Seperating out the routines that run only in the child (and putting them in child.c) is not a bad thing but this patch is difficult to review for several reasons:
1. The commit log did not mention the biggest change. Easy to intuit looking at the code, but it should have at least been mentioned in the commit log. 2. You broke the rule to not make more than one conceptual change to the code within a single commit. You moved code (to child.c) and made a function change (which pools were being used for what) in one commit. These changes should have been made as two seperate commits (or perhaps 3 or 4) to make review a bit less onorous. Bill > > Bill, > I am chewing on my tounge now to not be nasty. Where are you going with > this? Perhaps I missed it but was this change discussed on list? And the > commit log message describing the change is quite poor. > > Bill > > > -----Original Message----- > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] > > Sent: Monday, July 29, 2002 1:13 AM > > To: [EMAIL PROTECTED] > > Subject: cvs commit: httpd-2.0/server/mpm/winnt child.c mpm_winnt.c > > mpm_winnt.h > > > > > > wrowe 2002/07/28 22:12:50 > > > > Modified: server/mpm/winnt child.c mpm_winnt.c mpm_winnt.h > > Log: > > pconf global factors out nicely. The one other pconf appears to be > > eqivilant to pchild. > > > > Revision Changes Path > > 1.2 +3 -4 httpd-2.0/server/mpm/winnt/child.c > > >> > >
