Commit 7578b2f6 authored by Chuck Lever's avatar Chuck Lever

NFSD: Remove be32_to_cpu() from DRC hash function

Commit 7142b98d ("nfsd: Clean up drc cache in preparation for
global spinlock elimination"), billed as a clean-up, added
be32_to_cpu() to the DRC hash function without explanation. That
commit removed two comments that state that byte-swapping in the
hash function is unnecessary without explaining whether there was
a need for that change.

On some Intel CPUs, the swab32 instruction is known to cause a CPU
pipeline stall. be32_to_cpu() does not add extra randomness, since
the hash multiplication is done /before/ shifting to the high-order
bits of the result.

As a micro-optimization, remove the unnecessary transform from the
DRC hash function.
Signed-off-by: default avatarChuck Lever <chuck.lever@oracle.com>
parent 23a1a573
...@@ -87,7 +87,7 @@ nfsd_hashsize(unsigned int limit) ...@@ -87,7 +87,7 @@ nfsd_hashsize(unsigned int limit)
static u32 static u32
nfsd_cache_hash(__be32 xid, struct nfsd_net *nn) nfsd_cache_hash(__be32 xid, struct nfsd_net *nn)
{ {
return hash_32(be32_to_cpu(xid), nn->maskbits); return hash_32((__force u32)xid, nn->maskbits);
} }
static struct svc_cacherep * static struct svc_cacherep *
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment