[LLVMdev] target hexagon and sparcv9 lead to llc crack

Duncan Sands baldrick at free.fr
Thu Jul 19 00:26:48 PDT 2012


Hi Sebastian,

On 19/07/12 05:57, Sebastian Pop wrote:
> On Tue, Jul 3, 2012 at 9:27 PM, Sebastian Pop <spop at codeaurora.org> wrote:
>> Hi,
>>
>> On Tue, Jul 3, 2012 at 9:48 AM, Tony Linthicum <tlinth at codeaurora.org> wrote:
>>> On 7/3/2012 5:01 AM, Duncan Sands wrote:
>>>> Hi,
>>>>
>>>>> (4) llc -march=hexagon test.ll -o test.s
>>>>>
>>>>> '' is not a recognized processor for this target (ignoring processor)
>>>>> 0  llc 0x08c2512b
>>>>> Stack dump:
>>>>> 0.   Program arguments: llc -march=hexagon test.ll -o test.s
>>>>> 1.   Running pass 'Function Pass Manager' on module 'test.ll'.
>>>>> 2.   Running pass 'Hexagon DAG->DAG Pattern Instruction Selection' on
>>>>> function '@cmp'
>>>>> Segmentation fault (core dumped)
>>>>
>>>> if you provide an explicit choice of processor, eg -mcpu=hexagonv2, then this
>>>> works.  It looks like hexagon is missing a default choice of processor.
>>>>
>>>> Ciao, Duncan.
>>>> _______________________________________________
>>>> LLVM Developers mailing list
>>>> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
>>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>>>>
>>>
>>> We will set a default target for llc and upstream it shortly!
>>
>> Attached is the patch that I will commit: this should fix a bunch
>> of fails that we currently have in make check when configuring
>> with --target=hexagon.
>>
>
> Ping: is this patch ok to commit?

it seems fine to me.

Ciao, Duncan.

PS: I didn't realize you were looking for agreement before committing since you
are "Attached is the patch that I will commit" which I read as meaning you were
about to commit it.



More information about the llvm-dev mailing list