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] Add generic HAVE_RM_CTX implementation


> Joseph Myers wrote:
> On Fri, 23 May 2014, Wilco wrote:
> 
> > I think it should be possible to significantly reduce the complexity
> > by removing the _f and _l variants (AFAIK all FPUs have only one set
> > of FP status flags irrespectively of precision), and conditionally
> 
> The point of these variants is to allow x86_64 to manipulate
> floating-point state only for SSE for functions not using long double (for
> example).

The idea would be to keep SET_RESTORE_ROUNDL/SET_RESTORE_ROUND_53BIT, and
conditionally define these to SET_RESTORE_ROUND with generic libc_* calls.
If required, targets can then override these to use different functions.
This means we can remove all the defines, indirections and confusing names
from the generic code as well as the targets.

Wilco






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