[openib-general] [PATCH][RFC][0/4] InfiniBand userspace verbs implementation

Bernhard Fischer blist at aon.at
Wed May 18 10:05:02 PDT 2005


On Mon, Apr 04, 2005 at 03:09:00PM -0700, Roland Dreier wrote:
>Here is an initial implementation of InfiniBand userspace verbs.  I
>plan to commit this code to the OpenIB repository shortly, and submit
>it for inclusion during the 2.6.13 cycle, so I am posting it early for
>comments.
[snip]
>Please take a look at this code if you have a chance.  I would
>appreciate high-level criticism of the design and implementation as
>well as nitpicky complaints about coding style and typos.

Can we eventually have checks for NULL before dereferencing members? I'd
be happy if they would be wrapped in #ifdef DEBUG or the like, so
common use would not suffer the additional penalty those checks imply.

e.g.:
ibv_modify_qp(NULL,...) etc, etc i.e. ibv_() and also the API exposed by
mthca.
Something like
ibverbs:verbs.c:145: Error: ibv_modify_qp called with qp==NULL
Not pretty, agree, but would be convenient for testing new users of
those libraries.

Just a suggestion.

>
>In particular, the memory pinning code in in uverbs_mem.c could stand
>a looking over.  In addition, a sanity check of the write()-based
>scheme for passing commands into the kernel in uverbs_main.c and
>uverbs_cmd.c is probably worthwhile.



More information about the general mailing list