Ian Molton | 1 May 14:36 2004

Re: help me on IRDA driver.

On Sat, 24 Apr 2004 16:27:28 +0530
<raghavendra.hanumanthaiah <at> wipro.com> wrote:

> Am developing a FIRI driver (the h/w supports only FIR & MIR modes) for
> an application processor, which has ARM11 core.

Will this driver ever be released as open source?

--

-- 
Spyros lair: http://www.mnementh.co.uk/   ||||   Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup.

-------------------------------------------------------------------
Subscription options: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:       http://www.arm.linux.org.uk/armlinux/mlfaq.php
Etiquette: http://www.arm.linux.org.uk/armlinux/mletiquette.php

Ian Molton | 1 May 14:40 2004

Re: [RFC] New mailing list: linux-arm-historic

On Mon, 26 Apr 2004 12:06:58 -0400 (EDT)
Nicolas Pitre <nico <at> cam.org> wrote:

> 
> Guys, there is something you (and many others like you, especially the
> handhelds.org community) didn't understand yet.

Oi, would you mind not using such a wide brush please m8 ?

--

-- 
Spyros lair: http://www.mnementh.co.uk/   ||||   Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup.

-------------------------------------------------------------------
Subscription options: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:       http://www.arm.linux.org.uk/armlinux/mlfaq.php
Etiquette: http://www.arm.linux.org.uk/armlinux/mletiquette.php

Ian Molton | 1 May 14:41 2004

Re: [RFC] New mailing list: linux-arm-historic

On Mon, 26 Apr 2004 15:33:36 +0100
Philip Blundell <pb <at> nexus.co.uk> wrote:

> It would probably be better to suggest to these people that they upgrade
> their kernels and, failing that, just to ignore mail concerning versions
> that you don't have any interest in.

Someone else suggested a 2.4 and a 2.6 listr seperate. I like the idea.

--

-- 
Spyros lair: http://www.mnementh.co.uk/   ||||   Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup.

-------------------------------------------------------------------
Subscription options: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:       http://www.arm.linux.org.uk/armlinux/mlfaq.php
Etiquette: http://www.arm.linux.org.uk/armlinux/mletiquette.php

Andrew Victor | 2 May 20:40 2004

Re: Cogent CSB337 at91rm9200 questions (networking and compiler)

hi,

> > static void get_mac_address(struct net_device *dev) {
> > [snip]
> > 	hi = regs->EMAC_SA1H;	(0x23)
> > 	lo = regs->EMAC_SA1L;	(0x3137004b)
> > 
> > 	addr[0] = (lo & 0xff);	(0x4b)
> > 
> > Thus my claim that the code in get_mac_address is backwards and will only
> > appear to work if the last byte in your MAC address is even.
> > 
> 
> Ok, so someone needs to look up whether EMAC_SA1L bits 7-0 appear on
> the wire first, or EMAC_SA1H bits 15 to 8.  If the former, then the
> driver is correct, and whatever is loading the registers in the
> first place is obviously wrong.

The datasheet says:
  SA1H : Bits 47 .. 32
  SA1L : Bits 31 .. 0
"The first location contains the first four bytes of the address;  the
second location contains the last two bytes of the address stored in its
least significant byte positions."

Since the least-significant byte should be transmitted/received first, I
assume the driver is correct and that the CSB337 bootloader has stored
the value the wrong way around.

A workaround for the CSB337 will be added to get_mac_address() to load
(Continue reading)

Tony Lindgren | 3 May 03:46 2004

Re: size of dma_coherent alloc space.

* Tony Lindgren <tony <at> atomide.com> [040430 15:54]:
> * Woodruff, Richard <r-woodruff2 <at> ti.com> [040430 15:25]:
> >
> > It won't work for omap1's until their memory is rearranged.
> 
> I've just started tinkering to change the io addressing...

Finally got it booting again with modified io addressing. It was a 
bit of a pain, as some of the older code was assuming that 
paddr == vaddr :)

I will post it to the linux-omap BK with address change disabled 
for now until all the drivers are fixed too.

Regards,

Tony

-------------------------------------------------------------------
Subscription options: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:       http://www.arm.linux.org.uk/armlinux/mlfaq.php
Etiquette: http://www.arm.linux.org.uk/armlinux/mletiquette.php

Mukund JB. Choudary | 3 May 07:59 2004
Picon

How to built Cross Compiler for Omap-1510 patch against linux-2.4.18

Hi,

Thanks  to the linux world supporters in advance .

I am very new to linux and don't know how it internally works

So, please some on help me guide how to build the Cross Compiler for
Omap-1510 patch against linux-2.4.18.

Regards,

Mukund jampala

-------------------------------------------------------------------
Subscription options: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:       http://www.arm.linux.org.uk/armlinux/mlfaq.php
Etiquette: http://www.arm.linux.org.uk/armlinux/mletiquette.php

raghavendra.hanumanthaiah | 3 May 06:02 2004

RE: help me on IRDA driver.


-----Original Message-----
From: linux-arm-kernel-bounces <at> lists.arm.linux.org.uk
[mailto:linux-arm-kernel-bounces <at> lists.arm.linux.org.uk] On Behalf Of
Ian Molton
Sent: Saturday, May 01, 2004 6:06 PM
To: linux-arm-kernel <at> lists.arm.linux.org.uk
Subject: Re: help me on IRDA driver.

On Sat, 24 Apr 2004 16:27:28 +0530 <raghavendra.hanumanthaiah <at> wipro.com>
wrote:

> Am developing a FIRI driver (the h/w supports only FIR & MIR modes) 
> for an application processor, which has ARM11 core.

Will this driver ever be released as open source?
[Raghavendra ]: Yes of course, its comes with GPL. 

--

-- 
Spyros lair: http://www.mnementh.co.uk/   ||||   Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with
ketchup.

-------------------------------------------------------------------
Subscription options:
http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:       http://www.arm.linux.org.uk/armlinux/mlfaq.php
Etiquette: http://www.arm.linux.org.uk/armlinux/mletiquette.php

(Continue reading)

Tobias Arp | 3 May 11:10 2004
Picon

jffs2 on multiple flash sector size

Hi, 

is it possible to use a jffs2-partition on a mtd device with multiple sector size ?  
For example: 

Sector 0 - 62: 64 Kbytes 
Sector 63 - 70: 8 Kbytes 

jffs2 partition: starts at sector 56 and ends at sector 65. 

Can this partion work withouzt problems ? 

regards 

Tobias Arp 
____________________________________________________________________
Der WEB.DE Virenschutz schuetzt Ihr Postfach vor dem Wurm Sober.A-F!
Kostenfrei fuer FreeMail Nutzer. http://f.web.de/?mc=021158

-------------------------------------------------------------------
Subscription options: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:       http://www.arm.linux.org.uk/armlinux/mlfaq.php
Etiquette: http://www.arm.linux.org.uk/armlinux/mletiquette.php

Erik Mouw | 3 May 13:12 2004
Picon

Re: Porting kernel 2.6.4 to Lart

On Thu, Apr 29, 2004 at 10:09:31PM +0200, Diego Roversi wrote:
>   lart is supposed to be a supported platform, but when compiled with
> ide driver, I get some errors. So I made some simple correction to solve
> them, and to make ide interface work again (patch at bottom of message).

To be honest I haven't tried a LART with IDE on 2.6, it's on my ToDo
list.

> So it recognize correctly the interface and the CF card, but it can't read
> from it (partition table is readed correctly under 2.4.19-rmk7). Also I
> can't understand becouse irq1 is ignored by ide driver (irq1: nobody cared).
> Something wrong with ide_register_hw() (please, take a look at the patch
> below)? I look at other sa1100 machines, but it seems they do the same thing...
> 
> PS: I see that memory mapping for 'PCMCIA IO' was removed from kernel 2.6.x
> in lart.c. There was any reason for that?

It was in generic.c, AFAIK (don't have my laptop with me, so I can't
check), but IIRC it had to do with PCMCIA driver issues.

> diff -ur linux-2.6.4.orig/include/asm-arm/arch-sa1100/ide.h linux-2.6.4/include/asm-arm/arch-sa1100/ide.h
> --- linux-2.6.4.orig/include/asm-arm/arch-sa1100/ide.h	2004-03-11 03:55:21.000000000 +0100
> +++ linux-2.6.4/include/asm-arm/arch-sa1100/ide.h	2004-04-27 22:39:44.000000000 +0200
>  <at>  <at>  -113,16 +113,17  <at>  <at> 
>          hw_regs_t hw;
>  
>          /* Enable GPIO as interrupt line */
> -        GPDR &= ~LART_GPIO_IDE;
> -	set_irq_type(LART_IRQ_IDE, IRQT_RISING);
> +       GPDR &= ~GPIO_GPIO1; 
(Continue reading)

Peter Milne | 3 May 16:13 2004
Picon
Picon

Re: How to detect amount of physical memory ?

Jeff

your bootloader should calculate the memory size, possibly from SPD, and
fill in the appropriate tags (ATAG_MEM).

Two possible ways to get to this information are

1/ define a MACHINE_START FIXUP function and iterate the tags looking for
ATAG_MEM.

2/ replace the setup.c function

request_standard_resources(struct meminfo *mi, struct machine_desc *mdesc)

with your own function that makes use of the information in mi.

BTW: a general comment about setup.c, there is a lot of generic stuff in 
there
(lp, video etc) that a lot of systems won't have. It seemed to me that 
there was a good case for having a MACHINE_request_standard_resources() 
function to allow machine specific setups. I made my own, but it is 
uncomfortable making machine specific mods to this general part of the 
kernel.

/Peter.

Jeff Steele wrote:

>Hi,
>
(Continue reading)


Gmane