From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Steve Wise" Subject: RE: [PATCH for-next v3 0/2] iw_cxgb4 pr_debug() changes for 4.14-rc Date: Wed, 27 Sep 2017 09:13:31 -0500 Message-ID: <006301d3379a$cc4dac10$64e90430$@opengridcomputing.com> References: <1506497750-19972-1-git-send-email-bharat@chelsio.com> <1506519604.82712.19.camel@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8BIT Return-path: In-Reply-To: <1506519604.82712.19.camel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> Content-Language: en-us Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: 'Doug Ledford' , 'Potnuri Bharat Teja' Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-rdma@vger.kernel.org > > On Wed, 2017-09-27 at 13:05 +0530, Potnuri Bharat Teja wrote: > > Hi Doug, > > Current "v3" series is generated on top of your "for-next" branch > > overriding previous submissions. > > Please consider these for smoother application of subsequent patches. > > > > Changes since "v2" > > - Revert pr_debug()->pr_warn() change in fw4_ack(). > > > > Changes since original submission > > - Make patches on top of "for-next" branch. > > OK, I've taken your v3 of these patches. But something that jumped out > at me as a big issue, far bigger than I realized, is just how many > places in the iw_cxgb4 driver do BUG()s. Unless the official answer is > "This can never happen" (in which case we just need to remove the > conditional and the BUG() behind the conditional, because they can > never happen), someone needs to spend some time making the iw_cxgb4 > driver more tolerant of things that it might not like, but which > shouldn't bring the computer to a grinding halt. In particular, I > think things like if (skb_cloned(skb)) BUG() does not seem appropriate > to me. Are you really telling me that you can't find a way to > gracefully drop the skb you are working on and simply fail the current > operation instead of BUG()ing? Hey Doug, I agree. We will clean up the BUG()s and remove them if they really aren't needed, or perhaps make them WARN_ONCE(), and fail the operations w/o crashing. Steve. -- 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