Mask Issue?

Previous thread: SMP interrupt controller that only supports CPU affinity? by Kumar Gala on Wednesday, September 19, 2007 - 7:01 pm. (1 message)

Next thread: Arr! Linux 2.6.23-rc7 by Linus Torvalds on Wednesday, September 19, 2007 - 7:17 pm. (2 messages)
To: kernel list <linux-kernel@...>
Subject: Mask Issue?
Date: Wednesday, September 19, 2007 - 6:57 pm

Still being a little new at this I am not sure if this is an issue at
all or not but I noted that while building the 2.6.23-rc6-git8 kernel
this afternoon I received the following error message:

Building modules, stage 2.
MODPOST 1670 modules
WARNING: Can't handle masks in drivers/mtd/nand/cafe_nand:FFFF0

I have been building each of the kernels in the 2.6.23-rc series and had
not noted this message before. I apologize if this is a non-issue.

-

To: Chris Holvenstot <cholvenstot@...>
Cc: kernel list <linux-kernel@...>, <dwmw2@...>
Date: Monday, September 24, 2007 - 11:39 am

I noticed this in a -mm recently and never got around to kicking it
properly because I suspected it was a compiler issue. What version
of gcc and binutils are you using?

David Woodhouse added to cc in case this is a known problem.

--
Mel Gorman
Part-time Phd Student Linux Technology Center
University of Limerick IBM Dublin Software Lab
-

To: Mel Gorman <mel@...>
Cc: kernel list <linux-kernel@...>, <dwmw2@...>
Date: Monday, September 24, 2007 - 7:07 pm

Mel -

Thank you for the response. I was a little unsure if this was to
trivial to report.

In any case to answer your question I use version 4.1.2 of GCC and the
binutils package shows as 2.17.20070103cvs-0ubuntu2

Chris

-

To: Chris Holvenstot <cholvenstot@...>
Cc: kernel list <linux-kernel@...>, <dwmw2@...>
Date: Tuesday, September 25, 2007 - 5:20 am

It's never trivial but don't be discouraged if your report appears to
fall between the cracks - it happens from time to time particularly when

As David says the check will go away in the near future, I reckon it's safe
to ignore the problem (unless David objects) and disable it in your .config
and continue with your testing. As the driver is specific to the One Laptop
Per Child project, I'm guessing you are not affected.

Thanks for testing and reporting.

--
Mel Gorman
Part-time Phd Student Linux Technology Center
University of Limerick IBM Dublin Software Lab
-

To: Mel Gorman <mel@...>
Cc: Chris Holvenstot <cholvenstot@...>, kernel list <linux-kernel@...>
Date: Monday, September 24, 2007 - 11:50 am

The only reason for this is because once upon a time, the early FPGA
versions of the CAFÉ chip had the same PCI ident for all three
functions. The cluebat was applied; the chip was fixed long before
tape-out. We can just drop the whole class/mask thing, I suspect -- I
just want to boot it on and OLPC machine before I commit that.

--
dwmw2

-

Previous thread: SMP interrupt controller that only supports CPU affinity? by Kumar Gala on Wednesday, September 19, 2007 - 7:01 pm. (1 message)

Next thread: Arr! Linux 2.6.23-rc7 by Linus Torvalds on Wednesday, September 19, 2007 - 7:17 pm. (2 messages)