llvm dev - Sep 2013 - [LLVMdev] Fix crash in llvm_gcda_emit_arcs()

Hi Nick,

Thanks for looking over my patch.  I'd be happy to add the compare against
cur_buffer_size.  I'll try to write up such a change today and then send a
new patch to the list.

Joseph Kain


On Thu, Sep 5, 2013 at 1:37 AM, Nick Lewycky <nicholas at mxc.ca> wrote:
> Bill, could you take a look at this? I don't have a checkout of
> compiler-rt that builds.
>
> It seems obviously-correct to me, but I wish it did a compare against
> cur_buffer_size to make sure it's in range.
>
> Nick
>
> Joseph Kain wrote:
>
>> Hi,
>>
>> I've been seeing a crash in llvm_gcda_emit_arcs() on x86_64.  The
crash
>> occurs executing a movdqa instruction with an unaligned src address.
>>   The attached patch to the compiler-rt project fixes the problem by
>> using memcpy() to read data from the write_buffer[] in GCDAProfiling.c.
>>
>> This is my first patch submission to llvm so please let me know if
I've
>> missed any steps.  I'm not on the mailing list so please include me
on
>> any replies.
>>
>> I haven't run the tests on this change but would be happy to do so
if
>> someone could point me to documentation on how to do so.  I
couldn't
>> find testing instructions linked from
>>
http://llvm.org/docs/**DeveloperPolicy.html<http://llvm.org/docs/DeveloperPolicy.html>.
>>  I have built the change and
>> run it with the programs I was testing originally.  With this change
the
>> crash no longer occurs.
>>
>> Thanks,
>> Joseph Kain
>>
>>
>> ______________________________**_________________
>> LLVM Developers mailing list
>> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
>>
http://lists.cs.uiuc.edu/**mailman/listinfo/llvmdev<http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev>
>>
>
>

-- 
Joseph Kain
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20130905/f7bbb11a/attachment.html>

Hi Nick and Bill,

Attached is a new version of the patch that checks against cur_buffer_size.
This works with the cases I've been using but again I haven't run the
official tests on this change.

Thanks,
Joseph Kain



On Thu, Sep 5, 2013 at 9:46 AM, Joseph Kain <joekain at gmail.com> wrote:
> Hi Nick,
>
> Thanks for looking over my patch.  I'd be happy to add the compare
against
> cur_buffer_size.  I'll try to write up such a change today and then
send a
> new patch to the list.
>
> Joseph Kain
>
>
> On Thu, Sep 5, 2013 at 1:37 AM, Nick Lewycky <nicholas at mxc.ca>
wrote:
>
>> Bill, could you take a look at this? I don't have a checkout of
>> compiler-rt that builds.
>>
>> It seems obviously-correct to me, but I wish it did a compare against
>> cur_buffer_size to make sure it's in range.
>>
>> Nick
>>
>> Joseph Kain wrote:
>>
>>> Hi,
>>>
>>> I've been seeing a crash in llvm_gcda_emit_arcs() on x86_64. 
The crash
>>> occurs executing a movdqa instruction with an unaligned src
address.
>>>   The attached patch to the compiler-rt project fixes the problem
by
>>> using memcpy() to read data from the write_buffer[] in
GCDAProfiling.c.
>>>
>>> This is my first patch submission to llvm so please let me know if
I've
>>> missed any steps.  I'm not on the mailing list so please
include me on
>>> any replies.
>>>
>>> I haven't run the tests on this change but would be happy to do
so if
>>> someone could point me to documentation on how to do so.  I
couldn't
>>> find testing instructions linked from
>>>
http://llvm.org/docs/**DeveloperPolicy.html<http://llvm.org/docs/DeveloperPolicy.html>.
>>>  I have built the change and
>>> run it with the programs I was testing originally.  With this
change the
>>> crash no longer occurs.
>>>
>>> Thanks,
>>> Joseph Kain
>>>
>>>
>>> ______________________________**_________________
>>> LLVM Developers mailing list
>>> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
>>>
http://lists.cs.uiuc.edu/**mailman/listinfo/llvmdev<http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev>
>>>
>>
>>
>
>
> --
> Joseph Kain
>


-- 
Joseph Kain
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20130905/9f334b0f/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: memcpy-to-writebuffer-2.patch
Type: application/octet-stream
Size: 921 bytes
Desc: not available
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20130905/9f334b0f/attachment.obj>

Hi Joseph and Nick,

I'll take a look at this soon. :-)

-bw

On Sep 5, 2013, at 10:51 AM, Joseph Kain <joekain at gmail.com> wrote:
> Hi Nick and Bill,
> 
> Attached is a new version of the patch that checks against cur_buffer_size.
This works with the cases I've been using but again I haven't run the
official tests on this change.
> 
> Thanks,
> Joseph Kain
> 
> 
> 
> On Thu, Sep 5, 2013 at 9:46 AM, Joseph Kain <joekain at gmail.com>
wrote:
> Hi Nick,
> 
> Thanks for looking over my patch.  I'd be happy to add the compare
against cur_buffer_size.  I'll try to write up such a change today and then
send a new patch to the list.
> 
> Joseph Kain
> 
> 
> On Thu, Sep 5, 2013 at 1:37 AM, Nick Lewycky <nicholas at mxc.ca>
wrote:
> Bill, could you take a look at this? I don't have a checkout of
compiler-rt that builds.
> 
> It seems obviously-correct to me, but I wish it did a compare against
cur_buffer_size to make sure it's in range.
> 
> Nick
> 
> Joseph Kain wrote:
> Hi,
> 
> I've been seeing a crash in llvm_gcda_emit_arcs() on x86_64.  The crash
> occurs executing a movdqa instruction with an unaligned src address.
>   The attached patch to the compiler-rt project fixes the problem by
> using memcpy() to read data from the write_buffer[] in GCDAProfiling.c.
> 
> This is my first patch submission to llvm so please let me know if I've
> missed any steps.  I'm not on the mailing list so please include me on
> any replies.
> 
> I haven't run the tests on this change but would be happy to do so if
> someone could point me to documentation on how to do so.  I couldn't
> find testing instructions linked from
> http://llvm.org/docs/DeveloperPolicy.html.  I have built the change and
> run it with the programs I was testing originally.  With this change the
> crash no longer occurs.
> 
> Thanks,
> Joseph Kain
> 
> 
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
> 
> 
> 
> 
> -- 
> Joseph Kain
> 
> 
> 
> -- 
> Joseph Kain
> <memcpy-to-writebuffer-2.patch>