https://bugs.winehq.org/show_bug.cgi?id=36120
--- Comment #2 from Austin English austinenglish@gmail.com --- (In reply to Henri Verbeet from comment #1)
Commit 301a17d55649b0159f3ecc1e394a3dd6608ecc74 is likely to have helped here. (And not just because vertexdeclaration.c is gone, I think it should avoid the actual leak as well.)
Yeah. 36113 is still there, of course. And one I wasn't seeing before: ==19388== 4,096 bytes in 1 blocks are definitely lost in loss record 1,528 of 1,577 ==19388== at 0x6C2D6F1: ??? (in /usr/lib/libdrm_intel.so.1.0.0) ==19388== by 0x6C29378: drm_intel_bo_map (in /usr/lib/libdrm_intel.so.1.0.0) ==19388== by 0x68CEF46: ??? (in /usr/lib/dri/i965_dri.so) ==19388== by 0x68CF506: ??? (in /usr/lib/dri/i965_dri.so) ==19388== by 0x685CD38: ??? (in /usr/lib/dri/i965_dri.so) ==19388== by 0x6919AFB: ??? (in /usr/lib/dri/i965_dri.so) ==19388== by 0x684A2B9: ??? (in /usr/lib/dri/i965_dri.so) ==19388== by 0x685D70B: ??? (in /usr/lib/dri/i965_dri.so) ==19388== by 0x685DC71: ??? (in /usr/lib/dri/i965_dri.so) ==19388== by 0x685E661: ??? (in /usr/lib/dri/i965_dri.so) ==19388== by 0x683C823: ??? (in /usr/lib/dri/i965_dri.so) ==19388== by 0x65B3FED: ??? (in /usr/lib/dri/i965_dri.so) ==19388== by 0x638403D: shared_dispatch_stub_662 (in /usr/lib/libglapi.so.0.0.0) ==19388== by 0x4FDC95D: surface_depth_blt_fbo (surface.c:947) ==19388== by 0x4FDDC84: wined3d_surface_depth_blt (surface.c:1226) ==19388== by 0x4FF1824: wined3d_surface_blt (surface.c:6159) ==19388== by 0x4F494B6: resolve_depth_buffer (device.c:1895) ==19388== by 0x4F49702: wined3d_device_set_render_state (device.c:1924) ==19388== by 0x4982849: d3d9_device_SetRenderState (device.c:1671) ==19388== by 0x4CB4172: resz_test (visual.c:15179) ==19388==
let me know if you'd like a bug for that.