Re: [dpdk-dev] [PATCH v2 0/2] Fix file locking in EAL memory

2018-04-24 Thread Stephen Hemminger
On Tue, 24 Apr 2018 22:05:10 +0200 Thomas Monjalon wrote: > 24/04/2018 19:25, Burakov, Anatoly: > > On 24-Apr-18 5:32 PM, Stephen Hemminger wrote: > > > On Tue, 24 Apr 2018 16:54:21 +0100 > > > Anatoly Burakov wrote: > > > > > >> One of the ways to work around this was using OFD locks, > >

Re: [dpdk-dev] [PATCH v2 0/2] Fix file locking in EAL memory

2018-04-24 Thread Thomas Monjalon
24/04/2018 19:25, Burakov, Anatoly: > On 24-Apr-18 5:32 PM, Stephen Hemminger wrote: > > On Tue, 24 Apr 2018 16:54:21 +0100 > > Anatoly Burakov wrote: > > > >> One of the ways to work around this was using OFD locks, > >> but they are only supported on kernels 3.15+, > > > > Maybe time to move t

Re: [dpdk-dev] [PATCH v2 0/2] Fix file locking in EAL memory

2018-04-24 Thread Burakov, Anatoly
On 24-Apr-18 5:32 PM, Stephen Hemminger wrote: On Tue, 24 Apr 2018 16:54:21 +0100 Anatoly Burakov wrote: One of the ways to work around this was using OFD locks, but they are only supported on kernels 3.15+, Maybe time to move to next LTS kernel (3.16)? I'm all for it, but it's not my cal

Re: [dpdk-dev] [PATCH v2 0/2] Fix file locking in EAL memory

2018-04-24 Thread Stephen Hemminger
On Tue, 24 Apr 2018 16:54:21 +0100 Anatoly Burakov wrote: > One of the ways to work around this was using OFD locks, > but they are only supported on kernels 3.15+, Maybe time to move to next LTS kernel (3.16)?

[dpdk-dev] [PATCH v2 0/2] Fix file locking in EAL memory

2018-04-24 Thread Anatoly Burakov
This patchset replaces the fcntl()-based locking used in the original DPDK memory hotplug patchset, to an flock()- and lockfile-based implementation, due to numerous (well, one, really) problems with how fcntl() locks work. Long story short, fcntl() locks will be dropped if any fd referring to loc