Custom Query (37 matches)

Filters
 
Or
 
  
 
Columns

Show under each result:


Results (7 - 9 of 37)

1 2 3 4 5 6 7 8 9 10 11 12 13
Ticket Owner Reporter Resolution Summary
#50 [email protected] fixed fsync memory leak
Description

Whenever I call aefile_fsync on a valid file descriptor, the call leaks memory (details below).

==11238== 88 bytes in 1 blocks are possibly lost in loss record 57 of 106 ==11238== at 0x4C244E8: malloc (vg_replace_malloc.c:236) ==11238== by 0x447DAF: ae_opcache_get (opcache.c:329) ==11238== by 0x447C05: aethread_hint (aethread.c:177) ==11238== by 0x444F76: ae_worker_aefile_fsync (aefile.ae:88) ==11238== by 0x445BB0: aefile_fsync (aefile.ae:84)

#48 carns fixed set up automated Aesop tests in Jenkins
Description

We should add Aesop to MCS's Jenkins test environment for automatic regression testing. The aesop tests are sequential and do not require any MPI or job scheduling environment.

Testing can focus on the devel branch until it is merged back to trunk.

#43 carns dkimpe fixed Try compiling with clang and use static analyzer
Description
 >-Phil

> Oh, that reminds me!  I haven't tried it in a while, but when we first split
> aesop into its own tree and revised the aecc wrapper, one of the side
> effects (and test cases) was to compile aesop using clang.  I assume this
> still works.  Once Triton is fully building with the external aesop
> implementation, this means (in theory) that we can build all of Triton with
> clang as well.

> We have other priorities right now, but we should revisit that once we
> finish the triton conversion and have the various test programs working.  I
> think the clang static analyzer will turn up some interesting things to look
> at in the code.  It found a few things even when I ran it on the aesop tree
> by itself, and it is pretty easy to use.
1 2 3 4 5 6 7 8 9 10 11 12 13
Note: See TracQuery for help on using queries.