Hi Roland, I got another question about your code for mthca HCA. In the "mthca_provider.c", the functions "mthca_create_cq(struct ib_device *ibdev, int entries, struct ib_ucontext *context, struct ib_udata *udata)" and "mthca_alloc_pd(struct ib_device *ibdev, struct ib_ucontext *context, struct ib_udata *udata)" all have a parameter "ib_ucontext*", while in the "Verbs.c", the calling functions "ib_create_cq" and "ib_alloc_pd" all pass "NULL" to this parameter, is the " ib_ucontext *" only needed for fast path operation, in which situation the user application can write directly to user space? If it's true, how to set up the fast path operation in the driver environment?
Thanks! Yicheng Roland Dreier <[EMAIL PROTECTED]> 01/04/2008 02:45 PM To Yicheng Jia <[EMAIL PROTECTED]> cc [email protected], Jack Morgenstein <[EMAIL PROTECTED]> Subject Re: [ofa-general] synchronize commands issued to MTHCA > I'm using Duo-core Xeon and I just grep the source of "mmiowb()" in kernel > 2.6.23 include/asm-x86_64 /io.h and found that this function does nothing > on x86_64 platform, is it true? Yes -- this is why I kept referring to large SGI Altix systems. - R. _____________________________________________________________________________ Scanned by IBM Email Security Management Services powered by MessageLabs. For more information please visit http://www.ers.ibm.com _____________________________________________________________________________
_______________________________________________ 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
