This is self answer for another possibility.
Another possibility? ->original uClinux kernel treat TRAP_VEC15 as a SIGTRAP
therefore if gdb generate trap #15 as a breakpoint instruction then
i could avoid kernel modification for trap #1 catch.
But, Well, I can not avoid kernel patch for uCsimm, because
, without modification, uCsimm kernel route trap #15 exception to
somewhere at the bootloader (just like trap #1).
So, when I tried trap #15 instruction I met SIGILL (not SIGTRAP).
>I found another possibility for gdb on uClinux.
>GDB has definitions for TRAP vector for break point. (gdb/config/tm_m68k.h)
>Then, We could use gdbserver without kernel modification for trap vector.
>I'll try this.
>BTW, If all of you have no trouble with my gdbserver (after many
>I hope that we could see uClinux support on new gdb release.
>For that, maybe, we should provide low-uclinux.c and a patch for
>for auto-configuration (like ./configure --target=m68k-uclinux-coff)
>This message resent by the firstname.lastname@example.org list server
This message resent by the email@example.com list server http://www.uClinux.com/
This archive was generated by hypermail 2b30 : Sun Apr 07 2002 - 00:01:34 EST