On Thu Dec 4 17:43:17 2025 +0000, eric pouech wrote:
that's why I get with this MR (latest update) + dbghelp change on old wow configuration, running from the 64bit environemnt `dbghelp.c:983: Test failed: Wrong bitness aggregation count 0 0 ` `dbghelp.c:985: Test failed: Wrong kind aggregation count 0 0 ` `dbghelp.c:987: Test failed: Wrong directory aggregation count 0 0 ` `dbghelp.c:1045: Test failed: Wrong kind aggregation count 1 1 ` `dbghelp.c:1442: Test failed: [64/32 enum:+32bit wine_old_wow64]: Unexpected enum.count_exe 1 ` `dbghelp.c:1500: Test failed: [64/32 enum:+32bit wine_old_wow64]: Different sym/enum count32_bit (24/24) ` `dbghelp.c:1506: Test failed: [64/32 enum:+32bit wine_old_wow64]: Different sym/enum wowdir (23/23) ` `dbghelp.c:1508: Test failed: [64/32 enum:+32bit wine_old_wow64]: Different sym/enum exe (1/1) ` `dbghelp.c:1442: Test failed: [64/32 enum:default wine_old_wow64]: Unexpected enum.count_exe 0 ` `dbghelp.c:1455: Test failed: [64/32 enum:default wine_old_wow64]: Unexpected sym.count_exe 0` the first three entries with 'aggregation count 0 0' come from 0 module returned (IIRC) the next ones, stem from different system32 vs syswow64 report of modules (didn't check the details), but except for ntdll -- wine kernelbase doesn't return the same location as native does -- outcome should match what the dbghelp test do (otherwise you'll get the DLL in the wrong bitness) and I get an error from old wow, test run in 32bit dbghelp.c:1475: Test failed: \[32/32 enum:default wine_old_wow64\]: Unexpected event.count_wowdir 5 which doesn't correct either yes, this is the same thing i got. my next push will fix these problems.
-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9371#note_124733