On Fri, 2023-11-10 at 11:28 -0500, Chuck Lever wrote: > I've chased down the reported failures with krb5i/p. Essentially > the DRC was broken for krb5i/p because of the extra crap those > security flavors stick before and after the NFS headers. Huh, so the reported failures were due to retransmissions? > > These patches do address bugs in stable kernels, but they will need > to be massaged, tested, and applied by hand to get there. Voting now > open on whether it's worth the trouble for us to do it now, or wait > until someone complains about a particular LTS problem. > > --- > > Chuck Lever (3): > NFSD: Fix "start of NFS reply" pointer passed to nfsd_cache_update() > NFSD: Update nfsd_cache_append() to use xdr_stream > NFSD: Fix checksum mismatches in the duplicate reply cache > > > fs/nfsd/cache.h | 4 +-- > fs/nfsd/nfscache.c | 87 +++++++++++++++++++++++++++------------------- > fs/nfsd/nfssvc.c | 14 ++++++-- > 3 files changed, 65 insertions(+), 40 deletions(-) > > -- > Chuck Lever > Wow, nice work tracking those down. I'm sure that wasn't easy! Reviewed-by: Jeff Layton <jlayton@xxxxxxxxxx>