Re: [openib-general] [PATCH] sa_query: Remove debug print statement

On Fri, 2004-10-29 at 11:53, Roland Dreier wrote: Should we remove some of the debug output from mad.c? Sure. Are there specific ones you have in mind ? The ones that are KERN_DEBUG ? Any others ? -- Hal ___ openib-general mailing list [EMAIL

Re: [openib-general] [PATCH] sa_query: Remove debug print statement

Hal Sure. Are there specific ones you have in mind ? The ones Hal that are KERN_DEBUG ? Any others ? This alone will probably clean up my dmesg a lot: Index: infiniband/core/mad.c === --- infiniband/core/mad.c

Re: [openib-general] [PATCH] sa_query: Remove debug print statement

On Fri, 2004-10-29 at 12:21, Roland Dreier wrote: Actually how about this: Even better :-) Thanks. Applied. -- Hal ___ openib-general mailing list [EMAIL PROTECTED] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit

[openib-general] Latest IPoIB Status

Hi, I tested some very simple IP multicast test programs. The proper IB multicast groups are being joined. I used 239.0.0.1 as well as some others. This was something the gen1 implementation could not do as this hashed to the same ethernet address as 224.0.0.1 which is joined at ib network

Re: [Fwd: [openib-general] Latest IPoIB Status]

Hal I wrote a little too soon: On shutdown of the machine with Hal outstanding joins, the message: ib0: waiting on -5 multicast Hal groups is repeated and shutdown appears to hang. OK, looks like a reference counting bug. - R. ___

Re: [Fwd: [openib-general] Latest IPoIB Status]

On Fri, 2004-10-29 at 15:41, Roland Dreier wrote: OK, looks like a reference counting bug. A little more info which may help: It occured in conjunction with the multicast testing and there were some add membership failures at the socket layer. These might be the ones outstanding. It does not

Re: [Fwd: [openib-general] Latest IPoIB Status]

OK, I just fixed the bug below (left over in the port to new SA API). should work better now I hope. - R. Index: infiniband/ulp/ipoib/ipoib_multicast.c === --- infiniband/ulp/ipoib/ipoib_multicast.c (revision 1102) +++

[openib-general] [RFC] [PATCH] Remove redundant ib_qp_cap from 2 verb routines.

Hi, I know this changes the verbs interface a bit, but ... I don't see a value in the qp_cap being passed to different routines, when either ib_qp_attr or ib_qp_init_attr, both of which contain a qp_cap, are being passed at the same time. Above attached patch removes the qp_cap (saves stack

Re: [Fwd: [openib-general] Latest IPoIB Status]

On Fri, 2004-10-29 at 16:05, Roland Dreier wrote: OK, I just fixed the bug below (left over in the port to new SA API). should work better now I hope. Good find. It looks like that did the trick :-) Thanks for the fast turnaround. That makes things much nicer as rebooting is now more

[openib-general] [PATCH]code optimization in ib_register_mad_agent()

I am starting to look at the access layer code. Here is a code optimization patch in ib_register_mad_agent(). thanks Shirley Ma IBM Linux Technology Center 15300 SW Koll Parkway Beaverton, OR 97006-6063 Phone: (503) 578-7638 FAX: (503) 578-3228 access.mad.patch Description: Binary data

[openib-general] [PATCH]spinlock shouldn't be held while calling ib_post_send()

Here is the patch. thanks Shirley Ma IBM Linux Technology Center 15300 SW Koll Parkway Beaverton, OR 97006-6063 Phone: (503) 578-7638 FAX: (503) 578-3228 access.mad1.patch Description: Binary data ___ openib-general mailing list [EMAIL

Re: [openib-general] [PATCH]spinlock shouldn't be held while calling ib_post_send()

On Fri, 29 Oct 2004 18:06:47 -0600 Shirley Ma [EMAIL PROTECTED] wrote: Here is the patch. Note that my patch removes the lock when calling ib_post_send. But, holding the lock when calling ib_post_send() should be fine. Also, the current completion code assumes that the work requests are

Re: [openib-general] [PATCH]code optimization in ib_register_mad_agent()

On Fri, 29 Oct 2004 17:35:40 -0600 Shirley Ma [EMAIL PROTECTED] wrote: I am starting to look at the access layer code. Here is a code optimization patch in ib_register_mad_agent(). ib_mad_client_id must be incremented while holding the spinlock (or converted into an atomic). The rest of the

Re: [openib-general] atomic_read in ib_unregister_mad_agent()

Shirley It's better to use semaphore instead of atomic_read to Shirley check the reference count 0 in wait_event() in Shirley ib_unregister_mad_agent(). Agree? I don't see how one uses a semaphore to wait for a reference count to become zero (semaphores sleep until their count is

Re: [openib-general] [PATCH]code optimization in ib_register_mad_agent()

The inline patch was whitespace damaged and line wrapped. Is there any way to make your attachments have mime type text/x-patch? That way my client will display it inline. - R. ___ openib-general mailing list [EMAIL PROTECTED]