3rd-party/linux-kernel-module-cheat
1
0
Fork 0
mirror of https://github.com/cirosantilli/linux-kernel-module-cheat.git synced 2026-01-23 02:05:57 +01:00
Code Issues Releases Wiki Activity

readme: qemu failed interesting multicore debug attempt

Browse Source 
qemu cannot trace memory accesses
This commit is contained in:
Ciro Santilli
2018-05-02 17:07:39 +01:00
parent fe8844b3cd
commit bde13c3772
1 changed files with 19 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
README.adoc
View File

@@ -1399,6 +1399,18 @@ on many (all?) functions.
See also: https://github.com/cirosantilli/linux-kernel-module-cheat/issues/19
=== GDB step debug multicore
https://stackoverflow.com/questions/42800801/how-to-use-gdb-to-debug-qemu-with-smp-symmetric-multiple-processors
Modify the number of cores: <<number-of-cores>>
TODO: how to do something cool to see that in action?
I tried to play around with `taskset`, but when I have two CPUs the <<gdb-step-debug-userland-non-init,userland breakpoints>> don't work... Why?
We should also try it out with kernel modules: https://stackoverflow.com/questions/28347876/set-cpu-affinity-on-a-loadable-linux-kernel-module
== KGDB
TODO: only working with <<graphic-mode>>. Without it, nothing shows on the terminal. So likely something linked to the option `console=ttyS0`.
@@ -3535,7 +3547,13 @@ just appears to output both cores intertwined without any clear differentiation.
TODO: is is possible to show which instructions ran at each point in time, in addition to the address of the instruction with `exec_tb` shows? Hopefully dissembled, not just the instruction memory.
PANDA can list memory addresses, so I bet it can also decode the instructions: https://github.com/panda-re/panda/blob/883c85fa35f35e84a323ed3d464ff40030f06bd6/panda/docs/LINE_Censorship.md I wonder why they don't just upstream those things to QEMU's tracing.
PANDA can list memory addresses, so I bet it can also decode the instructions: https://github.com/panda-re/panda/blob/883c85fa35f35e84a323ed3d464ff40030f06bd6/panda/docs/LINE_Censorship.md I wonder why they don't just upstream those things to QEMU's tracing: https://github.com/panda-re/panda/issues/290
Memory access on vanilla seem impossible due to optimizations that QEMU does:
* https://lists.gnu.org/archive/html/qemu-devel/2015-06/msg07479.html
* https://lists.gnu.org/archive/html/qemu-devel/2014-04/msg02856.html
* https://lists.gnu.org/archive/html/qemu-devel/2012-08/msg03057.html
==== gem5 tracing