[PATCH 3/3] posix: New Linux posix_spawn{p} implementation

[Adhemerval Zanella]

On 26-02-2016 14:33, Joseph Myers wrote:
> On Fri, 26 Feb 2016, Adhemerval Zanella wrote:
> 
>> There is an inherent race condition in pipe2 usage for architectures 
>> that do not support the syscall directly.  In such cases the a pipe plus 
> 
> But we don't have any such architectures after the increase in the minimum 
> kernel for non-x86/x86_64.  The patch should be simplified accordingly.  
> (__ASSUME_PIPE2 is still relevant in non-Linux-specific files, so can't be 
> removed completely from glibc.)
> 

Indeed, I forgot the 3.2 clean up.  I think now we can now cleanup the alpha
kernel-features that still have the tests for 2.6.33.