[Mono-list] mono 1.2.6

[Geoff Norton]

Scott,

On 4-Dec-07, at 1:02 PM, Scott Fluto wrote:

> The problem is that they are so random I cant give any isolated tests.
> I have about 500 tests and it could happen on any of them.
>

This means you should file bugs for each of the crashes you have  
immediately with as much information as you have.  Otherwise you  
cannot expect 1.2.6 final to magically fix these bugs.  You can always  
add more information to a bug later as you get it.

> ex:
> file reflection.c: line 9694 (mono_reflection_lookup_dynamic_token):
> assertion failed: (obj).
> This error happens about 5% of the time running the tests and it can
> happen in random places.
>

Please file a bug.

> ** (process:10045): ERROR (recursed) **: file threads.c: line 386
> (mono_hazard_pointer_get): assertion failed: (current_thread &&
> current_thread->small_id >= 0)
> This error maybe happens 5% of the time.
>

Please file a bug.

> Im still trying to isolate the issues but the company is about the cut
> the time devoted to isolated the problems so Im really hoping 1.2.6
> release will make things better,
> if not we may reconsider our mono stategy but Im still looking into it
> anyhow.
>

Please see above.  We cannot fix things we're not aware of.  Please  
file individual bugs for each of the issues with as much information  
as you have available:

Architecture
Distro / version
Mono version
Test case that triggers the bug (even 5% of the time)
asserts/stack traces, etc

Thanks,

-g