This is the mail archive of the mailing list for the glibc project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]


On Fri, 21 Aug 2015, H.J. Lu wrote:

> My first patch is mechanical and there are no changes in generated code.

It's not mechanical, as evidenced by all the problems people have spotted 
looking at the code (presumably cases that don't apply on i386) and the 
random inclusion of *_hidden_* changes with no apparent relation to the 
main point of the patch.

> We can either revert my changes or fix my changes.
> I believe my change is an improvement and we should fix its shortcomings.

I think the fix process should not happen on master, but a separate 
branch, with the changes being reverted until they have consensus.

This isn't a matter of fixing isolated bugs that a reviewer missed - if it 
were just that, then fixing on master would be fine.  It's a matter of 
there being no consensus on the basic design of the changes or the overall 
shape of the new interfaces being introduced (and no detailed review of 
the changes either).

Joseph S. Myers

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]