[Patch/RFC] libstdc++/19495

[Paolo Carlini]

Hi everyone, hi Benjamin,

In order to "fix" for 4.0 this PR (basically, ext/array_allocator/2.cc
fails on powerpc), one possible approach is the following:

1- Generalize a bit ext/array_allocator to avoid using begin() and size(),
thus accepting also memory allocated via tr1/aligned_storage not only
tr1/array. Seems to me a nice idea, per se.

2- Tweak the ext/array_allocator testcases to actually use 
tr1/aligned_storage,
which (at variance with tr1/array) allows to specify the alignment.

For 4.1, or when the ABI can change, anyway, we can consider changing
basic_string to make use of memory aligned only as CharT requires, not
more, but, for the time being, the below seems to me a viable approach...

Opinions? Disagreements?

(regtested x86/x86_64)

Paolo.

////////////
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: draft_19495
URL: <http://gcc.gnu.org/pipermail/libstdc++/attachments/20050119/78dcd822/attachment.ksh>