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]

[Ada] fix bug in handling of Altivec


Manually tested on powerpc-aix
Tested on i686-linux, committed on trunk.

We used to ceil-round the entry address to the closest multiple of 16
on every target.  This makes no sense on targets where VECTOR_ALIGNMENT
is smaller than 16 because the address of a properly aligned vector could
be affected, which should never happen.

The fix applied here consists in using VECTOR_ALIGNMENT instead of a
harcoded 16 as the rounding factor.

2006-10-31  Olivier Hainque  <hainque@adacore.com>

	* g-alleve.adb (lvx, stvx): Ceil-Round the Effective Address to the
	closest multiple of VECTOR_ALIGNMENT and not the closest multiple of 16.

Attachment: difs
Description: Text document


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