* lib/base32.c, lib/base64.c: Include ialloc.h instad of stdlib.h.
Include intprops.h, verify.h.
* lib/base32.c (base32_encode, base32_encode_alloc, get_8, decode_8)
(base32_decode_ctx, base32_decode_alloc_ctx):
* lib/base32.h (struct base32_decode_context):
* lib/base64.c (base64_encode_fast, base
On 8/27/21 2:56 AM, Bruno Haible wrote:
So, to make the __access__ attribute [1] really consistent, they would need to
add a fifth access-mode 'any', that would not restrict the access pattern.
Yes, something like that might make sense. I would suggest something
more general, though, along th
Hi Paul,
Thanks for the explanations.
> any of the access modes 'read', 'write',
> 'read_write', 'none' are incorrect for some valid and reasonable regexec
> calls.
So, to make the __access__ attribute [1] really consistent, they would need to
add a fifth access-mode 'any', that would not rest
On 8/26/21 4:58 PM, Bruno Haible wrote:
Oops, my suggestion was not right. Instead, what needs to be done
in order to preserve good diagnostics with GCC 10 is to keep the
__access__ attribute, even with the VLA.
Yes, it appears I was too ambitious in my patch. I have reverted much of
it and th
