Skip to content

Conversation

@apache-hb
Copy link
Contributor

Adds version 0.5.2 of aklomps base64 library

)

cc = meson.get_compiler('c')
tests_opt = get_option('tests').disable_auto_if(meson.is_subproject())
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@apache-hb apache-hb force-pushed the base64-5.2.0 branch 2 times, most recently from 25e41d9 to 921ea29 Compare November 2, 2025 17:53
foreach feature, flags : arch_flags
present = is_x86 and cc.has_multi_arguments(flags)
config.set10('BASE64_WITH_' + feature, present)
present_flags += {
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You could simplify by removing present_flags entirely and doing arch_flags[feature] = [] if not present.

Comment on lines +47 to +49
config.set10('BASE64_WITH_SSSE3', is_x86)
config.set10('BASE64_WITH_SSE41', is_x86)
config.set10('BASE64_WITH_SSE42', is_x86)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

cc.has_multi_arguments([]) evaluates to true, so these are redundant.

c_args += ['-DBASE64_STATIC_DEFINE']
endif

neon32_supported = host_machine.cpu_family() == 'arm' and cc.has_argument(
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Couldn't arm and aarch64 use the arch_flags mechanism? I think the is_x86 check isn't actually necessary because the compiler should just fail to support the options that aren't relevant for this CPU.

]
},
"aklomp-base64": {
"_comment": "The test program has a bug that causes it to fail with a SIGILL on x86 platforms without avx512 (all CI runners)",
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Has this been reported upstream? At minimum we should link to the issue report so it's possible to follow up. Is there any way to work around the bug from meson.build without patching the source?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants