Date: Wed, 14 Jul 1999 18:51:21 -0600 From: Warner Losh <imp@village.org> To: Nate Williams <nate@mt.sri.com> Cc: mobile@FreeBSD.ORG Subject: Re: pccardd patches for review Message-ID: <199907150051.SAA59773@harmony.village.org> In-Reply-To: Your message of "Wed, 14 Jul 1999 18:17:59 MDT." <199907150017.SAA00008@mt.sri.com> References: <199907150017.SAA00008@mt.sri.com> <199907140636.AAA53085@harmony.village.org>
next in thread | previous in thread | raw e-mail | index | archive | help
In message <199907150017.SAA00008@mt.sri.com> Nate Williams writes:
: You asked. The addition of block paren's were they are un-necessary
: violates style(9), which is the format of the code. (I know because I
: was forced to re-indent the whole base as it had so many different style
: formats that it was difficult to follow.)
Those specific {} are allowed by style(9), speficically:
Closing and opening braces go on the same line as the else. Braces that
aren't necessary *may* be left out.
(I highlighed may). It doesn't say that they must be omitted. I have
plans for those specific if/else clauses.
Bruce would likely point out it a gratuitous change, which is
correct, to a point. Since I'll be adding more code in those specific
if/else clauses, I wanted to minimize the number of diffs between that
extremely green code and the code I'm trying to commit. At the
moment, that is more important to me.
There will also be much code movement from PAO in the near future if
things go well.
: Should logerr be used instead of logmsg when an error occurs?
Likely... I just found one and used it.
: Hosokawa-san should be asked about the delays, as many of them came from
: PAO vi him.
I plan to ask him about it. In the interrum, I've made the value a
command line tunable option (rather than a compile time constant).
For most of the cards I have, 500uS is a good value. However, I have
one modem card that needs 1.5s in order to work reliably (or actually
to keep from hanging the system). I think I may change the default to
2s rather than 5s with the suggestion that the install disk raise it
back to 5s to be sure.
In looking at the spec, there are signals that need to be waited for
which might be implemented as ioctls if we were going to keep pccardd
in userland long term. Since my plans are to move to an in-kernel
configuration of pccards, I'm going to not do that in this round of
cleanups....
Warner
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-mobile" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199907150051.SAA59773>