Linux 2.5.13-dj2

Merge in some fixes for various corruption possibilities. Drop some of the USB bits thanks to GregKH for going through those. Some of the other things reported as broken in -dj1 (such as cpqarray are still borked, I've not had chance to look at them yet).

As usual,..
Patch against 2.5.13 vanilla is available from:
ftp://ftp.kernel.org/pub/linux/kernel/people/davej/patches/2.5/

Merged patch archive: http://www.codemonkey.org.uk/patches/merged/

Check http://www.codemonkey.org.uk/Linux-2.5.html before reporting known bugs that are also in mainline.

-- Davej.

2.5.13-dj2

o   More fbdev updates/syncing.				(James Simmons)
o   Various VIA Rhine fixes.				(Ivan G)
o   Kill IDE PIO corruption bug.			(Osamu Tomita)
    | One down, n to go.
o   Fix up ext3 thinko.					(Milton Miller)
o   eepro region handling fix.				(William Stinson)
o   SCSI includes tweaks.				(Douglas Gilbert)

2.5.13-dj1

o   Merge 2.4.19pre7 & pre8
    | drop non-x86 archs, cpqarray update, watchdog bits, LARGE ips update
o   Back out bogus ext2_setup_super() change.
o   EFI RTC region handling cleanup.			(William Stinson)
o   Create stub for thermal_interrupt.			(Brian Gerst)
o   ADFS super_block cleanup.				(Brian Gerst)
o   UFS super_block cleanup.				(Brian Gerst)
o   IDE-49 changes.					(Martin Dalecki)
o   hpt34x compile fix.					(Martin Dalecki)
o   Add some missing MODULE_LICENSE tags to fs/		(Tomas Szepe)
o   Check request_irq in atari_lance			(William Stinson)
o   Update error handling for AHA1542			(Douglas Gilbert)
o   Small umem fixes/cleanup.				(Neil Brown)
o   Small ALSA compilation fix.				(Jaroslav Kysela)
o   Missing includes in sd/scsi_debug			(Russell King)
o   Add missing NIPQUAD conversion to nfsroot.		(Trond Myklebust)
o   IDE-50 and IDE-51.					(Martin Dalecki)
o   Missing MODULE_LICENSE for sisfb.			(John Tyner)
o   request_region cleanup for AHA1542			(William Stinson)
o   Sync framebuffer code with mainline.		(James Simmons)

| Dave Jones.        http://www.codemonkey.org.uk
| SuSE Labs

Hi Dave,

I got the following compile error in cpqfcTSinit.c:

<-- snip -->

...
gcc -D__KERNEL__ -I/home/bunk/linux/kernel-2.5/linux-2.5.13/include -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
-pipe -mpreferred-stack-boundary=2 -march=k6 -nostdinc -I
/usr/lib/gcc-lib/i386-linux/2.95.4/include -DKBUILD_BASENAME=cpqfcTSinit -c -o cpqfcTSinit.o cpqfcTSinit.c
cpqfcTSinit.c: In function `cpqfcTS_ioctl':
cpqfcTSinit.c:535: `CAP_RAW_IO' undeclared (first use in this function) cpqfcTSinit.c:535: (Each undeclared identifier is reported only once cpqfcTSinit.c:535: for each function it appears in.)
cpqfcTSinit.c: At top level:
cpqfcTSinit.c:1979: unknown field `reset' specified in initializer cpqfcTSinit.c:1979: duplicate initializer
cpqfcTSinit.c:1979: (near initialization for `driver_template.eh_device_reset_handler')
cpqfcTSinit.c:1979: unknown field `abort' specified in initializer make[3]: *** [cpqfcTSinit.o] Error 1
make[3]: Leaving directory `/home/bunk/linux/kernel-2.5/linux-2.5.13/drivers/scsi'

<-- snip -->

The first part of the error message is strange. 2.5.13-dj2 does exactly the following change to this file:

@@ -532,7 +532,7 @@
 
 	// must be super user to send stuff directly to the
 	// controller and/or physical drives...
-	if( !capable(CAP_SYS_ADMIN) )
+	if( !capable(CAP_RAW_IO) )
 	  return -EPERM;
 
 	// copy the caller's struct to our space.

This is the only place in the 2.5.13-dj2 patch where CAP_RAW_IO is mentioned and a grep showed that it's the only occurence of CAP_RAW_IO in the whole 2.5.13-dj2 kernel sources:

<-- snip -->

~/linux/kernel-2.5/linux-2.5.13$ grep -r CAP_RAW_IO *
drivers/scsi/cpqfcTSinit.c: if( !capable(CAP_RAW_IO) )
~/linux/kernel-2.5/linux-2.5.13$

<-- snip -->

cu
Adrian

You only think this is a free country. Like the US the UK spends a lot of
time explaining its a free country because its a police state.
								Alan Cox

Hi volks!

Here is my problem:
[ ... ]
rm -f math-emu.o
ld -r -o math-emu.o math.o qrnnd.o
make[2]: Leaving directory /root/linux-2.5.13/arch/alpha/math-emu' make[1]: Leaving directory /root/linux-2.5.13/arch/alpha/math-emu' ld -static -T arch/alpha/vmlinux.lds -N arch/alpha/kernel/head.o init/main.o init/version.o init/do_mounts.o \
--start-group \
arch/alpha/kernel/kernel.o arch/alpha/mm/mm.o kernel/kernel.o

mm/mm.o fs/fs.o ipc/ipc.o arch/alpha/math-emu/math-emu.o \
        /root/linux-2.5.13/arch/alpha/lib/lib.a

/root/linux-2.5.13/lib/lib.a /root/linux-2.5.13/arch/alpha/lib/lib.a \
drivers/base/base.o drivers/char/char.o drivers/block/block.o
drivers/misc/misc.o drivers/net/net.o drivers/media/media.o drivers/ide/idedriver.o drivers/scsi/scsidrv.o drivers/cdrom/driver.o drivers/pci/driver.o drivers/net/tulip/tulip_net.o drivers/pnp/pnp.o

drivers/video/video.o \
        net/network.o \
        --end-group \
        -o vmlinux

drivers/scsi/scsidrv.o: In function `sd_init':
drivers/scsi/scsidrv.o(.text+0x4d3d8): undefined reference to vmalloc' drivers/scsi/scsidrv.o(.text+0x4d3dc): undefined reference to vmalloc' drivers/scsi/scsidrv.o(.text+0x4d424): undefined reference to vmalloc' drivers/scsi/scsidrv.o(.text+0x4d428): undefined reference to vmalloc' drivers/scsi/scsidrv.o(.text+0x4d470): undefined reference to vmalloc' drivers/scsi/scsidrv.o(.text+0x4d474): more undefined references to vmalloc' follow
make: *** [vmlinux] Error 1
[ ... ]

Is there a patch, or anything? What makes video.o here??????????

Please help me!

-Oliver

Oliver Pitzeier wrote:
[ ... ]

OK. There is someone, who seems to help me...
See the thread from the linux-kernel-archive:
"Re: 2.5.13 on Alpha: undefined reference to vmalloc in scsidrv.o"

-Oliver