All of lore.kernel.org
 help / color / mirror / Atom feed
From: sagi@grimberg.me (Sagi Grimberg)
Subject: [PATCH 4.15-rc 1/3] nvme-core: Don't set nvme_wq as MEM_RECLAIM
Date: Thu, 21 Dec 2017 12:41:30 +0200	[thread overview]
Message-ID: <fd880cf7-e815-fa63-b999-6381b145ffd8@grimberg.me> (raw)
In-Reply-To: <20171221101741.GB17327@infradead.org>


>> From: Roy Shterman <roys at lightbitslabs.com>
>>
>> nvme_wq is not a MEM_RECLAIM workqueue because it
>> can allocate memory in some of the works it is executing.
> 
> But we need reset to work while in memory reclaim.

AFAIK, WQ_MEM_RECLAIM means that this workqueue can be drained
for memory reclaim, which means that a workqueue that hosts works
that are allocating memory cannot be such a workqueue.

How does this patch make reset not work in memory reclaim? memory
reclaim will drain workqueues that *are* reclaimable workqueues.

> So instead we'll need to make sure whatever memory allocation required
> (which ones, btw?) are marked GFP_NOIO.

namespace scannig can allocate new namespaces.

  reply	other threads:[~2017-12-21 10:41 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-21 10:07 [PATCH 4.15-rc 0/3] few fabrics fixes targeted to 4.15 Sagi Grimberg
2017-12-21 10:07 ` [PATCH 4.15-rc 1/3] nvme-core: Don't set nvme_wq as MEM_RECLAIM Sagi Grimberg
2017-12-21 10:17   ` Christoph Hellwig
2017-12-21 10:41     ` Sagi Grimberg [this message]
2017-12-21 13:00       ` Christoph Hellwig
2017-12-21 13:17         ` Sagi Grimberg
2017-12-21 13:54           ` Christoph Hellwig
2017-12-21 14:17             ` Sagi Grimberg
2017-12-29  9:35               ` Christoph Hellwig
2017-12-31  9:51                 ` Sagi Grimberg
2017-12-21 10:07 ` [PATCH 4.15-rc 2/3] nvme-core/loop/rdma: Host delete_work and reset_work on system workqueues Sagi Grimberg
2017-12-21 10:25   ` Christoph Hellwig
2017-12-21 10:43     ` Sagi Grimberg
2017-12-21 13:00       ` Christoph Hellwig
2017-12-22 17:39   ` James Smart
2017-12-24  8:55     ` Sagi Grimberg
2017-12-21 10:07 ` [PATCH 4.15-rc 3/3] nvme-fabrics: Protect against module unload during create_ctrl Sagi Grimberg
2017-12-21 10:25   ` Christoph Hellwig
2017-12-21 18:31     ` Sagi Grimberg

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=fd880cf7-e815-fa63-b999-6381b145ffd8@grimberg.me \
    --to=sagi@grimberg.me \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.