Re: [rvm-research] Potential issues with longjmp in bootImageRunner

Thanks for the fix. I was able to reproduce the same issue also in a
simple micro-benchmark I developed for the GC algorithms [1]. With the
commit for RVM-1027 [2], the issue doesn't appear anymore.

[1] https://sourceforge.net/projects/xpd/files/latest/download
[2] https://jira.codehaus.org/browse/RVM-1027

Regards,
Pradeeban.

On Mon, Apr 15, 2013 at 6:22 PM, Carl Ritson <[email protected]> wrote:
>>>> There are a number of approaches to dealing with this:
>>>>  1. assume allocation will always be top-down and ignore,
>>>>  2. explicitly allocate thread stacks (these can be passed to pthread_create),
>>>>  3. replace use of setjmp/longjmp with savecontext/swapcontext
>>>> (although these are deprecated in POSIX),
>>>>  4. implement a local version of longjmp without a check,
>>>>  5. bypass the __longjmp_chk by using glibc's internal __libc_longjmp().
>>>>
>>>> Disregarding (1), (5) is the smallest change set.
>>>
>>> I'd also disregard (1). (3) and (4) don't seem attractive to me; we
>>> ought to rely on non-deprecated methods from the system libraries as
>>> much as possible.
>>
>> My opinion is (3) the best option as (2) requires the bootImageRunner
>> to allocate memory at a specific place in memory and manage it
>> (because thread heap will need to be freed).  Instead of
>> reimplementing setjmp/longjmp completely, bootThread already has to be
>> implemented in machine specific assembly to pivot the stack when
>> entering the VM, the existing code should hence be extended to allow
>> for pivoting back onto the thread stack.  In effect bootThread should
>> return when sysThreadTerminate is called.
>
> It has been pointed out to me that where I said (3), I meant (4).  I
> agree that the use of deprecated APIs is not a good option.  An
> possible extension to (2) is to use the explicitly allocated stacks
> from inside the VM and entirely bypass the allocated of thread stacks
> in C/C++.  Although issues regarding the primordial thread would still
> need to be addressed.
>
> ------------------------------------------------------------------------------
> Precog is a next-generation analytics platform capable of advanced
> analytics on semi-structured data. The platform includes APIs for building
> apps and a phenomenal toolset for data science. Developers can use
> our toolset for easy data analysis & visualization. Get a free account!
> http://www2.precog.com/precogplatform/slashdotnewsletter
> _______________________________________________
> Jikesrvm-researchers mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers



-- 
Pradeeban Kathiravelu.
Postgraduate Student,
Erasmus Mundus European Master in Distributed Computing,
Instituto Superior Técnico, Lisbon, Portugal.

Blog: [Llovizna] http://kkpradeeban.blogspot.com/

------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_apr