Open joelreymont opened 1 year ago
This is not consistent as I have plenty of functions that were created and that start with CBZ.
Linearsweep of ARM/Thumb2 code is a very trick problem. If you have a binary you can share we might be able to tweak our heuristics.
Will ping you on Slack.
On Wed, Apr 19, 2023 at 5:29 PM Peter LaFosse @.***> wrote:
Linearsweep of ARM/Thumb2 code is a very trick problem. If you have a binary you can share we might be able to tweak our heuristics.
— Reply to this email directly, view it on GitHub https://github.com/Vector35/binaryninja-api/issues/4245#issuecomment-1514840682, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAESZZXUNKI5TEMKN4ZQOTXB7ZC5ANCNFSM6AAAAAAXEAM5Y4 . You are receiving this because you authored the thread.Message ID: @.***>
V35 should search for "Encourage Salesman Prompt Delay" to find the database.
Here's another example...
This
000c6218 78 x
000c6219 data_c6219:
000c6219 b1 6a b1 .j.
000c621c int32_t sub_c621c(int32_t* arg1, int32_t arg2, void* arg3)
000c621c 2429 cmp r1, #0x24
000c621e 0bd1 bne #0xc6238
000c6220 02f10801 add r1, r2, #8
000c6224 02f12c03 add r3, r2, #0x2c
000c6228 51f8042b ldr r2, [r1], #4
000c622c 9942 cmp r1, r3
000c622e 40f8042b str r2, [r0], #4
000c6232 f9d1 bne #0xc6228
000c6234 2420 movs r0, #0x24
000c6236 7047 bx lr
000c6238 0020 movs r0, #0
000c623a 7047 bx lr
should really be this
000c6218 void sub_c6218(int32_t* arg1, int32_t arg2, void* arg3)
000c6218 78b1 cbz r0, #0xc623a
000c621a 6ab1 cbz r2, #0xc6238
000c621c 2429 cmp r1, #0x24
000c621e 0bd1 bne #0xc6238
...
Version and Platform (required):
Bug Description:
Please see attached screenshots.
Screenshots: