svn commit: r284167 - head/sys/i386/i386

[Dimitry Andric]

On 09 Jun 2015, at 08:44, Bruce Evans <brde at optusnet.com.au> wrote:
> 
> On Mon, 8 Jun 2015, Dimitry Andric wrote:
> 
>> Log:
>> Merge r283870 from amd64:
>> 
>> Remove unneeded NULL checks in trap_fatal().
>> 
>> Since td_name is an array member of struct thread, it can never be NULL,
>> so the check can be removed.  In addition, curproc can never be NULL,
>> so remove the if statement, and splice the two printfs() together.
>> 
>> While here, remove the u_long cast, and use the correct printf format
>> specifier for curproc->p_pid.
>> 
>> Requested by:	jhb
> 
> Er. I gave a longer review which implicity requested not doing all of
> this.  The format was correct (it matched the cast), and the cast was
> less wrong than not casting.

Please read https://reviews.freebsd.org/D2695, where Kostik argued
"pid_t is int32_t on all arches", and I agreed with that.  The previous
obfuscation is unnecessary now.


> Both amd64/trap.c i386/trap.c still print pids portably (by casting
> to long) in one place.  They each had 2 unportable printings of pids;
> now they each have 3 unportable printings of pids.

I wasn't updating the other parts of the code, so I stayed out of there
for now.  Feel free to put a review in Phabricator to make everything
consistent.

-Dimitry

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 194 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://lists.freebsd.org/pipermail/svn-src-all/attachments/20150609/cd9630d2/attachment.sig>