Hi Nathan,
On 07/08/2014 07:04 AM, Nathan Scott wrote:
Hi Paul,
Small stuff only, as always - how does the attached patch look to you?
It has the following two little changes:
- the gfs2.control metrics seem to be defined as U32, so we should put
'em into the pmAtomValue "ul" field (this only becomes visible on big
endian boxes, benign otherwise). gfs2_control_check_value comment is
updated to match its existing return code behaviour.
- tweaks an explicit tracing file location to use the table, and also
get the compiler to work out the string size instead of hard-coding
it in the strncmp call (40).
Feel free to merge it into your tree if it looks ok & I'll pull from
there.
cheers.
The patch looks good to me and has been merged in my tree :). Have
checked and
also found that the other two metric pmAtomValue fields should have been
"ul"
in the fetch also, thanks for pointing that out.
--
Nathan
Cheers,
Paul
|