Re: [Qemu-devel] [PATCH 1/2] target/nios2: Fix bug in semihosted exit handling

[Laurent Vivier]

Le 21/08/2019 à 17:27, Sandra Loosemore a écrit :
> On 8/21/19 8:41 AM, Laurent Vivier wrote:
>> Le 21/08/2019 à 16:21, Sandra Loosemore a écrit :
>>> This patch fixes a bug that caused semihosted exit to always return
>>> status 0; it was incorrectly using the value of register R_ARG0 (which
>>> contains the HOSTED_EXIT request number) instead of register R_ARG1.
>>>
>>> Signed-off-by: Sandra Loosemore <address@hidden>
>>> ---
>>>   target/nios2/nios2-semi.c | 4 ++--
>>>   1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/target/nios2/nios2-semi.c b/target/nios2/nios2-semi.c
>>> index d7a80dd..06c0861 100644
>>> --- a/target/nios2/nios2-semi.c
>>> +++ b/target/nios2/nios2-semi.c
>>> @@ -215,8 +215,8 @@ void do_nios2_semihosting(CPUNios2State *env)
>>>       args = env->regs[R_ARG1];
>>>       switch (nr) {
>>>       case HOSTED_EXIT:
>>> -        gdb_exit(env, env->regs[R_ARG0]);
>>> -        exit(env->regs[R_ARG0]);
>>> +        gdb_exit(env, env->regs[R_ARG1]);
>>> +        exit(env->regs[R_ARG1]);
>>
>> It's weird: in line 215,  env->regs[R_ARG1] is args.
>>
>> Are you sure it's not something like:
>>
>>          GET_ARG(0)
>>          gdb_exit(env, arg0);
>>          exit(arg0);
>>
>> same for m68k.
>>
>> Did you check the kernel code?
> 
> It's not the kernel that's involved here, it's libgloss.  And yes, the
> HOSTED_EXIT case takes an immediate argument in the register rather than
> a pointer to an argument block.
> 
> Here's the documentation for nios2 semihosting.
> 
> https://www.sourceware.org/git/gitweb.cgi?p=newlib-cygwin.git;a=blob;f=libgloss/nios2/nios2-semi.txt;h=ded3a093c03dbae84cb95b4cd45bc3e0d751eda2;hb=HEAD
> 
> 
> And m68k:
> 
> https://www.sourceware.org/git/gitweb.cgi?p=newlib-cygwin.git;a=blob;f=libgloss/m68k/m68k-semi.txt;h=50520c15292aa7edf7eef28e09fd9202ce75b153;hb=HEAD
> 
> 
> Again, a lot of cutting and pasting involved here.  ;-)
> 
> -Sandra

Thank you for the details.

Could add this information in the commit messages of each patch?

Reviewed-by: Laurent Vivier <address@hidden>