Re: Request for -stable inclusion: time stamping fix for nfp

2018-05-18 Thread Guillaume Nault
On Thu, May 17, 2018 at 02:09:03PM -0400, David Miller wrote: > From: Guillaume Nault > Date: Thu, 17 May 2018 19:41:47 +0200 > > > On Thu, Nov 16, 2017 at 10:13:28AM +0900, David Miller wrote: > >> From: Guillaume Nault > >> Date: Wed, 15 Nov 2017 17:20:46 +0100 > >> > >> > Can you please queu

Re: Request for -stable inclusion: time stamping fix for nfp

2018-05-17 Thread Greg KH
On Thu, May 17, 2018 at 08:32:02PM +0200, Willy Tarreau wrote: > Adding Greg here. > > Greg, apparently a backport of 46f1c52e66db is needed in 4.9 according > to the thread below. It was merged in 4.13 so 4.14 already has it. Thanks for the notice, now queued up. greg k-h

Re: Request for -stable inclusion: time stamping fix for nfp

2018-05-17 Thread Willy Tarreau
Adding Greg here. Greg, apparently a backport of 46f1c52e66db is needed in 4.9 according to the thread below. It was merged in 4.13 so 4.14 already has it. Willy On Thu, May 17, 2018 at 02:09:03PM -0400, David Miller wrote: > From: Guillaume Nault > Date: Thu, 17 May 2018 19:41:47 +0200 > > >

Re: Request for -stable inclusion: time stamping fix for nfp

2018-05-17 Thread David Miller
From: Guillaume Nault Date: Thu, 17 May 2018 19:41:47 +0200 > On Thu, Nov 16, 2017 at 10:13:28AM +0900, David Miller wrote: >> From: Guillaume Nault >> Date: Wed, 15 Nov 2017 17:20:46 +0100 >> >> > Can you please queue commit 46f1c52e66db >> > ("nfp: TX time stamp packets before HW doorbell is

Re: Request for -stable inclusion: time stamping fix for nfp

2018-05-17 Thread Guillaume Nault
On Thu, Nov 16, 2017 at 10:13:28AM +0900, David Miller wrote: > From: Guillaume Nault > Date: Wed, 15 Nov 2017 17:20:46 +0100 > > > Can you please queue commit 46f1c52e66db > > ("nfp: TX time stamp packets before HW doorbell is rung") for -stable? > > We got hit but this bug in the late summer. W

Re: Request for -stable inclusion: time stamping fix for nfp

2017-11-15 Thread David Miller
From: Guillaume Nault Date: Wed, 15 Nov 2017 17:20:46 +0100 > Can you please queue commit 46f1c52e66db > ("nfp: TX time stamp packets before HW doorbell is rung") for -stable? > We got hit but this bug in the late summer. We run this fix internally > since a couple of months, but that'd be better

Re: Request for -stable inclusion (xt_TCPMSS)

2017-06-16 Thread David Miller
From: Guillaume Nault Date: Fri, 16 Jun 2017 09:45:17 +0200 > Hi David, > > Please queue 2638fd0f92d4 ("netfilter: xt_TCPMSS: add more sanity tests on > tcph->doff") > for -stable. > It fixes hard to debug crashes in production, that can be triggered by > remote peers. Pablo handles netfilter

Re: Request for -stable inclusion (xt_TCPMSS)

2017-06-16 Thread Guillaume Nault
On Fri, Jun 16, 2017 at 11:59:12AM +0200, Pablo Neira Ayuso wrote: > Hi Guillaume, > > On Fri, Jun 16, 2017 at 09:45:17AM +0200, Guillaume Nault wrote: > > Hi David, > > > > Please queue 2638fd0f92d4 ("netfilter: xt_TCPMSS: add more sanity tests on > > tcph->doff") > > for -stable. > > It fixes

Re: Request for -stable inclusion (xt_TCPMSS)

2017-06-16 Thread Pablo Neira Ayuso
Hi Guillaume, On Fri, Jun 16, 2017 at 09:45:17AM +0200, Guillaume Nault wrote: > Hi David, > > Please queue 2638fd0f92d4 ("netfilter: xt_TCPMSS: add more sanity tests on > tcph->doff") > for -stable. > It fixes hard to debug crashes in production, that can be triggered by > remote peers. What s

Re: request for stable inclusion

2015-09-28 Thread David Miller
From: Or Gerlitz Date: Wed, 16 Sep 2015 09:41:40 +0300 > Hi Dave, > > Commit 9293267 "net/mlx4_core: Capping number of requested MSIXs to > MAX_MSIX" fixes a bug under which the driver doesn't really starts > over a machine with > 32 cores. > > The bug was introduced in 4.2-rc1 but the fix miss