Is it possible that only one vector will be supported by the system, yielding the function pci_enable_msix to return 1?
If so, the stack will not work properly when using only 1 vector. Ron On Mon, Dec 22, 2008 at 10:13 AM, Yevgeny Petrilin <[email protected]> wrote: > Roland, > >> We encountered a problem that when a machine didn't support the >> required number of vectors (nvec), >> instead of trying to get 2 vectors like in the previous version, it >> didn't use MSI-X at all - causing a major performance degradation. >> Maybe in a case of failure we should try lowering the number of >> vectors to 2 (like in the previous version) or the return value of >> pci_enable_msix and goto no_msi only in case of a second failure. >> >> Ron > I agree with Ron on this issue, trying to get the number of vectors that was > returned by > pci_enable_msix seems to be the better solution in case of failure. > > As for the rest of the patch, I didn't find any major problems, except for a > small typo: >> + mdev->profile.prof[i].rx_ring_num = >> dev->caps.num_comp_vectors;; > The driver worked fine with the new addition. > > Thanks, > Yevgeny > > > _______________________________________________ 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
