From: "Hefty, Sean" <sean.hefty-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
To: "Shamis, Pavel" <shamisp-1Heg1YXhbW8@public.gmane.org>
Cc: Jack Morgenstein
<jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>,
"linux-rdma
(linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org)"
<linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"tziporet-VPRAkNaXOzVS1MOuV/RT9w@public.gmane.org"
<tziporet-VPRAkNaXOzVS1MOuV/RT9w@public.gmane.org>,
"dotanb-VPRAkNaXOzVS1MOuV/RT9w@public.gmane.org"
<dotanb-VPRAkNaXOzVS1MOuV/RT9w@public.gmane.org>,
"Jeff Squyres
(jsquyres-FYB4Gu1CFyUAvxtiuMwx3w@public.gmane.org)"
<jsquyres-FYB4Gu1CFyUAvxtiuMwx3w@public.gmane.org>,
"Shumilin,
Victor" <victor.shumilin-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
"Truschin,
Vladimir"
<vladimir.truschin-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
Devendar Bureddy
<bureddy-wPOY3OvGL++pAIv7I8X2sze48wsgrGvP@public.gmane.org>,
"mvapich-core-wPOY3OvGL++pAIv7I8X2sze48wsgrGvP@public.gmane.org"
<mvapich-core-wPOY3OvGL++pAIv7I8X2sze48wsgrGvP@public.gmane.org>
Subject: RE: [RFC] XRC upstream merge reboot
Date: Mon, 1 Aug 2011 15:03:22 +0000 [thread overview]
Message-ID: <1828884A29C6694DAF28B7E6B8A82373136F9075@ORSMSX101.amr.corp.intel.com> (raw)
In-Reply-To: <26AE60A9-D055-4D40-A830-5AADDBA20ED8-1Heg1YXhbW8@public.gmane.org>
> Actually I think it is really not so good idea manage reference counter
> across OOB communication.
But this is exactly what the current API *requires* that users of XRC do!!! And I agree, it's not a good idea. :)
> IMHO, I don't see a good reason to redefine existing API.
> I afraid, that such API change will encourage MPI developers to abandon XRC
> support.
I'm suggesting that most apps don't need to do anything special. No OOB connections, no special message exchanges carrying XRC numbers, no reference counting, no new verbs for xrc recv qps that do the same thing as the existing verbs. Using XRC becomes easier! The lifetime of the XRC receive QP is simply tied to the lifetime of the XRC domain, unless the creating process wants to destroy it sooner.
The only reason reference counting was added at all was to support a questionable usage model of sharing an xrc domain across jobs. I'm suggesting that *only those apps* that want that usage model can implement it over the provided APIs. They can continue to use OOB connections, pass XRC numbers, replace ibv_reg_recv_xrv() with atomic_inc(), and let the persistent server destroy the xrc recv qp when those processes are done. For everyone else (and it sounds like this is really everyone at this point), there's simply no need for any of that.
- Sean
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2011-08-01 15:03 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-05-16 21:13 [RFC] XRC upstream merge reboot Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373F7AB-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-05-18 14:54 ` Jack Morgenstein
[not found] ` <201105181754.33759.jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-05-18 15:27 ` Hefty, Sean
2011-06-22 7:17 ` Jack Morgenstein
[not found] ` <201106221017.06212.jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-06-22 16:14 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373029A95-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-06-22 17:03 ` Jack Morgenstein
[not found] ` <201106222003.50214.jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-06-22 18:19 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373029B3F-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-06-22 19:21 ` Jack Morgenstein
[not found] ` <201106222221.05993.jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-06-22 19:32 ` Tziporet Koren
2011-06-22 19:57 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373029BDE-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-06-23 6:11 ` Jack Morgenstein
2011-06-23 6:35 ` Jack Morgenstein
[not found] ` <201106230935.07425.jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-06-23 18:03 ` Hefty, Sean
2011-07-20 18:51 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373136F63B9-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-07-21 7:38 ` Jack Morgenstein
[not found] ` <201107211038.23000.jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-07-21 7:57 ` Jack Morgenstein
2011-07-21 11:58 ` Jeff Squyres
[not found] ` <D8276D45-5FE8-464C-B3A4-14404DE8C760-FYB4Gu1CFyUAvxtiuMwx3w@public.gmane.org>
2011-07-21 12:47 ` Jack Morgenstein
[not found] ` <201107211547.31850.jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-07-21 12:46 ` Jeff Squyres
2011-07-21 16:06 ` Hefty, Sean
2011-07-21 17:53 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373136F6691-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-07-26 20:04 ` Shamis, Pavel
[not found] ` <26AE60A9-D055-4D40-A830-5AADDBA20ED8-1Heg1YXhbW8@public.gmane.org>
2011-08-01 15:03 ` Hefty, Sean [this message]
[not found] ` <1828884A29C6694DAF28B7E6B8A82373136F9075-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-08-01 16:20 ` Shamis, Pavel
[not found] ` <AE625966-FD97-4DBF-A024-22B83B5F3E39-1Heg1YXhbW8@public.gmane.org>
2011-08-01 18:28 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373136F9194-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-08-02 10:44 ` Jack Morgenstein
[not found] ` <201108021344.25284.jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-08-02 16:29 ` Shamis, Pavel
[not found] ` <32D25205-3E9C-4757-B0AB-7117BDF3F2F7-1Heg1YXhbW8@public.gmane.org>
2011-08-03 10:37 ` Jack Morgenstein
[not found] ` <201108031337.24527.jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-08-10 22:20 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A8237316E3E55C-Q3cL8pyY+6ukrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-08-11 14:12 ` Shamis, Pavel
2011-08-21 14:42 ` Jack Morgenstein
[not found] ` <201108211742.18803.jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-08-22 15:46 ` Hefty, Sean
2011-08-02 19:08 ` Shamis, Pavel
[not found] ` <EABE213A-448A-45F8-B131-AE1EE3F9547F-1Heg1YXhbW8@public.gmane.org>
2011-08-02 21:25 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373136F962C-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-08-02 23:00 ` Shamis, Pavel
[not found] ` <DE779D97-F54F-45E4-B3D4-DBEB10F9302D-1Heg1YXhbW8@public.gmane.org>
2011-08-02 23:53 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373136F967E-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-08-03 13:47 ` Shamis, Pavel
[not found] ` <5C691E518F345F4882FAB9E9839E60BA0BCA4622F4-vxojlfkN5A++qDdrU24kdQ@public.gmane.org>
2011-08-03 15:52 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373136F97AC-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-08-03 20:22 ` Shamis, Pavel
[not found] ` <5C691E518F345F4882FAB9E9839E60BA0BCA462300-vxojlfkN5A++qDdrU24kdQ@public.gmane.org>
2011-08-03 20:49 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373136F9A13-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-08-03 21:16 ` Shamis, Pavel
[not found] ` <5C691E518F345F4882FAB9E9839E60BA0BCA462301-vxojlfkN5A++qDdrU24kdQ@public.gmane.org>
2011-08-03 21:36 ` Jason Gunthorpe
[not found] ` <20110803213642.GE28465-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2011-08-04 0:06 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373136F9A7E-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-08-04 4:05 ` Jason Gunthorpe
[not found] ` <20110804040503.GA13935-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2011-08-04 5:53 ` Hefty, Sean
2011-05-18 16:44 ` Roland Dreier
[not found] ` <BANLkTimWMU9ohSQGYEEnFR0HbBaypFR51A-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2011-05-18 17:02 ` Jason Gunthorpe
[not found] ` <20110518170226.GA2595-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2011-05-18 17:30 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373FBC7-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-05-18 18:05 ` Jason Gunthorpe
[not found] ` <20110518180519.GA11860-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2011-05-18 18:13 ` Hefty, Sean
[not found] ` <1828884A29C6694DAF28B7E6B8A82373FC13-P5GAC/sN6hmkrb+BlOpmy7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2011-05-18 18:22 ` Jason Gunthorpe
2011-05-18 19:22 ` Roland Dreier
[not found] ` <BANLkTi=cLjErM3pKzihyFtGWZ0kSu9BiPA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2011-05-19 5:29 ` Hefty, Sean
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=1828884A29C6694DAF28B7E6B8A82373136F9075@ORSMSX101.amr.corp.intel.com \
--to=sean.hefty-ral2jqcrhueavxtiumwx3w@public.gmane.org \
--cc=bureddy-wPOY3OvGL++pAIv7I8X2sze48wsgrGvP@public.gmane.org \
--cc=dotanb-VPRAkNaXOzVS1MOuV/RT9w@public.gmane.org \
--cc=jackm-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org \
--cc=jsquyres-FYB4Gu1CFyUAvxtiuMwx3w@public.gmane.org \
--cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=mvapich-core-wPOY3OvGL++pAIv7I8X2sze48wsgrGvP@public.gmane.org \
--cc=shamisp-1Heg1YXhbW8@public.gmane.org \
--cc=tziporet-VPRAkNaXOzVS1MOuV/RT9w@public.gmane.org \
--cc=victor.shumilin-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
--cc=vladimir.truschin-ral2JQCrhuEAvxtiuMwx3w@public.gmane.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.