Re: Linux 2.4.20-pre3

Re: Linux 2.4.20-pre3

[Pavel Machek]

Hi!

> Last attempt at documentation patches again against 2.4.20pre3 .

Feed them through trivial patch monkey?
								Pavel
-- 
Philips Velo 1: 1"x4"x8", 300gram, 60, 12MB, 40bogomips, linux, mutt,
details at http://atrey.karlin.mff.cuni.cz/~pavel/velo/index.html.

Linux 2.4.20-pre3

[Marcelo Tosatti]

Here goes -pre3.


Summary of changes from v2.4.20-pre2 to v2.4.20-pre3
============================================

<alex_williamson@hp.com>:
  o fix raid on GPT partitions

<andersen@codepoet.org>:
  o cdrom sane fallback vs 2.4.20-pre1

<garloff@suse.de>:
  o IBM 4000R needs LARGELUN

<gsromero@alumnos.euitt.upm.es>:
  o isofs multi volume compliance fix

<hch@lst.de>:
  o convert to yield() usage

Adrian Bunk <bunk@fs.tum.de>:
  o fix a typo in the description of CONFIG_BLK_STATS

Alan Cox <alan@lxorguk.ukuu.org.uk>:
  o PCI enable handling
  o fill in siginfo on alpha
  o ARM cpufreq hooks
  o update reiserfsprogs
  o document an acpi bogon
  o fix firestream gcc 3.1 warnings
  o fix io accounting on cpqarray
  o error check DAC960 copy*user
  o whoops
  o use proper macros in analog joystick
  o use proper tsc macros in random driver
  o add help button to sonypi driver, fix gcc 3.1 warnings
  o flip the vt ifdef the sane way around
  o fix the gcc warnings in cpqphp_nvram
  o make the st usb compile again
  o fix broken ifdef
  o Update CPIA driver (this has been in ac for a bit)
  o fix a gcc warning and a thinko
  o [resend] revert broken atarilance change
  o fix net/Config.in formatting, 83820 typo
  o fix eepro deeply broken code formatting (no other change)
  o latest mpt fusion update from author
  o update serverworks idents
  o fix extern->static inline on sbus
  o allow Zalon to be selected for HP
  o add hp tachyons to cpqfc driver + fix warnings
  o fix dpt_i2o warnings
  o fix esp driver to static inline
  o update scsi makefile for hp stuff
  o make NCR53c9x use inline not extern inline
  o update ncr asm assembler
  o switch scsi.h to static inline
  o fix incorrect read10 to read6 handling on error
  o sgiserial to static inline
  o usb to static inline
  o kaweth new ident (silicom usb)
  o ov511 driver updates
  o update pwc to new map handing too
  o update se401 the same way
  o same again for stv
  o update ohci driver to handle strange natsemi bits
  o update usbvideo and vicam
  o use static inline in clgenfb
  o allow people to select befs
  o make the dnotify cache consistent with other naming
  o make the fasync_cache also named in accordance tonorms
  o make file_lock cache also match default format
  o fix a problem where fsync of an nfs dir gives wrong code
  o small nls tidying
  o update to new ldm partition code
  o add sem_getcount to stop people poking in semaphore
  o allow DMA0 on isapnp
  o allow reporting of 3rd/4th codec
  o ALi 5455 audio
  o files_init - set file limit based on ram
  o get the types right on lib/inflate.c constants
  o add down_read_trylock/write_trylock
  o add befs maintainer
  o update hfs maintainer
  o remove dead url
  o format fix
  o add parisc directories to build
  o remainign minor atm bits
  o fix gcc 3 warning
  o make semaphores gcc 3.1 safe
  o the wbinvd is safe anyway
  o this fixes the remaining vm86/tf screwups
  o add HIL to serio
  o headers for SOM (HP) binary format
  o missed earlier - header change for sonypi
  o second item I found testing - misse scsi.h macro add
  o first bits of ps2esdi cleanup
  o fix depca warnings
  o revert wrong change to isicom
  o first pass xd.c clean up

Andi Kleen <ak@muc.de>:
  o pageattr for 2.4

Andrew Morton <akpm@zip.com.au>:
  o copy_strings speedup
  o fix lru_cache_add vs activate_page race

Bjorn Wesen <bjorn.wesen@axis.com>:
  o arch/cris update

Douglas Gilbert <dougg@torque.net>:
  o lk2.4.19 sg driver header file synchronization
  o scsi_debug driver update

Geert Uytterhoeven <geert@linux-m68k.org>:
  o m68k: PCI DMA updates
  o m68k: add page_to_phys()

jack_hammer@adaptec.com <Jack_Hammer@adaptec.com>:
  o ServeRAID driver update

Jean Tourrilhes <jt@bougret.hpl.hp.com>:
  o ir240_usb_fix_greg.diff
  o ir240_usb_disconnect-3.diff

Jens Axboe <axboe@suse.de>:
  o elevator seek accounting fixes

Kai Germaschewski <kai@tp1.ruhr-uni-bochum.de>:
  o Add support for ISDN card Eicon Diva 2.02
  o Move PCI device id to include/linux/pci_ids.h
  o Add support for ISDN card Formula-n enter:now, a.k.a. Gerdes Power ISDN
  o Add in-kernel ISAPnP support to HiSax driven ISDN cards
  o Doc changes / Cleanup for ISDN ISAPnP changes
  o ISDN MPPP crash fix
  o Update README.HiSax for the added card
  o Update README.HiSax for the added card
  o ISDN: LED support for netjet driver
  o ISDN: Add Data Over Voice support
  o ISDN: Cisco HDLC update
  o ISDN: Fix DoV (Data over Voice)

Marc Boucher <marc@mbsi.ca>:
  o yet another VAIO dmi_blacklist entry

Marcelo Tosatti <marcelo@plucky.distro.conectiva>:
  o Revert 2.4.19's AMD Athlon prefetch workaround
  o Revert APIC error message silencing: APIC errors can be fatal
  o Makefile

Martin Mares <mj@ucw.cz>:
  o pci.ids for 2.4.20-pre2

Neil Brown <neilb@cse.unsw.edu.au>:
  o resend - Enable NFS over TCP via config option

Stephen Rothwell <sfr@canb.auug.org.au>:
  o [2.4.20-pre1] File lease fixes

Steven Cole <elenstev@mesatop.com>:
  o 2.4.20-pre2 update Documentation/sysctl/vm.txt
  o 2.4.20-pre2 remove 8 duplicate help texts from

Trond Myklebust <trond.myklebust@fys.uio.no>:
  o Fix NFS locking bug
  o Teach RPC client to send pages rather than iovecs. [1/3]
  o Teach RPC client to send pages rather than iovecs. [2/3]
  o Teach RPC client to send pages rather than iovecs. [3/3]
  o Fix typo in the RPC reconnect code
  o Clean up RPC receive code [1/2]
  o Clean up RPC receive code [2/2]
  o fixup conflict between NFS kmap patches and 2.4.20-pre

Urban Widmark <urban@teststation.com>:
  o smbfs poll

Re: Linux 2.4.20-pre3

[Marcelo Tosatti]

On Sat, 17 Aug 2002, Skidley wrote:

> check.c: In function `devfs_register_disc':
> check.c:328: structure has no member named `number'
> check.c:329: structure has no member named `number'
> check.c: In function `devfs_register_partitions':
> check.c:361: structure has no member named `number'
> make[3]: *** [check.o] Error 1
> make[3]: Leaving directory
> `/home/skidley/kernel/linux-2.4.20-pre3/fs/partitions'
> make[2]: *** [first_rule] Error 2
> make[2]: Leaving directory
> `/home/skidley/kernel/linux-2.4.20-pre3/fs/partitions'
> make[1]: *** [_subdir_partitions] Error 2
> make[1]: Leaving directory `/home/skidley/kernel/linux-2.4.20-pre3/fs'
> make: *** [_dir_fs] Error 2

Yeah, I forgot to apply the fix to this one, sorry.

Here it is:

Subject: [PATCH] fix current BK tree compilation with devfs enabled


Not that I care for devfs, but there was at least one report on lkml.

I tried to also put the devfs_handle_t under CONFIG_DEVFS_FS, but the
devfs wrappers require it.  And yes, I'm seriously pissed that devfs
puts wordsize objects everywhere even if not enabled.


--- linux-2.4.20-bk-20020810/include/linux/genhd.h	Sat Aug 10 14:37:16 2002
+++ linux/include/linux/genhd.h	Mon Aug 12 23:40:37 2002
@@ -62,7 +62,9 @@ struct hd_struct {
 	unsigned long start_sect;
 	unsigned long nr_sects;
 	devfs_handle_t de;              /* primary (master) devfs entry  */
-
+#ifdef CONFIG_DEVFS_FS
+	int number;
+#endif /* CONFIG_DEVFS_FS */
 #ifdef CONFIG_BLK_STATS
 	/* Performance stats: */
 	unsigned int ios_in_flight;

Linux 2.4.20-pre3

[Skidley]

gcc -D__KERNEL__ -I/home/skidley/kernel/linux-2.4.20-pre3/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
-fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=i686
-nostdinc -I /usr/lib/gcc-lib/i386-linux/2.95.4/include
-DKBUILD_BASENAME=check  -DEXPORT_SYMTAB -c check.c
check.c: In function `devfs_register_disc':
check.c:328: structure has no member named `number'
check.c:329: structure has no member named `number'
check.c: In function `devfs_register_partitions':
check.c:361: structure has no member named `number'
make[3]: *** [check.o] Error 1
make[3]: Leaving directory
`/home/skidley/kernel/linux-2.4.20-pre3/fs/partitions'
make[2]: *** [first_rule] Error 2
make[2]: Leaving directory
`/home/skidley/kernel/linux-2.4.20-pre3/fs/partitions'
make[1]: *** [_subdir_partitions] Error 2
make[1]: Leaving directory `/home/skidley/kernel/linux-2.4.20-pre3/fs'
make: *** [_dir_fs] Error 2

-- 
"I mean they are gonna kill ya so like if ya give em a quick, short, sharp,
shock they won't do it again. Dig it! I mean he got off lightly cuz I would 
have given him a thrashing. I only hit him once. It was only a difference of 
opinion but really... I mean good manners don't cost nothin do they. Eh?"

Re: Linux 2.4.20-pre3

[Adrian Bunk]

The compile error in efi.c is still present in -pre3:

<--  snip  -->

...
gcc -D__KERNEL__
-I/home/bunk/linux/kernel-2.4/linux-2.4.19-modular/include -Wal
l -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
-pipe -mpreferred-stack-boundary=2 -march=k6   -nostdinc -I
/usr/lib/gcc-lib/i386-linu
x/2.95.4/include -DKBUILD_BASENAME=efi  -c -o efi.o efi.c
efi.c: In function `add_gpt_partitions':
efi.c:728: `NULL_GUID' undeclared (first use in this function)
efi.c:728: (Each undeclared identifier is reported only once
efi.c:728: for each function it appears in.)
make[3]: *** [efi.o] Error 1
make[3]: Leaving directory
`/home/bunk/linux/kernel-2.4/linux-2.4.19-modular/fs/partitions'

<--  snip  -->

Matt suggested to move include/asm-ia64/efi.h to include/linux/efi.h. Is
it possible to do this before 2.4.20-final?


TIA
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

Re: Linux 2.4.20-pre3

[Stephane Wirtel]

about the patch of devfs, why don't add directly this patch in the new 2.4.20-pre3
?



On Sat, Aug 17, 2002 at 01:18:17AM -0300, Marcelo Tosatti wrote:
> 
> 
> On Sat, 17 Aug 2002, Skidley wrote:
> 
> > check.c: In function `devfs_register_disc':
> > check.c:328: structure has no member named `number'
> > check.c:329: structure has no member named `number'
> > check.c: In function `devfs_register_partitions':
> > check.c:361: structure has no member named `number'
> > make[3]: *** [check.o] Error 1
> > make[3]: Leaving directory
> > `/home/skidley/kernel/linux-2.4.20-pre3/fs/partitions'
> > make[2]: *** [first_rule] Error 2
> > make[2]: Leaving directory
> > `/home/skidley/kernel/linux-2.4.20-pre3/fs/partitions'
> > make[1]: *** [_subdir_partitions] Error 2
> > make[1]: Leaving directory `/home/skidley/kernel/linux-2.4.20-pre3/fs'
> > make: *** [_dir_fs] Error 2
> 
> Yeah, I forgot to apply the fix to this one, sorry.
> 
> Here it is:
> 
> Subject: [PATCH] fix current BK tree compilation with devfs enabled
> 
> 
> Not that I care for devfs, but there was at least one report on lkml.
> 
> I tried to also put the devfs_handle_t under CONFIG_DEVFS_FS, but the
> devfs wrappers require it.  And yes, I'm seriously pissed that devfs
> puts wordsize objects everywhere even if not enabled.
> 
> 
> --- linux-2.4.20-bk-20020810/include/linux/genhd.h	Sat Aug 10 14:37:16 2002
> +++ linux/include/linux/genhd.h	Mon Aug 12 23:40:37 2002
> @@ -62,7 +62,9 @@ struct hd_struct {
>  	unsigned long start_sect;
>  	unsigned long nr_sects;
>  	devfs_handle_t de;              /* primary (master) devfs entry  */
> -
> +#ifdef CONFIG_DEVFS_FS
> +	int number;
> +#endif /* CONFIG_DEVFS_FS */
>  #ifdef CONFIG_BLK_STATS
>  	/* Performance stats: */
>  	unsigned int ios_in_flight;
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

-- 
Stephane Wirtel <stephane.wirtel@belgacom.net>
Web : www.linux-mons.be	 "Linux Is Not UniX !!!"

Re: Linux 2.4.20-pre3

[Mel]

Last attempt at documentation patches again against 2.4.20pre3 .

http://www.csn.ul.ie/~mel/projects/vm/patches/2.4.20pre3/2.4.20pre3_documentation_numa
http://www.csn.ul.ie/~mel/projects/vm/patches/2.4.20pre3/2.4.20pre3_page_alloc_commentry
http://www.csn.ul.ie/~mel/projects/vm/patches/2.4.20pre3/2.4.20pre3_paging_documentation
http://www.csn.ul.ie/~mel/projects/vm/patches/2.4.20pre3/2.4.20pre3_slab_commentry
http://www.csn.ul.ie/~mel/projects/vm/patches/2.4.20pre3/2.4.20pre3_vmalloc_commentry

All of them but page_alloc is the same as the pre2 ones and it applied
cleanly with offsets.

-- 
Mel Gorman
MSc Student, University of Limerick
http://www.csn.ul.ie/~mel

Re: Linux 2.4.20-pre3

[Geert Uytterhoeven]

On Fri, 16 Aug 2002, Marcelo Tosatti wrote:
> Alan Cox <alan@lxorguk.ukuu.org.uk>:
>   o files_init - set file limit based on ram

Add missing prototype (cfr. 2.5.x).
BTW, the one is 2.5.x is wrong because it lacks the __init

--- linux-2.4.20-pre3/include/linux/fs.h	Sat Aug 17 14:11:08 2002
+++ linux-m68k-2.4.20-pre3/include/linux/fs.h	Sat Aug 17 15:58:51 2002
@@ -206,6 +206,7 @@
 extern void buffer_init(unsigned long);
 extern void inode_init(unsigned long);
 extern void mnt_init(unsigned long);
+extern void __init files_init(unsigned long mempages);
 
 /* bh state bits */
 enum bh_state_bits {

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: Linux 2.4.20-pre3

[Christoph Hellwig]

On Sat, Aug 17, 2002 at 04:12:03PM +0200, Geert Uytterhoeven wrote:
> On Fri, 16 Aug 2002, Marcelo Tosatti wrote:
> > Alan Cox <alan@lxorguk.ukuu.org.uk>:
> >   o files_init - set file limit based on ram
> 
> Add missing prototype (cfr. 2.5.x).
> BTW, the one is 2.5.x is wrong because it lacks the __init

The prototype doesn't need the __init, and consensus is to not add it.

Re: Linux 2.4.20-pre3

[Geert Uytterhoeven]

On Sat, 17 Aug 2002, Christoph Hellwig wrote:
> On Sat, Aug 17, 2002 at 04:12:03PM +0200, Geert Uytterhoeven wrote:
> > On Fri, 16 Aug 2002, Marcelo Tosatti wrote:
> > > Alan Cox <alan@lxorguk.ukuu.org.uk>:
> > >   o files_init - set file limit based on ram
> > 
> > Add missing prototype (cfr. 2.5.x).
> > BTW, the one is 2.5.x is wrong because it lacks the __init
> 
> The prototype doesn't need the __init, and consensus is to not add it.

IC. Then please apply this one instead:

--- linux-2.4.20-pre3/include/linux/fs.h	Sat Aug 17 14:11:08 2002
+++ linux-m68k-2.4.20-pre3/include/linux/fs.h	Sat Aug 17 15:58:51 2002
@@ -206,6 +206,7 @@
 extern void buffer_init(unsigned long);
 extern void inode_init(unsigned long);
 extern void mnt_init(unsigned long);
+extern void files_init(unsigned long mempages);
 
 /* bh state bits */
 enum bh_state_bits {

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: Linux 2.4.20-pre3

[Skidley]

Not much a bug report but when I boot 2.4.20-pre3 I only get as far as
this :

Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket

Just hangs there no OOPs or anything. Please let me know any info you
might need like hardware or .config options. 2.4.20-pre2 is fine with 
same .config, etc.

-- 
"I mean they are gonna kill ya so like if ya give em a quick, short, sharp,
shock they won't do it again. Dig it! I mean he got off lightly cuz I would 
have given him a thrashing. I only hit him once. It was only a difference of 
opinion but really... I mean good manners don't cost nothin do they. Eh?"