graphene: fix alignment of graphene_simd4f_t on armhf and s390x #1105
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
On systems with the armhf ABI as used by Debian and Ubuntu (amdv7 + VFP, but without NEON), graphene falls back to the scalar definition of graphene_simd4f_t, which has the same alignment as its base type, float. On s390x, the underlying type is a generic float __attribute__((vector_size(16)). Surprinsingly, the s390x ABI does not mandate alignment bigger than 8 for the vector types. Without these patches, the layout tests fail on those architectures. The armhf case is notable because there are a fair number of Raspberry Pis out there that use that ABI, either because their SoC isn't 64-bit ready or just because they have low memory. Ideally, we could specifiy this in the Gir.toml file, but it doesn't seem like the tool is flexible enough to express this kind of nuances,
|
Based on #1107 there is also a problem on armel and i386 in addition. The main problem here is that See e.g. |
|
From that issue it sounds like |
|
Any updates here? |
|
Sorry, this fell off my radar entirely. I'll try to get back into it
this week.
|
|
@schopin-pro any updates on this one? |
|
Still on my todolist, still haven't found the time to work on it. Feel free
to close out the PR though, since the final implementation will likely be
different anyway :)
…On Fri, Oct 27, 2023 at 2:45 PM Bilal Elmoussaoui ***@***.***> wrote:
@schopin-pro <https://github.com/schopin-pro> any updates on this one?
—
Reply to this email directly, view it on GitHub
<#1105 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUXZP7MOGRBWZY5SMDENHRDYBOUGNAVCNFSM6AAAAAAYQLWER2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBSHA2TMMZWGE>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
On systems with the armhf ABI as used by Debian and Ubuntu (amdv7 + VFP, but without NEON), graphene falls back to the scalar definition of graphene_simd4f_t, which has the same alignment as its base type, float.
On s390x, the underlying type is a generic float
attribute((vector_size(16)). Surprinsingly, the s390x ABI does not mandate alignment bigger than 8 for the vector types.
Without these patches, the layout tests fail on those architectures. The armhf case is notable because there are a fair number of Raspberry Pis out there that use that ABI, either because their SoC isn't 64-bit ready or just because they have low memory.
Ideally, we could specifiy this in the Gir.toml file, but it doesn't seem like the tool is flexible enough to express this kind of nuances,