[parisc-linux] parisc-linux Digest, Vol 2, Issue 33

[parisc-linux-request@lists.parisc-linux.org]

[-- Attachment #1: Type: text/plain, Size: 5957 bytes --]

Send parisc-linux mailing list submissions to
	parisc-linux@lists.parisc-linux.org

To subscribe or unsubscribe via the World Wide Web, visit
	http://lists.parisc-linux.org/mailman/listinfo/parisc-linux
or, via email, send a message with subject or body 'help' to
	parisc-linux-request@lists.parisc-linux.org

You can reach the person managing the list at
	parisc-linux-owner@lists.parisc-linux.org

When replying, please edit your Subject line so it is more specific
than "Re: Contents of parisc-linux digest..."


Today's Topics:

   1. Re: A fix for B2k and CONFIG_PDC_CONSOLE pb (Matthew Wilcox)
   2. Re: A fix for B2k and CONFIG_PDC_CONSOLE pb (Joel Soete)
   3. Re: A fix for B2k and CONFIG_PDC_CONSOLE pb (Joel Soete)


----------------------------------------------------------------------

Message: 1
Date: Mon, 26 Jan 2004 17:25:05 +0000
From: Matthew Wilcox <willy@debian.org>
Subject: Re: [parisc-linux] A fix for B2k and CONFIG_PDC_CONSOLE pb
To: Joel Soete <soete.joel@tiscali.be>
Cc: Christoph Plattner <christoph.plattner@gmx.at>,	Grant Grundler
	<grundler@parisc-linux.org>,
parisc-linux@lists.parisc-linux.org
Message-ID: <20040126172505.GQ11844@parcelfarce.linux.theplanet.co.uk>
Content-Type: text/plain; charset=us-ascii

On Mon, Jan 26, 2004 at 06:00:53PM +0100, Joel Soete wrote:
>  
>  	/* Bail if no console input device. */
> -	if (!PAGE0->mem_kbd.iodc_io)
> +	if ((PAGE0->mem_cons.cl_class != CL_DUPLEX) &&
!PAGE0->mem_kbd.iodc_io)
>  		return 0;
>  	
>  	/* wait for a keyboard (rs232)-input */
>  	spin_lock_irqsave(&pdc_lock, flags);
> -	real32_call(PAGE0->mem_kbd.iodc_io,
> -		    (unsigned long)PAGE0->mem_kbd.hpa, ENTRY_IO_CIN,
> -		    PAGE0->mem_kbd.spa, __pa(PAGE0->mem_kbd.dp.layers), 
> -		    __pa(iodc_retbuf), 0, __pa(iodc_dbuf), 1, 0);
> +	if (PAGE0->mem_cons.cl_class == CL_DUPLEX)

This doesn't make sense.  PAGE0->mem_cons.cl_class has to be CL_DUPLEX
otherwise the test above would have failed.  Do you perhaps mean that &&
above to be an || ?  If so, this would all read better as ...

	/* wait for a keyboard (rs232)-input */
	spin_lock_irqsave(&pdc_lock, flags);
	if (PAGE0->mem_cons.cl_class == CL_DUPLEX) {

		real32_call(PAGE0->mem_cons.iodc_io,
			    (unsigned long)PAGE0->mem_cons.hpa,
ENTRY_IO_CIN,
			    PAGE0->mem_cons.spa,
__pa(PAGE0->mem_cons.dp.layers), 
			    __pa(iodc_retbuf), 0, __pa(iodc_dbuf), 1,
0);
	} else if (PAGE0->mem_kbd.iodc_io) {
		real32_call(PAGE0->mem_kbd.iodc_io,
			    (unsigned long)PAGE0->mem_kbd.hpa,
ENTRY_IO_CIN,
			    PAGE0->mem_kbd.spa,
__pa(PAGE0->mem_kbd.dp.layers), 
			    __pa(iodc_retbuf), 0, __pa(iodc_dbuf), 1,
0);
	} else {
		spin_unlock_irqrestore(&pdc_lock, flags);
		return 0;
	}

Slightly ugly to have two unlocks for the same lock, but that's better
than duplicating the test.  I haven't looked at the code, maybe we could
use a goto instead.

-- 
"Next the statesmen will invent cheap lies, putting the blame upon 
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and
refuse
to examine any refutations of them; and thus he will by and by convince 
himself that the war is just, and will thank God for the better sleep 
he enjoys after this process of grotesque self-deception." -- Mark Twain


------------------------------

Message: 2
Date: Mon, 26 Jan 2004 19:20:26 +0100
From: "Joel Soete" <soete.joel@tiscali.be>
Subject: Re: [parisc-linux] A fix for B2k and CONFIG_PDC_CONSOLE pb
To: "Matthew Wilcox" <willy@debian.org>
Cc: Christoph Plattner <christoph.plattner@gmx.at>,	Grant Grundler
	<grundler@parisc-linux.org>,
parisc-linux@lists.parisc-linux.org
Message-ID: <40153110000001C0@ocpmta2.freegates.net>
Content-Type: text/plain; charset="ISO-8859-1"



> +	if ((PAGE0->mem_cons.cl_class != CL_DUPLEX) &&
!PAGE0->mem_kbd.iodc_io)
>  		return 0;
Right Matthew this case is nearly impossible (may be excepted if
disconected
the usb kbd during the boot), it would means that my pdc_console is not
a
serial one (ie graphic([01])) but no usb kdb found, so in that case pdc
would
switch console automaticaly to serial.

I would just still check the case when I have an usb kbd (I would have
to
recover one) but the pdc console is setup to serial port (eg no graphic
screen
available) just for an output screen without input kbd? 

Thanks for your attention and remarks,
    Joel


------------------------------------------------------------------------
-
Tiscali ADSL: 12 mois à 29,50 EUR/mois! L'Internet rapide, c'est pour
tout
le monde.
http://reg.tiscali.be/default.asp?lg=fr




------------------------------

Message: 3
Date: Mon, 26 Jan 2004 19:38:50 +0100
From: "Joel Soete" <soete.joel@tiscali.be>
Subject: Re: [parisc-linux] A fix for B2k and CONFIG_PDC_CONSOLE pb
To: "Matthew Wilcox" <willy@debian.org>
Cc: Christoph Plattner <christoph.plattner@gmx.at>,	Grant Grundler
	<grundler@parisc-linux.org>,
parisc-linux@lists.parisc-linux.org
Message-ID: <4015311000000204@ocpmta2.freegates.net>
Content-Type: text/plain; charset="ISO-8859-1"

Sorry for auto reply, but I just find back the right kdb and:

>I would just still check the case when I have an usb kbd (I would have
to
>recover one) but the pdc console is setup to serial port (e
>no graphic screen available) just for an output screen without input
kbd?


That case is also impossible, the console setup as serial_1 (eg) is
always
CL_DUPLEX :(

I will so review my patch.

Thanks for all,
    Joel


------------------------------------------------------------------------
-
Tiscali ADSL: 12 mois à 29,50 EUR/mois! L'Internet rapide, c'est pour
tout
le monde.
http://reg.tiscali.be/default.asp?lg=fr




------------------------------

_______________________________________________
parisc-linux mailing list
parisc-linux@lists.parisc-linux.org
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux


End of parisc-linux Digest, Vol 2, Issue 33
*******************************************

[-- Attachment #2: Type: text/html, Size: 12781 bytes --]