On Mon, Sep 16, 2024 at 08:18:16PM +0200, Salvatore Bonaccorso wrote: > > > > On Mon, Sep 16, 2024 at 01:54:35PM +0200, Steve Dickson wrote: > > > > > It did because it was not in the appropriate format... The patch > > > > > was an attachment, not in-line, no Signed-off-by: line and > > > > > the patch was not create by git format-patch command (which > > > > > adds PATCH in the subject line). > > > > (not even by reference to the Linux kernel tree). > > > > > > Thank you all for looking into it. Steve, do you need to have it > > > re-submitted in a git format-patch format? At least a Signed-off-by > > > line by Sergio would be needed in my understanding. > > > > I guess otherwise we can use soemthing like the following though a > > Signed-off-by is probably not right here, or is it enough if I say > > Reported-by: and Patch-originally-by: although the later is not an > > official kernel doc mentioned tag? I don't object to (nor do I require) your adding a Signed-off-by: Sergio Gelato <Sergio.Gelato@xxxxxxxxxxx> (I am indeed the author of the patch, and it's too trivial to be copyrightable anyway). I guess Neil's comment earlier can be taken as equivalent to an Acked-by:. By the way, I have wondered whether to change the other two instances of "goto out" in dirscancb() as well. For the first one (calloc() failure) I think this would be counterproductive (one is likely to get the same error on all subsequent iterations), the second (directory open()) is more debatable (but failure of the directory open() should be so rare that it doesn't matter). > And then one seems to be able to do so many mistakes around one patch. > Here is an improved version (v2), fixing a typo in the Link reference > and using a Closes: tag after the Reported-by: > > This now should be better than the previous one. > > Regards, > Salvatore > > From 46989ce210d27dbb3ad89095030c687db70c78be Mon Sep 17 00:00:00 2001 > From: Sergio Gelato <Sergio.Gelato@xxxxxxxxxxx> > Date: Tue, 4 Jun 2024 16:02:59 +0200 > Subject: [PATCH v2] rpc.idmapd: nfsopen() failures should not be fatal > > dirscancb() loops over all clnt* subdirectories of /run/rpc_pipefs/nfs/. > Some of these directories contain /idmap files, others don't. nfsopen() > returns -1 for the latter; we then want to skip the directory, not abort > the entire scan. > > Reported-by: Sergio Gelato <sergio.gelato@xxxxxxxxxxx> > Closes: https://lore.kernel.org/linux-nfs/ZmCB_zqdu2cynJ1M@xxxxxxxxxxx/ > Link: https://bugs.debian.org/1072573 > Patch-originally-by: Sergio Gelato <sergio.gelato@xxxxxxxxxxx> > Signed-off-by: Salvatore Bonaccorso <carnil@xxxxxxxxxx> > --- > Changes in v2: > - Fix typo in URL for Debian bug reference > - Reorder patch tags and use Closes tag after Reported-by > > utils/idmapd/idmapd.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/utils/idmapd/idmapd.c b/utils/idmapd/idmapd.c > index cd9a965f8fbc..5231f56d24a8 100644 > --- a/utils/idmapd/idmapd.c > +++ b/utils/idmapd/idmapd.c > @@ -556,7 +556,7 @@ dirscancb(int fd, short UNUSED(which), void *data) > if (nfsopen(ic) == -1) { > close(ic->ic_dirfd); > free(ic); > - goto out; > + continue; > } > > if (verbose > 2) > -- > 2.45.2 >
