On Fri, 2007-06-08 at 12:48 +0100, Alan Cox wrote:
> > It works fine. The only problem is that if I set a _standard_ baud rate
> > with BOTHER and then read it back with something that doesn't grok
> > BOTHER, I get it back just as I set it.
>
> That seemed to me to be the right thing to do.
>
>
> It works fine. The only problem is that if I set a _standard_ baud rate
> with BOTHER and then read it back with something that doesn't grok
> BOTHER, I get it back just as I set it.
That seemed to me to be the right thing to do.
> It might be better if it was returning B38400, rather than BOTH
On Fri, 08 Jun 2007 12:14:49 +0100
David Woodhouse <[EMAIL PROTECTED]> wrote:
> The uart_set_termios() function will bail out early without bothering to
> touch the hardware, if it decides that nothing "relevant" has changed.
> Unfortunately, its idea of "relevant" doesn't include c_[io]speed. So
On Wed, 2007-06-06 at 13:30 +0100, David Woodhouse wrote:
> + printk("No relevant change\n");
Oops, that wasn't supposed to sneak into the final patch. I'll send a
new one.
--
dwmw2
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message
On Thu, 2007-06-07 at 23:54 +0100, Alan Cox wrote:
> If it doesn't only involve editing the header files for this case (and
> maybe needing a define to indicate old==new) then the tty layer wants
> fixing to sort that out. Its on my todo list.
It works fine. The only problem is that if I set a _s
On Thu, 07 Jun 2007 23:15:17 +0100
David Woodhouse <[EMAIL PROTECTED]> wrote:
> On Thu, 2007-06-07 at 22:55 +0100, Alan Cox wrote:
> > Umm if your struct termios has the c_ispeed/c_ospeed fields then you
> > don't need to add the new ioctls to the PPC either - the Alpha is the
> > same here.
>
>
On Thu, 2007-06-07 at 22:55 +0100, Alan Cox wrote:
> Umm if your struct termios has the c_ispeed/c_ospeed fields then you
> don't need to add the new ioctls to the PPC either - the Alpha is the
> same here.
Well, OK -- if it only involves editing patch files I might _send_ a
patch tonight but I'l
On Thu, 2007-06-07 at 22:55 +0100, Alan Cox wrote:
> Umm if your struct termios has the c_ispeed/c_ospeed fields then you
> don't need to add the new ioctls to the PPC either - the Alpha is the
> same here.
Ah, OK. I hadn't previously noticed that setting TERMIOS_OLD only
actually affected the co
On Thu, 07 Jun 2007 16:50:21 +0100
David Woodhouse <[EMAIL PROTECTED]> wrote:
> On Thu, 2007-06-07 at 16:38 +0100, Alan Cox wrote:
> > If your termios and termios2 structures differ in size then you need to
> > copy the right number of bytes or you won't get speed values into the
> > kernel. If th
On Thu, 2007-06-07 at 16:38 +0100, Alan Cox wrote:
> If your termios and termios2 structures differ in size then you need to
> copy the right number of bytes or you won't get speed values into the
> kernel. If they are the same size it wont matter.
+/* Yay. A third identical definition of the same
On Wed, 06 Jun 2007 18:29:58 +0100
David Woodhouse <[EMAIL PROTECTED]> wrote:
> On Wed, 2007-06-06 at 16:03 +0100, Alan Cox wrote:
> > Yep - and there are some other changes needed as well once everyone
> > gets their ports properly lined up (notably handing back the actual
> > speed).
>
> Yeah,
On Wed, 2007-06-06 at 16:03 +0100, Alan Cox wrote:
> Yep - and there are some other changes needed as well once everyone
> gets their ports properly lined up (notably handing back the actual
> speed).
Yeah, probably. This was was required just to get the speed thing to
pass basic testing though.
On Wed, 06 Jun 2007 13:30:10 +0100
David Woodhouse <[EMAIL PROTECTED]> wrote:
> The uart_set_termios() function will bail out early without bothering to
> touch the hardware, if it decides that nothing "relevant" has changed.
> Unfortunately, its idea of "relevant" doesn't include c_[io]speed. So
13 matches
Mail list logo
