patches for DPDK stable branches
 help / color / mirror / Atom feed
* [PATCH] common/mlx5: adjust fork call with the new kernel API
@ 2023-05-24 12:01 erezf
  2023-05-24 14:50 ` Stephen Hemminger
                   ` (2 more replies)
  0 siblings, 3 replies; 7+ messages in thread
From: erezf @ 2023-05-24 12:01 UTC (permalink / raw)
  To: dev; +Cc: viacheslavo, matan, rasland, Erez Ferber, stable

From: Erez Ferber <erezf@nvidia.com>

While doing process fork() the operating system remaps all the parent
process's memory to the address space of the child process and activates
the Copy-on-Write mechanics - it duplicates physical pages once memory
writing happens in the child process. Sometimes memory duplication is
not allowed - for example, if the page contains hardware queue
descriptors. To handle similar issues the rdma-core library should be
prepared for forking.

The ibv_fork_init() prepares the library to track all the related memory
and prevent it from forking using madvise() system API. This approach
allows fork, but not all the memory is forked to the child process and,
application should care not to touch pages where the parent application
allocated the rdma-core objects.

The newer kernels propose an option of copy-on-fork for DMA pages and
tracking all the memory and disabling it for the forking is no longer
needed. The new API routine ibv_is_fork_initialized() should be involved
to decide if library initialization for forking is required.

Fixes: 0e83b8e536 ("net/mlx5: move rdma-core calls to separate file")
Cc: stable@dpdk.org
Signed-off-by: Erez Ferber <erezf@nvidia.com>
---
 drivers/common/mlx5/linux/meson.build | 2 ++
 drivers/common/mlx5/linux/mlx5_glue.c | 4 ++++
 2 files changed, 6 insertions(+)

diff --git a/drivers/common/mlx5/linux/meson.build b/drivers/common/mlx5/linux/meson.build
index 96a6c6c9be..c1fcb36fdd 100644
--- a/drivers/common/mlx5/linux/meson.build
+++ b/drivers/common/mlx5/linux/meson.build
@@ -219,6 +219,8 @@ has_sym_args = [
             'ibv_import_device' ],
         [ 'HAVE_MLX5DV_DR_ACTION_CREATE_DEST_ROOT_TABLE', 'infiniband/mlx5dv.h',
             'mlx5dv_dr_action_create_dest_root_table' ],
+	[ 'HAVE_IBV_FORK_UNNEEDED', 'infiniband/verbs.h',
+		'ibv_is_fork_initialized'],
 ]
 if  libmtcr_ul_found
     has_sym_args += [
diff --git a/drivers/common/mlx5/linux/mlx5_glue.c b/drivers/common/mlx5/linux/mlx5_glue.c
index 702eb36b62..88b99fe029 100644
--- a/drivers/common/mlx5/linux/mlx5_glue.c
+++ b/drivers/common/mlx5/linux/mlx5_glue.c
@@ -19,6 +19,10 @@
 static int
 mlx5_glue_fork_init(void)
 {
+#ifdef HAVE_IBV_FORK_UNNEEDED
+	if (ibv_is_fork_initialized() == IBV_FORK_UNNEEDED)
+		return 0; /* ibv_fork_init() not needed */
+#endif
 	return ibv_fork_init();
 }
 
-- 
2.14.1


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH] common/mlx5: adjust fork call with the new kernel API
  2023-05-24 12:01 [PATCH] common/mlx5: adjust fork call with the new kernel API erezf
@ 2023-05-24 14:50 ` Stephen Hemminger
  2023-05-25  8:10   ` Slava Ovsiienko
  2023-05-24 16:05 ` Slava Ovsiienko
  2023-06-22 12:19 ` Raslan Darawsheh
  2 siblings, 1 reply; 7+ messages in thread
From: Stephen Hemminger @ 2023-05-24 14:50 UTC (permalink / raw)
  To: erezf; +Cc: dev, viacheslavo, matan, rasland, stable

On Wed, 24 May 2023 15:01:40 +0300
<erezf@nvidia.com> wrote:

> From: Erez Ferber <erezf@nvidia.com>
> 
> While doing process fork() the operating system remaps all the parent
> process's memory to the address space of the child process and activates
> the Copy-on-Write mechanics - it duplicates physical pages once memory
> writing happens in the child process. Sometimes memory duplication is
> not allowed - for example, if the page contains hardware queue
> descriptors. To handle similar issues the rdma-core library should be
> prepared for forking.
> 
> The ibv_fork_init() prepares the library to track all the related memory
> and prevent it from forking using madvise() system API. This approach
> allows fork, but not all the memory is forked to the child process and,
> application should care not to touch pages where the parent application
> allocated the rdma-core objects.
> 
> The newer kernels propose an option of copy-on-fork for DMA pages and
> tracking all the memory and disabling it for the forking is no longer
> needed. The new API routine ibv_is_fork_initialized() should be involved
> to decide if library initialization for forking is required.
> 
> Fixes: 0e83b8e536 ("net/mlx5: move rdma-core calls to separate file")
> Cc: stable@dpdk.org
> Signed-off-by: Erez Ferber <erezf@nvidia.com>

I don't think DPDK applications should fork(), and lots other
parts of the shared huge pages will break if an application does this.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* RE: [PATCH] common/mlx5: adjust fork call with the new kernel API
  2023-05-24 12:01 [PATCH] common/mlx5: adjust fork call with the new kernel API erezf
  2023-05-24 14:50 ` Stephen Hemminger
@ 2023-05-24 16:05 ` Slava Ovsiienko
  2023-06-22 12:19 ` Raslan Darawsheh
  2 siblings, 0 replies; 7+ messages in thread
From: Slava Ovsiienko @ 2023-05-24 16:05 UTC (permalink / raw)
  To: Erez Ferber, dev; +Cc: Matan Azrad, Raslan Darawsheh, stable

> -----Original Message-----
> From: Erez Ferber <erezf@nvidia.com>
> Sent: Wednesday, May 24, 2023 3:02 PM
> To: dev@dpdk.org
> Cc: Slava Ovsiienko <viacheslavo@nvidia.com>; Matan Azrad
> <matan@nvidia.com>; Raslan Darawsheh <rasland@nvidia.com>; Erez Ferber
> <erezf@nvidia.com>; stable@dpdk.org
> Subject: [PATCH] common/mlx5: adjust fork call with the new kernel API
> 
> From: Erez Ferber <erezf@nvidia.com>
> 
> While doing process fork() the operating system remaps all the parent
> process's memory to the address space of the child process and activates the
> Copy-on-Write mechanics - it duplicates physical pages once memory writing
> happens in the child process. Sometimes memory duplication is not allowed -
> for example, if the page contains hardware queue descriptors. To handle
> similar issues the rdma-core library should be prepared for forking.
> 
> The ibv_fork_init() prepares the library to track all the related memory and
> prevent it from forking using madvise() system API. This approach allows fork,
> but not all the memory is forked to the child process and, application should
> care not to touch pages where the parent application allocated the rdma-core
> objects.
> 
> The newer kernels propose an option of copy-on-fork for DMA pages and
> tracking all the memory and disabling it for the forking is no longer needed.
> The new API routine ibv_is_fork_initialized() should be involved to decide if
> library initialization for forking is required.
> 
> Fixes: 0e83b8e536 ("net/mlx5: move rdma-core calls to separate file")
> Cc: stable@dpdk.org
> Signed-off-by: Erez Ferber <erezf@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>



^ permalink raw reply	[flat|nested] 7+ messages in thread

* RE: [PATCH] common/mlx5: adjust fork call with the new kernel API
  2023-05-24 14:50 ` Stephen Hemminger
@ 2023-05-25  8:10   ` Slava Ovsiienko
  2023-05-25 15:27     ` Stephen Hemminger
  0 siblings, 1 reply; 7+ messages in thread
From: Slava Ovsiienko @ 2023-05-25  8:10 UTC (permalink / raw)
  To: Stephen Hemminger, Erez Ferber; +Cc: dev, Matan Azrad, Raslan Darawsheh, stable


> -----Original Message-----
> From: Stephen Hemminger <stephen@networkplumber.org>
> Sent: Wednesday, May 24, 2023 5:50 PM
> To: Erez Ferber <erezf@nvidia.com>
> Cc: dev@dpdk.org; Slava Ovsiienko <viacheslavo@nvidia.com>; Matan Azrad
> <matan@nvidia.com>; Raslan Darawsheh <rasland@nvidia.com>;
> stable@dpdk.org
> Subject: Re: [PATCH] common/mlx5: adjust fork call with the new kernel API
> 
> On Wed, 24 May 2023 15:01:40 +0300
> <erezf@nvidia.com> wrote:
> 
> > From: Erez Ferber <erezf@nvidia.com>
> >
> > While doing process fork() the operating system remaps all the parent
> > process's memory to the address space of the child process and
> > activates the Copy-on-Write mechanics - it duplicates physical pages
> > once memory writing happens in the child process. Sometimes memory
> > duplication is not allowed - for example, if the page contains
> > hardware queue descriptors. To handle similar issues the rdma-core
> > library should be prepared for forking.
> >
> > The ibv_fork_init() prepares the library to track all the related
> > memory and prevent it from forking using madvise() system API. This
> > approach allows fork, but not all the memory is forked to the child
> > process and, application should care not to touch pages where the
> > parent application allocated the rdma-core objects.
> >
> > The newer kernels propose an option of copy-on-fork for DMA pages and
> > tracking all the memory and disabling it for the forking is no longer
> > needed. The new API routine ibv_is_fork_initialized() should be
> > involved to decide if library initialization for forking is required.
> >
> > Fixes: 0e83b8e536 ("net/mlx5: move rdma-core calls to separate file")
> > Cc: stable@dpdk.org
> > Signed-off-by: Erez Ferber <erezf@nvidia.com>
> 
Hi,

> I don't think DPDK applications should fork(), and lots other parts of the
> shared huge pages will break if an application does this.

I agree - application should not, we have the secondary/primary processes approach.
Nonetheless, we have the real use case - DPDK application does fork() and works well.
Without mlx5 PMD 😊. With mlx5 it ran into some troubles. Now we have the solution.

With best regards,
Slava

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH] common/mlx5: adjust fork call with the new kernel API
  2023-05-25  8:10   ` Slava Ovsiienko
@ 2023-05-25 15:27     ` Stephen Hemminger
  2023-05-26  8:05       ` Slava Ovsiienko
  0 siblings, 1 reply; 7+ messages in thread
From: Stephen Hemminger @ 2023-05-25 15:27 UTC (permalink / raw)
  To: Slava Ovsiienko; +Cc: Erez Ferber, dev, Matan Azrad, Raslan Darawsheh, stable

On Thu, 25 May 2023 08:10:03 +0000
Slava Ovsiienko <viacheslavo@nvidia.com> wrote:

> > -----Original Message-----
> > From: Stephen Hemminger <stephen@networkplumber.org>
> > Sent: Wednesday, May 24, 2023 5:50 PM
> > To: Erez Ferber <erezf@nvidia.com>
> > Cc: dev@dpdk.org; Slava Ovsiienko <viacheslavo@nvidia.com>; Matan Azrad
> > <matan@nvidia.com>; Raslan Darawsheh <rasland@nvidia.com>;
> > stable@dpdk.org
> > Subject: Re: [PATCH] common/mlx5: adjust fork call with the new kernel API
> > 
> > On Wed, 24 May 2023 15:01:40 +0300
> > <erezf@nvidia.com> wrote:
> >   
> > > From: Erez Ferber <erezf@nvidia.com>
> > >
> > > While doing process fork() the operating system remaps all the parent
> > > process's memory to the address space of the child process and
> > > activates the Copy-on-Write mechanics - it duplicates physical pages
> > > once memory writing happens in the child process. Sometimes memory
> > > duplication is not allowed - for example, if the page contains
> > > hardware queue descriptors. To handle similar issues the rdma-core
> > > library should be prepared for forking.
> > >
> > > The ibv_fork_init() prepares the library to track all the related
> > > memory and prevent it from forking using madvise() system API. This
> > > approach allows fork, but not all the memory is forked to the child
> > > process and, application should care not to touch pages where the
> > > parent application allocated the rdma-core objects.
> > >
> > > The newer kernels propose an option of copy-on-fork for DMA pages and
> > > tracking all the memory and disabling it for the forking is no longer
> > > needed. The new API routine ibv_is_fork_initialized() should be
> > > involved to decide if library initialization for forking is required.
> > >
> > > Fixes: 0e83b8e536 ("net/mlx5: move rdma-core calls to separate file")
> > > Cc: stable@dpdk.org
> > > Signed-off-by: Erez Ferber <erezf@nvidia.com>  
> >   
> Hi,
> 
> > I don't think DPDK applications should fork(), and lots other parts of the
> > shared huge pages will break if an application does this.  
> 
> I agree - application should not, we have the secondary/primary processes approach.
> Nonetheless, we have the real use case - DPDK application does fork() and works well.
> Without mlx5 PMD 😊. With mlx5 it ran into some troubles. Now we have the solution.

The problem is you are allowing fork(). And many other libraries may break.
Imagine a DPDK library which has some local mutex and hugepages.
If two forked processes use it then the locks won't work and hugepage data
will be corrupted.


^ permalink raw reply	[flat|nested] 7+ messages in thread

* RE: [PATCH] common/mlx5: adjust fork call with the new kernel API
  2023-05-25 15:27     ` Stephen Hemminger
@ 2023-05-26  8:05       ` Slava Ovsiienko
  0 siblings, 0 replies; 7+ messages in thread
From: Slava Ovsiienko @ 2023-05-26  8:05 UTC (permalink / raw)
  To: Stephen Hemminger; +Cc: Erez Ferber, dev, Matan Azrad, Raslan Darawsheh, stable

> -----Original Message-----
> From: Stephen Hemminger <stephen@networkplumber.org>
> Sent: Thursday, May 25, 2023 6:28 PM
> To: Slava Ovsiienko <viacheslavo@nvidia.com>
> Cc: Erez Ferber <erezf@nvidia.com>; dev@dpdk.org; Matan Azrad
> <matan@nvidia.com>; Raslan Darawsheh <rasland@nvidia.com>;
> stable@dpdk.org
> Subject: Re: [PATCH] common/mlx5: adjust fork call with the new kernel API
> 
> On Thu, 25 May 2023 08:10:03 +0000
> Slava Ovsiienko <viacheslavo@nvidia.com> wrote:
> 
> > > -----Original Message-----
> > > From: Stephen Hemminger <stephen@networkplumber.org>
> > > Sent: Wednesday, May 24, 2023 5:50 PM
> > > To: Erez Ferber <erezf@nvidia.com>
> > > Cc: dev@dpdk.org; Slava Ovsiienko <viacheslavo@nvidia.com>; Matan
> > > Azrad <matan@nvidia.com>; Raslan Darawsheh <rasland@nvidia.com>;
> > > stable@dpdk.org
> > > Subject: Re: [PATCH] common/mlx5: adjust fork call with the new
> > > kernel API
> > >
> > > On Wed, 24 May 2023 15:01:40 +0300
> > > <erezf@nvidia.com> wrote:
> > >
> > > > From: Erez Ferber <erezf@nvidia.com>
> > > >
> > > > While doing process fork() the operating system remaps all the
> > > > parent process's memory to the address space of the child process
> > > > and activates the Copy-on-Write mechanics - it duplicates physical
> > > > pages once memory writing happens in the child process. Sometimes
> > > > memory duplication is not allowed - for example, if the page
> > > > contains hardware queue descriptors. To handle similar issues the
> > > > rdma-core library should be prepared for forking.
> > > >
> > > > The ibv_fork_init() prepares the library to track all the related
> > > > memory and prevent it from forking using madvise() system API.
> > > > This approach allows fork, but not all the memory is forked to the
> > > > child process and, application should care not to touch pages
> > > > where the parent application allocated the rdma-core objects.
> > > >
> > > > The newer kernels propose an option of copy-on-fork for DMA pages
> > > > and tracking all the memory and disabling it for the forking is no
> > > > longer needed. The new API routine ibv_is_fork_initialized()
> > > > should be involved to decide if library initialization for forking is
> required.
> > > >
> > > > Fixes: 0e83b8e536 ("net/mlx5: move rdma-core calls to separate
> > > > file")
> > > > Cc: stable@dpdk.org
> > > > Signed-off-by: Erez Ferber <erezf@nvidia.com>
> > >
> > Hi,
> >
> > > I don't think DPDK applications should fork(), and lots other parts
> > > of the shared huge pages will break if an application does this.
> >
> > I agree - application should not, we have the secondary/primary processes
> approach.
> > Nonetheless, we have the real use case - DPDK application does fork() and
> works well.
> > Without mlx5 PMD 😊. With mlx5 it ran into some troubles. Now we have
> the solution.
> 
> The problem is you are allowing fork(). And many other libraries may break.
> Imagine a DPDK library which has some local mutex and hugepages.
> If two forked processes use it then the locks won't work and hugepage data
> will be corrupted.

IMO, this problem is not fork() specific - applications/libraries supporting 
secondary/primary process approach should be developed with the similar
precautions.

And, IIRC, fork() is neither disallowed nor discouraged in DPDK documentation.
Moreover, some library unit tests ensure fork() works. No wonder users
may develop applications using fork(). They do, and app works well, beside mlx5 PMD.
Also, the recommendations to re-design with pri/sec approach was given.

With best regards,
Slava
 



^ permalink raw reply	[flat|nested] 7+ messages in thread

* RE: [PATCH] common/mlx5: adjust fork call with the new kernel API
  2023-05-24 12:01 [PATCH] common/mlx5: adjust fork call with the new kernel API erezf
  2023-05-24 14:50 ` Stephen Hemminger
  2023-05-24 16:05 ` Slava Ovsiienko
@ 2023-06-22 12:19 ` Raslan Darawsheh
  2 siblings, 0 replies; 7+ messages in thread
From: Raslan Darawsheh @ 2023-06-22 12:19 UTC (permalink / raw)
  To: Erez Ferber, dev; +Cc: Slava Ovsiienko, Matan Azrad, stable

Hi,

> -----Original Message-----
> From: Erez Ferber <erezf@nvidia.com>
> Sent: Wednesday, May 24, 2023 3:02 PM
> To: dev@dpdk.org
> Cc: Slava Ovsiienko <viacheslavo@nvidia.com>; Matan Azrad
> <matan@nvidia.com>; Raslan Darawsheh <rasland@nvidia.com>; Erez Ferber
> <erezf@nvidia.com>; stable@dpdk.org
> Subject: [PATCH] common/mlx5: adjust fork call with the new kernel API
> 
> From: Erez Ferber <erezf@nvidia.com>
> 
> While doing process fork() the operating system remaps all the parent
> process's memory to the address space of the child process and activates
> the Copy-on-Write mechanics - it duplicates physical pages once memory
> writing happens in the child process. Sometimes memory duplication is
> not allowed - for example, if the page contains hardware queue
> descriptors. To handle similar issues the rdma-core library should be
> prepared for forking.
> 
> The ibv_fork_init() prepares the library to track all the related memory
> and prevent it from forking using madvise() system API. This approach
> allows fork, but not all the memory is forked to the child process and,
> application should care not to touch pages where the parent application
> allocated the rdma-core objects.
> 
> The newer kernels propose an option of copy-on-fork for DMA pages and
> tracking all the memory and disabling it for the forking is no longer
> needed. The new API routine ibv_is_fork_initialized() should be involved
> to decide if library initialization for forking is required.
> 
> Fixes: 0e83b8e536 ("net/mlx5: move rdma-core calls to separate file")
> Cc: stable@dpdk.org
> Signed-off-by: Erez Ferber <erezf@nvidia.com>

Patch applied to next-net-mlx,

Kindest regards,
Raslan Darawsheh

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2023-06-22 12:19 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-05-24 12:01 [PATCH] common/mlx5: adjust fork call with the new kernel API erezf
2023-05-24 14:50 ` Stephen Hemminger
2023-05-25  8:10   ` Slava Ovsiienko
2023-05-25 15:27     ` Stephen Hemminger
2023-05-26  8:05       ` Slava Ovsiienko
2023-05-24 16:05 ` Slava Ovsiienko
2023-06-22 12:19 ` Raslan Darawsheh

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).