Closed juxeii closed 1 year ago
The crash happened in function read_memory_wrapper
, which copies data from the mmap-ed core file to the input buffer. I would guess the disk IO or the core file has an issue. Could you repeat the crash in your environment?
Hi,
no, it is a cross debug session. At startup there are a lot of these warnings:
warning: Can't open file /dev/shm/CCS-AACONFIG_RAD_TAG_FILE-0-NO_ENV_NAME during file-backed mapping note processing
warning: Can't open file /dev/shm/CCS-AAFILE_STDIO_SHMEM_ROM_PATH-0-NO_ENV_NAME during file-backed mapping note processing
warning: Can't open file /dev/shm/CCS-AAFILE_STDIO_SHMEM_RAM_PATH-0-NO_ENV_NAME during file-backed mapping note processing
So, I guess, if I am not on the real target, these mappings won't work, and so the tool fails?
I see. It is an interesting usage. My guess is that mapping doesn't support the memory device well because its content changes.
Too bad :( Thx for your help and this nice tool!
Hi,
I have compiled GDB 12.1 and tried to work with a coredump.
Architecture is
The target architecture is set to "auto" (currently "i386:x86-64").
Any hints on where to start the investigation?