Linux-RDMA Archive on lore.kernel.org
 help / color / Atom feed
From: Jason Gunthorpe <jgg@mellanox.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Doug Ledford <dledford@redhat.com>,
	"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [GIT PULL] Please pull RDMA subsystem changes
Date: Tue, 30 Apr 2019 12:53:16 +0000
Message-ID: <20190430125310.GH3562@mellanox.com> (raw)
In-Reply-To: <CAHk-=whWh+5-jGmgw2HPM8XhzXzvGcLo19UGLtqYe_teJsRgRA@mail.gmail.com>

On Sun, Apr 28, 2019 at 05:09:08PM -0700, Linus Torvalds wrote:
> On Sun, Apr 28, 2019 at 4:49 PM Jason Gunthorpe <jgg@mellanox.com> wrote:
> >
> > It is for high availability - we have situations where the hardware
> > can fault and needs some kind of destructive recovery. For instance a
> > firmware reboot, or a VM migration.
> >
> > In these designs there may be multiple cards in the system and the
> > userspace application could be using both. Just because one card
> > crashed we can't send SIGBUS and kill the application, that breaks the
> > HA design.
> 
> Why can't this magical application that is *so* special that it is HA
> and does magic mmap's of special rdma areas just catch the SIGBUS?
> 
> Honestly, the whole "it's for HA" excuse stinks. It stinks because you
> now silently just replace the mapping with *garbage*. That's not HA,
> that's just random.

This should only used in cases where user space only writes to the BAR
page (it is an interrupt to the device essentially), so it doesn't
care that the pages are now garbage, we just need to redirect the
writes away from the bar.

However I think someone later on added a readable counter BAR pages to
one of the devices :( So even that ideal wasn't respected.

> Wouldn't it be a lot better to just get the SIGBUS, and then that
> magical application knows that "oh, it's gone", and it could - in its
> SIGBUS handler - just do the dummy anonymous mmap() with /dev/zero it
> if it wants to?

This does sound more appealing, and probably should have been done
instead. All this VMA stuff has been a big pain in the long run

Thanks,
Jason

  reply index

Thread overview: 103+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-28 11:52 Jason Gunthorpe
2019-04-28 16:59 ` Linus Torvalds
2019-04-28 23:49   ` Jason Gunthorpe
2019-04-29  0:09     ` Linus Torvalds
2019-04-30 12:53       ` Jason Gunthorpe [this message]
2019-04-28 18:05 ` pr-tracker-bot
2019-04-29  6:09 ` Heiko Carstens
2019-04-29  8:40   ` Jason Gunthorpe
2019-04-29  9:00     ` Michal Kubecek
2019-04-29  9:19       ` Leon Romanovsky
2019-04-29 15:42     ` Doug Ledford
2019-04-29 16:29       ` Doug Ledford
  -- strict thread matches above, loose matches on Subject: below --
2020-05-29 14:15 Jason Gunthorpe
2020-05-29 21:10 ` pr-tracker-bot
2020-05-15 19:13 Jason Gunthorpe
2020-05-15 20:15 ` pr-tracker-bot
2020-04-28 19:59 Jason Gunthorpe
2020-04-28 20:40 ` pr-tracker-bot
2020-04-01 23:01 Jason Gunthorpe
2020-04-02  2:05 ` pr-tracker-bot
2020-03-26 17:31 Jason Gunthorpe
2020-03-26 17:50 ` pr-tracker-bot
2020-03-08  0:07 Jason Gunthorpe
2020-03-08  2:00 ` pr-tracker-bot
2020-02-15 21:32 Jason Gunthorpe
2020-02-16  1:20 ` pr-tracker-bot
2020-01-31 15:16 Jason Gunthorpe
2020-01-31 23:10 ` pr-tracker-bot
2020-01-10  2:16 Jason Gunthorpe
2020-01-11 14:45 ` pr-tracker-bot
2019-11-27  0:24 Jason Gunthorpe
2019-11-27 18:19 ` Linus Torvalds
2019-11-27 19:23   ` Jason Gunthorpe
2019-11-27 18:45 ` pr-tracker-bot
2019-11-14 15:16 Jason Gunthorpe
2019-11-14 16:55 ` pr-tracker-bot
2019-10-31 18:22 Jason Gunthorpe
2019-11-01 17:10 ` pr-tracker-bot
2019-10-09 14:28 Jason Gunthorpe
2019-10-09 19:30 ` pr-tracker-bot
2019-09-19 16:34 Jason Gunthorpe
2019-09-21 17:40 ` pr-tracker-bot
2019-07-30 12:15 Jason Gunthorpe
2019-07-30 20:40 ` pr-tracker-bot
2019-07-15 15:26 Jason Gunthorpe
2019-07-16  4:35 ` pr-tracker-bot
2019-06-06 20:14 Jason Gunthorpe
2019-06-07 16:45 ` pr-tracker-bot
2019-05-15  0:46 Jason Gunthorpe
2019-05-15  4:05 ` pr-tracker-bot
2019-05-09 13:37 Jason Gunthorpe
2019-05-09 16:25 ` pr-tracker-bot
2019-04-10 18:46 Jason Gunthorpe
2019-04-10 19:55 ` pr-tracker-bot
2019-03-18  1:04 Jason Gunthorpe
2019-03-19 18:15 ` pr-tracker-bot
2019-03-07  1:34 Jason Gunthorpe
2019-03-10  1:40 ` pr-tracker-bot
2019-02-21 23:07 Jason Gunthorpe
2019-02-22 18:40 ` pr-tracker-bot
2019-02-01 17:41 Jason Gunthorpe
2019-02-01 18:45 ` pr-tracker-bot
2019-01-18  3:56 Jason Gunthorpe
2019-01-18  5:35 ` pr-tracker-bot
2019-01-04  5:00 Jason Gunthorpe
2019-01-06  2:40 ` pr-tracker-bot
2018-12-24 22:16 Jason Gunthorpe
2018-12-29  1:30 ` pr-tracker-bot
2018-11-29 22:52 Jason Gunthorpe
2018-10-25 21:21 Jason Gunthorpe
2018-10-26 14:50 ` Linus Torvalds
2018-09-27 18:24 Jason Gunthorpe
2018-09-27 20:17 ` Greg Kroah-Hartman
2018-09-10 23:04 Jason Gunthorpe
2018-08-22 21:44 Jason Gunthorpe
2018-08-16 21:57 Jason Gunthorpe
2018-08-17 19:31 ` Linus Torvalds
2018-08-17 19:44   ` Linus Torvalds
2018-08-17 20:50     ` Linus Torvalds
2018-08-17 21:16       ` Jason Gunthorpe
2018-08-17 23:56         ` Linus Torvalds
2018-08-17 20:15   ` Jason Gunthorpe
2018-08-17 20:27     ` Linus Torvalds
2018-08-17 21:27       ` Jason Gunthorpe
2018-08-03 16:11 Jason Gunthorpe
2018-07-13 17:13 Jason Gunthorpe
2018-06-20 21:00 Jason Gunthorpe
2018-06-06 21:42 Jason Gunthorpe
2018-06-01 16:47 Jason Gunthorpe
2018-05-24 20:51 Jason Gunthorpe
2018-04-06 16:05 Jason Gunthorpe
2018-05-16 17:39 ` Eugene Syromiatnikov
2018-05-16 17:49   ` Jason Gunthorpe
2018-05-16 18:01     ` Eugene Syromiatnikov
2018-03-29 19:17 Jason Gunthorpe
2018-03-20 23:17 Jason Gunthorpe
2018-01-31 17:47 Jason Gunthorpe
     [not found] ` <20180131174735.GA18568-uk2M96/98Pc@public.gmane.org>
2018-01-31 20:11   ` Linus Torvalds
     [not found]     ` <CA+55aFxmnW-iu1Na3QC8Ci8Q_Qdfn2Ak_9wDB6+A564R=Xn9Ag-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2018-01-31 21:04       ` Jason Gunthorpe
     [not found]         ` <20180131210457.GE23352-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2018-02-01 19:12           ` Linus Torvalds
     [not found]             ` <CA+55aFxySLoCHcmZgx2PiF6jEazVOSy=2idccRsWqOzmyK2gaw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2018-02-01 20:33               ` Jason Gunthorpe
2017-12-28 21:03 Jason Gunthorpe
2017-12-16 17:12 Jason Gunthorpe

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190430125310.GH3562@mellanox.com \
    --to=jgg@mellanox.com \
    --cc=dledford@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=torvalds@linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Linux-RDMA Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-rdma/0 linux-rdma/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-rdma linux-rdma/ https://lore.kernel.org/linux-rdma \
		linux-rdma@vger.kernel.org
	public-inbox-index linux-rdma

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-rdma


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git