4 Dec
2023
4 Dec
'23
7:14 a.m.
On Sat Dec 2 07:42:43 2023 +0000, Alex Henrie wrote:
changed this line in [version 2 of the diff](/wine/wine/-/merge_requests/4539/diffs?diff_id=87632&start_sha=3faf40b879c2d2ac0c86799f507327949da060b9#3044797f8b7a2b6b857401bed33085d76589aeb3_4270_4269)
There is no reason not to continue with the function even on realloc failure, the ret->buffer value is still valid in that case. Actually, returning like that would leak the memory (in this case technically it wouldn't but it will skip reinitializing current_parameter_block).
This is all quite academic, since if the shrinking realloc fails it's very likely that the process won't last much longer. But...