[PATCH] Avoid mapping past end of shared object (BZ #18685)

[Carlos O'Donell]

On 07/17/2015 08:00 AM, pinskia@gmail.com wrote:
>> Is this really a bug?
>>
>> The cost of the checks is more than just performance, but also maintenance
>> and support of that code into the future. It is about testing that code,
>> making sure we exercise those weird conditional failure paths which we
>> thought would be a good thing to add. Each of these checks is an additional
>> complication, added branch, and conditional in already complicated code.
>> I would like to see the core loader as simple as possible and to expect
>> as correct as possible input.
> 
> Yes but the cost is cheaper than someone accidentally doing a dlopen
> of a file and causing a crash rather than retuning malformed elf
> file. I suspect that is a way how to reproduce this failure.

Andrew,

Sorry, could you please clarify: Cheaper for the glibc community or
cheaper for the user? Could you clarify what you mean by "returning
malformed elf file?"

Cheers,
Carlos.