From: Stephen Hemminger <stephen@networkplumber.org>
To: Alexander Duyck <alexander.duyck@gmail.com>
Cc: dev@dpdk.org, Avi Kivity <avi@scylladb.com>,
hjk@hansjkoch.de, gregkh@linux-foundation.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 0/2] uio_msi: device driver
Date: Thu, 1 Oct 2015 17:04:03 -0700 [thread overview]
Message-ID: <20151001170403.1ba4dad3@urahara> (raw)
In-Reply-To: <560DC51B.7030507@gmail.com>
On Thu, 1 Oct 2015 16:43:23 -0700
Alexander Duyck <alexander.duyck@gmail.com> wrote:
> Yes, but in the case of something like a VF it is going to just make a
> bigger mess of things since INTx doesn't work. So what would you expect
> your driver to do in that case? Also we have to keep in mind that the
> MSI-X failure case is very unlikely.
>
> One other thing that just occurred to me is that you may want to try
> using the range allocation call instead of a hard set number of
> interrupts. Then if you start running short on vectors you don't hard
> fail and instead just allocate what you can.
I tried that but the bookkeeping gets messy since there is no good
way to communicate that back to userspace and have it adapt.
WARNING: multiple messages have this Message-ID (diff)
From: Stephen Hemminger <stephen@networkplumber.org>
To: Alexander Duyck <alexander.duyck@gmail.com>
Cc: Avi Kivity <avi@scylladb.com>,
dev@dpdk.org, hjk@hansjkoch.de, gregkh@linux-foundation.org,
linux-kernel@vger.kernel.org
Subject: Re: [dpdk-dev] [PATCH 0/2] uio_msi: device driver
Date: Thu, 1 Oct 2015 17:04:03 -0700 [thread overview]
Message-ID: <20151001170403.1ba4dad3@urahara> (raw)
In-Reply-To: <560DC51B.7030507@gmail.com>
On Thu, 1 Oct 2015 16:43:23 -0700
Alexander Duyck <alexander.duyck@gmail.com> wrote:
> Yes, but in the case of something like a VF it is going to just make a
> bigger mess of things since INTx doesn't work. So what would you expect
> your driver to do in that case? Also we have to keep in mind that the
> MSI-X failure case is very unlikely.
>
> One other thing that just occurred to me is that you may want to try
> using the range allocation call instead of a hard set number of
> interrupts. Then if you start running short on vectors you don't hard
> fail and instead just allocate what you can.
I tried that but the bookkeeping gets messy since there is no good
way to communicate that back to userspace and have it adapt.
next prev parent reply other threads:[~2015-10-02 0:03 UTC|newest]
Thread overview: 80+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-09-30 22:28 [PATCH 0/2] uio_msi: device driver Stephen Hemminger
2015-09-30 22:28 ` Stephen Hemminger
2015-09-30 22:28 ` [PATCH 1/2] uio: add support for ioctls Stephen Hemminger
2015-09-30 22:28 ` Stephen Hemminger
2015-09-30 22:28 ` [PATCH 2/2] uio: new driver to support PCI MSI-X Stephen Hemminger
2015-09-30 22:28 ` Stephen Hemminger
2015-10-01 8:33 ` Michael S. Tsirkin
2015-10-01 8:33 ` Michael S. Tsirkin
2015-10-01 10:37 ` Michael S. Tsirkin
2015-10-01 10:37 ` Michael S. Tsirkin
2015-10-01 16:06 ` Michael S. Tsirkin
2015-10-01 16:06 ` Michael S. Tsirkin
2015-10-01 14:50 ` Stephen Hemminger
2015-10-01 14:50 ` Stephen Hemminger
2015-10-01 15:22 ` Michael S. Tsirkin
2015-10-01 15:22 ` Michael S. Tsirkin
2015-10-01 16:31 ` Michael S. Tsirkin
2015-10-01 16:31 ` Michael S. Tsirkin
2015-10-01 17:26 ` Stephen Hemminger
2015-10-01 17:26 ` Stephen Hemminger
2015-10-01 18:25 ` Michael S. Tsirkin
2015-10-01 18:25 ` Michael S. Tsirkin
2015-10-05 21:54 ` Michael S. Tsirkin
2015-10-05 21:54 ` Michael S. Tsirkin
2015-10-05 22:09 ` Vladislav Zolotarov
2015-10-05 22:49 ` Michael S. Tsirkin
2015-10-05 22:49 ` [dpdk-dev] " Michael S. Tsirkin
2015-10-06 7:33 ` Stephen Hemminger
2015-10-06 7:33 ` [dpdk-dev] " Stephen Hemminger
2015-10-06 12:15 ` Avi Kivity
2015-10-06 12:15 ` [dpdk-dev] " Avi Kivity
2015-10-06 14:07 ` Michael S. Tsirkin
2015-10-06 15:41 ` Avi Kivity
2015-10-06 15:41 ` [dpdk-dev] " Avi Kivity
2015-10-16 17:11 ` Thomas Monjalon
2015-10-16 17:11 ` [dpdk-dev] " Thomas Monjalon
2015-10-16 17:20 ` Stephen Hemminger
2015-10-16 17:20 ` [dpdk-dev] " Stephen Hemminger
2015-10-06 13:42 ` Michael S. Tsirkin
2015-10-06 13:42 ` [dpdk-dev] " Michael S. Tsirkin
2015-10-06 8:23 ` Vlad Zolotarov
2015-10-06 8:23 ` [dpdk-dev] " Vlad Zolotarov
2015-10-06 13:58 ` Michael S. Tsirkin
2015-10-06 13:58 ` [dpdk-dev] " Michael S. Tsirkin
2015-10-06 14:49 ` Vlad Zolotarov
2015-10-06 15:00 ` Michael S. Tsirkin
2015-10-06 15:00 ` [dpdk-dev] " Michael S. Tsirkin
2015-10-06 16:40 ` Vlad Zolotarov
2015-10-06 16:40 ` [dpdk-dev] " Vlad Zolotarov
2015-10-01 23:40 ` Alexander Duyck
2015-10-01 23:40 ` [dpdk-dev] " Alexander Duyck
2015-10-02 0:01 ` Stephen Hemminger
2015-10-02 0:01 ` [dpdk-dev] " Stephen Hemminger
2015-10-02 1:21 ` Alexander Duyck
2015-10-02 1:21 ` [dpdk-dev] " Alexander Duyck
2015-10-02 0:04 ` Stephen Hemminger
2015-10-02 2:33 ` Alexander Duyck
2015-10-02 2:33 ` [dpdk-dev] " Alexander Duyck
2015-10-01 8:36 ` [PATCH 0/2] uio_msi: device driver Michael S. Tsirkin
2015-10-01 8:36 ` Michael S. Tsirkin
2015-10-01 10:59 ` Avi Kivity
2015-10-01 10:59 ` [dpdk-dev] " Avi Kivity
2015-10-01 14:57 ` Stephen Hemminger
2015-10-01 19:48 ` Alexander Duyck
2015-10-01 19:48 ` [dpdk-dev] " Alexander Duyck
2015-10-01 22:00 ` Stephen Hemminger
2015-10-01 22:00 ` [dpdk-dev] " Stephen Hemminger
2015-10-01 23:03 ` Alexander Duyck
2015-10-01 23:03 ` [dpdk-dev] " Alexander Duyck
2015-10-01 23:39 ` Stephen Hemminger
2015-10-01 23:39 ` [dpdk-dev] " Stephen Hemminger
2015-10-01 23:43 ` Alexander Duyck
2015-10-02 0:04 ` Stephen Hemminger [this message]
2015-10-02 0:04 ` Stephen Hemminger
2015-10-02 1:39 ` Alexander Duyck
2015-10-04 16:49 ` Vlad Zolotarov
2015-10-04 16:49 ` [dpdk-dev] " Vlad Zolotarov
2015-10-04 19:03 ` Greg KH
2015-10-04 20:49 ` Vlad Zolotarov
2015-10-04 20:49 ` [dpdk-dev] " Vlad Zolotarov
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=20151001170403.1ba4dad3@urahara \
--to=stephen@networkplumber.org \
--cc=alexander.duyck@gmail.com \
--cc=avi@scylladb.com \
--cc=dev@dpdk.org \
--cc=gregkh@linux-foundation.org \
--cc=hjk@hansjkoch.de \
--cc=linux-kernel@vger.kernel.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
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.