On 10/23/2016 06:20 AM, Richard Henderson wrote:
> On 10/22/2016 08:01 PM, Marek Vasut wrote:
>>> For signed division, you have to protect against 0x80000000 / -1 as
>>> well, which raises an overflow exception on the x86 host.
>>
>> You mean similar to what mips does on OPC_DIV vs OPC_DIVU , right ?
>
> Yes.
>
>>> No CPU_LOG_TB_IN_ASM disassembly? I thought patch 1 added a nios2
>>> disassembler.
>>
>> Nope, I removed it in V4, maybe i misunderstood the review comment?
>> "
>>> + /* Dump the CPU state to the log */
>>> + if (qemu_loglevel_mask(CPU_LOG_TB_IN_ASM)) {
>>> + qemu_log("--------------\n");
>>> + log_cpu_state(cs, 0);
>>> + }
>>
>> Cpu state is dumped by -d cpu generically.
>> "
>
> CPU state != disassembly. I mean
>
> if (qemu_loglevel_mask(CPU_LOG_TB_IN_ASM)
> && qemu_log_in_addr_range(pc_start)) {
> qemu_log("IN: %s\n", lookup_symbol(pc_start));
> log_target_disas(cs, pc_start, ctx.pc - pc_start, 1);
> qemu_log("\n");
> }
>
>
> r~
>
Both added, thanks!
--
Best regards,
Marek Vasut
