| 771c18a5 |
22-Oct-2024 |
mrg <mrg@NetBSD.org> |
sort-of-fix Mesa on most 32-bit platforms.
the new sparse_array implementation in Mesa 21 (not present in Mesa 19,
as shipped with netbsd-10) triggers it asking for an __atomic_load_8()
function, alongside it's __atomic_fetch_add_8/__atomic_fetch_sub_8/
__sync_val_compare_and_swap_8 versions on 32-bit platforms, so add a
__atomic_load_8() workaround that uses the same method as the others
to provide sync. (i did check, and almost the accesses in mesa that
this accessor is used with do go via the API, the only ones not used
are as part setup/clean up.
tested by rin@ on macppc. |
| e4ae85df |
20-Dec-2014 |
mrg <mrg@NetBSD.org> |
fix my prior hack here to only apply on platforms that need it.
namely, any current netbsd 64 bit platform besides x86. (i meant
to do this before merging.)
the problem here is the "Node" type is only 4 bytes, an allocation
is assigned to a "Node *node", and then "node + 1" is returned.
this returns a 4-byte aligned pointer, and some callers expect to
be able to store 64-bit (pointer, at least) values there and demand
that the structure be properly aligned. this change forces Node to
be 8-byte aligned on all-but-x86 64 bit netbsd platforms. |