Bump lldb-mi commit hash

[ylatuya]

Fix #6874

[WardenGnaw]

This alone wont upgrade the version of lldb-mi used. Someone will need to re-run the internal lldb-mi and update the release.

[bobbrow]

This alone wont upgrade the version of lldb-mi used. Someone will need to re-run the internal lldb-mi and update the release.

Will you be able to do that @WardenGnaw?

[WardenGnaw]

Theres a couple issues that I do not have the cycles to fix.

1. cpptoolsMacM1pool seems to have been removed and lldb-mi.template.yml fails to build LLVM on the public macos-13-arm64
2. 0d44201451f03ba907cdb268ddddfc3fa38a0ebd seems to be too old for macos-latest and is failing to build for x64 lldb-mi.

@ylatuya Were you able to build lldb-mi locally with LLVM commit 0d44201451f03ba907cdb268ddddfc3fa38a0ebd?

[ylatuya]

@ylatuya Were you able to build lldb-mi locally with LLVM commit 0d44201451f03ba907cdb268ddddfc3fa38a0ebd?

To fix the lldb-mi bug, I built it using main, which at that time was at commit:

commit 296671f059198185bedc3eba80056ebf4bc7ad30 (HEAD -> main, origin/main, origin/HEAD)
Author: Nikita Popov <npopov@redhat.com>
Date:   Wed Sep 27 11:59:44 2023 +0200

    [LoopUnroll] Store more information in UnrollCostEstimator (NFCI)
    
    Instead of having ApproximateLoopSize() use a bunch of out parameters,
    from which we later construct an UnrollCostEstimator, directly
    construct UnrollCostEstimator which holds all the information
    derived from loop analysis. This makes it easier to add additional
    metrics in the future.

[H-G-Hristov]

@ylatuya Sorry for off-topic do you know how to fix this issue or how to approach fixing it: lldb-tools/lldb-mi#101
This renders debugging on macOS useless basically. And I wonder if it is related to your fix in anyway (or at least the other reprorted issues with cpptools).

[ylatuya]

@ylatuya Sorry for off-topic do you know how to fix this issue or how to approach fixing it: lldb-tools/lldb-mi#101 This renders debugging on macOS useless basically. And I wonder if it is related to your fix in anyway (or at least the other reprorted issues with cpptools).

I don't know how to fix it, sorry. This was my first contribution and it was a trivial fix.

[H-G-Hristov]

@ylatuya Sorry for off-topic do you know how to fix this issue or how to approach fixing it: lldb-tools/lldb-mi#101 This renders debugging on macOS useless basically. And I wonder if it is related to your fix in anyway (or at least the other reprorted issues with cpptools).

I don't know how to fix it, sorry. This was my first contribution and it was a trivial fix.

@ylatuya There seems to be a proposed fix for the memory leak issue. Could you please update this PR to the latest commit: a6c8c66 here: https://github.com/lldb-tools/lldb-mi/commits/main/

@WardenGnaw @bobbrow Is there any change updating this PR and moving it forward?

[H-G-Hristov]

Can we get this updated and merged finally, please!

[sunshaoce]

Hi @H-G-Hristov, lldb-mi added some useful fixes after your last commit, can you update your lldb-mi commit hash?

[ylatuya]

Updated lldb-mi to the latest version.