* Re: [dpdk-stable] [PATCH] mem: fix free segment when using huge-unlink option
[not found] ` <CAJ5KnGAWmNnWdSrzyGF2hfxYJrUEn_df8SWn1yKhG_swBrivOw@mail.gmail.com>
@ 2021-03-01 10:44 ` Burakov, Anatoly
2021-03-25 13:34 ` David Marchand
0 siblings, 1 reply; 3+ messages in thread
From: Burakov, Anatoly @ 2021-03-01 10:44 UTC (permalink / raw)
To: Roy Shterman; +Cc: dev, Yuval Mintz, Aviv Ben David, dpdk stable
On 28-Feb-21 1:21 PM, Roy Shterman wrote:
>
>
> On Mon, Feb 22, 2021 at 5:53 PM Burakov, Anatoly
> <anatoly.burakov@intel.com <mailto:anatoly.burakov@intel.com>> wrote:
>
> On 22-Feb-21 10:41 AM, Roy Shterman wrote:
> > When using huge_unlink we unlink the segment right
> > after allocation. Although we unlink the file we keep
> > the fd in fd_list so file still exist just the path deleted.
> > When freeing the hugepage we need to close the fd and assign
> > it with (-1) in fd_list for the page to be released.
> >
> > The current flow fails rte_malloc in the following flow when working
> > with --huge-unlink option:
> > 1. alloc_seg() for segment A -
> > We allocate segment, unlink the path to the segment
> > and keep the file descriptor in fd_list.
> > 2. free_seg() for segment A -
> > We clear the segment metadata and return - without closing fd
> > or assigning (-1) in fd list.
> > 3. alloc_seg() for segment A again -
> > We find segment A as available, try to allocate it,
> > find the old fd in fd_list try to unlink it
> > as part of alloc_seg() but failed because path doesn't exist.
> >
> > The impact of such error is falsly failing rte_malloc()
> > although we have hugepages available.
> >
> > Fixes: d435aad37da7 ("mem: support --huge-unlink mode")
> >
> > Signed-off-by: Roy Shterman <roy.shterman@vastdata.com
> <mailto:roy.shterman@vastdata.com>>
>
> Cc: stable@dpdk.org <mailto:stable@dpdk.org>
>
> Provisionally, patch looks fine, but i'll have to have a closer look.
>
>
> Hi Anatoly,
>
> Do I need to send this patch also to stable or it will happen
> automatically if the patch will reach the next release candidate?
Not automatically, generally you should add a CC to stable for the patch
to be considered as part of stable release. That said, usually Fixed:
tag is also used to determine whether a patch is backport-able, so
having a correctly identified Fixed: tag is a mandatory minimum :)
> Also I wonder if you had more time to review this one?
Not yet, but i'll try to make some time this week.
>
>
> --
> Thanks,
> Anatoly
>
>
> Thanks,
> Roy
--
Thanks,
Anatoly
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [dpdk-stable] [PATCH] mem: fix free segment when using huge-unlink option
2021-03-01 10:44 ` [dpdk-stable] [PATCH] mem: fix free segment when using huge-unlink option Burakov, Anatoly
@ 2021-03-25 13:34 ` David Marchand
0 siblings, 0 replies; 3+ messages in thread
From: David Marchand @ 2021-03-25 13:34 UTC (permalink / raw)
To: Burakov, Anatoly
Cc: Roy Shterman, dev, Yuval Mintz, Aviv Ben David, dpdk stable
Hello,
On Mon, Mar 1, 2021 at 11:44 AM Burakov, Anatoly
<anatoly.burakov@intel.com> wrote:
>
> On 28-Feb-21 1:21 PM, Roy Shterman wrote:
> >
> >
> > On Mon, Feb 22, 2021 at 5:53 PM Burakov, Anatoly
> > <anatoly.burakov@intel.com <mailto:anatoly.burakov@intel.com>> wrote:
> >
> > On 22-Feb-21 10:41 AM, Roy Shterman wrote:
> > > When using huge_unlink we unlink the segment right
> > > after allocation. Although we unlink the file we keep
> > > the fd in fd_list so file still exist just the path deleted.
> > > When freeing the hugepage we need to close the fd and assign
> > > it with (-1) in fd_list for the page to be released.
> > >
> > > The current flow fails rte_malloc in the following flow when working
> > > with --huge-unlink option:
> > > 1. alloc_seg() for segment A -
> > > We allocate segment, unlink the path to the segment
> > > and keep the file descriptor in fd_list.
> > > 2. free_seg() for segment A -
> > > We clear the segment metadata and return - without closing fd
> > > or assigning (-1) in fd list.
> > > 3. alloc_seg() for segment A again -
> > > We find segment A as available, try to allocate it,
> > > find the old fd in fd_list try to unlink it
> > > as part of alloc_seg() but failed because path doesn't exist.
> > >
> > > The impact of such error is falsly failing rte_malloc()
> > > although we have hugepages available.
> > >
> > > Fixes: d435aad37da7 ("mem: support --huge-unlink mode")
> > >
> > > Signed-off-by: Roy Shterman <roy.shterman@vastdata.com
> > <mailto:roy.shterman@vastdata.com>>
> >
> > Cc: stable@dpdk.org <mailto:stable@dpdk.org>
> >
> > Provisionally, patch looks fine, but i'll have to have a closer look.
> >
> >
> > Hi Anatoly,
> >
> > Do I need to send this patch also to stable or it will happen
> > automatically if the patch will reach the next release candidate?
>
> Not automatically, generally you should add a CC to stable for the patch
> to be considered as part of stable release. That said, usually Fixed:
> tag is also used to determine whether a patch is backport-able, so
> having a correctly identified Fixed: tag is a mandatory minimum :)
>
> > Also I wonder if you had more time to review this one?
>
> Not yet, but i'll try to make some time this week.
Any update?
Thanks.
--
David Marchand
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [dpdk-stable] [dpdk-dev] [PATCH] mem: fix free segment when using huge-unlink option
[not found] ` <63d59764-3a86-6b7d-97c3-28a8d1a6eebb@intel.com>
@ 2021-04-07 9:09 ` David Marchand
0 siblings, 0 replies; 3+ messages in thread
From: David Marchand @ 2021-04-07 9:09 UTC (permalink / raw)
To: Roy Shterman
Cc: dev, Burakov, Anatoly, Yuval Mintz, Aviv Ben David, dpdk stable
On Thu, Apr 1, 2021 at 1:07 PM Burakov, Anatoly
<anatoly.burakov@intel.com> wrote:
>
> On 22-Feb-21 10:41 AM, Roy Shterman wrote:
> > When using huge_unlink we unlink the segment right
> > after allocation. Although we unlink the file we keep
> > the fd in fd_list so file still exist just the path deleted.
> > When freeing the hugepage we need to close the fd and assign
> > it with (-1) in fd_list for the page to be released.
> >
> > The current flow fails rte_malloc in the following flow when working
> > with --huge-unlink option:
> > 1. alloc_seg() for segment A -
> > We allocate segment, unlink the path to the segment
> > and keep the file descriptor in fd_list.
> > 2. free_seg() for segment A -
> > We clear the segment metadata and return - without closing fd
> > or assigning (-1) in fd list.
> > 3. alloc_seg() for segment A again -
> > We find segment A as available, try to allocate it,
> > find the old fd in fd_list try to unlink it
> > as part of alloc_seg() but failed because path doesn't exist.
> >
> > The impact of such error is falsly failing rte_malloc()
falsely*
> > although we have hugepages available.
> >
> > Fixes: d435aad37da7 ("mem: support --huge-unlink mode")
Cc: stable@dpdk.org
> >
> > Signed-off-by: Roy Shterman <roy.shterman@vastdata.com>
> Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Applied, thanks.
--
David Marchand
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2021-04-07 9:10 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <20210222104131.11979-1-roy.shterman@vastdata.com>
[not found] ` <ba9df18a-036f-dbd4-b19f-2f5faec43acc@intel.com>
[not found] ` <CAJ5KnGAWmNnWdSrzyGF2hfxYJrUEn_df8SWn1yKhG_swBrivOw@mail.gmail.com>
2021-03-01 10:44 ` [dpdk-stable] [PATCH] mem: fix free segment when using huge-unlink option Burakov, Anatoly
2021-03-25 13:34 ` David Marchand
[not found] ` <63d59764-3a86-6b7d-97c3-28a8d1a6eebb@intel.com>
2021-04-07 9:09 ` [dpdk-stable] [dpdk-dev] " David Marchand
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).