| Message ID | 20170903114024.13183-1-samuel.thibault@ens-lyon.org |
|---|---|
| State | New |
| Headers | show |
| Series | [hurd,commited] hurd: mach/hurd/i386 also needs mach/hurd/x86 | expand |
On Sun, 3 Sep 2017, Samuel Thibault wrote:
> * sysdeps/mach/hurd/i386/Implies: Add mach/hurd/x86.
Is there an x86_64 Hurd port planned? Because x86 sysdeps directories are
for things shared between i386 and x86_64; there's nothing *wrong* with
having a mach/hurd/x86 directory in the absence of an x86_64 port, but it
doesn't seem that useful without such a port either (and without the port
I'd expect it to be hard to tell what is i386-specific and what applies to
both x86_64 and i386 and so belongs in an x86 sysdeps directory).
Joseph Myers, on lun. 04 sept. 2017 15:42:18 +0000, wrote: > On Sun, 3 Sep 2017, Samuel Thibault wrote: > > * sysdeps/mach/hurd/i386/Implies: Add mach/hurd/x86. > > Is there an x86_64 Hurd port planned? It's long way, but it's planned, yes :) Samuel
Joseph Myers, on lun. 04 sept. 2017 15:42:18 +0000, wrote: > (and without the port I'd expect it to be hard to tell what is > i386-specific and what applies to both x86_64 and i386 and so belongs > in an x86 sysdeps directory). Well, it's mostly meant to contain the equivalent of what the Linux port puts there, so we have things looking good, e.g. sysdeps/mach/hurd/x86/dl-sysdep.c which just includes sysdeps/x86/cpu-tunables.c before the arch-specific dl-sysdep.c Samuel
diff --git a/ChangeLog b/ChangeLog index a43030fe3d..ebabb12a8e 100644 --- a/ChangeLog +++ b/ChangeLog @@ -29,6 +29,7 @@ * sysdeps/x86/fpu/include/bits/fenv.h [NO_HIDDEN]: Redirect __feraiseexcept_renamed to feraiseexcept instead of __GI_feraiseexcept. + * sysdeps/mach/hurd/i386/Implies: Add mach/hurd/x86. 2017-09-01 Joseph Myers <joseph@codesourcery.com> diff --git a/sysdeps/mach/hurd/i386/Implies b/sysdeps/mach/hurd/i386/Implies new file mode 100644 index 0000000000..94db5e92ef --- /dev/null +++ b/sysdeps/mach/hurd/i386/Implies @@ -0,0 +1 @@ +mach/hurd/x86