On Tue, Dec 19, 2023 at 05:21:39AM +0100, Christoph Hellwig wrote: > On Mon, Dec 18, 2023 at 02:39:02PM -0800, Darrick J. Wong wrote: > > > - trace_xfs_attr_sf_lookup(args); > > > - > > > ASSERT(ifp->if_format == XFS_DINODE_FMT_LOCAL); > > > sfe = &sf->list[0]; > > > for (i = 0; i < sf->hdr.count; > > > @@ -905,6 +903,9 @@ xfs_attr_shortform_getvalue( > > > int i; > > > > > > ASSERT(args->dp->i_af.if_format == XFS_DINODE_FMT_LOCAL); > > > + > > > + trace_xfs_attr_sf_lookup(args); > > > > Shouldn't this get renamed to trace_xfs_attr_shortform_getvalue to match > > the function? Especially since xfs_attr_shortform_lookup disappears > > later, AFAICT. > > If we value accurate naming over being able to use a historical > trace point: yes. Although in that case I'd probably structure it > as a patch adding the new xfs_attr_shortform_getvalue tracepoint only, > and removing the xfs_attr_sf_lookup one with the function. Eh, tracepoint names are greppable enough. Reviewed-by: Darrick J. Wong <djwong@xxxxxxxxxx> --D
