build, qt: Drop support for i686-linux-android host #23744
Conversation
|
friendly ping @icota |
Guix builds: |
|
The following sections might be updated with supplementary metadata relevant to reviewers and maintainers. ConflictsReviewers, this pull request conflicts with the following ones:
If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first. |
|
Concept ACK. I don't think this architecture (32-bit x86 android) is worth spending effort on, I expect there to be zero actual devices that use that. |
|
Concept ACK with a suggestion to drop |
hebasto
force-pushed
the
211211-android32
branch
from
66a20a5 to
86afe77
Compare
|
Updated 66a20a5 -> 86afe77 (pr23744.01 -> pr23744.02, diff).
|
hebasto
changed the title
i686-linux-android host
Guix builds: |
How can we verify this? IIRC, aren't some modern devices with AArch64-capable chips still running a 32-bit ARM OS anyway? |
You seem to be right. To protest Pokemon Go dropping 32-bit someone compiled a list of phones sold as recently as 2019 running 32-bit Android (even though most 64-bit capable). Having said that,
So I would still recommend dropping |
|
I don't know how i feel about this Arm has recently announced that all of there cpu's will be 64 bit starting in 2023. In there announcement, they state that 2% of mobile devices around the world are 32 bit. Going off that, it seems like this would be ok to do. But, some cheaper android phones have a 64 bit CPU yet run a 32 bit android OS. An example of such device is the Xiaomi Redmi 8a which was released in 2019. Additionally, these 32 bit androids are most likely the cheapest devices in poorer markets. It seems to me that it's not yet time to cut out 32 bit android support. |
|
We still support some archs that are not in our releases, e.g., Assuming that Dropping of Also https://android.googlesource.com/platform/bionic/+/master/docs/32-bit-abi.md, though not sure if our code has been affected. |
|
~0 about dropping 32 bit for Android. In general, ARM and RISC-V 32 bit are still feasible architectures so please don't drop support for building for them project-wide (FWIW it's definitely possible to run a node on 32-bit ARM). That said, maybe not for Android. |
|
We can always drop 32-bit ARM Android support later. But from that list, right now it sounds like it's still common. |
|
Having read the comments and having spoken to @hebasto today I agree we keep |
hebasto
force-pushed
the
211211-android32
branch
from
86afe77 to
66a20a5
Compare
hebasto
changed the title
|
It's nice we have a consensus about keeping support of So reverted back to 66a20a5. The OP and the title has been also reverted back. |
66a20a5 build, qt: Drop support for `i686-linux-android` host (Hennadii Stepanov) Pull request description: There are no reasons to keep support for `i686-linux-android` host, which is actually broken in master (50c502f), and this fact has been unnoticed for months :) bitcoin#23675 (comment): > I'm surprised `i686-linux-android` ABI is still supported. I would love to drop it... bitcoin#23675 (comment) > What is `i686-linux-android`? 32-bit x86 android? is that really a thing? ACKs for top commit: prusnak: utACK 66a20a5 Tree-SHA512: 211f794de2fc569f0ade2a4da805b8bfd4ce2ab0930c5d427acc4f5d015fcdc4911f02fc64f6401197f7641aed79944a9594be80c817547be3269cdd721cf79b
d1b7bcb qt: Drop no longer supported Android architecture (Hennadii Stepanov) Pull request description: The `i686-linux-android` arch support has been dropped since bitcoin/bitcoin#23744. ACKs for top commit: katesalazar: ACK d1b7bcb icota: utACK d1b7bcb prusnak: Approach ACK d1b7bcb Tree-SHA512: 13689ec8c63c92b9a52a3c25edc35536b8e51ff583f57c45b168515f928d020d6bb85d03db9efd8d5efd57b944dfd313a89f5ff8a52f99982ccc8d9671f6e7a9
There are no reasons to keep support for
i686-linux-androidhost, which is actually broken in master (50c502f), and this fact has been unnoticed for months :)#23675 (comment):
#23675 (comment)