Hi all, I think I've found a memory leak in cairo 1.4.10, file cairo-pattern.c, function _cairo_pattern_create_solid. In that function, variable *pattern is malloc()'ed but has not free() yet. I am not so surely this is a bug because I am quite new to C and cairo. Please excuse me If I am wrong. As you can see in the following details, I've detected this issue with valgrind. If this bug is real, then cairo possiblely contains many similar problems in other files/functions. I also include a diff of origin src/cairo- pattern.c and the a ( possible ) FIXED cairo-pattern.c Please see the following for details. $cat test/test.cpp.ONLY.rairo #include <cairo.h> int main (int argc, char *argv[]) {         cairo_surface_t *surface;         cairo_t *cr;         surface = cairo_image_surface_create (CAIRO_FORMAT_ARGB32, 240, 80);         cr = cairo_create (surface);         cairo_select_font_face (cr, "serif", CAIRO_FONT_SLANT_ITALIC, CAIRO_FONT_WEIGHT_BOLD); #if 0         cairo_set_font_size (cr, 32.0);         cairo_set_source_rgb (cr, 0.0, 0.0, 1.0);         cairo_move_to (cr, 10.0, 50.0);         cairo_show_text (cr, "Hello, world"); #endif         cairo_destroy (cr);         cairo_surface_write_to_png (surface, "hello.png");         cairo_surface_destroy (surface);         return 0; } valgrind --leak-check=full -v --show-reachable=yes ./testImage [snip] ==31669== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 36 from 1) --31669-- --31669-- supp:   36 Ubuntu-stripped-ld.so ==31669== malloc/free: in use at exit: 132 bytes in 1 blocks. ==31669== malloc/free: 29 allocs, 28 frees, 362,305 bytes allocated. ==31669== ==31669== searching for pointers to 1 not-freed blocks. ==31669== checked 370,744 bytes. ==31669== ==31669== 132 bytes in 1 blocks are still reachable in loss record 1 of 1 ==31669==    at 0x4004405: malloc (vg_replace_malloc.c:149) ==31669==    by 0x805B7BC: _cairo_pattern_create_solid (cairo-pattern.c:309) ==31669==    by 0x804EA54: _cairo_gstate_init (cairo-gstate.c:94) ==31669==    by 0x804A073: cairo_create (cairo.c:210) ==31669==    by 0x8049F86: main (in testImage) ==31669== ==31669== LEAK SUMMARY: ==31669==    definitely lost: 0 bytes in 0 blocks. ==31669==      possibly lost: 0 bytes in 0 blocks. ==31669==    still reachable: 132 bytes in 1 blocks. ==31669==         suppressed: 0 bytes in 0 blocks. --31669--  memcheck: sanity checks: 47 cheap, 2 expensive --31669--  memcheck: auxmaps: 0 auxmap entries (0k, 0M) in use --31669--  memcheck: auxmaps: 0 searches, 0 comparisons --31669--  memcheck: secondaries: 45 issued (2880k, 2M) --31669--  memcheck: secondaries: 81 accessible and distinguished (5184k, 5M) --31669--     tt/tc: 7,958 tt lookups requiring 8,215 probes --31669--     tt/tc: 7,958 fast-cache updates, 3 flushes --31669-- translate: new        3,761 (81,413 -> 1,340,177; ratio 164:10) [0 scs] --31669-- translate: dumped     0 (0 -> ??) --31669-- translate: discarded  9 (216 -> ??) --31669-- scheduler: 2,383,621 jumps (bb entries). --31669-- scheduler: 47/4,707 major/minor sched events. --31669--    sanity: 48 cheap, 2 expensive checks. --31669--    exectx: 30,011 lists, 62 contexts (avg 0 per list) --31669--    exectx: 93 searches, 31 full compares (333 per 1000) --31669--    exectx: 0 cmp2, 76 cmp4, 0 cmpAll [vuhung@ cairo-1.4.10]$diff src/cairo- pattern.c src/cairo-pattern.c.2007-10-12.orig 317,321c317 <     cairo_solid_pattern_t tmp_pattern; <       memcpy(&tmp_pattern, pattern, sizeof(pattern)); <       if (pattern ) free(pattern); <     return &tmp_pattern.base; <     --- >     return &pattern->base; [vuhung@ cairo-1.4.10]$ [vuhung@ Linux]$uname -a Linux  2.6.9-42.ELsmp #1 SMP Wed Jul 12 23:27:17 EDT 2006 i686 GNU/Linux [vuhung@ Linux]$cat /etc/redhat-release Red Hat Enterprise Linux ES release 4 (Nahant Update 4) [vuhung@ Linux]$gcc --version gcc (GCC) 3.4.6 20060404 (Red Hat 3.4.6-3) [vuhung@ Linux]$rpm -qa | grep glibc glibc-headers-2.3.4-2.25 glibc-common-2.3.4-2.25 glibc-devel-2.3.4-2.25 glibc-2.3.4-2.25 glibc-kernheaders-2.4-9.1.98.EL [vuhung@ ]$rpm -qa | grep valgr valgrind-callgrind-0.10.1-2.EL4 valgrind-3.1.1-1.EL4 -- Best Regards, Nguyen Hung Vu <a href="mailto:vuhung16plus{remove}@gmail.dot.com">vuhung16plus{remove}@gmail.dot.com</a> An inquisitive look at Harajuku <a href="http://www.flickr.com/photos/vuhung/sets/72157600109218238/"> http://www.flickr.com/photos/vuhung/sets/72157600109218238/</a>