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]

Re: [PATCH v2] handle sem_t with ILP32 and __HAVE_64B_ATOMICS

On Tue, 27 Jan 2015, Carlos O'Donell wrote:

> > I am certainly willing to use 32-bit atomics here for now.  The absolutely
> > minimal change is to change tilegx32 to set __HAVE_64B_ATOMICS to 0.
> > Obviously the AArch64 ILP32 mode (out of tree) will want to do the same.
> > And I don't know whether m68k or mips require a similar change or if
> > they can tolerate unaligned 64-bit atomic operations.  So arguably if
> > we can decide that my proposed re-aligning change is best, then it also
> > automatically handles all the other platforms as well.
> I would like several more rounds of review of your realigning code.
> You have my permission to checkin the minimal change for tilegx32.
> Similarly other ILP32 with 64-bit atomic arches should do the same,
> and we can continue this conversation without the pressure of the
> time boxed release.

Is __HAVE_64B_ATOMICS used *only* for semaphores?  Because if there are 
other effects of changing it, doing so seems risky - it would be better to 
have a define such as SEM_USE_64B_ATOMICS, and make that default to 0 for 
ILP32 architectures.

Joseph S. Myers

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