Re: [PATCH] Shut up valgrind

Hi Francois,

I just committed a large patch to fix this issue among other malloc related inconsistencies.
valgrind was right, it was a bug!
qe_mallocz would have cleared the allocated structure, but you used malloc in haiku.cpp to allocate it :(
valgrind rules!

I added macros to prevent spurious uses of malloc/free.
Can you verify the I did not break the compilation?

Thanks

Chqrlie

On 21 janv. 2014, at 22:33, François Revol <[email protected]> wrote:

> This shuts up valgrind which seems to think QEFont::refcount isn't
> initialized (while in theory qe_mallocz() sets it to 0 but well...
> 
> ==29937== Conditional jump or move depends on uninitialised value(s)
> ==29937==    at 0x40ECC1: print_at_byte (display.h:259)
> ==29937==    by 0x40F111: put_status (qe.c:4521)
> ==29937==    by 0x415E2A: qe_init (qe.c:7987)
> ==29937==    by 0x41F66E: url_main_loop (unix.c:305)
> ==29937==    by 0x409028: main (qe.c:8040)
> ==29937==
> ==29937== Conditional jump or move depends on uninitialised value(s)
> ==29937==    at 0x40C213: get_line_height (display.h:259)
> ==29937==    by 0x4134E5: do_refresh (qe.c:6472)
> ==29937==    by 0x415E37: qe_init (qe.c:7992)
> ==29937==    by 0x41F66E: url_main_loop (unix.c:305)
> ==29937==    by 0x409028: main (qe.c:8040)
> ==29937==
> 
> 
> François.
> <display-valgrind-001.patch>_______________________________________________
> Qemacs-devel mailing list
> [email protected]
> https://lists.nongnu.org/mailman/listinfo/qemacs-devel