Re: [PATCH] S390: Sync ptrace.h with kernel. [BZ #21539]

[Stefan Liebler <stli@linux.vnet.ibm.com>]

On 06/19/2017 03:26 PM, Dmitry V. Levin wrote:
> On Mon, Jun 19, 2017 at 03:10:57PM +0200, Stefan Liebler wrote:
>> On 06/13/2017 10:05 PM, Dmitry V. Levin wrote:
>>> On Tue, Jun 06, 2017 at 12:17:33PM +0200, Stefan Liebler wrote:
>>> [...]
>>>> diff --git a/sysdeps/unix/sysv/linux/s390/tst-ptrace-singleblock.c b/sysdeps/unix/sysv/linux/s390/tst-ptrace-singleblock.c
>>>> new file mode 100644
>>   >> [...]
>>>> +      /* Get information about tracee: gprs, last breaking address.  */
>>>> +      parea.len = sizeof (regs);
>>>> +      parea.process_addr = (unsigned long) &regs;
>>>> +      parea.kernel_addr = 0;
>>>> +      ptrace (PTRACE_PEEKUSR_AREA, pid, &parea);
>>>
>>> Note that you can verify whether PTRACE_PEEKUSR_AREA has returned
>>> the expected result by comparing registers with those returned
>>> by PTRACE_GETREGSET.  The latter is implemented on s390 since
>>> linux 2.6.27 so its use in glibc is safe.
>>>
>> Okay. Now the gprs are obtained by PTRACE_PEEKUSR_AREA and
>> PTRACE_GETREGSET. Afterwards I use memcmp to check whether the values
>> are the same.
>>>> +      ptrace (PTRACE_GET_LAST_BREAK, pid, NULL, &last_break);
>>>
>>> As these ptrace calls are expected to succeed,
>>> you might want to check their return code.
>>>
>> Done with several usages of TEST_VERIFY_EXIT.
>>
>>>> +
>>>> +      printf ("child IA: %p last_break: %p\n",
>>>> +	      (void *) regs[1], (void *) last_break);
>>>> +
>>>> +      /* Execute tracee until next taken branch.
>>>> +
>>>> +	 Note:
>>>> +	 Before the commit which introduced this testcase,
>>>> +	 <glibc>/sysdeps/unix/sysv/linux/s390/sys/ptrace.h
>>>> +	 uses ptrace-request 12 for PTRACE_GETREGS,
>>>> +	 but <kernel>/include/uapi/linux/ptrace.h
>>>> +	 uses 12 for PTRACE_SINGLEBLOCK.
>>>> +
>>>> +	 The s390 kernel has no support for PTRACE_GETREGS!
>>>> +	 Thus glibc ptrace.h is adjusted to match kernel ptrace.h.
>>>> +
>>>> +	 This test ensures, that PTRACE_SINGLEBLOCK defined in glibc
>>>> +	 works as expected.  If the kernel would interpret it as
>>>> +	 PTRACE_GETREGS, then the tracee will not make any progress
>>>> +	 and this testcase will time out.  */
>>>> +      ptrace (req_singleblock, pid, NULL, NULL);
>>>
>>> Likewise.
>>
>> I've attached the patch with the mentioned changes and the NEWS entry
>> requested by Andreas.
>>
>> Is this okay?
> 
> Looks fine, thanks.
> 
> 
Committed.

Thanks.
Stefan