Re: [PATCH v4 7/9] IB/core: generic RDMA READ/WRITE API

On 03/10/2016 08:21 AM, Christoph Hellwig wrote:

I see the use case for enabling MRs for debugging, but I still don't
think the ULP is the right place.  I also don't see where you
cna do this currently: for srpt, the only driver converted in this
series there is no support for registrations at all currently,
and with the the series you can force it to register with a one
line code change in rdma_rw_use_mr.  For iSERt, which I plan to
convert for 4.7 (and which has some WIP code in git) MR support
exists, but is only enabled if the device supports the vendor
specific signature handover extension (aka if running on mlx5 hardware).
With the future conversion you'll still have a simply compile time
switch to always enabled it, and I'd be fine with adding a module
parameter to force it as well.

Hello Christoph,

The above is fine with me. But when I ran a test with rdma_rw_use_mr() changed into "return true" the following error messages appeared in the kernel log:

[  364.460709] ib_srpt 0x1: parsing SRP descriptor table failed.
[  383.604809] ib_srpt 0x0: parsing SRP descriptor table failed.
[  383.605627] ib_srpt 0x2: parsing SRP descriptor table failed.
[  386.702905] ib_srpt 0x3: parsing SRP descriptor table failed.
[  386.703092] ib_srpt 0x4: parsing SRP descriptor table failed.
[  386.703242] ib_srpt 0x5: parsing SRP descriptor table failed.
[  386.703411] ib_srpt 0x6: parsing SRP descriptor table failed.

Is this expected? I ran this test on a server equipped with two mlx4 HCAs with latest firmware (2.36.5000). I installed git commit c4c65482b56a433a82bc5b63db8ba125727e9f80 of the rdma-rw-api merged with v4.5-rc7. Initiator and target drivers were running on the same server and were communicating with each other via loopback. Before I modified rdma_rw_use_mr() the same test passed on the same setup.

Do you perhaps want me to look further into this?

Thanks,

Bart.
--
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