Hi,
Moving on with the 2.4.24-pre tree, here is 2.4.25-pre4.
It contains an ext2/3 update (mostly forward compatibility related), the
usual architecture updates (this time S390, PPC64/32, SH), osst update,
TG3 bugfixes, amongst others.
Some of the fixes listed in this changelog (the rtc fixes, the IrDA "log
buster" fix and the netfilter MASQUERADE oops) were already in other
-pre's, they got removed and re added for technical BK reasons.
Summary of changes from v2.4.24-pre3 to v2.4.25-pre4
============================================
<bjorn.helgaas:hp.com>:
o Fix 2.4 EFI RTC oops
<lethal:unusual.internal.linux-sh.org>:
o sh/sh64: Add CONFIG_OOM_KILLER entries
o sh: Add EXPEVT to pt_regs
o sh64: Add dma.o to export-objs
o sh64: shwdt updates
<marcelo.tosatti:cyclades.com>:
o Andrea Arcangeli: malicious users of mremap() syscall can gain priviledges
<marcelo:logos.cnet>:
o Harald Welte: Fix ipchains MASQUERADE oops
o Change EXTRAVERSION to 2.4.24-rc1
o Cset exclude: [email protected]|ChangeSet|20031218183339|13120
o Cset exclude: [email protected]|ChangeSet|20031210203050|36304
o Cset exclude: [email protected]|ChangeSet|20031213132008|01226
o Cset exclude: [email protected]|ChangeSet|20031204183256|31723
o Change Makefile to 2.4.24-rc1
<trini:mvista.com>:
o /dev/rtc can leak parts of kernel memory to unpriviledged users
David Engebretsen:
o [PPC64] Store and use the ibm,phandle device-tree property from OF
o [PPC64] Export Logical Partitioning config data to userspace
David S. Miller:
o [TG3]: Update version and reldate
Erik Andersen:
o fix broken 2.4.x rt_sigprocmask error handling
Fran?ois Romieu:
o [TG3]: Fix bogus return value in tg3_init_one()
Herbert Xu:
o Handle j_commit_interval == 0
Hollis Blanchard:
o [PPC64] Recognize new-style device-tree nodes for virtual terminals
Jean Tourrilhes:
o IrDA kernel log buster
Kai Makisara:
o SCSI tape bug fix (variable block mode,
Linus Torvalds:
o Daniel Tram Lux: IDE timeout race fix
Martin Schwidefsky:
o S390 base fixes
o S390 common i/o layer fixes
o S390: 31 bit compat bug fixes
o S390: ctc network driver update
o S390: xpram device driver
o S390: DASD update
Oleg Drokin:
o Fix megaraid leak survived by latest update
Olof Johansson:
o [PPC64] Rename some RTAS-specific constants to avoid name clashes
Paul Mackerras:
o [PPC64] Remove references to KDB since it isn't in the official tree
o [PPC64] Fix compilation with CONFIG_SMP=n
o [PPC64] Add include/asm-ppc64/iSeries/vio.h which was missed before
o [PPC64] Add support for the VMX (aka Altivec) unit on the PPC970
o [PPC64] Add CPU feature bits to indicate presence of breakpoint registers
o [PPC64] Fix a few compile warnings and remove some dead code
o [PPC64] Fix a bug in starting kernel threads
o [PPC64] Set ELF_HWCAP to something useful: a bitmap of CPU features
o [PPC64] Fix for periodic interrupts on iSeries with shared processors
o [PPC64] Cope with slow RTC chips
o [PPC64] Better handling of machine checks
o [PPC64] Don't create /proc/rtas files for unimplemented services
o [PPC64] Fix up bug in setting up the firmware features bitmap
o [PPC64] Fix a compile error introduced with some recent changes
Theodore Y. T'so:
o EXT2/3 Updates: Reclaim pages in truncate
o EXT2/3 Updates: 2.6 EA symlink compatibility
o EXT2/3 Updates: forward-compatibility: online resizing
o EXT2/3 Updates: Allow filesystems with expanded inodes to be mounted
Tom Rini:
o PPC32: Add support for the OpenPIC register set to be in BE mode
o PPC32: Fix the floppy driver, on CONFIG_NOT_COHERENT_CACHE
o PPC32: Fix a typo in two files
o PPC32: Fix memory detection of PReP machines with OF
Willem Riede:
o OnStream tape driver update
Hi,
Trying to compile $subj with following config (these options seem to
cause the problem, full config attached):
CONFIG_SCSI_MEGARAID=y
CONFIG_SCSI_MEGARAID2=y
gives me this result:
gcc -D__KERNEL__ -I/disk2/usr/src/Kernel/linux-2.4.24/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
-fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=i586
-nostdinc -iwithprefix include -DKBUILD_BASENAME=sd -c -o sd.o sd.c
ld -m elf_i386 -r -o sd_mod.o sd.o
rm -f scsidrv.o
ld -m elf_i386 -e stext -r -o scsidrv.o scsi_mod.o dpt_i2o.o
aacraid/aacraid.o aic7xxx/aic7xxx.o aic7xxx/aic79xx.o ips.o megaraid.o
megaraid2.o gdth.o 3w-xxxx.o sd_mod.o
megaraid2.o: In function `megaraid_info':
megaraid2.o(.text+0x1e70): multiple definition of `megaraid_info'
megaraid.o(.text+0x27f8): first defined here
ld: Warning: size of symbol `megaraid_info' changed from 71 to 62 in
megaraid2.o
make[3]: *** [scsidrv.o] Error 1
make[3]: Leaving directory
`/disk2/usr/src/Kernel/linux-2.4.24/drivers/scsi'
make[2]: *** [first_rule] Error 2
make[2]: Leaving directory
`/disk2/usr/src/Kernel/linux-2.4.24/drivers/scsi'
make[1]: *** [_subdir_scsi] Error 2
make[1]: Leaving directory `/disk2/usr/src/Kernel/linux-2.4.24/drivers'
make: *** [_dir_drivers] Error 2
Is this a known issue and megaraids can't live together, or am I
supposed to be able to compile both drivers in and this is a bug?
On Tue, 2004-01-06 at 15:14, Marcelo Tosatti wrote:
> Hi,
>
> Moving on with the 2.4.24-pre tree, here is 2.4.25-pre4.
> at http://www.tux.org/lkml/
--
Jan Kokoska
> Trying to compile $subj with following config (these options seem to
> cause the problem, full config attached):
>
> CONFIG_SCSI_MEGARAID=y
> CONFIG_SCSI_MEGARAID2=y
>
> Is this a known issue and megaraids can't live together, or am I
> supposed to be able to compile both drivers in and this is a bug?
yes, this is known and expected. You can build both as modules, but
they're not intended to both be loaded simultaneously (either built-in
or as modules). They're mutually exclusive.
Thanks,
Matt
--
Matt Domsch
Sr. Software Engineer, Lead Engineer
Dell Linux Solutions http://www.dell.com/linux
Linux on Dell mailing lists @ http://lists.us.dell.com
Matt Domsch <[email protected]> ha scritto:
>> Trying to compile $subj with following config (these options seem to
>> cause the problem, full config attached):
>>
>> CONFIG_SCSI_MEGARAID=y
>> CONFIG_SCSI_MEGARAID2=y
>>
>> Is this a known issue and megaraids can't live together, or am I
>> supposed to be able to compile both drivers in and this is a bug?
>
> yes, this is known and expected. You can build both as modules, but
> they're not intended to both be loaded simultaneously (either built-in
> or as modules). They're mutually exclusive.
Ok, what about this patch (against 2.4.25-pre4):
--- linux-2.4/drivers/scsi/Config.in.orig Tue Jan 6 18:11:10 2004
+++ linux-2.4/drivers/scsi/Config.in Tue Jan 6 18:23:29 2004
@@ -66,8 +66,13 @@
dep_tristate 'AdvanSys SCSI support' CONFIG_SCSI_ADVANSYS $CONFIG_SCSI
dep_tristate 'Always IN2000 SCSI support' CONFIG_SCSI_IN2000 $CONFIG_SCSI
dep_tristate 'AM53/79C974 PCI SCSI support' CONFIG_SCSI_AM53C974 $CONFIG_SCSI $CONFIG_PCI
-dep_tristate 'AMI MegaRAID support' CONFIG_SCSI_MEGARAID $CONFIG_SCSI
-dep_tristate 'AMI MegaRAID2 support' CONFIG_SCSI_MEGARAID2 $CONFIG_SCSI
+
+if [ "$CONFIG_SCSI_MEGARAID2" == "n" -o "$CONFIG_SCSI_MEGARAID2" == "" ]; then
+ dep_tristate 'AMI MegaRAID support' CONFIG_SCSI_MEGARAID $CONFIG_SCSI
+fi
+if [ "$CONFIG_SCSI_MEGARAID" == "n" -o "$CONFIG_SCSI_MEGARAID" == "" ]; then
+ dep_tristate 'AMI MegaRAID2 support' CONFIG_SCSI_MEGARAID2 $CONFIG_SCSI
+fi
dep_tristate 'BusLogic SCSI support' CONFIG_SCSI_BUSLOGIC $CONFIG_SCSI
if [ "$CONFIG_SCSI_BUSLOGIC" != "n" ]; then
I'm not very familiar with 2.4 config language, maybe there's a better
way to do it.
Luca
--
Reply-To: [email protected]
Home: http://kronoz.cjb.net
Carpe diem, quam minimum credula postero. (Q. Horatius Flaccus)
On Tue, Jan 06, 2004 at 12:14:23PM -0200, Marcelo Tosatti wrote:
> It contains an ext2/3 update (mostly forward compatibility related), the
Do you plan to merge htree?
On Tue, 6 Jan 2004, Mike Fedyk wrote:
> On Tue, Jan 06, 2004 at 12:14:23PM -0200, Marcelo Tosatti wrote:
> > It contains an ext2/3 update (mostly forward compatibility related), the
>
> Do you plan to merge htree?
Yes, in the next -pre.
On Tue, Jan 06, 2004 at 05:16:24PM +0100, Jan Kokoska wrote:
> Hi,
Hi Jan,
> Trying to compile $subj with following config (these options seem to
> cause the problem, full config attached):
>
> CONFIG_SCSI_MEGARAID=y
> CONFIG_SCSI_MEGARAID2=y
>
> gives me this result:
>...
> Is this a known issue and megaraids can't live together, or am I
> supposed to be able to compile both drivers in and this is a bug?
>...
They can't live together.
Below is a patch to tell the config system not to alllow illegal
configurations.
> Jan Kokoska
>...
cu
Adrian
--- linux-2.4.25-pre4-full/drivers/scsi/Config.in.old 2004-01-06 18:06:29.000000000 +0100
+++ linux-2.4.25-pre4-full/drivers/scsi/Config.in 2004-01-06 18:08:00.000000000 +0100
@@ -67,7 +67,16 @@
dep_tristate 'Always IN2000 SCSI support' CONFIG_SCSI_IN2000 $CONFIG_SCSI
dep_tristate 'AM53/79C974 PCI SCSI support' CONFIG_SCSI_AM53C974 $CONFIG_SCSI $CONFIG_PCI
dep_tristate 'AMI MegaRAID support' CONFIG_SCSI_MEGARAID $CONFIG_SCSI
-dep_tristate 'AMI MegaRAID2 support' CONFIG_SCSI_MEGARAID2 $CONFIG_SCSI
+if [ "$CONFIG_SCSI_MEGARAID" = "y" ]; then
+ define_tristate CONFIG_SCSI_MEGARAID2_DEP n
+else
+ if [ "$CONFIG_SCSI_MEGARAID" = "m" ]; then
+ define_tristate CONFIG_SCSI_MEGARAID2_DEP m $CONFIG_SCSI
+ else
+ define_tristate CONFIG_SCSI_MEGARAID2_DEP $CONFIG_SCSI
+ fi
+fi
+dep_tristate 'AMI MegaRAID2 support' CONFIG_SCSI_MEGARAID2 $CONFIG_SCSI_MEGARAID2_DEP
dep_tristate 'BusLogic SCSI support' CONFIG_SCSI_BUSLOGIC $CONFIG_SCSI
if [ "$CONFIG_SCSI_BUSLOGIC" != "n" ]; then
Andrzej Krzysztofowicz wrote:
>From [email protected] Wed Jan 7 16:31:11 2004
Return-Path: <[email protected]>
Received: from sunrise.pg.gda.pl ([email protected] [153.19.40.230])
by green.mif.pg.gda.pl (8.12.10/8.12.6) with ESMTP id i07FVBo0032429
for <[email protected]>; Wed, 7 Jan 2004 16:31:11 +0100
Received: from sunrise.pg.gda.pl (localhost [127.0.0.1])
by sunrise.pg.gda.pl (8.12.10/8.12.9) with ESMTP id i07FV6DJ005180
for <[email protected]>; Wed, 7 Jan 2004 16:31:06 +0100 (CET)
Received: (from ankry@localhost)
by sunrise.pg.gda.pl (8.12.10/8.12.9/Submit) id i07FV4VW005179
for [email protected]; Wed, 7 Jan 2004 16:31:04 +0100 (CET)
From: Andrzej Krzysztofowicz <[email protected]>
Message-Id: <[email protected]>
Subject: Re: Linux 2.4.25-pre4 (fwd)
To: [email protected]
Date: Wed, 7 Jan 104 16:31:04 +0100 (CET)
Content-Type: text
"F wrote:"
>From [email protected] Tue Jan 6 18:34:02 2004
X-BrightmailFiltered: true
Date: Tue, 6 Jan 2004 18:30:17 +0100
From: Kronos <[email protected]>
To: [email protected]
Cc: Matt Domsch <[email protected]>,
Marcelo Tosatti <[email protected]>
Subject: Re: Linux 2.4.25-pre4
Message-ID: <[email protected]>
Reply-To: [email protected]
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <[email protected]>
User-Agent: Mutt/1.4i
Sender: [email protected]
Precedence: bulk
X-Mailing-List: [email protected]
X-Spam-Checker-Version: SpamAssassin 2.70-cvs (1.218-2003-11-09-exp) on
mordred.oi.pg.gda.pl
X-Spam-Level:
X-Spam-Status: No, hits=0.3 required=5.0 tests=UPPERCASE_25_50 autolearn=no
version=2.70-cvs
Matt Domsch <[email protected]> ha scritto:
>> Trying to compile $subj with following config (these options seem to
>> cause the problem, full config attached):
>>
>> CONFIG_SCSI_MEGARAID=y
>> CONFIG_SCSI_MEGARAID2=y
>>
>> Is this a known issue and megaraids can't live together, or am I
>> supposed to be able to compile both drivers in and this is a bug?
>
> yes, this is known and expected. You can build both as modules, but
> they're not intended to both be loaded simultaneously (either built-in
> or as modules). They're mutually exclusive.
Ok, what about this patch (against 2.4.25-pre4):
--- linux-2.4/drivers/scsi/Config.in.orig Tue Jan 6 18:11:10 2004
+++ linux-2.4/drivers/scsi/Config.in Tue Jan 6 18:23:29 2004
@@ -66,8 +66,13 @@
dep_tristate 'AdvanSys SCSI support' CONFIG_SCSI_ADVANSYS $CONFIG_SCSI
dep_tristate 'Always IN2000 SCSI support' CONFIG_SCSI_IN2000 $CONFIG_SCSI
dep_tristate 'AM53/79C974 PCI SCSI support' CONFIG_SCSI_AM53C974 $CONFIG_SCSI $CONFIG_PCI
-dep_tristate 'AMI MegaRAID support' CONFIG_SCSI_MEGARAID $CONFIG_SCSI
-dep_tristate 'AMI MegaRAID2 support' CONFIG_SCSI_MEGARAID2 $CONFIG_SCSI
+
+if [ "$CONFIG_SCSI_MEGARAID2" == "n" -o "$CONFIG_SCSI_MEGARAID2" == "" ]; then
+ dep_tristate 'AMI MegaRAID support' CONFIG_SCSI_MEGARAID $CONFIG_SCSI
+fi
+if [ "$CONFIG_SCSI_MEGARAID" == "n" -o "$CONFIG_SCSI_MEGARAID" == "" ]; then
+ dep_tristate 'AMI MegaRAID2 support' CONFIG_SCSI_MEGARAID2 $CONFIG_SCSI
+fi
dep_tristate 'BusLogic SCSI support' CONFIG_SCSI_BUSLOGIC $CONFIG_SCSI
if [ "$CONFIG_SCSI_BUSLOGIC" != "n" ]; then
I'm not very familiar with 2.4 config language, maybe there's a better
way to do it.
Luca
--
Reply-To: [email protected]
Home: http://kronoz.cjb.net
Carpe diem, quam minimum credula postero. (Q. Horatius Flaccus)
Sorry for previous bogus messge.
> --- linux-2.4/drivers/scsi/Config.in.orig Tue Jan 6 18:11:10 2004
> +++ linux-2.4/drivers/scsi/Config.in Tue Jan 6 18:23:29 2004
> @@ -66,8 +66,13 @@
> dep_tristate 'AdvanSys SCSI support' CONFIG_SCSI_ADVANSYS $CONFIG_SCSI
> dep_tristate 'Always IN2000 SCSI support' CONFIG_SCSI_IN2000 $CONFIG_SCSI
> dep_tristate 'AM53/79C974 PCI SCSI support' CONFIG_SCSI_AM53C974 $CONFIG_SCSI $CONFIG_PCI
> -dep_tristate 'AMI MegaRAID support' CONFIG_SCSI_MEGARAID $CONFIG_SCSI
> -dep_tristate 'AMI MegaRAID2 support' CONFIG_SCSI_MEGARAID2 $CONFIG_SCSI
> +
> +if [ "$CONFIG_SCSI_MEGARAID2" == "n" -o "$CONFIG_SCSI_MEGARAID2" == "" ]; then
> + dep_tristate 'AMI MegaRAID support' CONFIG_SCSI_MEGARAID $CONFIG_SCSI
> +fi
> +if [ "$CONFIG_SCSI_MEGARAID" == "n" -o "$CONFIG_SCSI_MEGARAID" == "" ]; then
> + dep_tristate 'AMI MegaRAID2 support' CONFIG_SCSI_MEGARAID2 $CONFIG_SCSI
> +fi
>
> dep_tristate 'BusLogic SCSI support' CONFIG_SCSI_BUSLOGIC $CONFIG_SCSI
> if [ "$CONFIG_SCSI_BUSLOGIC" != "n" ]; then
This way you disallow using both as modules.
2.4 config language generally does not support back references, so it is
safer to do sth like this:
dep_tristate 'AMI MegaRAID support' CONFIG_SCSI_MEGARAID $CONFIG_SCSI
if [ "$CONFIG_SCSI_MEGARAID" = "m" ]
define_tristate CONFIG_SCSI_MEGARAID2_DEP m
else
define_tristate CONFIG_SCSI_MEGARAID2_DEP y
fi
if [ "$CONFIG_SCSI_MEGARAID" != "y" ]
dep_tristate 'AMI MegaRAID2 support' CONFIG_SCSI_MEGARAID2 $CONFIG_SCSI_MEGARAID2_DEP $CONFIG_SCSI
fi
Using the same variable name in many interactive definitions breaks xconfig.
--
=======================================================================
Andrzej M. Krzysztofowicz [email protected]
phone (48)(58) 347 14 61
Faculty of Applied Phys. & Math., Gdansk University of Technology
excuse me for my english bad :) !
alloc_netdev func can return NULL and I think that is
not right to use dev pointer in this case without a
test.
so maybe it is better ? !
---------------------------------------------
--- knl_src/net/ipv4/ipmr.c 2003-11-25
16:02:59.000000000 -0300
+++ ipmr_patch.c 2004-01-08 10:21:22.000000000 -0300
@@ -205,7 +205,7 @@
dev = alloc_netdev(sizeof(struct net_device_stats),
"pimreg",
reg_vif_setup);
- if (register_netdevice(dev)) {
+ if (dev&®ister_netdevice(dev)) {
kfree(dev);
return NULL;
}
----------------------------------------------
sorry if is a dumb question and thanks for atention
pub 1024D/5139533E Joilnen Batista Leite
F565 BD0B 1A39 390D 827E 03E5 0CD4 0F20 5139 533E
__________________________________
Do you Yahoo!?
Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes
http://hotjobs.sweepstakes.yahoo.com/signingbonus
--- ipmr.c.org 2004-01-09 12:07:45.271713432 +0530
+++ ipmr.c 2004-01-09 12:09:39.389364912 +0530
@@ -205,6 +205,9 @@ static struct net_device *ipmr_reg_vif(v
dev = alloc_netdev(sizeof(struct net_device_stats), "pimreg",
reg_vif_setup);
+ if( dev == NULL)
+ return NULL;
+
if (register_netdevice(dev)) {
kfree(dev);
return NULL;
On Fri, 9 Jan 2004, Raj wrote:
> I would prefer this way of checking the NULL
>
> That would be more consistent with ip_gre.c and ipip.c
>
> any suggestions ?
Yes, this is much better form.
- James
--
James Morris
<[email protected]>
On Fri, 09 Jan 2004 12:11:39 +0530
Raj <[email protected]> wrote:
> I would prefer this way of checking the NULL
>
> That would be more consistent with ip_gre.c and ipip.c
Patch applied, thank you.
Marcelo Tosatti <[email protected]> wrote:
> On Tue, 6 Jan 2004, Mike Fedyk wrote:
>> On Tue, Jan 06, 2004 at 12:14:23PM -0200, Marcelo Tosatti wrote:
>> > It contains an ext2/3 update (mostly forward compatibility related), the
>>
>> Do you plan to merge htree?
>
> Yes, in the next -pre.
Hm. Afaict this has not happened yet (-pre8), is it still planned for
.25?
cu andreas
On Fri, 2004-01-30 at 15:57, Andreas Metzler wrote:
> Marcelo Tosatti <[email protected]> wrote:
> > On Tue, 6 Jan 2004, Mike Fedyk wrote:
> >> On Tue, Jan 06, 2004 at 12:14:23PM -0200, Marcelo Tosatti wrote:
> >> > It contains an ext2/3 update (mostly forward compatibility related), the
> >>
> >> Do you plan to merge htree?
> >
> > Yes, in the next -pre.
>
> Hm. Afaict this has not happened yet (-pre8), is it still planned for
> .25?
I think he meant for the 2.4.26-pre tree.
Markus
On Fri, 30 Jan 2004, Markus H?stbacka wrote:
> On Fri, 2004-01-30 at 15:57, Andreas Metzler wrote:
> > Marcelo Tosatti <[email protected]> wrote:
> > > On Tue, 6 Jan 2004, Mike Fedyk wrote:
> > >> On Tue, Jan 06, 2004 at 12:14:23PM -0200, Marcelo Tosatti wrote:
> > >> > It contains an ext2/3 update (mostly forward compatibility related), the
> > >>
> > >> Do you plan to merge htree?
> > >
> > > Yes, in the next -pre.
> >
> > Hm. Afaict this has not happened yet (-pre8), is it still planned for
> > .25?
> I think he meant for the 2.4.26-pre tree.
And it wont happen anymore. After saying that I would merge it, I decided
(based on input from sct and tytso) that we don't want htree in 2.4.x.
On Sat, Jan 31, 2004 at 08:47:32PM -0200, Marcelo Tosatti wrote:
> On Fri, 30 Jan 2004, Markus H?stbacka wrote:
>> On Fri, 2004-01-30 at 15:57, Andreas Metzler wrote:
>>> Marcelo Tosatti <[email protected]> wrote:
>>>> On Tue, 6 Jan 2004, Mike Fedyk wrote:
>>>>> On Tue, Jan 06, 2004 at 12:14:23PM -0200, Marcelo Tosatti wrote:
>>>>>> It contains an ext2/3 update (mostly forward compatibility
>>>>>> related), the
>>>>> Do you plan to merge htree?
>>>> Yes, in the next -pre.
>>> Hm. Afaict this has not happened yet (-pre8), is it still planned for
>>> .25?
>> I think he meant for the 2.4.26-pre tree.
> And it wont happen anymore. After saying that I would merge it, I decided
> (based on input from sct and tytso) that we don't want htree in 2.4.x.
Ok. Looks like I will have to switch to 2.6 some time. ;-) Thanks for the
information (and all the work).
cu andreas