Closed v1X3Q0 closed 2 years ago
Does this reproduce on the latest development branch? There have been a large number of changes there and we're very close to releasing a new stable.
It does, though instigating it seems a little different due to the interface changes. I'll see if i can post a crash of that one as well.
Thanks -- can you share the binary triggering it? Does the crash happen on multiple files or just one?
i cannot share the binary triggering it, but i'll see if i can replicate it with another binary
The development branch doesn't seem to be rendering any strings, so my searches aren't working... I'll prioritize getting you a binary and bndb.
Hmm, I don't recall anything that intentionally changed that. That might be a separate issue. I can definitely search disassembly for "call" on both dev and stable and get similar results.
(there's less in dev because of a fix that caused the calling convention to go from fastcall to regparm)
One other thing to check might be to try launching with. -p
which will disable plugins temporarily so you can make sure no plugin is impacting that (it shouldn't but easy enough to check)
here is the dev branch crash and call stack
Working on that binary
Can you copy that as text? Much easier to rebase appropriately. And confirm the exact version number please (we've been pumping out a bunch of dev builds the last few days)
(61ec.4240): Access violation - code c0000005 (first chance)
First chance exceptions are reported before any exception handling.
This exception may be expected and handled.
binaryninjacore!BNSetBinaryReaderEndianness+0x7c116:
00007fff`e72cbf46 440fb65001 movzx r10d,byte ptr [rax+1] ds:0000026e`befa1000=??
0:000> k
# Child-SP RetAddr Call Site
00 00000089`2352a840 00007fff`e7297297 binaryninjacore!BNSetBinaryReaderEndianness+0x7c116
01 00000089`2352a890 00007fff`e72ce82f binaryninjacore!BNSetBinaryReaderEndianness+0x47467
02 00000089`2352a9b0 00007ff8`4a5f0314 binaryninjacore!BNFindAllDataWithProgress+0x1bf
03 00000089`2352ab30 00007ff8`4a535d18 binaryninjaui!CrossReferenceWidget::wheelEvent+0x7ba74
04 00000089`2352ac40 00007ff6`7a82f920 binaryninjaui!View::findAllData+0x68
05 00000089`2352ac90 00007ff6`7a838d0a binaryninja+0x7f920
06 00000089`2352b010 00007ff8`0c3083db binaryninja+0x88d0a
07 00000089`2352b0a0 00007ff8`0c306aee Qt6Core!QMetaCallEvent::placeMetaCall+0x3b
08 00000089`2352b0e0 00007ff8`0cd8bec6 Qt6Core!QObject::event+0x18e
09 00000089`2352b300 00007ff8`0cd51e0d Qt6Widgets!QWidget::event+0xe76
0a 00000089`2352b3e0 00007ff8`0cd50f5b Qt6Widgets!QApplicationPrivate::notify_helper+0x10d
0b 00000089`2352b410 00007ff8`0c2c6894 Qt6Widgets!QApplication::notify+0x190b
0c 00000089`2352b8e0 00007ff8`0c2c8c57 Qt6Core!QCoreApplication::notifyInternal2+0xc4
0d 00000089`2352b950 00007ff8`0765aeaf Qt6Core!QCoreApplicationPrivate::sendPostedEvents+0x227
0e 00000089`2352ba30 00007ff8`0c40cf92 Qt6Gui!QWindowsGuiEventDispatcher::sendPostedEvents+0xf
0f 00000089`2352ba60 00007ff8`0765ae89 Qt6Core!QEventDispatcherWin32::processEvents+0x72
10 00000089`2352eba0 00007ff8`0c2cb4fe Qt6Gui!QWindowsGuiEventDispatcher::processEvents+0x19
11 00000089`2352ebd0 00007ff8`0c2c496b Qt6Core!QEventLoop::exec+0x19e
12 00000089`2352ec70 00007ff6`7a7f4168 Qt6Core!QCoreApplication::exec+0x15b
13 00000089`2352ecd0 00007ff6`7a9de9a7 binaryninja+0x44168
14 00000089`2352f810 00007ff6`7a9dd972 binaryninja+0x22e9a7
15 00000089`2352f8a0 00007ff8`551b7034 binaryninja+0x22d972
16 00000089`2352f8e0 00007ff8`56122651 KERNEL32!BaseThreadInitThunk+0x14
17 00000089`2352f910 00000000`00000000 ntdll!RtlUserThreadStart+0x21
Nothing stands out from the crash-log. Do you have a repeatable set of actions that can crash it on any other file?
The set of actions were as bare as opening the file, ctrl+f and searching for "call", i tried to randomize the bytes not of the string in the specific file so that i could share it with you, but then it doesn't crash anymore... I'm gonna see if i can share the specific bndb with you, no promises.
If this is still crashing for you can you please post a crash log and the exact version you're testing it on (either dev or stable is fine)
I've tried this on a bunch of binaries and databases on Windows 11 with 3.0.3321-dev and haven't had it crash on me. Since we don't have a test case we can reproduce the issue with, and there hasn't been any additional information in 2 months, I'm going to close this. Feel free to re-open if there's anything additional we can use to track down the issue, though.
Binary Ninja Version: 2.4.2846 Personal, df7c027e Platform: Windows 10 Version 2009 I'm searching for the string "call" in my generated strings, and binary ninja seems to be crashing.