3 Mar
2010
3 Mar
'10
5:12 p.m.
Eric Blake wrote:
According to Jim Meyering on 3/3/2010 2:29 AM:
ACK, looks fine !
I pushed this, and then (sorry I missed it the first time) noticed that we'd rather avoid that new use of errno. errno is not defined for a shorter-than-expected read, so including strerror(errno) in the diagnostic would be misleading.
Good point. Were you planning on the followup patch for this, or do I need to pick up the slack?
I figured you'd do it while addressing the malicious input problem a few lines below.