From: Long Li <longli@microsoft.com>
To: Stephen Hemminger <stephen@networkplumber.org>
Cc: "longli@linuxonhyperv.com" <longli@linuxonhyperv.com>,
Ferruh Yigit <ferruh.yigit@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>,
Ajay Sharma <sharmaajay@microsoft.com>,
Stephen Hemminger <sthemmin@microsoft.com>
Subject: RE: [PATCH 01/17] net/mana: add basic driver, build environment and doc
Date: Tue, 5 Jul 2022 19:50:27 +0000 [thread overview]
Message-ID: <PH7PR21MB326375FEBDB737CC2C5FF529CE819@PH7PR21MB3263.namprd21.prod.outlook.com> (raw)
In-Reply-To: <20220704094117.340d9b46@hermes.local>
> Subject: Re: [PATCH 01/17] net/mana: add basic driver, build environment and
> doc
>
> On Sun, 3 Jul 2022 07:56:18 +0000
> Long Li <longli@microsoft.com> wrote:
>
> > > Subject: Re: [PATCH 01/17] net/mana: add basic driver, build
> > > environment and doc
> > >
> > > On Fri, 1 Jul 2022 02:02:31 -0700
> > > longli@linuxonhyperv.com wrote:
> > >
> > > > + uint64_t max_mr_size;
> > > > + rte_rwlock_t mr_list_lock;
> > > > +};
> > >
> > > Reader/Writer locks are slower for the usual uncontended case.
> > > Unless you have a reader holding onto the lock for a long time,
> > > better to use spin lock.
> > >
> > > This is Linux wisdom (thank you paulmck), Windows seems to love
> > > reader/writer locks.
> >
> > This lock is used mostly uncontended for readers and allow them concurrently
> to access the list. It's rare that a writer might want to access it.
> >
> > But changing it to spinlock will prevent multiple readers from accessing the list
> at the same time. In this use case, is it beneficial to use spinlocks?
>
> https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.re
> searchgate.net%2Ffigure%2FCosts-of-Simple-Spinlock-and-Distributed-Reader-
> Writer-Spinlock-To-write-acquire-
> the_fig1_244490342&data=05%7C01%7Clongli%40microsoft.com%7C51cc
> 3d29340347b87b8508da5ddc06c0%7C72f988bf86f141af91ab2d7cd011db47%7
> C1%7C0%7C637925496841806674%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC
> 4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C
> %7C%7C&sdata=bWePngCuLBWN%2FFasdYpTuuCJW6Y8kX6WVTInLYvd1E
> c%3D&reserved=0
>
> Conclusion
>
> Breakeven curves for the locking primitives were
> computed, and the breakeven between simple spinlock
> and distributed reader-writer spinlock supports the rule of
> thumb that states that simple spinlock should be used in
> cases where the update ratio f is greater than the
> reciprocal of the number of CPUs, 1/nm.
Thank you for the pointer!
I think this driver should use spinlocks. I'm sending v2 with this change.
Long
next prev parent reply other threads:[~2022-07-05 19:50 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-07-01 9:02 [PATCH 00/17] Introduce Microsoft Azure Network Adatper (MANA) PMD longli
2022-07-01 9:02 ` [PATCH 01/17] net/mana: add basic driver, build environment and doc longli
2022-07-01 16:29 ` Stephen Hemminger
2022-07-01 16:44 ` Stephen Hemminger
2022-07-01 16:45 ` Stephen Hemminger
2022-07-01 16:47 ` Stephen Hemminger
2022-07-03 7:56 ` Long Li
2022-07-04 16:35 ` Stephen Hemminger
2022-07-04 16:41 ` Stephen Hemminger
2022-07-05 19:50 ` Long Li [this message]
2022-07-01 16:49 ` Stephen Hemminger
2022-07-01 19:40 ` Long Li
2022-07-01 9:02 ` [PATCH 02/17] net/mana: add device configuration and stop longli
2022-07-01 16:34 ` Stephen Hemminger
2022-07-01 9:02 ` [PATCH 03/17] net/mana: add function to report support ptypes longli
2022-07-01 9:02 ` [PATCH 04/17] net/mana: add link update longli
2022-07-01 9:02 ` [PATCH 05/17] net/mana: add function for device removal interrupts longli
2022-07-01 9:02 ` [PATCH 06/17] net/mana: add device info longli
2022-07-01 9:02 ` [PATCH 07/17] net/mana: add function to configure RSS longli
2022-07-01 9:02 ` [PATCH 08/17] net/mana: add function to configure RX queues longli
2022-07-01 9:02 ` [PATCH 09/17] net/mana: add function to configure TX queues longli
2022-07-01 9:02 ` [PATCH 10/17] net/mana: implement memory registration longli
2022-07-01 16:37 ` Stephen Hemminger
2022-07-01 9:02 ` [PATCH 11/17] net/mana: implement the hardware layer operations longli
2022-07-01 16:38 ` Stephen Hemminger
2022-07-01 9:02 ` [PATCH 12/17] net/mana: add function to start/stop TX queues longli
2022-07-01 16:39 ` Stephen Hemminger
2022-07-01 9:02 ` [PATCH 13/17] net/mana: add function to start/stop RX queues longli
2022-07-01 16:40 ` Stephen Hemminger
2022-07-02 8:42 ` Long Li
2022-07-01 9:02 ` [PATCH 14/17] net/mana: add function to receive packets longli
2022-07-01 9:02 ` [PATCH 15/17] net/mana: add function to send packets longli
2022-07-01 9:02 ` [PATCH 16/17] net/mana: add function to start/stop device longli
2022-07-01 9:02 ` [PATCH 17/17] net/mana: add function to report queue stats longli
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=PH7PR21MB326375FEBDB737CC2C5FF529CE819@PH7PR21MB3263.namprd21.prod.outlook.com \
--to=longli@microsoft.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
--cc=longli@linuxonhyperv.com \
--cc=sharmaajay@microsoft.com \
--cc=stephen@networkplumber.org \
--cc=sthemmin@microsoft.com \
/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.