| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
advantages:
- memory is trackable by insure
- a very simple talloc_realloc() is possible (I've added it)
It is slower than the previous talloc code, but I don't think that
is going to be a problem. If it is a problem then there are
some ways we can make it faster but I'd like to leave those
until we have tested this a bit and can see what performance
problems might show up in profiling
(This used to be commit 1ab63cf3a6f62c3e4ca19a549bde78dad0eb4340)
|
|
pool is getting bloated. Also added a talloc_zero function to return zeroed memory.
Added debug in rpc_server/srv_pipe_hnd.c so we know when a talloc pool is being
freed. Syncup with srv_pipe_hnd.c from 2.2 so we are freeing memory at the same time.
Jeremy.
(This used to be commit d3a56c6042acf037bbd53de88d7636a5803ead20)
|
|
memory access.
Tridge, I don't think using 0xdeadbeef for size allocations
of 0 is going to work. I ended up having to use NULL
as much code that works on UNISTR checks to see if the buffer ptr
is NULL. So valid code ends up with a seg fault.
Rather than rewriting it all, I added a DEBUG_TALLOC
#ifdef in talloc.h that sets a macro BAD_PTR.
This is the value assigned to ptr for an allocation of 0 bytes.
jerry
(This used to be commit 233b9cffa2350552d9f775f791d5d5e0464a1ed4)
|
|
in the RPC code. This change was prompted by trying to save a long (>256)
character comment in the printer properties page.
The new system associates a TALLOC_CTX with the pipe struct, and frees
the pool on return of a complete PDU.
A global TALLOC_CTX is used for the odd buffer allocated in the BUFFERxx
code, and is freed in the main loop.
This code works with insure, and seems to be free of memory leaks and
crashes (so far) but there are probably the occasional problem with
code that uses UNISTRxx structs on the stack and expects them to contain
storage without doing a init_unistrXX().
This means that rpcclient will probably be horribly broken.
A TALLOC_CTX also needed associating with the struct cli_state also,
to make the prs_xx code there work.
The main interface change is the addition of a TALLOC_CTX to the
prs_init calls - used for dynamic allocation in the prs_XXX calls.
Now this is in place it should make dynamic allocation of all RPC
memory on unmarshall *much* easier to fix.
Jeremy.
(This used to be commit 0ff2ce543ee54f7364e6d839db6d06e7ef1edcf4)
|
|
lp_string() bug properly.
we still need to add lp_talloc_free() calls in all the main event
loops, I've only put it in smbd and nmbd thus far.
(This used to be commit aa7f81552540f5dca2c146f5edd805611d5b390f)
|
