 +static void iomap_fail_buffered_read(struct bio *bio)
 +{
 +	unsigned long flags;
 +
 +	/*
 +	 * Bounce I/O errors to a workqueue to avoid nested i_lock acquisitions
 +	 * in the fserror code.  The caller no longer owns the bio reference
 +	 * after the spinlock drops.
 +	 */
 +	spin_lock_irqsave(&failed_read_lock, flags);
 +	if (bio_list_empty(&failed_read_list))
 +		WARN_ON_ONCE(!schedule_work(&failed_read_work));
 +	bio_list_add(&failed_read_list, bio);
 +	spin_unlock_irqrestore(&failed_read_lock, flags);
 +}
 +
  static void iomap_read_end_io(struct bio *bio)
  {
 -	__iomap_read_end_io(bio, blk_status_to_errno(bio->bi_status));
 +	if (bio->bi_status) {
 +		iomap_fail_buffered_read(bio);
 +		return;
 +	}
 +
- 	__iomap_read_end_io(bio);
++	__iomap_read_end_io(bio, 0);
+ }
+ 
+ u32 iomap_finish_ioend_buffered_read(struct iomap_ioend *ioend)
+ {
+ 	return __iomap_read_end_io(&ioend->io_bio, ioend->io_error);
  }
  
- static void iomap_bio_submit_read(struct iomap_read_folio_ctx *ctx)
+ static void iomap_bio_submit_read(const struct iomap_iter *iter,
+ 		struct iomap_read_folio_ctx *ctx)
  {
  	struct bio *bio = ctx->read_ctx;
  

> 
> --D
> 
> > diff --cc fs/iomap/bio.c
> > index edd908183058f,f989ffcaac96d..0000000000000
> > --- a/fs/iomap/bio.c
> > +++ b/fs/iomap/bio.c
> > @@@ -8,66 -9,33 +9,78 @@@
> >   #include "internal.h"
> >   #include "trace.h"
> >   
> >  +static DEFINE_SPINLOCK(failed_read_lock);
> >  +static struct bio_list failed_read_list = BIO_EMPTY_LIST;
> >  +
> > - static void __iomap_read_end_io(struct bio *bio)
> > + static u32 __iomap_read_end_io(struct bio *bio, int error)
> >   {
> > - 	int error = blk_status_to_errno(bio->bi_status);
> >   	struct folio_iter fi;
> > + 	u32 folio_count = 0;
> >   
> > - 	bio_for_each_folio_all(fi, bio)
> > + 	bio_for_each_folio_all(fi, bio) {
> >   		iomap_finish_folio_read(fi.folio, fi.offset, fi.length, error);
> > + 		folio_count++;
> > + 	}
> > + 	if (bio_integrity(bio))
> > + 		fs_bio_integrity_free(bio);
> >   	bio_put(bio);
> > + 	return folio_count;
> >   }
> >   
> >  +static void
> >  +iomap_fail_reads(
> >  +	struct work_struct	*work)
> >  +{
> >  +	struct bio		*bio;
> >  +	struct bio_list		tmp = BIO_EMPTY_LIST;
> >  +	unsigned long		flags;
> >  +
> >  +	spin_lock_irqsave(&failed_read_lock, flags);
> >  +	bio_list_merge_init(&tmp, &failed_read_list);
> >  +	spin_unlock_irqrestore(&failed_read_lock, flags);
> >  +
> >  +	while ((bio = bio_list_pop(&tmp)) != NULL) {
> > - 		__iomap_read_end_io(bio);
> > ++		__iomap_read_end_io(bio, blk_status_to_errno(bio->bi_status));
> >  +		cond_resched();
> >  +	}
> >  +}
> >  +
> >  +static DECLARE_WORK(failed_read_work, iomap_fail_reads);
> >  +
> >  +static void iomap_fail_buffered_read(struct bio *bio)
> >  +{
> >  +	unsigned long flags;
> >  +
> >  +	/*
> >  +	 * Bounce I/O errors to a workqueue to avoid nested i_lock acquisitions
> >  +	 * in the fserror code.  The caller no longer owns the bio reference
> >  +	 * after the spinlock drops.
> >  +	 */
> >  +	spin_lock_irqsave(&failed_read_lock, flags);
> >  +	if (bio_list_empty(&failed_read_list))
> >  +		WARN_ON_ONCE(!schedule_work(&failed_read_work));
> >  +	bio_list_add(&failed_read_list, bio);
> >  +	spin_unlock_irqrestore(&failed_read_lock, flags);
> >  +}
> >  +
> >   static void iomap_read_end_io(struct bio *bio)
> >   {
> >  -	__iomap_read_end_io(bio, blk_status_to_errno(bio->bi_status));
> >  +	if (bio->bi_status) {
> >  +		iomap_fail_buffered_read(bio);
> >  +		return;
> >  +	}
> >  +
> > - 	__iomap_read_end_io(bio);
> > ++	__iomap_read_end_io(bio, 0);
> >   }
> >   
> > - static void iomap_bio_submit_read(struct iomap_read_folio_ctx *ctx)
> > ++
> > + u32 iomap_finish_ioend_buffered_read(struct iomap_ioend *ioend)
> > + {
> > + 	return __iomap_read_end_io(&ioend->io_bio, ioend->io_error);
> > + }
> > + 
> > + static void iomap_bio_submit_read(const struct iomap_iter *iter,
> > + 		struct iomap_read_folio_ctx *ctx)
> >   {
> >   	struct bio *bio = ctx->read_ctx;
> >   
> 
> 

Re: linux-next: manual merge of the vfs-brauner tree with the vfs-brauner-fixes tree

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 807 bytes --]

On Thu, Mar 26, 2026 at 02:24:26PM +0100, Christian Brauner wrote:
> On Wed, Mar 25, 2026 at 08:13:56AM -0700, Darrick J. Wong wrote:

> > That looks correct to me, thanks for pointing out the merge conflict. :)

> Wait wait, the vfs.fixes tree is merged into the vfs.all tree which
> should have the merge conflict resolution:

I'm still merging the version from next-20260312, see my report from
yesterday:

  https://lore.kernel.org/linux-next/acPV19IY3Gna6Ira@sirena.org.uk/T/#u

The initially reported issue was fixed, but there's a rather similar new
issue showed up, I suspect it was masked by the prior issue since it's
in a separate kselftest directory (this is namespaces, prior one was in
filesystems) and the kselftest build system is less good at surfacing
all errors than the main kernel one.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]