Re: re.Fix hang in mpu401_uart.c patch

[Takashi Iwai <tiwai@suse.de>]

At Wed, 17 May 2006 22:45:21 +0100,
Alan Horstmann wrote:
> 
> On Tuesday 16 May 2006 18:50, you wrote:
> > Ok. I signed up to alsa-devel and read the thread...
> >
> > On 5/15/06, Alan Horstmann <gineera@aspect135.co.uk> wrote:
> > > > > > However, as the 3rd augument is '0' ie don't check ACK, OK is
> > > > > > always true in snd_mpu401_uart_cmd and there is surely no point to
> > > > > > the extra check lines?
> >
> > IIRC I added the extra checks you refer to for completeness (it's bad
> > form not to check return), but the error path in snd_mpu401_uart_cmd
> > is getting called on my system when it's called during init -
> > printk()ing a warning is useless, the function should return an error.
> >
> > What's the problem anyway that you're experiencing?
> 
> It's no big deal -it's not causing any problems, and I understand entirely why 
> the function should be able to return error, and that the return should be 
> checked.  However I just noticed that in the case of the 2 'close' functions, 
> calling snd_mpu401_uart_cmd with 3rd arg '0', an error could never be 
> returned (ok always=1), and so there is no point in checking for it.  As 
> Takashi is normally keen to omit any unnecessary lines, I thought I would 
> mention it.

Yeah it's a small dead code right now.

But in this case it's better to keep the error check for possible
future changes (for example, snd_mpu401_uart_cmd() returns an error
when tx timeout occurs).  If we remove the error check there, this
change would have no effect, too.


Takashi


-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642