Hello. Guess this goes mostly to Raster and Cedric. Or anyone else who feels familiar with the eina mp one-big code.
I had a "fix" sittign here for a some time where I never was sure if it is correct or not. I somehow have the feeling that I missed some of the code logic and this is actually all correct and how it should be. We should add a coment in the code for this then. :) Specifically I talk about this: http://git.enlightenment.org/core/efl.git/commit/?h=devs/stefan/eina-mp-fix&id=cdfceacb24f1b157d47beeb02933302248e388ad retry_smaller: eina_error_set(0); - mem = malloc(sizeof(Eina_Inlist) + pool->item_size); + mem = calloc(1, sizeof(Eina_Inlist) + pool->item_size); if (!mem) eina_error_set(EINA_ERROR_OUT_OF_MEMORY); else { pool->over++; - memset(mem, 0, sizeof(Eina_Inlist)); pool->over_list = eina_inlist_append(pool->over_list, (Eina_Inlist *)mem); mem = ((unsigned char *)mem) + sizeof(Eina_Inlist); The memory allocated and assigned to mem is pool->item_size bigger than the region we memset later. It that on purpose? For me it looks like an oversight and the calloc should be better suited here. regards Stefan Schmidt ------------------------------------------------------------------------------ This SF.net email is sponsored by Windows: Build for Windows Store. http://p.sf.net/sfu/windows-dev2dev _______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
