Hi!
I suspect that using resmgr_attach and resmgr_detach can cause memory leaks.
In other words - resmgr_detach does not free all memory that resmgr_attach
allocates.
What i mean:
In my test app i periodically create and destroy thread with resmgr. Memory
size is slowly growing with time (i use psin). More, i link app with simple
debug versions of malloc(), realloc() and free(). These just print
parameters and return values of functions. As i can see
=== START… ===
…
server: DEBUG resmgr_attach:
MY_MALLOC( 28 ) = 8052d38 <<<
MY_MALLOC( 36 ) = 8053800 <<<
MY_REALLOC( 0, 64) = MY_MALLOC( 64 ) = 805316c = <<<
805316c
MY_MALLOC( 32 ) = 8052d5c
…
=== STOP… ===
…
server: DEBUG resmgr_detach:
MY_FREE( 8052d5c )
…
So there are no free() calls for marked malloc()'s. I tried to use malloc_g
also, but it seems that malloc_g don’t report memory leaks for
malloc(size<=64).
I hope i missed or misunderstood something, but now i don’t know what to do
with it 
. Any help would be appreciated…
Thank you,
Artem
Artem Smolenkov <iskraDEL@delufacom.ru> wrote:
Hi!
I suspect that using resmgr_attach and resmgr_detach can cause memory leaks.
In other words - resmgr_detach does not free all memory that resmgr_attach
allocates.
What i mean:
In my test app i periodically create and destroy thread with resmgr. Memory
size is slowly growing with time (i use psin). More, i link app with simple
debug versions of malloc(), realloc() and free(). These just print
parameters and return values of functions. As i can see
This problem has been addressed and will be fixed in an upcoming update/
release.
Thanks for reporting the issue.
Thomas
\
Thomas (toe-mah) Fletcher QNX Software Systems
thomasf@qnx.com Core OS Technology Group
(613)-591-0931 http://www.qnx.com/