On 04/19/10 02:02, matthew green wrote:
XXX Should kernel rev be bumped?
did you change any ABIs? ie, will any old modules of the same version
now fail to load due to your change, or fail to work properly?
No
it seems unlikely, but i didn't read the big diff.
It fixes a bug that coul
On Mon, 19 Apr 2010, Joerg Sonnenberger wrote:
> On Sun, Apr 18, 2010 at 10:51:35AM +, Alan Barrett wrote:
> > Log Message:
> > Bitwise operations on signed types are well-defined if the values
> > happen to be positive, and indeed the values here were guaranteed
> > to be positive, but some co
On Mon, Apr 19, 2010 at 03:02:05AM +, Christos Zoulas wrote:
>> In that case use an explicit cast to the unsigned type. What you did is
>> penalize the code for no good reason.
>
> Any compiler these days knows how to do these micro-optimizations.
Furthermore, as swab() is fairly useless,
In article <20100419021334.ga8...@britannica.bec.de>,
Joerg Sonnenberger wrote:
>On Sun, Apr 18, 2010 at 10:51:35AM +, Alan Barrett wrote:
>> Log Message:
>> Bitwise operations on signed types are well-defined if the values
>> happen to be positive, and indeed the values here were guaranteed
On Sun, Apr 18, 2010 at 10:51:35AM +, Alan Barrett wrote:
> Log Message:
> Bitwise operations on signed types are well-defined if the values
> happen to be positive, and indeed the values here were guaranteed
> to be positive, but some compilers complained anyway, so convert
> the bitwise opera
XXX Should kernel rev be bumped?
did you change any ABIs? ie, will any old modules of the same version
now fail to load due to your change, or fail to work properly?
it seems unlikely, but i didn't read the big diff.
.mrg.
On Apr 18, 11:01am, mar...@duskware.de (Martin Husemann) wrote:
-- Subject: Re: CVS commit: src/lib/libc/string
| On Sat, Apr 17, 2010 at 09:14:21PM -0400, Christos Zoulas wrote:
| > Yes, it does not make sense, but that is what POSIX specifies.
|
| Yeah, but wouldn't ignoring this part (keeping
On Sat, Apr 17, 2010 at 09:14:21PM -0400, Christos Zoulas wrote:
> Yes, it does not make sense, but that is what POSIX specifies.
Yeah, but wouldn't ignoring this part (keeping it size_t) still work
for all valid calls, i.e. differing from the obviously stupid standard
be a harmless thing?
Martin
