[Bug 40945] New: Post provider setup of WMI core 1.5 installer needs
https://bugs.winehq.org/show_bug.cgi?id=40945 Bug ID: 40945 Summary: Post provider setup of WMI core 1.5 installer needs Product: Wine Version: 0.9.34. Hardware: x86 URL: https://web.archive.org/web/*/http://download.microsof t.com/download/platformsdk/wmi9x/1.5/W9X/EN-US/wmi9x.e xe OS: Linux Status: NEW Keywords: download, Installer, patch Severity: minor Priority: P2 Component: advapi32 Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com Distribution: --- Follow up to bug 37429 You'll need to grab the exe from archive.org. Set your wineprefix to win98, and install. It will crash late in the process, with: Unhandled exception: page fault on read access to 0x00000004 in 32-bit code (0x7da6799f). Register dump: CS:0023 SS:002b DS:002b ES:002b FS:0063 GS:006b EIP:7da6799f ESP:0033c590 EBP:0033c5c8 EFLAGS:00210202( R- -- I - - - ) EAX:7d895869 EBX:7da81000 ECX:0033c580 EDX:7ffdf000 ESI:01419738 EDI:00000000 Stack dump: 0x0033c590: 7d98a460 014192b8 00000000 014192e0 0x0033c5a0: 0033c630 01419b50 0000c5d0 7d98a460 0x0033c5b0: 00000020 7da67946 0033c5dc 00b81aa0 0x0033c5c0: 00000000 0033cbe0 0033c5f8 6206ccb5 0x0033c5d0: 00000000 0033cbe0 6206ccb5 01419b50 0x0033c5e0: 00000000 0033c630 00e90690 00000000 Backtrace: =>0 0x7da6799f class_object_SpawnInstance+0x6f(iface=<couldn't compute location>, lFlags=<couldn't compute location>, ppNewInstance=<couldn't compute location>) [/home/austin/wine-git/dlls/wbemprox/class.c:221] in wbemprox (0x0033c5c8) 1 0x6206ccb5 in mofd (+0xccb4) (0x0033c5f8) 2 0x6206e94a in mofd (+0xe949) (0x0033c680) 3 0x62062199 in mofd (+0x2198) (0x0033cbec) 4 0x62074d0a in mofd (+0x14d09) (0x0033d1bc) 5 0x003428bc in wbemupgd (+0x28bb) (0x0033d408) 6 0x003410ac in wbemupgd (+0x10ab) (0x0033d9a4) 7 0x00343528 in wbemupgd (+0x3527) (0x0033d9b8) 8 0x100150d0 in glc96e7.tmp (+0x150cf) (0x0033e3d8) 9 0x1000433e in glc96e7.tmp (+0x433d) (0x0033f1c8) 0x7da6799f class_object_SpawnInstance+0x6f [/home/austin/wine-git/dlls/wbemprox/class.c:221] in wbemprox: movl 0x4(%edi),%eax 221 if (!(record->fields = heap_alloc( table->num_cols * sizeof(struct field) ))) wine-1.9.14-15-g1bcc7ce austin(a)austin2:~$ sha1sum wmi9x.exe 62752e9c1b879688c26f205eebf07d3783906c3e wmi9x.exe austin(a)austin2:~$ du -sh wmi9x.exe 3.0M wmi9x.exe -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=40945 Austin English <austinenglish(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|Post provider setup of WMI |Post provider setup of WMI |core 1.5 installer needs |core 1.5 installer fails in | |wbemprox -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=40945 Nikolay Sivov <bunglehead(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Keywords|patch | Component|advapi32 |wmi&wbemprox -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=40945 winetest(a)luukku.com changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |winetest(a)luukku.com --- Comment #1 from winetest(a)luukku.com --- Still crashes wine 2.4 and staging 2.4. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=40945 --- Comment #2 from winetest(a)luukku.com --- Still valid wine 2.5-git. I tried few hacks around source. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=40945 Gijs Vermeulen <gijsvrm(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Severity|minor |normal URL|https://web.archive.org/web |https://web.archive.org/web |/*/http://download.microsof |/20160401071609/http://down |t.com/download/platformsdk/ |load.microsoft.com/download |wmi9x/1.5/W9X/EN-US/wmi9x.e |/platformsdk/wmi9x/1.5/W9X/ |xe |EN-US/wmi9x.exe --- Comment #3 from Gijs Vermeulen <gijsvrm(a)gmail.com> --- This is still present in wine-5.14. I've debugged this a bit and I think there are multiple issues here. 1) The installer calls IWbemServices::GetObject to query "__NAMESPACE". get_object calls create_instance_enum which calls exec_query. This all succeeds, but will cause view->table_count to be 0. It seems IWbemServices::GetObject should maybe use get_propval to go to get_system_propval so the right things are returned. 2) The actual cause of the crash is that after create_instance_enum returns, get_object calls create_class_object and the app will use the returned IWbemClassObject to call IWbemClassObject::SpawnInstance. SpawnInstance calls create_record with a NULL table. Since table->num_cols is 0, record->fields won't be properly alloc'd, ultimately leading to a crash. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=40945 Zebediah Figura <z.figura12(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |z.figura12(a)gmail.com --- Comment #4 from Zebediah Figura <z.figura12(a)gmail.com> --- (In reply to Gijs Vermeulen from comment #3)
This is still present in wine-5.14.
I've debugged this a bit and I think there are multiple issues here.
1) The installer calls IWbemServices::GetObject to query "__NAMESPACE". get_object calls create_instance_enum which calls exec_query. This all succeeds, but will cause view->table_count to be 0.
It seems IWbemServices::GetObject should maybe use get_propval to go to get_system_propval so the right things are returned.
2) The actual cause of the crash is that after create_instance_enum returns, get_object calls create_class_object and the app will use the returned IWbemClassObject to call IWbemClassObject::SpawnInstance. SpawnInstance calls create_record with a NULL table. Since table->num_cols is 0, record->fields won't be properly alloc'd, ultimately leading to a crash.
Oh hey, I may have just fixed that: https://source.winehq.org/git/wine.git/commitdiff/a0c9aab5deffcdf30826574ba3... -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=40945 --- Comment #5 from Gijs Vermeulen <gijsvrm(a)gmail.com> --- (In reply to Zebediah Figura from comment #4)
(In reply to Gijs Vermeulen from comment #3)
Oh hey, I may have just fixed that:
https://source.winehq.org/git/wine.git/commitdiff/ a0c9aab5deffcdf30826574ba331ea8f18616ce3
This certainly did help. I did see this patch beforehand, but of course I didn't put two and two together... Unfortunately, this commit does not fix all the crash but evolves 2) into a crash in Put(), for which I've just sent patches. The installer now ends with an error message, but doesn't crash anymore. I suspect it will need fixes for 1) and related issues for it to work fully. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=40945 --- Comment #6 from Gijs Vermeulen <gijsvrm(a)gmail.com> --- 2) is now fixed after: <https://source.winehq.org/git/wine.git/commit/a0c9aab5deffcdf30826574ba331ea8f18616ce3> <https://source.winehq.org/git/wine.git/commit/11f2057768ed2d5416fad30d6e3e18683ffc17cd> <https://source.winehq.org/git/wine.git/commit/c3bd08b02299e3ac89b708ff3bd4bda262ec6bde> The installer doesn't crash anymore, but finishes with errors. To fix those errors (and probably also the program in general), I suspect 1) will need to be fixed. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
participants (2)
-
wine-bugs＠winehq.org -
WineHQ Bugzilla