[openib-general] [openfabrics-ewg] RHEL5 and OFED ...

[Doug Ledford]

On Fri, 2006-10-20 at 17:01 +0200, Michael S. Tsirkin wrote:

> > OK.  The total diffstat is large enough that it's going to be a very
> > hard sell to include.  I'll try, but it'll be hard.
> 
> Actually, the only large patch by far is the ipath-fixes.patch:
> 
> Some of these patches are already in 2.6.18.
> 
> wc -l *.patch

This doesn't equate to the update I had to do.  Keep in mind that the
2.6.18.1 kernel didn't have either the SDP or ehca modules, and as such
our previous kernel grabbed those two from the OFED 1.0 code.  So, the
diffstat just to bring them both up to OFED 1.1 status so that the
remaining smaller patches would apply was large.  In addition, the
previous kernel was missing the rdma_ucma module entirely, so that's
another largish diff prior to applying these patches.  So, my diffstat
is much larger than this.

>      39 cm_add_mra_timeout_limit.patch
>      72 cma_established1.patch
>      18 cma_increase_max_cm_retries.patch
>      24 cma_list_init.patch
>      20 cma_mem_leak.patch
>      38 cma_race_fix.patch
>      33 cma_tavor_quirk.patch
>      18 cm_cleanup_timewait.patch
>      21 ib_sa_names.patch
>   11273 ipath-fixes.patch
>     194 ipath-limit-packets-sent-without-ack.patch
>     169 ipath-memcpy_cachebypass.patch
>      13 ipath-x86_64.patch
>      44 ipoib_issue3.patch
>      47 ipoib_mcast_join_mask.patch
>      28 ipoib_mcast_restart.patch
>      74 ipoib_selector_updated.patch
>      66 ishai_srp_attributes.patch
>      57 ishai_srp_remove_reconnect.patch
>      71 ishai_srp_wa_post_send.patch
>      20 lockdep_header.patch
>      18 mthca_av_statrate.patch
>     314 mthca_catas_reset.patch
>      18 mthca_mad_traps.patch
>      18 mthca_query_port.patch
>      38 mthca_query_qp_portnum.patch
>      17 mthca_query_qp_statrate_bits.patch
>      29 mthca_use_uar2.patch
>      96 robert-ipath-diagpkt-init-fixup.patch
>      39 sdp_credits_by_seq.patch
>      52 sdp_post_credits.patch
>     157 sean_cma_establish.patch
>     104 sean_cma_hotplug.patch
>      17 sean_cma_typo_fix.patch
>      68 sean_cm_drep_on_not_found.patch
>     101 sean_cm_randomize_psn.patch
>     199 sean_cm_unload_crash.patch
>      83 srp_1_recreate_at_reconnect.patch
>     112 srp_2_use_multiple_initiator_ports.patch
>     121 svnehca_0015_1.patch
>      12 svnehca_0015_2.patch
> 
> and despite the name it is actually adding a new features -
> such as mmap of kernel buffers into user space - not purely a bug fix.
> So, how about pushing all the rest for starters?
> 
> 
-- 
Doug Ledford <dledford at redhat.com>
              GPG KeyID: CFBFF194
              http://people.redhat.com/dledford

Infiniband specific RPMs available at
              http://people.redhat.com/dledford/Infiniband
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.openfabrics.org/pipermail/general/attachments/20061020/54c6ad35/attachment.sig>