Re: [PATCH for-6.1 v5 00/15] tcg: breakpoint reorg

[Richard Henderson]

On 7/19/21 3:17 PM, Richard Henderson wrote:
> Patches lacking review are all new:
>    03-target-alpha-Drop-goto_tb-path-in-gen_call_pal.patch
>    08-hw-core-Introduce-TCGCPUOps.debug_check_breakpoin.patch
>    09-target-arm-Implement-debug_check_breakpoint.patch
>    10-target-i386-Implement-debug_check_breakpoint.patch
>    11-accel-tcg-Merge-tb_find-into-its-only-caller.patch
>    12-accel-tcg-Move-breakpoint-recognition-outside-tra.patch
>    13-accel-tcg-Remove-TranslatorOps.breakpoint_check.patch
>    15-accel-tcg-Record-singlestep_enabled-in-tb-cflags.patch

... and then there's a bit of code in translate/avr that's a bit odd:

    /*
     * This is due to some strange GDB behavior
     * Let's assume main has address 0x100:
     * b main   - sets breakpoint at address 0x00000100 (code)
     * b *0x100 - sets breakpoint at address 0x00800100 (data)
     *
     * The translator driver has already taken care of the code pointer.
     */
    if (!ctx->base.singlestep_enabled &&
        cpu_breakpoint_test(cs, OFFSET_DATA + ctx->base.pc_next, BP_ANY)) {
        gen_breakpoint(ctx);
        return;
    }

I guess the thing to do is to allow the address to be frobbed in cpu_breakpoint_insert, so 
that we squash this phantom address early.


r~