p54 build failure

p54 build failure

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 196 bytes --]

p54_parse_eeprom must not be static, or it fails to build on powerpc:

drivers/net/wireless/p54/p54common.c:482: error:
__ksymtab_p54_parse_eeprom causes a section type conflict


johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: p54 build failure

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 737 bytes --]

can you turn off html?

On Mon, 2008-10-06 at 22:19 +0200, Christian Lamparter wrote:
> On Monday 06 October 2008 21:48:21 Johannes Berg wrote:
> 
> > p54_parse_eeprom must not be static, or it fails to build on
> powerpc:
> 
> >
> 
> > drivers/net/wireless/p54/p54common.c:482: error:
> 
> > __ksymtab_p54_parse_eeprom causes a section type conflict
> 
> >
> 
> Kind of funny, remember the sparse warning?!
> 
> .../p54common.c:313:5: warning: symbol 'p54_parse_eeprom' was not
> declared.
> Shou

Well, if it's exported then clearly it must be declared in some header
file; therefore the file needs to include that header file. Mindlessly
fixing sparse warnings has never brought anything forward.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] p54: fix build failure

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 232 bytes --]

still HTML and as such not really useful.

However,


> +int p54_parse_eeprom(struct ieee80211_hw *dev, void *eeprom, int
> len);

How about just killing the export instead? It isn't used anywhere else
anyway?

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] p54: fix build failure

[Christian Lamparter]

On Monday 06 October 2008 22:47:45 Johannes Berg wrote:
> still HTML and as such not really useful.
>
if only kmail let me disable it! :-)
Is it off now? If not, I'm really sorry, for all html ****.
I'll fix it, but it'll take another while.

> However,
>
> > +int p54_parse_eeprom(struct ieee80211_hw *dev, void *eeprom, int
> > len);
>
> How about just killing the export instead? It isn't used anywhere else
> anyway?

True.

on the other hand: what about stlc45xx (p54spi)?
These ?SoC? chips don't have any eeprom attached to the I2C, but they
need calibration data and output power limits like the usb/pci parts as well.

Regards,
	Chr 

Re: [PATCH] p54: fix build failure

[Johannes Berg]

Christian Lamparter wrote:
> if only kmail let me disable it! :-)
> Is it off now? If not, I'm really sorry, for all html ****.
> I'll fix it, but it'll take another while.

Seems to be ok now.

>> However,
>>
>> > +int p54_parse_eeprom(struct ieee80211_hw *dev, void *eeprom, int
>> > len);
>>
>> How about just killing the export instead? It isn't used anywhere else
>> anyway?
>
> True.
>
> on the other hand: what about stlc45xx (p54spi)?
> These ?SoC? chips don't have any eeprom attached to the I2C, but they
> need calibration data and output power limits like the usb/pci parts as
> well.

Yeah, but it hardly matters now, we can always re-export it when needed.

johannes

p54 build failure

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 390 bytes --]

On Fri, 2009-01-16 at 06:15 +0100, Cron Daemon wrote:
> 
> +drivers/net/wireless/p54/p54spi.h:28:11: error: unable to open 'mach/board.h'
> +make[2]: *** [drivers/net/wireless/p54/p54spi.o] Error 1
> +make[1]: *** [drivers/net/wireless/p54] Error 2
> +make: *** [_module_drivers/net/wireless] Error 2

p54spi only depends on SPI_MASTER, but includes arm-specific files.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]