[PATCH] Implement ARM EHABI

[Renato Golin]

Hi folks,

It's great to see such interest in ARM EHABI, and I agree with both of you
that you should merge all patches into one. Ultimately, Anton is the final
approval, since that's his area, so you should always wait for his review.

I'd also suggest that you run the regression test in the test-suite
(SingleSource/Regression/C++) by re-enabling EH in the Makefile and running
only that test (with --only-test) and seeing what this patch fixes and what
not (and why).

I'm not expecting the patch to fix much (especially not everything), but
would be good to know how are we progressing, and that would make the patch
more trustworthy.

cheers,
--renato


On 15 April 2013 07:08, WenHan Gu (谷汶翰) <wenhan.gu at gmail.com> wrote:

> Hi, Cho,
>
> Thanks for your response.
> It's important to clarify that I won't insist that who's patch will be
> accepted, just FYI. :)
>
> ARM EHABI has been implemented at most part, and I think functions by this
> patch should be review asap.
> Maybe you can contact Logan Chien (owner of the other patch) and derive a
> refined patch, and ask Anton to help review.
> More people included makes the result more better, that is just what I
> want to said.
>
> Thanks for your working on this!
>
>
> 2013/4/15 Ilyong Cho <ilyoan at gmail.com>
>
>> Thank you for informing me.
>>
>> I didn't know that there was already a patch about the same issue. And I
>> didn't want to step on anybody's toe.
>>
>> My opinion about that patch? Great!
>> I've juat looked the patch and I think it looks good but I found the
>> implementation of that patch is different from gas's. I have a weird use
>> case (hand made ARM assembly with a complex function prologue) that the
>> patch on D414 couldn't cover but gas implementation(and mine) could cover.
>>
>> But I believe it'll work better with normal situation than mine because
>> it implement .vsave that I fogot to implement :(
>>  2013. 4. 15. 오후 1:21에 "WenHan Gu (谷汶翰)" <wenhan.gu at gmail.com>님이 작성:
>>
>> Hi,
>>>
>>> Actually, ARM EHABI has been patch sumbutted at March.
>>> However, Anton is too busy to review that.
>>>
>>> http://llvm-reviews.chandlerc.com/D414
>>>
>>> What do you think on this patch?
>>> Thanks.
>>>
>>>
>>> 2013/4/15 ilyong cho <ily.cho at samsung.com>
>>>
>>>> Hi all,
>>>>
>>>> I've just found a bug in this patch and fixed it.
>>>> The bug was that the code miscalculated the number of words on
>>>> EmitHandlerData()
>>>>
>>>> You can find the fixed patch on attachment and also
>>>> http://llvm-reviews.chandlerc.com/D662
>>>>
>>>> Review please.
>>>>
>>>> ILYONG CHO, ilyoan at gmail.com
>>>>
>>>>
>>>> > Hi all,
>>>>
>>>> > This patch is about implemention of ARM EHABI.
>>>> > There are three "Not Implemented" functions on
>>>> lib/Target/ARM/MCTargetDesc/ARMELFStreamer.cpp. This patch implements it
>>>> and .movsp directive.
>>>> > This patch based on ARM IHI 0038A and gas implemantation.
>>>>
>>>> > I'd be appreciative review comments. Patch is available on
>>>> http://llvm-reviews.chandlerc.com/D662
>>>>
>>>> > Thanks.
>>>>
>>>> > ILYONG CHO
>>>>
>>>> _______________________________________________
>>>> llvm-commits mailing list
>>>> llvm-commits at cs.uiuc.edu
>>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>>>>
>>>>
>>>
>>>
>>> --
>>> Best Regards,
>>> WenHan Gu (Nowar)
>>>
>>
>
>
> --
> Best Regards,
> WenHan Gu (Nowar)
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20130415/c7b36785/attachment.html>