[Patch mach-o 3/3] section directives needed to support GCC
Iain Sandoe
developer@sandoe-acoustics.co.uk
Tue Dec 13 10:01:00 GMT 2011
More information about the Binutils mailing list
Tue Dec 13 10:01:00 GMT 2011
- Previous message (by thread): [Patch mach-o 3/3] section directives needed to support GCC
- Next message (by thread): [Patch mach-o 3/3] section directives needed to support GCC
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Hi Tristan, Thanks for reviewing the three patches - apologies that the first one was so large. ... I would normally try to make smaller bites - but I guess I felt that it did all belong together (a lot of the bulk is in repetitive table stuff). Do you wish me to repost after rebasing and addressing your points - (or I could send to you as a diff, if you intend to apply it). On 13 Dec 2011, at 09:50, Tristan Gingold wrote:e > > On Dec 13, 2011, at 3:23 AM, Iain Sandoe wrote: > >> >> mach-o defaults to starting up with only the text section defined - >> the data and bss sections are created on demand and not emitted >> otherwise. >> >> further, it is possible to suppress even the text section, such >> that an object file can contain only specialized sections >> (TODO to implement the c/l switch to do this). > > As this touches common code, I can't approve it. > > An alternative is to remove these sections unless they are not empty > (ok, slightly different semantic). Here is what I wrote to achieve > that: I didn't do that because ISTM that we can't tell the difference between gas inserting them by default - and the User intentionally defining them as empty (for some devious reason). I suppose we could keep track of whether the User does define them.... WDYT? Iain
- Previous message (by thread): [Patch mach-o 3/3] section directives needed to support GCC
- Next message (by thread): [Patch mach-o 3/3] section directives needed to support GCC
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Binutils mailing list