[llvm-commits] [llvm] r141440 - /llvm/trunk/lib/MC/ELFObjectWriter.cpp

[Chandler Carruth]

On Sat, Oct 8, 2011 at 1:02 AM, Nick Lewycky <nicholas at mxc.ca> wrote:

> Chandler Carruth wrote:
>
>> On Fri, Oct 7, 2011 at 4:29 PM, Nick Lewycky <nicholas at mxc.ca
>> <mailto:nicholas at mxc.ca>> wrote:
>>
>>    Author: nicholas
>>    Date: Fri Oct  7 18:29:53 2011
>>    New Revision: 141440
>>
>>    URL: http://llvm.org/viewvc/llvm-**project?rev=141440&view=rev<http://llvm.org/viewvc/llvm-project?rev=141440&view=rev>
>>    <http://llvm.org/viewvc/llvm-**project?rev=141440&view=rev<http://llvm.org/viewvc/llvm-project?rev=141440&view=rev>
>> >
>>    Log:
>>    Don't emit the symbol table entry for the .symtab_shndx section either.
>>
>>
>> Test case?
>>
>
> The test case is guaranteed to exceed 65535 sections, at a bare minimum of
> one section per line. It is also wildly unlikely that this would regress,
> short of doing something drastic enough that the programmer would
> ignore/delete any tests too. Are you sure?


But there are all sorts of other bugs that such a test case might catch: n^2
algorithms per-section, other overflow, silly amounts of memory allocation
per-section.

Also, I suspect that you can write the test case in less than 64k lines by
either making it a preprocessed assembly file or by use of fancy shell bits
in the RUN line....
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20111008/45a538ae/attachment.html>