This is the mail archive of the libc-alpha@sourceware.org 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]

Re: [PATCH] Use inline syscalls for non-cancellable versions


On Tue, 26 May 2015, Chris Metcalf wrote:

> Perhaps an argument for -Wshadow?  I don't know how terrible
> the fallout of another -W option would be, though :-)  Adding
> -Wshadow to my reduced testcase shows the bug nice and clearly,
> though of course I only thought to check this after I had already
> finished tracking down the bug:

See <https://sourceware.org/ml/libc-alpha/2014-10/msg00336.html>.  I think 
we should try to find a subset of Roland's changes that don't affect 
stripped installed shared libraries at all, and so are trivially safe to 
go in - then changes that only affect assertion line numbers, then look at 
those with more substantial code generation effects to understand why such 
changes occur.  (Then of course there would be further work to do before 
-Wshadow could be added by default without -Wno-error=shadow.)

-- 
Joseph S. Myers
joseph@codesourcery.com


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