On Thu, June 25, 2009 10:00 pm, [email protected] wrote: > Send general mailing list submissions to > [email protected] > > To subscribe or unsubscribe via the World Wide Web, visit > http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general > or, via email, send a message with subject or body 'help' to > [email protected] > > You can reach the person managing the list at > [email protected] > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of general digest..." > > > Today's Topics: > > 1. Re: Re: [PATCH v2 RESEND] rdma_cm: Add debugfs entries to > monitor rdma_cm connections (Moni Shoua) > 2. Re: Re: [PATCH v2 RESEND] rdma_cm: Add debugfs entries to > monitor rdma_cm connections (Moni Shoua) > 3. OFEM installation problem (Boyan Lazarov) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Thu, 25 Jun 2009 18:23:27 +0300 > From: Moni Shoua <[email protected]> > Subject: Re: [ofa-general] Re: [PATCH v2 RESEND] rdma_cm: Add debugfs > entries to monitor rdma_cm connections > To: Jason Gunthorpe <[email protected]> > Cc: Roland Dreier <[email protected]>, [email protected], > OpenFabrics General <[email protected]> > Message-ID: <[email protected]> > Content-Type: text/plain; charset=us-ascii > > Jason Gunthorpe wrote: >> On Wed, Jun 24, 2009 at 06:55:43PM +0300, Moni Shoua wrote: >> >>> I believe that Jason and I still disagree but... >> >>> Jason suggests that I implement this feature with netlink. This >>> approach might have an advantage but if I understand it right this >>> approach requires a patch also to some user application in order to >>> take advantage of this patch. >> >> The kernel devs have made it clear that the preferred way to export >> this information is through netlink and/or a file in /proc/net/.. >> >> You never got an answer if /proc/net is truely discouraged or not. > As I remember I got opinions against /proc/net. I'm willing to move the > virtual files to there if this is acceptable by all. Otherwise I will > find myself in the same situation. > > Roland, Sean - what do you think? > >> >> debugfs is completely useless because it cannot be used for actual >> end-user interrogation and the rules are it should not be used as a >> stable kernel-user interface. (ie is NOT a new dumping ground like >> /proc/ was) >> > I don't know about that. The only fault of debugfs that I'm aware of is > that distros don't mount it by default > >>> Also, I think that there is a value for a virtual text file under >>> debugfs to monitor rdma_cm connections easily and without any other >>> special app besides 'cat' (just like in IPoIB). >> >> If you have the netlink based program there really is no point in >> including something under debugfs. It is just bloat. >> >>> Finally, this implementation doesn't contradict netlink >>> implementation in the future (and it won't be the first time). >> >> As I said before, the stack has reached a level of maturity that new >> stuff going in should meaningfully 'move the ball forward' toward a >> mature and complete stack. In the context of state reporting that >> means a stable user interface that programs like lsof, netstat, etc, >> can rely upon. >> >> Putting something under debugfs clearly does nothing to advance that >> goal. >> >> Further, there is clearly alot of state information we should be >> exporting to userspace. RDMA-CM stuff is only a tiny portion. netlink >> is the kernel devs answer to all of this. >> > I just want to emphasize that my goal for now is to supply a simple way to > monitor rdma_cm connections, nothing more and nothing less. I don't want > it to depend on special userspace application and in fact, I just want to > add a capability like we have in many other modules. This is a feature > that we always get request for from customers and developers. > > > > > ------------------------------ > > Message: 2 > Date: Thu, 25 Jun 2009 18:24:35 +0300 > From: Moni Shoua <[email protected]> > Subject: Re: [ofa-general] Re: [PATCH v2 RESEND] rdma_cm: Add debugfs > entries to monitor rdma_cm connections > To: Thierry <[email protected]> > Cc: Roland Dreier <[email protected]>, [email protected], > OpenFabrics General <[email protected]> > Message-ID: <[email protected]> > Content-Type: text/plain; charset=UTF-8 > > Thanks :) > > > > ------------------------------ > > Message: 3 > Date: Thu, 25 Jun 2009 20:40:38 +0200 > From: Boyan Lazarov <[email protected]> > Subject: [ofa-general] OFEM installation problem > To: [email protected] > Message-ID: <[email protected]> > Content-Type: text/plain; charset="us-ascii" > > Hello, > > We have installed OFED 1.4.1 on Scientific Linux 53 (RH derivative). When > I > run the osu_* tests (OpenMPI) as root it seems that everything is running > fine and the mpi is using openib for communications. However if I run the > test as an ordinary user I am getting the following: > > WARNING: There was an error initializing an OpenFabrics device. > > and openmpi is using tcp for communications - the speed is much lower. > > Anyone with similar problem and any suggestions? > > Best regards, > Boyan > > > ------------------------------ > > _______________________________________________ > general mailing list > [email protected] > http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general > > End of general Digest, Vol 29, Issue 96 > *************************************** >
-- http://www.wbitt.com Computer bugs are like ship-wrecks. They are not found, unless they want to be found. _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
