On 10 Apr 2014 15:56, "Brian Paul" <<a href="mailto:brianp@vmware.com">brianp@vmware.com</a>> wrote: > > On 04/10/2014 01:43 AM, Jakob Bornecrantz wrote: >> >> On Thu, Apr 10, 2014 at 3:26 AM, Brian Paul <<a href="mailto:brianp@vmware.com">brianp@vmware.com</a>> wrote: >>> >>> We were using REALLOC() from u_memory.h but FREE() from imports.h. >>> This mismatch caused us to trash the heap on Windows after we >>> deleted a texture object. >>> >>> This fixes a regression from commit 6c59be7776e4d. >>> --- >>> src/mesa/state_tracker/st_cb_texture.c | 2 +- >>> src/mesa/state_tracker/st_texture.c | 12 ++++++++++++ >>> src/mesa/state_tracker/st_texture.h | 3 +++ >>> 3 files changed, 16 insertions(+), 1 deletion(-) >> >> >> Is this patch obsoleted by the FREE/MALLOC remove series? > > > Not exactly. If I try to use FREE(stObj->sampler_views) in st_cb_texture.c I need to include "u_memory.h". But then I get preprocessor warnings about MALLOC_STRUCT() and CALLOC_STRUCT() being redefined. So then I need to insert some #undef MALLOC_STRUCT lines. > It's just messy. Ah fair enough. Cheers, Jakob.