Re: [PATCH V2 6/8] lzma: Make lzma available to non initramfs/initrd code

[Albin Tonnerre <albin.tonnerre@free-electrons.com>]

On Wed, 06 Jan 2010 14:04 -0800, Andrew Morton wrote :
> On Fri, 11 Dec 2009 01:34:08 +0000
> Phillip Lougher <phillip@lougher.demon.co.uk> wrote:
> 
> > Add a config option DECOMPRESS_LZMA_NEEDED which allows subsystems to
> > specify they need the unlzma code.  Normally decompress_unlzma.c is
> > compiled with __init and unlzma is not exported to modules.
> > 
> > Move INIT definition into separate header files for bzip2/lzma/inflate
> > so it can be defined differently for each decompressor.
> > 
> 
> This patch (which is in linux-next) breaks
> lib-add-support-for-lzo-compressed-kernels.patch, below.  The
> definition of INIT is no longer available in lib/decompress_unlzo.c, and
> 
> lib/decompress_unlzo.c: In function 'unlzo':
> lib/decompress_unlzo.c:106: error: 'error' undeclared (first use in this function)
> lib/decompress_unlzo.c:106: error: (Each undeclared identifier is reported only once
> lib/decompress_unlzo.c:106: error: for each function it appears in.)
> lib/decompress_unlzo.c:111: error: implicit declaration of function 'error'
> 
> 
> I'm planning on merging 
> 
> zlib-optimize-inffast-when-copying-direct-from-output.patch
> lib-add-support-for-lzo-compressed-kernels.patch
> arm-add-support-for-lzo-compressed-kernels.patch
> x86-add-support-for-lzo-compressed-kernels.patch
> add-lzo-compression-support-for-initramfs-and-old-style-initrd.patch
> 
> into 2.6.33.  I don't immediately remember why I decided that - perhaps
> because the patches did arrive in time for .33, but got stalled because
> people were screwing around in other trees.
> 
> So if I go ahead with that merge, linux-next will need fixing.  And I
> didn't get down and work what the appropriate fix is, and I don't want
> to break linux-next in serious ways.
> 
> 
> So what to do?  I guess I could go ahead with the mainline merge, and
> Stephen drops <whatever that tree was> from linux-next until it has
> been fixed up?

I'd go for that one, although I'm obviously biaised. IMHO the linux-next fix
should be easy enough not to postpone the above five patches until 2.6.34.

Regards,
-- 
Albin Tonnerre, Free Electrons
Kernel, drivers and embedded Linux development,
consulting, training and support.
http://free-electrons.com