Re: [PATCH 1/2] Add an optional "alias" attribute to syscall entries.

[John Baldwin <jhb@FreeBSD.org>]

On 10/11/18 11:13 PM, Kevin Buettner wrote:
> Hi John,
> 
> On Wed,  3 Oct 2018 10:30:04 -0700
> John Baldwin <jhb@FreeBSD.org> wrote:
> 
>> When setting a syscall catchpoint by name, catch syscalls whose name
>> or alias matches the requested string.
>>
>> When the ABI of a system call is changed in the FreeBSD kernel, this
>> is implemented by leaving a compatability system call using the old
>> ABI at the existing "slot" and allocating a new system call for the
>> version using the new ABI.  For example, new fields were added to the
>> 'struct kevent' used by the kevent() system call in FreeBSD 12.  The
>> previous kevent() system call in FreeBSD 12 kernels is now called
>> freebsd11_kevent() and is still used by older binaries compiled
>> against the older ABI.  The freebsd11_kevent() system call can be
>> tagged with an "alias" attribute of "kevent" permitting 'catch syscall
>> kevent' to catch both system calls and providing the expected user
>> behavior for both old and new binaries.  It also provides the expected
>> behavior if GDB is compiled on an older host (such as a FreeBSD 11
>> host).
> 
> Very nice.
> 
> I read through your patch.  The only problem I found was this
> use of a GNU extension involving the use of the ternary ?: operator
> without the middle operand.
> 
>>  			     char *groups)
>>  {
>> -  syscall_desc *sysdesc = new syscall_desc (number, name);
>> +  syscall_desc *sysdesc = new syscall_desc (number, name, alias ?: "");
>>  
>>    syscalls_info->syscalls.emplace_back (sysdesc);
>>  
> 
> In addition, the GDB coding standard specifies that pointer variables
> should have explicit comparisons against NULL or nullptr.  So, even if
> it weren't a GNU extension, GDB's coding standard would force you to
> write that expression using an explicit comparison - which in turn would
> necessitate adding the middle argument.  (Which is kind of
> unfortunate, because I like the compactness of that expression.)

Oof, I'll expand it out instead.
-- 
John Baldwin