[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20081103080340.GC18445@roarinelk.homelinux.net>
Date: Mon, 3 Nov 2008 09:03:40 +0100
From: Manuel Lauss <mano@...rinelk.homelinux.net>
To: Mike Frysinger <vapier@...too.org>
Cc: netdev@...r.kernel.org
Subject: Re: smc91x: commit 51ac3beffd4afaea4350526cf01fe74aaff25eff breaks
compilation
On Mon, Nov 03, 2008 at 02:48:52AM -0500, Mike Frysinger wrote:
> On Mon, Nov 3, 2008 at 2:35 AM, Manuel Lauss wrote:
> > On Mon, Nov 03, 2008 at 02:30:01AM -0500, Mike Frysinger wrote:
> >> On Mon, Nov 3, 2008 at 2:17 AM, Manuel Lauss wrote:
> >> > (Resend with correct netdev addr)
> >> >
> >> > Hi Mike,
> >> >
> >> > your commit 51ac3beffd4afaea4350526cf01fe74aaff25eff causes this
> >> > compile failure on my SH and MIPS testboards (2.6.28-rc3):
> >> >
> >> > CC drivers/net/smc91x.o
> >> > /mnt/work/sh7760/kernel/linux-2.6.git/drivers/net/smc91x.c: In function 'smc_request_attrib':
> >> > /mnt/work/sh7760/kernel/linux-2.6.git/drivers/net/smc91x.c:2067: error: 'lp' undeclared (first use in this function)
> >> > /mnt/work/sh7760/kernel/linux-2.6.git/drivers/net/smc91x.c:2067: error: (Each undeclared identifier is reported only once
> >> > /mnt/work/sh7760/kernel/linux-2.6.git/drivers/net/smc91x.c:2067: error: for each function it appears in.)
> >> > /mnt/work/sh7760/kernel/linux-2.6.git/drivers/net/smc91x.c: In function 'smc_release_attrib':
> >> > /mnt/work/sh7760/kernel/linux-2.6.git/drivers/net/smc91x.c:2079: error: 'lp' undeclared (first use in this function)
> >> > make[2]: *** [drivers/net/smc91x.o] Error 1
> >> > make[1]: *** [drivers/net/smc91x.o] Error 2
> >> >
> >> > Reverting this commit fixes it.
> >>
> >> are you using modifiied sources or something ? there's clearly no lp
> >> reference in mainline:
> >> ...
> >> static int smc_request_attrib(struct platform_device *pdev,
> >> struct net_device *ndev)
> >> {
> >> struct resource * res = platform_get_resource_byname(pdev,
> >> IORESOURCE_MEM, "smc91x-attrib");
> >>
> >> if (!res)
> >> return 0;
> >>
> >> if (!request_mem_region(res->start, ATTRIB_SIZE, CARDNAME))
> >> return -EBUSY;
> >>
> >> return 0;
> >> }
> >>
> >> static void smc_release_attrib(struct platform_device *pdev,
> >> struct net_device *ndev)
> >> {
> >> struct resource * res = platform_get_resource_byname(pdev,
> >> IORESOURCE_MEM, "smc91x-attrib");
> >>
> >> if (res)
> >> release_mem_region(res->start, ATTRIB_SIZE);
> >> }
> >
> > No changes, pristine 2.6.28-rc3. Note that I don't have any board entries
> > in smc91x.h; just platform_data.
>
> then you'll need to generate the preprocessed output and figure out
> where the lp reference is coming from.
> make drivers/net/smc91x.i
To be more precise:
smc91x configured purely through platform_data:
ATTRIB_SIZE is ((64*1024) << SMC_IO_SHIFT)
with SMC_IO_SHIFT being "lp->io_shift" (smc91x.h line 450).
So please revert this patch.
Thanks,
Manuel Lauss
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists