Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Verification failures on macos-14-arm64 #294

Closed
6 tasks done
darkk opened this issue Sep 11, 2024 · 0 comments
Closed
6 tasks done

Verification failures on macos-14-arm64 #294

darkk opened this issue Sep 11, 2024 · 0 comments

Comments

@darkk
Copy link
Contributor

darkk commented Sep 11, 2024 •
edited
Loading

macOS Build highlights several verification failures:

  • superfast - Verification value 0x0C80403A ....... FAIL! (Expected 0x6306A6FE)
  • nmhash32 - Verification value 0x9905BCB1 ....... FAIL! (Expected 0x12A30553)
  • nmhash32x - Verification value 0x74B572DA ....... FAIL! (Expected 0xA8580227)
  • k-hashv32 - Verification value 0x9FBDD2B1 ....... FAIL! (Expected 0xB69DF8EB)
  • k-hashv64 - Verification value 0x70E67C61 ....... FAIL! (Expected 0xA6B7E55B)
  • polymur - Verification value 0xA612032C ....... FAIL! (Expected 0x4F894810)
Current runner version: '2.319.1'
Operating System
  macOS
  14.6.1
  23G93
Runner Image
  Image: macos-14-arm64
  Version: 20240827.4
  Included Software: https://github.com/actions/runner-images/blob/macos-14-arm64/20240827.4/images/macos/macos-14-arm64-Readme.md
  Image Release: https://github.com/actions/runner-images/releases/tag/macos-14-arm64%2F20240827.4
Runner Image Provisioner
  2.0.384.1+6d6c56aa16f1b9c7dd7935df5d63980397e44def
@darkk darkk mentioned this issue Sep 11, 2024
Merged
darkk added a commit to darkk/smhasher that referenced this issue Sep 25, 2024
@darkk
SuperFastHash: codify an UB-based Verification value :-)
b023e3b 
It's unclear to me if it should be `0` (SKIP) or just a value that
is quite common across platforms and compilers.

See rurban#294
darkk added a commit to darkk/smhasher that referenced this issue Sep 26, 2024
@darkk
polymur: fix arm builds, including macos-14-arm64
3c5b93b 
polymur Verification value 0xA612032C means that polymur_seed_init()
was not called. Currently, polymur is compiled in unconditionally,
so it should be initialized unconditionally as well. See rurban#294.
darkk added a commit to darkk/smhasher that referenced this issue Sep 26, 2024
@darkk
polymur: fix arm builds, including macos-14-arm64
cde8f69 
polymur Verification value 0xA612032C means that polymur_seed_init()
was not called. Currently, polymur is compiled in unconditionally,
so it should be initialized unconditionally as well.

See rurban#294, rurban#296, rurban#297, rurban#298 and rurban#299.
darkk added a commit to darkk/smhasher that referenced this issue Sep 26, 2024
@darkk
k-hashv32, k-hashv64: fix arm builds, including macos-14-arm64
6cc35a3 
Verification value 0x9FBDD2B1 for k-hashv32 and 0x70E67C61 for k-hashv64
signal, that the has was not properly seeded. Currently, k-hashv32
and k-hashv64 are compiled depending on HAVE_KHASHV from CMakeLists.txt,
so initialization should be done depending on same macro as well.

See rurban#294
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
nmhash32: add minimal test-case for the hash implementation being broken
f0f9c00 
It's not 100% clear to me if it's UB, CC bug or lack of some kind
of memory barrier, but the very same quite innocent-looking code leads
to different results on macos-arm64 platform and breaks nmhash32

See rurban#294
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
SuperFastHash: codify an UB-based Verification value :-)
1474c90 
It's unclear to me if it should be `0` (SKIP) or just a value that
is quite common across platforms and compilers.

See rurban#294
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
polymur: add missing polymur_seed_init() call
9d52b78 
It fixes arm builds, including macos-14-arm64

polymur Verification value 0xA612032C means that polymur_seed_init()
was not called. Currently, polymur is compiled in unconditionally,
so it should be initialized unconditionally as well.

See rurban#294, rurban#296, rurban#297, rurban#298 and rurban#299.
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
k-hashv32, k-hashv64: add missing initialization
d7e2da7 
It fixes arm builds, including macos-14-arm64

Verification value 0x9FBDD2B1 for k-hashv32 and 0x70E67C61 for k-hashv64
signal, that the has was not properly seeded. Currently, k-hashv32
and k-hashv64 are compiled depending on HAVE_KHASHV from CMakeLists.txt,
so initialization should be done depending on same macro as well.

See rurban#294
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
SuperFastHash: codify an UB-based Verification value :-)
15dc620 
It's unclear to me if it should be `0` (SKIP) or just a value that
is quite common across platforms and compilers.

See rurban#294
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
polymur: add missing polymur_seed_init() call
57e2993 
It fixes arm builds, including macos-14-arm64

polymur Verification value 0xA612032C means that polymur_seed_init()
was not called. Currently, polymur is compiled in unconditionally,
so it should be initialized unconditionally as well.

See rurban#294, rurban#296, rurban#297, rurban#298 and rurban#299.
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
k-hashv32, k-hashv64: add missing initialization
1284d78 
It fixes arm builds, including macos-14-arm64

Verification value 0x9FBDD2B1 for k-hashv32 and 0x70E67C61 for k-hashv64
signal, that the has was not properly seeded. Currently, k-hashv32
and k-hashv64 are compiled depending on HAVE_KHASHV from CMakeLists.txt,
so initialization should be done depending on same macro as well.

See rurban#294
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
nmhash32: add minimal test-case for the hash implementation being broken
076285a 
It's not 100% clear to me if it's UB, CC bug or lack of some kind
of memory barrier, but the very same quite innocent-looking code leads
to different results on macos-arm64 platform and breaks nmhash32

See rurban#294
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
SuperFastHash: codify an UB-based Verification value :-)
650f02f 
It's unclear to me if it should be `0` (SKIP) or just a value that
is quite common across platforms and compilers.

See rurban#294
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
polymur: add missing polymur_seed_init() call
dcd1153 
It fixes arm builds, including macos-14-arm64

polymur Verification value 0xA612032C means that polymur_seed_init()
was not called. Currently, polymur is compiled in unconditionally,
so it should be initialized unconditionally as well.

See rurban#294, rurban#296, rurban#297, rurban#298 and rurban#299.
darkk added a commit to darkk/smhasher that referenced this issue Sep 27, 2024
@darkk
k-hashv32, k-hashv64: add missing initialization
80f0945 
It fixes arm builds, including macos-14-arm64

Verification value 0x9FBDD2B1 for k-hashv32 and 0x70E67C61 for k-hashv64
signal, that the has was not properly seeded. Currently, k-hashv32
and k-hashv64 are compiled depending on HAVE_KHASHV from CMakeLists.txt,
so initialization should be done depending on same macro as well.

See rurban#294
@darkk darkk mentioned this issue Sep 27, 2024
Merged
rurban pushed a commit that referenced this issue Sep 28, 2024
@darkk @rurban
nmhash32: add minimal test-case for the hash implementation being broken
e63b800 
It's not 100% clear to me if it's UB, CC bug or lack of some kind
of memory barrier, but the very same quite innocent-looking code leads
to different results on macos-arm64 platform and breaks nmhash32

See #294
rurban pushed a commit that referenced this issue Sep 28, 2024
@darkk @rurban
SuperFastHash: codify an UB-based Verification value :-)
410ce6e 
It's unclear to me if it should be `0` (SKIP) or just a value that
is quite common across platforms and compilers.

See #294
rurban pushed a commit that referenced this issue Sep 28, 2024
@darkk @rurban
polymur: add missing polymur_seed_init() call
52391b9 
It fixes arm builds, including macos-14-arm64

polymur Verification value 0xA612032C means that polymur_seed_init()
was not called. Currently, polymur is compiled in unconditionally,
so it should be initialized unconditionally as well.

See #294, #296, #297, #298 and #299.
rurban pushed a commit that referenced this issue Sep 28, 2024
@darkk @rurban
k-hashv32, k-hashv64: add missing initialization
64e94d5 
It fixes arm builds, including macos-14-arm64

Verification value 0x9FBDD2B1 for k-hashv32 and 0x70E67C61 for k-hashv64
signal, that the has was not properly seeded. Currently, k-hashv32
and k-hashv64 are compiled depending on HAVE_KHASHV from CMakeLists.txt,
so initialization should be done depending on same macro as well.

See #294
@darkk darkk closed this as completed Sep 28, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@darkk