On 6/10/21 3:06 AM, Stefan Hajnoczi wrote:
> On Wed, Jun 09, 2021 at 04:03:55PM -0500, Mike Christie wrote:
>> On 6/7/21 10:19 AM, Stefan Hajnoczi wrote:
>>> My concern is that threads should probably accounted against
>>> RLIMIT_NPROC and max_threads rather than
On 6/23/21 11:34 PM, kernel test robot wrote:
>>> kernel/kthread.c:466:6: warning: function 'kthread_create_for_user' might
>>> be a candidate for 'gnu_printf' format attribute
>>> [-Wsuggest-attribute=format]
> 466 | namefmt, args);
> | ^~~
>
Will add a __printf(4,
On 6/24/21 3:26 AM, kernel test robot wrote:
>>> drivers/vhost/vhost.c:599:57: sparse: sparse: dereference of noderef
>>> expression
> vim +599 drivers/vhost/vhost.c
>
>581
>582/* Caller should have device mutex */
>583long vhost_dev_set_owner(struct vhost_dev
On 6/7/21 10:19 AM, Stefan Hajnoczi wrote:
> My concern is that threads should probably accounted against
> RLIMIT_NPROC and max_threads rather than something indirect like 128 *
> RLIMIT_NOFILE (a userspace process can only have RLIMIT_NOFILE
> vhost-user file descriptors open).
>
Ah ok, I see
On 5/5/21 3:38 PM, Steven Rostedt wrote:
> The new trace-cmd 3.0 (which is almost ready to be released) allows for
> tracing between host and guests with timestamp synchronization such that
> the events on the host and the guest can be interleaved in the proper order
> that they occur. KernelShark
On 6/3/21 5:13 AM, Stefan Hajnoczi wrote:
> On Tue, May 25, 2021 at 01:05:51PM -0500, Mike Christie wrote:
>> Results:
>>
>> When running with the null_blk driver and vhost-scsi I can get 1.2
>> million IOPs by just running a simple
>>
>> fio --f
The following patches apply over linus's tree and mst's vhost branch.
The patches are just some flush cleanups and a patch to reduce flush
calls and some coding style fixups.
My worker threading patches are built over these patches, but they seem
like patches that would be ok even if the
the
vhost_work_dev_flush call flushed.
Signed-off-by: Mike Christie
Reviewed-by: Stefan Hajnoczi
Acked-by: Jason Wang
---
drivers/vhost/scsi.c | 8
1 file changed, 8 deletions(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 051a7f8dadba..2f9633ef26aa 100644
--- a/drivers/vhost/scsi.c
vhost_scsi_flush will flush everything, so we can clear the backends then
flush, then destroy. We don't need to flush before each vq destruction
because after the flush we will have made sure there can be no new cmds
started and there are no running cmds.
Signed-off-by: Mike Christie
vhost_work_flush doesn't do anything with the work arg. This patch drops
it and then renames vhost_work_flush to vhost_work_dev_flush to reflect
that the function flushes all the works in the dev and not just a
specific queue or work item.
Signed-off-by: Mike Christie
Acked-by: Jason Wang
We use 3 coding styles in this struct. Switch to just tabs.
Signed-off-by: Mike Christie
Reviewed-by: Chaitanya Kulkarni
Reviewed-by: Stefan Hajnoczi
Acked-by: Jason Wang
---
drivers/vhost/vhost.h | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/vhost
Switch from a mix of tabs and spaces to just tabs.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index 575c8180caad..7d5306d1229d 100644
--- a/drivers/vhost/vhost.h
The following patches apply over linus's tree or mst's vhost branch
and my cleanup patchset:
https://lists.linuxfoundation.org/pipermail/virtualization/2021-May/054354.html
These patches allow us to support multiple vhost workers per device. I
ended up just doing Stefan's original idea where
to the drivers, so the next patch will
convert that code.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 136 --
drivers/vhost/vhost.h | 4 +-
2 files changed, 94 insertions(+), 46 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
This is just a prep patch. It moves the worker related fields to a new
vhost_worker struct and moves the code around to create some helpers that
will be used in the next patches.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 94 +--
drivers
This patch allows userspace to create workers and bind them to vqs, so you
can have N workers per dev and also share N workers with M vqs. The next
patch will allow sharing across devices.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c| 94
This allows vhost_polls to use the worker the vq the poll is associated
with.
This also exports a helper vhost_vq_work_queue which is used here
internally, and will be used in the vhost-scsi patches.
Signed-off-by: Mike Christie
---
drivers/vhost/net.c | 6 --
drivers/vhost/vhost.c | 19
here. If the
tools are using the newer ioctls that have already created the workers
then we also detect that here and do nothing.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 23 ---
1 file changed, 16 insertions(+), 7 deletions(-)
diff --git a/drivers/vhost/vhost.c
The next patch allows a vhost_worker to handle different devices. To
prepare for that, this patch adds a pointer to the device on the work so
we can get to the different mms in the vhost_worker thread.
Signed-off-by: Mike Christie
---
drivers/vhost/scsi.c | 7 ---
drivers/vhost/vhost.c
that have the same owner (dev->mm value)?
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 16 +++-
drivers/vhost/vhost.h | 1 -
2 files changed, 7 insertions(+), 10 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index eb16eb2bbee0..c32f72b1901c 100
completions running on different threads/CPUs.
Signed-off-by: Mike Christie
---
drivers/vhost/scsi.c | 48 +++-
1 file changed, 25 insertions(+), 23 deletions(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index e585f2180457..b607bff41074 100644
On 6/3/21 9:37 AM, Stefan Hajnoczi wrote:
> On Tue, May 25, 2021 at 01:05:51PM -0500, Mike Christie wrote:
>> The following patches apply over linus's tree or mst's vhost branch
>> and my cleanup patchset:
>>
>> https://lists.linuxfoundation.org/pipermail/virtuali
The vhost driver will create a kthread when userspace does a
VHOST_SET_OWNER ioctl, but the thread is charged to the kthreadd thread.
We can then end up violating the userspace process's RLIMIT_NPROC. This
patchset allows drivers to pass in the user to charge/check.
The patches were made over
a user_struct and pass it to
copy_process. The next patches will then convert the fork/cred code.
Signed-off-by: Mike Christie
---
include/linux/kthread.h| 5
include/linux/sched/task.h | 2 ++
kernel/kthread.c | 58 --
3 files changed, 56
This allows kthread to pass copy_process the user we want to check for the
RLIMIT_NPROC limit for and also charge for the new process. It will be used
by vhost where userspace has that driver create threads but the kthreadd
thread is checked/charged.
Signed-off-by: Mike Christie
---
include
This has vhost pass in the user to the kthread API, so the process doing
the ioctl has its RLIMIT_NPROC checked and its processes count
incremented.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/vhost
On 6/29/21 8:04 AM, Christian Brauner wrote:
> On Wed, Jun 23, 2021 at 10:08:03PM -0500, Mike Christie wrote:
>> This allows kthread to pass copy_process the user we want to check for the
>> RLIMIT_NPROC limit for and also charge for the new process. It will be used
>> by vhos
vhost_scsi_flush will flush everything, so we can clear the backends then
flush, then destroy. We don't need to flush before each vq destruction
because after the flush we will have made sure there can be no new cmds
started and there are no running cmds.
Signed-off-by: Mike Christie
We use like 3 coding styles in this struct. Switch to just tabs.
Signed-off-by: Mike Christie
Reviewed-by: Chaitanya Kulkarni
Reviewed-by: Stefan Hajnoczi
Acked-by: Jason Wang
---
drivers/vhost/vhost.h | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers
This is just a prep patch. It moves the worker related fields to a new
vhost_worker struct and moves the code around to create some helpers that
will be used in the next patches.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 94 +--
drivers
here. If the
tools are using the newer ioctls that have already created the workers
then we also detect that here and do nothing.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 23 ---
1 file changed, 16 insertions(+), 7 deletions(-)
diff --git a/drivers/vhost/vhost.c
completions running on different threads/CPUs.
Signed-off-by: Mike Christie
---
drivers/vhost/scsi.c | 48 +++-
1 file changed, 25 insertions(+), 23 deletions(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 96462032a972..2f84cf602ab3 100644
to the driver, so the next patch will
convert that code.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 136 --
drivers/vhost/vhost.h | 4 +-
2 files changed, 94 insertions(+), 46 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
This allows vhost_polls to use the worker the vq the poll is associated
with.
This also exports a helper vhost_vq_work_queue which is used here
internally, and will be used in the vhost-scsi patches.
Signed-off-by: Mike Christie
---
drivers/vhost/net.c | 6 --
drivers/vhost/vhost.c | 19
This allows a worker to handle multiple device's vqs.
TODO:
- How to handle if the devices are in different cgroups/VMs.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 18 +++---
drivers/vhost/vhost.h | 1 -
2 files changed, 7 insertions(+), 12 deletions(-)
diff --git
their responses when the tmf's work is run.
So this patch has vhost-scsi flush the IO vqs on other worker threads
before we send the tmf response.
Signed-off-by: Mike Christie
---
drivers/vhost/scsi.c | 17 ++---
drivers/vhost/vhost.c | 6 ++
drivers/vhost/vhost.h | 1 +
3 files
This patch allows userspace to create workers and bind them to vqs, so you
can have N workers per dev and also share N workers with M vqs. The next
patch will allow sharing across devices.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c| 95
Add an ioctl cmd to allow userspace to figure out the vq's worker.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 10 ++
include/uapi/linux/vhost.h | 3 ++-
2 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index
The next patch allows a vhost_worker to handle different devices. To
prepare for that, this patch adds a pointer to the device on the work so
we can get to the different mms in the vhost_worker thread.
Signed-off-by: Mike Christie
---
drivers/vhost/scsi.c | 7 ---
drivers/vhost/vhost.c
The following patches apply over mst's vhost branch and were tested
againt that branch and also mkp's 5.13 branch which has some vhost-scsi
changes.
These patches allow us to support multiple vhost workers per device. I
ended up just doing Stefan's original idea where userspace has the
kernel
vhost_work_flush doesn't do anything with the work arg. This patch drops
it and then renames vhost_work_flush to vhost_work_dev_flush to reflect
that the function flushes all the works in the dev and not just a
specific queue or work item.
Signed-off-by: Mike Christie
Acked-by: Jason Wang
the
vhost_work_dev_flush call flushed.
Signed-off-by: Mike Christie
Reviewed-by: Stefan Hajnoczi
Acked-by: Jason Wang
---
drivers/vhost/scsi.c | 8
1 file changed, 8 deletions(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 0fd596da1834..b3e6fe9b1767 100644
--- a/drivers/vhost/scsi.c
ng old kernel by just checking for EPERM. Does this require
a feature?
Signed-off-by: Mike Christie
---
hw/scsi/vhost-scsi.c | 85 +++-
hw/virtio/vhost-backend.c| 8 ++
include/hw/virtio/vhost-backend.h| 4 +
include/hw/virtio/
On 5/4/21 10:56 AM, Stefano Garzarella wrote:
> Hi Mike,
>
> On Wed, Apr 28, 2021 at 05:36:59PM -0500, Mike Christie wrote:
>> The following patches apply over mst's vhost branch and were tested
>> againt that branch and also mkp's 5.13 branch which has some
On 5/4/21 10:33 AM, Stefano Garzarella wrote:
> On Wed, Apr 28, 2021 at 05:37:12PM -0500, Mike Christie wrote:
>> The next patch allows a vhost_worker to handle different devices. To
>> prepare for that, this patch adds a pointer to the device on the work so
>> we can get
On 5/4/21 10:30 AM, Stefano Garzarella wrote:
> On Wed, Apr 28, 2021 at 05:37:11PM -0500, Mike Christie wrote:
>> This patch allows userspace to create workers and bind them to vqs, so you
>> can have N workers per dev and also share N workers with M vqs. The next
>> pat
I just realized I forgot to cc the virt list so adding now.
Christian see the very bottom for a different fork patch.
On 7/12/21 7:05 AM, Stefan Hajnoczi wrote:
> On Fri, Jul 09, 2021 at 11:25:37AM -0500, Mike Christie wrote:
>> Hi,
>>
>> The goal of this email is to try a
The following patches were made over linus's tree and also apply over
Jens's 5.15 io_ring branch and Michael's vhost branch.
The patchset allows the vhost layer to do a copy_process on the thread
that does the VHOST_SET_OWNER ioctl like how io_uring does a copy_process
against its userspace app
The kthread API creates threads that ignore all signals by default so
modules like vhost that will move from that API to kernel_copy_process
will not be expecting them. This patch exports ignore_signals so those
modules can keep their existing behavior.
Signed-off-by: Mike Christie
---
kernel
and io_uring in later patches in this set.
This patch also exports __set_task_comm and wake_up_new_task which is
needed by modules to use the new helper. io_uring calls these functions
already but its always built into the kernel so was not needed before.
Signed-off-by: Mike Christie
---
fs
option so these worker threads can prevent
taking an extra refcount on themselves.
Signed-off-by: Mike Christie
---
include/linux/sched/task.h | 3 ++-
kernel/fork.c | 14 +++---
2 files changed, 13 insertions(+), 4 deletions(-)
diff --git a/include/linux/sched/task.h b/include/li
Convert io_uring and io-wq to use kernel_copy_process.
Signed-off-by: Mike Christie
---
fs/io-wq.c | 9 +++--
fs/io_uring.c | 5 -
include/linux/sched/task.h | 1 -
kernel/fork.c | 22 --
4 files changed, 11 insertions
that owns the device.
Note this patch converts us. The next patch will remove the code that is
no longer needed.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 49 +++
drivers/vhost/vhost.h | 7 ++-
2 files changed, 37 insertions(+), 19
This is just a prep patch. It moves the worker related fields to a new
vhost_worker struct and moves the code around to create some helpers that
will be used in the next patches.
Signed-off-by: Mike Christie
Reviewed-by: Stefan Hajnoczi
---
drivers/vhost/vhost.c | 98
The vhost worker threads need the same frame setup as io_uring's worker
threads, but do not handle signals and may not even be executing IO. This
moves the setup part to a new flag.
Signed-off-by: Mike Christie
---
arch/x86/kernel/process.c | 4 ++--
include/linux/sched.h | 1 +
include
We inherit v1 and v2 cgroups from copy_process now, so we can drop the v1
only code.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 26 --
1 file changed, 26 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 6e58417b13fc
This patchset adds a couple new options to kernel_clone_args for IO thread
like/related users. Instead of adding new fields to kernel_clone_args for
each option, this moves us to a flags based approach by first converting
io_thread.
Signed-off-by: Mike Christie
---
include/linux/sched/task.h
that function can see.
Signed-off-by: Mike Christie
---
arch/alpha/kernel/process.c | 2 +-
arch/arc/kernel/process.c| 2 +-
arch/arm/kernel/process.c| 3 ++-
arch/arm64/kernel/process.c | 3 ++-
arch/csky/kernel/process.c | 3 ++-
arch/h8300/kernel/process.c | 3
reads can prevent
taking an extra refcount on themselves.
Signed-off-by: Mike Christie
---
include/linux/sched/task.h | 1 +
kernel/fork.c | 11 +--
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/include/linux/sched/task.h b/include/linux/sched/tas
and io_uring in later patches in this set.
Signed-off-by: Mike Christie
---
include/linux/sched/task.h | 6 -
kernel/fork.c | 48 ++
2 files changed, 53 insertions(+), 1 deletion(-)
diff --git a/include/linux/sched/task.h b/include/linux
The vhost worker threads need the same frame setup as io_uring's worker
threads, but handle signals differently and do not need the same
scheduling behavior. This patch separate's the frame setup parts of
PF_IO_WORKER into a kernel_clone_args flag, KERN_WORKER_USER.
Signed-off-by: Mike Christie
The kthread API creates threads that ignore all signals by default so
modules like vhost that will move from that API to kernel_worker will
not be expecting them. This patch adds a worker flag that tells
kernel_worker to setup the task to ignore signals.
Signed-off-by: Mike Christie
---
include
Convert io_uring and io-wq to use kernel_worker.
Signed-off-by: Mike Christie
---
fs/io-wq.c | 15 ---
fs/io_uring.c | 11 +--
include/linux/sched/task.h | 1 -
3 files changed, 13 insertions(+), 14 deletions(-)
diff --git a/fs/io-wq.c b/fs/io
that owns the device.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 68 ---
drivers/vhost/vhost.h | 7 -
2 files changed, 31 insertions(+), 44 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index c9a1f706989c
This is just a prep patch. It moves the worker related fields to a new
vhost_worker struct and moves the code around to create some helpers that
will be used in the next patches.
Signed-off-by: Mike Christie
Reviewed-by: Stefan Hajnoczi
---
drivers/vhost/vhost.c | 98
The following patches were made over Linus's tree but also apply over
Jens's 5.16 io_uring branch and Michaels' vhost/next branch.
This is version 2 of the patchset and should handle all the review
comments posted in V1 here:
The vhost worker threads need the same frame setup as io_uring's worker
threads, but handle signals differently and do not need the same
scheduling behavior. This patch separate's the frame setup parts of
PF_IO_WORKER into a new PF flag PF_USER_WORKER.
Signed-off-by: Mike Christie
---
arch
that owns the device.
Signed-off-by: Mike Christie
Acked-by: Michael S. Tsirkin
---
drivers/vhost/vhost.c | 65 +++
drivers/vhost/vhost.h | 7 -
2 files changed, 28 insertions(+), 44 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost
This is just a prep patch. It moves the worker related fields to a new
vhost_worker struct and moves the code around to create some helpers that
will be used in the next patches.
Signed-off-by: Mike Christie
Reviewed-by: Stefan Hajnoczi
Acked-by: Michael S. Tsirkin
---
drivers/vhost/vhost.c
The following patches apply over Linus's, Jens's or mst's trees, but were
made over linux-next because patch 6:
io_uring: switch to kernel_worker
has merge conflicts with Jens Axboe's for-next branch and Paul Moore's
selinux tree's next branch.
This is V4 of the patchset. It should handle all
This patchset adds a couple new options to kernel_clone_args for IO thread
like/related users. Instead of adding new fields to kernel_clone_args for
each option, this moves us to a flags based approach by first converting
io_thread.
Signed-off-by: Mike Christie
Suggested-by: Christian Brauner
reads can prevent
taking an extra refcount on themselves.
Signed-off-by: Mike Christie
Acked-by: Christian Brauner
---
include/linux/sched/task.h | 1 +
kernel/fork.c | 11 +--
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/include/linux/sched/task.h b/incl
and io_uring in later patches in this set.
[added flag validation code from Christian Brauner's patch]
Signed-off-by: Mike Christie
Acked-by: Christian Brauner
---
include/linux/sched/task.h | 4 +++
kernel/fork.c | 71 ++
2 files changed, 75
py_signals() to ease the
transition from kthreads to this new api.
Signed-off-by: Christian Brauner
Signed-off-by: Mike Christie
---
include/linux/sched/task.h | 1 +
kernel/fork.c | 3 +++
2 files changed, 4 insertions(+)
diff --git a/include/linux/sched/task.h b/include/linux/sc
Convert io_uring and io-wq to use kernel_worker.
Signed-off-by: Mike Christie
Reviewed-by: Christian Brauner
Reviewed-by: Jens Axboe
---
Note: To avoid patch application conflicts this patch was made over
linux-next which has Jens Axboe's block tree's for-next branch and Paul
Moore's selinux
On 10/5/21 7:50 AM, Christian Brauner wrote:
> On Mon, Oct 04, 2021 at 02:21:24PM -0500, Mike Christie wrote:
>> The vhost layer has similar requirements as io_uring where its worker
>> threads need to access the userspace thread's memory, want to inherit the
>> parents's c
This patchset adds a couple new options to kernel_clone_args for IO thread
like/related users. Instead of adding new fields to kernel_clone_args for
each option, this moves us to a flags based approach by first converting
io_thread.
Signed-off-by: Mike Christie
Suggested-by: Christian Brauner
The following patches were made over Linus's tree but also apply over
Jens's for-next io_uring branch and Michaels' vhost/next branch.
This is V3 of the patchset. It should handle all the review comments
posted in V1 and V2. If I missed a comment, please let me know.
This patchset allows the
that function can see.
Signed-off-by: Mike Christie
Acked-by: Christian Brauner
Acked-by: Geert Uytterhoeven
---
arch/alpha/kernel/process.c | 2 +-
arch/arc/kernel/process.c| 2 +-
arch/arm/kernel/process.c| 3 ++-
arch/arm64/kernel/process.c | 3 ++-
arch/csky/kernel/process.c
reads can prevent
taking an extra refcount on themselves.
Signed-off-by: Mike Christie
Acked-by: Christian Brauner
---
include/linux/sched/task.h | 1 +
kernel/fork.c | 11 +--
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/include/linux/sched/task.h b/incl
and io_uring in later patches in this set.
Signed-off-by: Mike Christie
Acked-by: Christian Brauner
---
include/linux/sched/task.h | 6 -
kernel/fork.c | 48 ++
2 files changed, 53 insertions(+), 1 deletion(-)
diff --git a/include/linux
This is just a prep patch. It moves the worker related fields to a new
vhost_worker struct and moves the code around to create some helpers that
will be used in the next patches.
Signed-off-by: Mike Christie
Reviewed-by: Stefan Hajnoczi
Acked-by: Michael S. Tsirkin
---
drivers/vhost/vhost.c
The vhost worker threads need the same frame setup as io_uring's worker
threads, but handle signals differently and do not need the same
scheduling behavior. This patch separate's the frame setup parts of
PF_IO_WORKER into a kernel_clone_args flag, KERN_WORKER_USER.
Signed-off-by: Mike Christie
The kthread API creates threads that ignore all signals by default so
modules like vhost that will move from that API to kernel_worker will
not be expecting them. This patch adds a worker flag that tells
kernel_worker to setup the task to ignore signals.
Signed-off-by: Mike Christie
Acked
that owns the device.
Signed-off-by: Mike Christie
Acked-by: Michael S. Tsirkin
---
drivers/vhost/vhost.c | 68 ---
drivers/vhost/vhost.h | 7 -
2 files changed, 31 insertions(+), 44 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost
Convert io_uring and io-wq to use kernel_worker.
Signed-off-by: Mike Christie
Reviewed-by: Christian Brauner
---
fs/io-wq.c | 15 ---
fs/io_uring.c | 11 +--
include/linux/sched/task.h | 1 -
3 files changed, 13 insertions(+), 14 deletions
The following patches apply over linus's tree and this patchset
https://lore.kernel.org/all/20211007214448.6282-1-michael.chris...@oracle.com/
which allows us to check the vhost owner thread's RLIMITs:
It looks like that patchset has been ok'd by all the major parties
and just needs a small
ODO:
- Better support for sharing workers where we bind based on ISR to vq
mapping.
Signed-off-by: Mike Christie
---
hw/scsi/vhost-scsi.c | 65
hw/virtio/vhost-backend.c| 8 +++
include/hw/virtio/vhost-backend.h| 4 ++
incl
This adds a helper to check if a vq has work pending and converts
vhost-net to use it.
Signed-off-by: Mike Christie
---
drivers/vhost/net.c | 2 +-
drivers/vhost/vhost.c | 6 +++---
drivers/vhost/vhost.h | 2 +-
3 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/vhost
This has the drivers pass in their poll to vq mapping and then converts
the core poll code to use the vq based helpers.
Signed-off-by: Mike Christie
---
drivers/vhost/net.c | 6 --
drivers/vhost/vhost.c | 10 ++
drivers/vhost/vhost.h | 4 +++-
3 files changed, 13 insertions
This patchset allows userspace to map vqs to different workers. This
patch adds a worker pointer to the vq so we can store that info.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 24 +---
drivers/vhost/vhost.h | 1 +
2 files changed, 14 insertions(+), 11
completions running on different threads/CPUs.
Signed-off-by: Mike Christie
Reviewed-by: Stefan Hajnoczi
---
drivers/vhost/scsi.c | 48 +++-
1 file changed, 25 insertions(+), 23 deletions(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index
Convert from vhost dev based helpers to vq ones.
Signed-off-by: Mike Christie
---
drivers/vhost/vsock.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c
index 938aefbc75ec..c50c60d0955e 100644
--- a/drivers/vhost/vsock.c
in the next patches.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 30 +-
drivers/vhost/vhost.h | 1 +
2 files changed, 22 insertions(+), 9 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index f2c6981297d0..e4d765f6c821 100644
--- a/drivers
in the next patches.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 44 +++
drivers/vhost/vhost.h | 1 +
2 files changed, 29 insertions(+), 16 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index f6bfa1c481df..f2c6981297d0
vhost_work_queue and vhost_work_dev_flush are no longer used, so drop
them.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c | 12
drivers/vhost/vhost.h | 2 --
2 files changed, 14 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index c84024afefff
Convert from vhost dev based helpers to vq ones.
Signed-off-by: Mike Christie
---
drivers/vhost/scsi.c | 27 +--
1 file changed, 13 insertions(+), 14 deletions(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 0d85ddb68420..08beba73ada4 100644
vqs before sending the TMF
response.
Signed-off-by: Mike Christie
---
drivers/vhost/scsi.c | 22 +++---
1 file changed, 19 insertions(+), 3 deletions(-)
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 08beba73ada4..29d9adcdb4fc 100644
--- a/drivers/vhost/scsi.c
This patch allows userspace to create workers and bind them to vqs. You
can have N workers per dev and also share N workers with M vqs.
Signed-off-by: Mike Christie
---
drivers/vhost/vhost.c| 99
drivers/vhost/vhost.h| 2 +-
include/uapi
py_signals() to ease the
transition from kthreads to this new api.
Signed-off-by: Christian Brauner
Signed-off-by: Mike Christie
Reviewed-by: Christoph Hellwig
---
include/linux/sched/task.h | 1 +
kernel/fork.c | 3 +++
2 files changed, 4 insertions(+)
diff --git a/include/linux/sc
1 - 100 of 407 matches
Mail list logo
