This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC 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 [FSF mainline] darwin patch to fix size problem with generic vectors



On Sep 1, 2004, at 11:54 AM, Geoffrey Keating wrote:



Paolo is right, though, this should really read


   : (TREE_CODE (STRUCT) == VECTOR_TYPE \
      && ALTIVEC_VECTOR_MODE (TYPE_MODE (STRUCT))) \

because TARGET_ALTIVEC shouldn't change the ABI.

I agree that we should remove test for TARGET_ALTIVEC. But Paolo was also pointing to using
"int_size_for_bytes (STRUCT) == 16 instead of checking for mode. But I am not yet convinced that
this is necessary. The only case I see is for generic vectors of size 16. But then I see that the alignment
for these such vectors are set correctly with and without -maltivec with my patch as well. To conclude,
I will use above patch (after testing), unless I hear otherwise.


- fariborz




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