[llvm-dev] Linux/ARM: Segfault issue when we build clang sources including __thread variable using -O2 flag

[Geunsik Lim via llvm-commits]

>  I think I've spotted what's wrong with
>  libcoreclr.so: it looks like a bunch of PC-relative constant pools
>  have been collapsed down to a single one. So in objdump -d we see, for
>  example:
>
>    28b428:       f8df 06d4       ldr.w   r0, [pc, #1748] ; 28bb00
>    28b42c:       4478            add     r0, pc
>    28b42e:       f5a6 e8c6       blx     315bc <__tls_get_addr at plt>
>    [...]
>
BTW, How could you get the  "*****  <__tls_get_addr at plt>" content?
When I ran the below commands, I could not get the  "*****
 <__tls_get_addr at plt>" line.
u14.04$> cd ./runtime-arm-u1404-debug-mode-20160503-global-dynamic-O2/
u14.04$> objdump -d ./libcoreclr.so  | grep tls_get_addr



On Sun, May 8, 2016 at 12:54 AM, Tim Northover <t.p.northover at gmail.com>
wrote:

> On 6 May 2016 at 23:00, Geunsik Lim <leemgs at gmail.com> wrote:
> > To understand litpools in more detail, is there a technical document
> besides
> > the below "ARM Compiler armasm User Guide"?
>
> Not that I know of, I'm afraid. It's not really the kind of thing
> people write about.
>
> > On the other hand,
> > it seems that this issue can be reproduced on the AArch64 port of LLVM.
>
> I don't think so. AArch64 doesn't even have the custom litpool entry
> class that was affected. Of course, it may segfault anyway but if so
> I'd expect the reasons to be entirely different.
>
> Cheers.
>
> Tim.
>



-- 
http://leemgs.fedorapeople.org
Don't try to avoid pain if you fail.
If you decided to face the challenges in life,
you can gain a lot by giving your best.
Cheolsang Jeong's Book & life
--
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160510/ccd1c7e9/attachment.html>