From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: From: Luis Chamberlain Message-ID: <20200619204626.GK11244@42.do-not-panic.com> References: <20200610154923.27510-1-mcgrof@kernel.org> <20200617174348.70710c3ecb14005fb1b9ec39@linux-foundation.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20200617174348.70710c3ecb14005fb1b9ec39@linux-foundation.org> Subject: Re: [Bridge] [PATCH 0/5] kmod/umh: a few fixes List-Id: Linux Ethernet Bridging List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Date: Fri, 19 Jun 2020 20:46:30 -0000 Cc: ast@kernel.org, jarkko.sakkinen@linux.intel.com, linux-kernel@vger.kernel.org, dhowells@redhat.com, keyrings@vger.kernel.org, christian.brauner@ubuntu.com, bridge@lists.linux-foundation.org, jmorris@namei.org, kuba@kernel.org, serge@hallyn.com, keescook@chromium.org, nikolay@cumulusnetworks.com, roopa@cumulusnetworks.com, josh@joshtriplett.org, slyfox@gentoo.org, viro@zeniv.linux.org.uk, axboe@kernel.dk, chainsaw@gentoo.org, ravenexp@gmail.com, gregkh@linuxfoundation.org, philipp.reisner@linbit.com, linux-security-module@vger.kernel.org, linux-fsdevel@vger.kernel.org, lars.ellenberg@linbit.com, davem@davemloft.net On Wed, Jun 17, 2020 at 05:43:48PM -0700, Andrew Morton wrote: > On Wed, 10 Jun 2020 15:49:18 +0000 "Luis R. Rodriguez" wrote: > > > Tiezhu Yang had sent out a patch set with a slew of kmod selftest > > fixes, and one patch which modified kmod to return 254 when a module > > was not found. This opened up pandora's box about why that was being > > used for and low and behold its because when UMH_WAIT_PROC is used > > we call a kernel_wait4() call but have never unwrapped the error code. > > The commit log for that fix details the rationale for the approach > > taken. I'd appreciate some review on that, in particular nfs folks > > as it seems a case was never really hit before. > > > > This goes boot tested, selftested with kmod, and 0-day gives its > > build blessings. > > Any thoughts on which kernel version(s) need some/all of these fixes? Well, in so far as fixes, this is the real important part: * request_module() used to fail with an error code of 256 when a module was not found. Now it properly returns 1. * fs/nfsd/nfs4recover.c: we never were disabling the upcall as the error code of -ENOENT or -EACCES was *never* properly checked for error code Since the request_module() fix is only affecting userspace for the kmod tests, through the kmod test driver, ie, we don't expose this to userspace in any other place, I don't see that as critical. Let me be clear, we have a test_kmod driver which exposes knobs and one of the knobs lets userspace query the return value of a request_module() call, and we use this test_kmod driver to stress test kmod loader. Let us also recall that the fix is *iff* an error *did* occur. I *cannot* think of a reason why this would be critical to merge to older stable kernels for this reason for request_module()'s sake. Bruce, Chuck: But... for NFS... I'd like the NFS folks to really look at that and tell us is some folks really should care about that. I also find it perplexing there was a comment in place there to *ensure* the error was checked for, and so it seemed someone cared for that condition. > > drivers/block/drbd/drbd_nl.c | 20 +++++------ > > fs/nfsd/nfs4recover.c | 2 +- > > include/linux/sched/task.h | 13 ++++++++ > > kernel/kmod.c | 5 ++- > > kernel/umh.c | 4 +-- > > lib/test_kmod.c | 2 +- > > net/bridge/br_stp_if.c | 10 ++---- > > security/keys/request_key.c | 2 +- > > tools/testing/selftests/kmod/kmod.sh | 50 +++++++++++++++++++++++----- > > I'm not really sure who takes kmod changes - I'll grab these unless > someone shouts at me. Greg usually takes it, but as usual, thanks for picking up the slack ;) Luis