On Wed, Feb 17, 2021 at 03:05:07PM +0100, Miklos Szeredi wrote:
> On Mon, Jan 25, 2021 at 4:31 PM Alessio Balsini wrote:
> >
> > Enabling FUSE passthrough for mmap-ed operations not only affects
> > performance, but has also been shown as mandatory for the correct
&g
On Wed, Mar 24, 2021 at 03:43:12PM +0800, Rokudo Yan wrote:
> On 1/26/21 12:46 AM, Alessio Balsini wrote:
> > On Mon, Jan 25, 2021 at 03:30:50PM +, Alessio Balsini wrote:
> > > OverlayFS implements its own function to translate iocb flags into rw
> > > flags, so t
On Thu, Mar 18, 2021 at 10:15:43PM +0100, Arnd Bergmann wrote:
> On Thu, Mar 18, 2021 at 5:13 PM Alessio Balsini wrote:
> > On Tue, Mar 16, 2021 at 07:53:06PM +0100, Arnd Bergmann wrote:
> > > On Mon, Jan 25, 2021 at 4:48 PM Alessio Balsini
> > > wrote:
> > >
d Bergmann
Signed-off-by: Alessio Balsini
---
fs/fuse/dev.c | 7 ++-
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index c0fee830a34e..a5ceccc5ef00 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -2233,11 +2233,8 @@ static long fuse_dev_ioctl(struct
On Tue, Mar 16, 2021 at 07:53:06PM +0100, Arnd Bergmann wrote:
> On Mon, Jan 25, 2021 at 4:48 PM Alessio Balsini wrote:
> >
> > With a 64-bit kernel build the FUSE device cannot handle ioctl requests
> > coming from 32-bit user space.
> > This is due to the ioc
On Fri, Feb 19, 2021 at 09:40:21AM +0100, Miklos Szeredi wrote:
> On Fri, Feb 19, 2021 at 8:05 AM Peng Tao wrote:
> >
> > On Wed, Feb 17, 2021 at 9:41 PM Miklos Szeredi wrote:
>
> > > What I think would be useful is to have an explicit
> > > FUSE_DEV_IOC_PASSTHROUGH_CLOSE ioctl, that would need
On Wed, Feb 17, 2021 at 11:21:04AM +0100, Miklos Szeredi wrote:
> On Mon, Jan 25, 2021 at 4:31 PM Alessio Balsini wrote:
> >
> > With a 64-bit kernel build the FUSE device cannot handle ioctl requests
> > coming from 32-bit user space.
> > This is due to the ioc
On Fri, Feb 05, 2021 at 05:23:56PM +0800, Peng Tao wrote:
> On Mon, Jan 25, 2021 at 11:31 PM Alessio Balsini wrote:
> >
> > When using FUSE passthrough, read/write operations are directly
> > forwarded to the lower file system file through VFS, but there is no
> > guar
Hi all,
I'm more than happy to change the interface into something that is
objectively better and accepted by everyone.
I would really love to reach the point at which we have a "stable-ish"
UAPI as soon as possible.
I've been thinking about a few possible approaches to fix the issue, yet
to pres
On Wed, Jan 27, 2021 at 08:15:19PM +0800, qxy wrote:
> Hi Allesio,
>
> Thank you for your supply for 32-bit user space.
> Actually we have already met this issue on our product and we resolved it
> like this:
>
> Project *platform/external/libfuse*
> diff --git a/include/fuse_kernel.h b/include/f
the
passthrough performance benefits to mmap-ed operations.
[1] https://lore.kernel.org/lkml/20210119110654.11817-1-wu-...@tcl.com/
Signed-off-by: Alessio Balsini
---
fs/fuse/file.c| 3 +++
fs/fuse/fuse_i.h | 1 +
fs/fuse/passthrough.c | 41
the passthrough functions for the set-up and tear-down of the
data structures and locks that will be used both when fuse_conns and
fuse_files are created/deleted.
Signed-off-by: Alessio Balsini
---
fs/fuse/Makefile | 1 +
fs/fuse/dev.c | 12
fs/fuse/dir.c
to be targeted
for this feature.
Passthrough comes with another limitation: no further file system
stacking is allowed for those FUSE file systems using passthrough.
Signed-off-by: Alessio Balsini
---
fs/fuse/inode.c | 5 +++
fs/fuse/passthrough.
l the FUSE daemon responsible for enabling or not
the passthrough feature at file open time, and should enable the feature
only after appropriate access policy checks.
Signed-off-by: Alessio Balsini
---
fs/fuse/fuse_i.h | 5 -
fs/fuse/passthrough.c | 11 +++
2 files change
n.
Reduce the OverlayFS code by making the flag conversion function generic
and reusable.
Signed-off-by: Alessio Balsini
---
fs/overlayfs/file.c | 23 +--
include/linux/fs.h | 5 +
2 files changed, 10 insertions(+), 18 deletions(-)
diff --git a/fs/overlayfs/file.c b/fs
On Mon, Jan 25, 2021 at 03:30:50PM +, Alessio Balsini wrote:
> OverlayFS implements its own function to translate iocb flags into rw
> flags, so that they can be passed into another vfs call.
> With commit ce71bfea207b4 ("fs: align IOCB_* flags with RWF_* flags")
> Jens
Use the fs_stack_depth to prevent further stacking and a minor fix.
[Proposed by Jann Horn]
Changes in v2:
* Changed the feature name to passthrough from stacked_io.
[Proposed by Linus Torvalds]
Alessio Balsini (8):
fs: Generic function to convert iocb to rw flags
fuse: 32-bit user space ioctl co
onous operations, yet covering
the majority of the use cases.
Signed-off-by: Alessio Balsini
---
fs/fuse/file.c| 8 --
fs/fuse/fuse_i.h | 2 ++
fs/fuse/passthrough.c | 57 +++
3 files changed, 65 insertions(+), 2 deletions(-)
diff --git a/fs
number from the ioctl command to enable
32-bit user space compatibility on 64-bit kernel builds.
Signed-off-by: Alessio Balsini
---
fs/fuse/dev.c | 29 ++---
include/uapi/linux/fuse.h | 3 ++-
2 files changed, 20 insertions(+), 12 deletions(-)
diff --git a
-by: Alessio Balsini
---
fs/fuse/passthrough.c | 89 +--
1 file changed, 78 insertions(+), 11 deletions(-)
diff --git a/fs/fuse/passthrough.c b/fs/fuse/passthrough.c
index d949ca07a83b..c7fa1eeb7639 100644
--- a/fs/fuse/passthrough.c
+++ b/fs/fuse
On Tue, Jan 19, 2021 at 12:34:23PM +, Alessio Balsini wrote:
> On Tue, Jan 19, 2021 at 07:06:54PM +0800, Rokudo Yan wrote:
> > on Mon, Jan 18, 2021 at 5:27 PM Alessio Balsini wrote:
> > >
> > > This is the 11th version of the series, rebased on top of v5.11
On Tue, Jan 19, 2021 at 07:06:54PM +0800, Rokudo Yan wrote:
> on Mon, Jan 18, 2021 at 5:27 PM Alessio Balsini wrote:
> >
> > This is the 11th version of the series, rebased on top of v5.11-rc4.
> > Please find the changelog at the bottom of this cover letter.
> >
>
On Mon, Jan 18, 2021 at 07:27:45PM +, Alessio Balsini wrote:
> Implement the FUSE passthrough ioctl() that associates the lower
> (passthrough) file system file with the fuse_file.
>
> The file descriptor passed to the ioctl() by the FUSE daemon is used to
> access the relati
On Tue, Jan 19, 2021 at 08:33:16AM +0200, Amir Goldstein wrote:
> On Mon, Jan 18, 2021 at 9:28 PM Alessio Balsini wrote:
> >
> > Expose the FUSE_PASSTHROUGH interface to user space and declare all the
> > basic data structures and functions as the skeleton on top of which the
number from the ioctl command to enable
32-bit user space compatibility on 64-bit kernel builds.
Signed-off-by: Alessio Balsini
---
fs/fuse/dev.c | 29 ++---
include/uapi/linux/fuse.h | 3 ++-
2 files changed, 20 insertions(+), 12 deletions(-)
diff --git a/fs/fuse
n.
Reduce the OverlayFS code by making the flag conversion function generic
and reusable.
Signed-off-by: Alessio Balsini
---
fs/overlayfs/file.c | 23 +--
include/linux/fs.h | 5 +
2 files changed, 10 insertions(+), 18 deletions(-)
diff --git a/fs/overlayfs/file.c b/fs
, add the passthrough functions for the set-up and tear-down of the
data structures and locks that will be used both when fuse_conns and
fuse_files are created/deleted.
Signed-off-by: Alessio Balsini
---
fs/fuse/Makefile | 1 +
fs/fuse/dev.c | 12
fs/fuse/dir.c
l the FUSE daemon responsible for enabling or not the
passthrough feature at file open time, and should enable the feature only
after appropriate access policy checks.
Signed-off-by: Alessio Balsini
---
fs/fuse/fuse_i.h | 5 -
fs/fuse/passthrough.c | 11 +++
2 files change
onous operations, yet covering the majority
of the use cases.
Signed-off-by: Alessio Balsini
---
fs/fuse/file.c| 8 --
fs/fuse/fuse_i.h | 2 ++
fs/fuse/passthrough.c | 57 +++
3 files changed, 65 insertions(+), 2 deletions(-)
diff --git a/fs
-by: Alessio Balsini
---
fs/fuse/passthrough.c | 89 +--
1 file changed, 78 insertions(+), 11 deletions(-)
diff --git a/fs/fuse/passthrough.c b/fs/fuse/passthrough.c
index 8f6882a31a0b..da71a74014d7 100644
--- a/fs/fuse/passthrough.c
+++ b/fs/fuse
ations.
Signed-off-by: Alessio Balsini
---
fs/fuse/inode.c | 5 +++
fs/fuse/passthrough.c | 87 ++-
2 files changed, 90 insertions(+), 2 deletions(-)
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index d5c46eafb419..bc327789f25d 100644
--- a/f
:
* Use the fs_stack_depth to prevent further stacking and a minor fix.
[Proposed by Jann Horn]
Changes in v2:
* Changed the feature name to passthrough from stacked_io.
[Proposed by Linus Torvalds]
Alessio Balsini (7):
fs: Generic function to convert iocb to rw flags
fuse: 32-bit user sp
On Wed, Dec 16, 2020 at 09:32:51PM +0800, wu-...@tcl.com wrote:
> Hi Alessio,
>
> It may cause file reference counter leak in fuse_passthrough_open. If the
> passthrough_filp
>
> not implement read_iter/write_iter or passthrough struct allocated failed,
> the reference counter get in fget(pro->fd
Hi Tao,
On Sat, Nov 28, 2020 at 09:57:31AM +0800, Peng Tao wrote:
> On Fri, Nov 27, 2020 at 9:41 PM Alessio Balsini wrote:
> >
> > Hi Peng,
> >
> > Thanks for the heads up!
> >
> > On Thu, Nov 26, 2020 at 09:33:34PM +0800, Peng Tao wrote:
> > >
On Sat, Nov 28, 2020 at 10:10:37AM +0800, Peng Tao wrote:
> On Tue, Oct 27, 2020 at 1:00 AM Alessio Balsini wrote:
> >
> > This is the 10th version of the series. Please find the changelog at the
> > bottom of this cover letter.
> >
> > Add support for file syste
Hi Peng,
Thanks for the heads up!
On Thu, Nov 26, 2020 at 09:33:34PM +0800, Peng Tao wrote:
> On Tue, Oct 27, 2020 at 12:19 AM Alessio Balsini wrote:
> > [...]
> > int fuse_passthrough_setup(struct fuse_conn *fc, struct fuse_file *ff,
> >
-by: Alessio Balsini
---
fs/fuse/passthrough.c | 85 ++-
1 file changed, 76 insertions(+), 9 deletions(-)
diff --git a/fs/fuse/passthrough.c b/fs/fuse/passthrough.c
index 5a78cb336db4..10b6872cdaa7 100644
--- a/fs/fuse/passthrough.c
+++ b/fs/fuse
l the FUSE daemon responsible for enabling or not the
passthrough feature at file open time, and should enable the feature only
after appropriate access policy checks.
Signed-off-by: Alessio Balsini
---
fs/fuse/fuse_i.h | 5 -
fs/fuse/passthrough.c | 11 +++
2 files change
v2:
* Changed the feature name to passthrough from stacked_io.
[Proposed by Linus Torvalds]
Alessio Balsini (5):
fuse: Definitions and ioctl() for passthrough
fuse: Passthrough initialization and release
fuse: Introduce synchronous read and write for passthrough
fuse: Handle asynchron
onous operations, yet covering the majority
of the use cases.
Signed-off-by: Alessio Balsini
---
fs/fuse/file.c| 8 +++--
fs/fuse/fuse_i.h | 2 ++
fs/fuse/passthrough.c | 70 +++
3 files changed, 78 insertions(+), 2 deletions(-)
diff --git a/fs/fuse/f
, add the passthrough functions for the set-up and tear-down of the
data structures and locks that will be used both when fuse_conns and
fuse_files are created/deleted.
Signed-off-by: Alessio Balsini
---
fs/fuse/Makefile | 1 +
fs/fuse/dev.c | 40
system stacking
is allowed for those FUSE file systems using passthrough.
Signed-off-by: Alessio Balsini
---
fs/fuse/inode.c | 5 +++
fs/fuse/passthrough.c | 80 +--
2 files changed, 83 insertions(+), 2 deletions(-)
diff --git a/fs/fuse/inode.c b/f
On Wed, Sep 30, 2020 at 08:54:03PM +0200, Miklos Szeredi wrote:
> On Thu, Sep 24, 2020 at 3:13 PM Alessio Balsini wrote:
> >
> > Extend the passthrough feature by handling asynchronous IO both for read
> > and write operations.
> >
> > When an AIO request is r
On Wed, Sep 30, 2020 at 08:50:46PM +0200, Miklos Szeredi wrote:
> > [...]
> > +ssize_t fuse_passthrough_read_iter(struct kiocb *iocb_fuse,
> > + struct iov_iter *iter)
> > +{
> > + ssize_t ret;
> > + const struct cred *old_cred;
> > + struct file *
On Wed, Sep 30, 2020 at 03:16:20PM -0400, Antonio SJ Musumeci wrote:
> On 9/30/2020 2:45 PM, Miklos Szeredi wrote:
> > On Thu, Sep 24, 2020 at 3:13 PM Alessio Balsini wrote:
> > > Add a reference to the FUSE daemon credentials, so that they can be used
> > > to
>
On Wed, Sep 30, 2020 at 05:44:54PM +0200, Miklos Szeredi wrote:
> On Thu, Sep 24, 2020 at 3:13 PM Alessio Balsini wrote:
> >
> > Introduce the new FUSE passthrough ioctl(), which allows userspace to
> > specify a direct connection between a FUSE file and a lower file syst
Hi Miklos, all,
After being stuck with some strange and hard to reproduce results from my SSD,
I finally decided to overcome the biggest chunk of inconsistencies by
forgetting about the SSD and switching to a RAM block device to host my lower
file system.
Getting rid of the discrete storage devic
On Wed, Sep 30, 2020 at 05:33:30PM +0200, Miklos Szeredi wrote:
> On Thu, Sep 24, 2020 at 3:13 PM Alessio Balsini wrote:
>
> > The first benchmarks were done by running FIO (fio-3.21) with:
> > - bs=4Ki;
> > - file size: 50Gi;
> > - ioengine: sync;
> > - fsync
Hi,
I noticed the following fixup suggested by Amir slipped from this
submission.
Thanks,
Alessio
---8<---
diff --git a/fs/fuse/passthrough.c b/fs/fuse/passthrough.c
index b7d1a5517ffd..eba26196be92 100644
--- a/fs/fuse/passthrough.c
+++ b/fs/fuse/passthrough.c
@@ -185,7 +185,7 @@ int fuse_passt
On Tue, Sep 22, 2020 at 07:08:48PM +0300, Amir Goldstein wrote:
>
> I am hearing about a lot of these projects.
> I think that FUSE_PASSTHROUGH is a very useful feature.
> I have an intention to explore passthrough to directory fd for
> directory modifications. I sure hope you will beat me to it ;
passthrough is active, instead, the read/write operation is directly
applied to the lower file system, so to keep the same behavior as before,
the calling process temporarily receives the same credentials as the
daemon, that should be removed as soon as the operation completes.
Signed-off-by: Alessio
read_ and write_iter file operations.
This extra check avoids special pseudofiles to be targets for this feature.
An additional enforced limitation is that when FUSE passthrough is enabled,
no further file system stacking is allowed.
Signed-off-by: Alessio Balsini
---
fs/fuse/Makefile
by Mike Shal]
Changes in v3 and v4:
* Use the fs_stack_depth to prevent further stacking and a minor fix.
[Proposed by Jann Horn]
Changes in v2:
* Changed the feature name to passthrough from stacked_io.
[Proposed by Linus Torvalds]
Alessio Balsini (4):
fuse: Definitions and ioctl() f
-off-by: Alessio Balsini
---
fs/fuse/passthrough.c | 64 +--
1 file changed, 62 insertions(+), 2 deletions(-)
diff --git a/fs/fuse/passthrough.c b/fs/fuse/passthrough.c
index f70c0ef6945b..b7d1a5517ffd 100644
--- a/fs/fuse/passthrough.c
+++ b/fs/fuse
, and should enable the feature only after
appropriate access policy checks.
This change only implements synchronous requests in passthrough, returning
an error in the case of ansynchronous operations, yet covering the majority
of the use cases.
Signed-off-by: Alessio Balsini
---
fs/fuse/f
On Fri, Sep 18, 2020 at 10:59:16PM +0300, Amir Goldstein wrote:
> On Fri, Sep 18, 2020 at 7:33 PM Alessio Balsini wrote:
> [...]
> > > ... for example:
> > >
> > >if (fs_stack_depth && passthrough_sb->s_type == fuse_fs_type) {
> > >
On Fri, Sep 11, 2020 at 11:23:22AM -0600, Jens Axboe wrote:
> On 9/11/20 10:34 AM, Alessio Balsini wrote:
> > Extend the passthrough feature by handling asynchronous IO both for read
> > and write operations.
> > When an AIO request is received, targeting a FUSE
Hi Amir,
On Sat, Sep 12, 2020 at 12:55:35PM +0300, Amir Goldstein wrote:
> On Fri, Sep 11, 2020 at 7:34 PM Alessio Balsini wrote:
> > +ssize_t fuse_passthrough_read_iter(struct kiocb *iocb_fuse,
> > + struct iov_iter *iter)
> > +{
&g
Hi Amir,
Thanks again for your feedback.
On Sat, Sep 12, 2020 at 02:06:02PM +0300, Amir Goldstein wrote:
> On Fri, Sep 11, 2020 at 7:34 PM Alessio Balsini wrote:
> [...]
> > diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
> > index bba747520e9b..eb223130a917 100644
> &g
your feedback,
Alessio
On Fri, Sep 11, 2020 at 02:46:04PM -0400, Antonio SJ Musumeci wrote:
> On 9/11/2020 12:34 PM, Alessio Balsini via fuse-devel wrote:
> > Add support for file system passthrough read/write of files when enabled in
> > userspace through the option FUSE_PASSTHRO
completion
callback as well.
Signed-off-by: Alessio Balsini
---
fs/fuse/passthrough.c | 66 +--
1 file changed, 63 insertions(+), 3 deletions(-)
diff --git a/fs/fuse/passthrough.c b/fs/fuse/passthrough.c
index 44a78e02f45d..87b57b26fd8a 100644
--- a/fs/fuse
s_stack_depth to prevent further stacking and a minor fix.
[Proposed by Jann Horn]
Changes in v2:
* Changed the feature name to passthrough from stacked_io.
[Proposed by Linus Torvalds]
Alessio Balsini (3):
fuse: Definitions and ioctl() for passthrough
fuse: Introduce synchronous read and wr
write operation is completed, the file stats change is notified (and
propagated) to the lower file system.
This change only implements synchronous requests in passthrough, returning
an error in the case of ansynchronous operations, yet covering the majority
of the use cases.
Signed-off-by: Alessio Ba
read_ and write_iter file operations.
This extra check avoids special pseudofiles to be targets for this feature.
An additional enforced limitation is that when FUSE passthrough is enabled,
no further file system stacking is allowed.
Signed-off-by: Alessio Balsini
---
fs/fuse/Makefile
Thanks all for the comments.
I have a patchset ready that hopefully wraps together the extendability
suggested by Nikolaus, that I agree is a good idea.
The way I tried to make it more flexible is first of all transitioning to a
ioctl(), as suggested by both Jann and Miklos, and by using a data
st
Thanks Jann,
Sorry for the late reply, I wanted to better understand the problems you
mentioned and explore the ioctl solution before coming back to discussion.
I have a new patch ready which addresses the feedbacks received, and that
has been converted to using a new ioctl. And I have to say I'm
nd.
>
> I recommend looking at ovl_write_iter() in overlayfs to see how
> they're dealing with this case.
That was a great suggestion, I hopefully picked the right things from
overlayfs, without overlooking some security aspects.
Thanks again,
Alessio
---8<---
>From 5f0e16
his seems like your area of expertise.]
>
> On Wed, Aug 12, 2020 at 6:15 PM Alessio Balsini wrote:
> > Add support for filesystem passthrough read/write of files when enabled in
> > userspace through the option FUSE_PASSTHROUGH.
>
> Oh, I remember this old series... v5 was from 2
y: Nikhilesh Reddy
Signed-off-by: Alessio Balsini
--
Performance
What follows has been performed with this change rebased on top of a
vanilla v5.8 Linux kernel, using a custom passthrough_hp FUSE daemon that
enables pass-through for each file that is opened during both “open” and
“create”. Te
Ops, please forgive the wrong in-reply-to messge id :)
Cheers,
Alessio
On Wed, Oct 23, 2019 at 06:17:36PM +0100, Alessio Balsini wrote:
> [ Upstream commit fdbe4eeeb1aac219b14f10c0ed31ae5d1123e9b8 ]
>
> Enabling Direct I/O with loop devices helps reducing memory usage by
> avo
.
This patch fixes the compatibility issue mentioned above by exporting
LOOP_SET_DIRECT_IO as additional lo_compat_ioctl() entry.
The input argument for this ioctl is a single long converted to a 1-bit
boolean, so compatibility is preserved.
Cc: Jens Axboe
Signed-off-by: Alessio Balsini
Signed-off-by
ruct definition when navigating the code.
Signed-off-by: Alessio Balsini
Cc: Luis Chamberlain
Cc: Kees Cook
---
include/linux/sysctl.h | 6 ++
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 6df477329b76e..02fa84493f237 100644
Hi Peter,
While testing your series (peterz/sched/wip-deadline 7a9e91d3fe951), I ended up
in a panic at boot on a x86_64 kvm guest, would you please have a look? Here
attached the backtrace.
Happy to test any suggestion that fixes the issue.
Thanks,
Alessio
---
-->8--
[0.798326]
only looks at utilization and density, a very
large period can allow a very large runtime, which in turn can
incur a very large latency to lower priority tasks.
- for very short periods we can end up spending more time programming
the hardware timer than actually running the task.
Mitigate t
On Mon, Aug 05, 2019 at 01:53:09PM +0200, Peter Zijlstra wrote:
>
> Like so?
>
Yes, that's exactly what I meant!
What about this refactoring?
Thanks,
Alessio
---
>From 459d5488acb3fac938b0f35f480a81a6e401ef92 Mon Sep 17 00:00:00 2001
From: Alessio Balsini
Date: Thu, 22
Oops, you are totally right, I was still looking at the latest
mlx5_ib_create_qp_resp struct while backporting these patches :)
Sorry for that,
Alessio
On Mon, Aug 12, 2019 at 04:23:16PM +0200, Greg KH wrote:
> On Mon, Aug 12, 2019 at 11:55:03AM +0100, Alessio Balsini wrote:
> > Fr
-by: Jason Gunthorpe
Signed-off-by: Alessio Balsini
---
drivers/infiniband/hw/mlx5/qp.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index a7bc89f5dae7..89357d9e489d 100644
--- a/drivers/infiniband/hw/mlx5/
From: Jason Gunthorpe
mlx5_ib_create_qp_resp was never initialized and only the first 4 bytes
were written.
Fixes: 41d902cb7c32 ("RDMA/mlx5: Fix definition of mlx5_ib_create_qp_resp")
Cc:
Acked-by: Leon Romanovsky
Signed-off-by: Jason Gunthorpe
Signed-off-by: Alessio Balsini
--
From: Jason Gunthorpe
mlx5_ib_create_qp_resp was never initialized and only the first 4 bytes
were written.
Fixes: 41d902cb7c32 ("RDMA/mlx5: Fix definition of mlx5_ib_create_qp_resp")
Cc:
Acked-by: Leon Romanovsky
Signed-off-by: Jason Gunthorpe
Signed-off-by: Alessio Balsini
--
Oops, thanks!
On Thu, Aug 8, 2019 at 2:00 AM Greg KH wrote:
>
> On Thu, Aug 08, 2019 at 03:28:19AM +0100, Alessio Balsini wrote:
> > From: xiao jin
> >
> > commit 54648cf1ec2d7f4b6a71767799c45676a138ca24 upstream.
> >
> > We find the memory use-after-free iss
fail case of
blk_init_allocated_queue().
Fixes: commit 7c94e1c157a2 ("block: introduce blk_flush_queue to drive flush
machinery")
Cc:
Reviewed-by: Ming Lei
Reviewed-by: Bart Van Assche
Signed-off-by: xiao jin
Signed-off-by: Jens Axboe
Signed-off-by: Alessio Balsini
---
block/blk-core.c | 1 +
1 fil
Hi Peter,
This is indeed an important missing piece.
I think it would be worth having some simple checks, e.g.:
- period_min <= period_max;
- period_min >= (1ULL << DL_SCALE).
To be even more picky, I'm wondering if it would be a good practice to
fail the write operation if there are already dl-
d
Cc: sta...@vger.kernel.org
Signed-off-by: Richard Weinberger
Signed-off-by: Alessio Balsini
---
arch/um/os-Linux/file.c | 1 +
arch/um/os-Linux/signal.c | 2 ++
arch/x86/um/stub_segv.c | 1 +
3 files changed, 4 insertions(+)
diff --git a/arch/um/os-Linux/file.c b/arch/um/os-Linux/file.c
index 26
Hi Luca,
On Mon, May 06, 2019 at 06:48:31AM +0200, Luca Abeni wrote:
> From: luca abeni
>
> extern void dl_change_utilization(struct task_struct *p, u64 new_bw);
> @@ -785,6 +786,8 @@ struct root_domain {
> unsigned long max_cpu_capacity;
> unsigned long min_cpu
Fix two non-critical typos in the documentation of the dl_entity_overflow
function:
- "rather then" --> "rather than";
- "in such way that" --> "in such a way that".
Signed-off-by: Alessio Balsini
---
kernel/sched/deadline.c | 4 ++--
1 file change
Hello Patrick,
What do you think about the following minor changes:
On Tue, Jan 15, 2019 at 10:14:58AM +, Patrick Bellasi wrote:
> /* SCHED_ISO: reserved but not implemented yet */
> #define SCHED_IDLE 5
> #define SCHED_DEADLINE 6
> +/* Must be the last entry: used
: Juri Lelli
Cc: Tommaso Cucinotta
Cc: Luca Abeni
Cc: Claudio Scordino
Cc: Daniel Bristot de Oliveira
Cc: Patrick Bellasi
Cc: Ingo Molnar
Cc: Peter Zijlstra
Tested-by: Joel Fernandes (Google)
Reviewed-by: Joel Fernandes (Google)
Signed-off-by: Alessio Balsini
---
Having a precise means for
Hi,
On 19/12/17 16:16, Peter Zijlstra wrote:
[...]
This inaccuracy was introduced in 06a76fe08d4d, that moved the function
from core.c to deadline.c. Before that, it was making more sense to
access sched_priority, either if the function name __getparam_dl() was
misleading.
OK, I'm dense, what?
is out of scope and is
removed.
This inaccuracy was introduced in 06a76fe08d4d, that moved the function
from core.c to deadline.c. Before that, it was making more sense to
access sched_priority, either if the function name __getparam_dl() was
misleading.
Signed-off-by: Alessio Balsini
Reviewed-by:
Remove the adjective "relative" associated to the period of a periodic
task, since the period is an absolute value.
Rephrased the comment with the aim of making it even more clear.
Signed-off-by: Alessio Balsini
Cc: Ingo Molnar
Cc: Peter Zijlstra
Cc: Juri Lelli
Cc: Daniel Bristot d
-off-by: Luca Abeni
Cc: Tommaso Cucinotta
Cc: Juri Lelli
Cc: Daniel Bristot de Oliveira
Cc: Steven Rostedt
Cc: Ingo Molnar
Cc: Peter Zijlstra
Signed-off-by: Alessio Balsini
---
include/linux/sched.h| 13 +-
kernel/sched/autogroup.c | 4 +-
kernel/sched/core.c | 86
available bandwidth is the given runtime/period multiplied by M.
Signed-off-by: Andrea Parri
Signed-off-by: Luca Abeni
Cc: Tommaso Cucinotta
Cc: Juri Lelli
Cc: Daniel Bristot de Oliveira
Cc: Steven Rostedt
Cc: Ingo Molnar
Cc: Peter Zijlstra
Signed-off-by: Alessio Balsini
---
kernel/sched
n SCHED_DL).
The mechanism has been evaluated on several synthetic tests with RT groups,
containing RT tasks with different priorities and the groups bandwidths are
limited as expected. If anyone is interested in the specific tests, we can
share some code.
Alessio Balsini (3):
sched/r
Lelli
Cc: Daniel Bristot de Oliveira
Cc: Steven Rostedt
Cc: Ingo Molnar
Cc: Peter Zijlstra
Signed-off-by: Alessio Balsini
---
kernel/sched/core.c | 63 +---
kernel/sched/deadline.c | 27 --
kernel/sched/debug.c| 3 -
kernel/sched/rt.c | 752
95 matches
Mail list logo
