Skip to content

L
linux
Switch branch/tag
  1. 18 Dec, 2020 5 commits
    • Chuck Lever's avatar
      NFSD: Restore NFSv4 decoding's SAVEMEM functionality · 7b723008
      Chuck Lever authored 
      While converting the NFSv4 decoder to use xdr_stream-based XDR
processing, I removed the old SAVEMEM() macro. This macro wrapped
a bit of logic that avoided a memory allocation by recognizing when
the decoded item resides in a linear section of the Receive buffer.
In that case, it returned a pointer into that buffer instead of
allocating a bounce buffer.

The bounce buffer is necessary only when xdr_inline_decode() has
placed the decoded item in the xdr_stream's scratch buffer, which
disappears the next time xdr_inline_decode() is called with that
xdr_stream. That happens only if the data item crosses a page
boundary in the receive buffer, an exceedingly rare occurrence.

Allocating a bounce buffer every time results in a minor performance
regression that was introduced by the recent NFSv4 decoder overhaul.
Let's restore the previous behavior. On average, it saves about 1.5
kmalloc() calls per COMPOUND.
Signed-off-by: default avatarChuck Lever <chuck.lever@oracle.com>
      7b723008
    • Chuck Lever's avatar
      SUNRPC: Handle TCP socket sends with kernel_sendpage() again · 4a85a6a3
      Chuck Lever authored 
      Daire Byrne reports a ~50% aggregrate throughput regression on his
Linux NFS server after commit da1661b9 ("SUNRPC: Teach server to
use xprt_sock_sendmsg for socket sends"), which replaced
kernel_send_page() calls in NFSD's socket send path with calls to
sock_sendmsg() using iov_iter.

Investigation showed that tcp_sendmsg() was not using zero-copy to
send the xdr_buf's bvec pages, but instead was relying on memcpy.
This means copying every byte of a large NFS READ payload.

It looks like TLS sockets do indeed support a ->sendpage method,
so it's really not necessary to use xprt_sock_sendmsg() to support
TLS fully on the server. A mechanical reversion of da1661b9 is
not possible at this point, but we can re-implement the server's
TCP socket sendmsg path using kernel_sendpage().
Reported-by: default avatarDaire Byrne <daire@dneg.com>
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=209439Signed-off-by: default avatarChuck Lever <chuck.lever@oracle.com>
      4a85a6a3
    • Chuck Lever's avatar
      NFSD: Fix sparse warning in nfssvc.c · d6c9e436
      Chuck Lever authored 
      fs/nfsd/nfssvc.c:36:6: warning: symbol 'inter_copy_offload_enable' was not declared. Should it be static?

The parameter was added by commit ce0887ac ("NFSD add nfs4 inter
ssc to nfsd4_copy"). Relocate it into the source file that uses it,
and make it static. This approach is similar to the
nfs4_disable_idmapping, cltrack_prog, and cltrack_legacy_disable
module parameters.
Signed-off-by: default avatarChuck Lever <chuck.lever@oracle.com>
      d6c9e436
    • Trond Myklebust's avatar
      nfsd: Don't set eof on a truncated READ_PLUS · b68f0cbd
      Trond Myklebust authored 
      If the READ_PLUS operation was truncated due to an error, then ensure we
clear the 'eof' flag.

Fixes: 9f0b5792 ("NFSD: Encode a full READ_PLUS reply")
Signed-off-by: default avatarTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: default avatarChuck Lever <chuck.lever@oracle.com>
      b68f0cbd
    • Trond Myklebust's avatar
      nfsd: Fixes for nfsd4_encode_read_plus_data() · 72d78717
      Trond Myklebust authored 
      Ensure that we encode the data payload + padding, and that we truncate
the preallocated buffer to the actual read size.

Fixes: 528b8493 ("NFSD: Add READ_PLUS data support")
Signed-off-by: default avatarTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: default avatarChuck Lever <chuck.lever@oracle.com>
      72d78717
  3. 16 Dec, 2020 35 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7