Linux 2.4.21-pre5

Linux 2.4.21-pre5

[Marcelo Tosatti]

So here goes -pre5.


Summary of changes from v2.4.21-pre4 to v2.4.21-pre5
============================================

<ajoshi@kernel.crashing.org>:
  o rivafb 0.9.4 update

<alex_williamson@hp.com[helgaas]>:
  o ia64: fix typo in ia32_support.c

<andrew.wood@ivarch.com>:
  o USB: USB-MIDI support for Roland SC8820

<arun.sharma@intel.com[helgaas]>:
  o ia64: ia32 emulation layer bug fix

<bbosch@iphase.com>:
  o [netdrvr ns83820] big endian fixes

<benh@zion.wanadoo.fr>:
  o Fix a bug in the workaround for closed P2P bridge IO windows which could actually break bridges that didn't need fixing
  o Export atomic_{clear,set}_mask for modules
  o Request Open Firmware to open all "display" devices instead limiting us to the first one. This helps getting all cards properly POSTed
  o Prevent the stack from growing on reads. This works around a problem with the mount syscall calling copy_mount_options() which can trigger a fault via copy_from_user() between the last core VMA and the stack.
  o Properly fixup the Winbond W83C553 IDE on Longtrail and BriQ's so the controller is switched to fully native mode and interrupts are configured properly
  o Fix serial table for BriQ hardware (different base clock) and make sure it works with CONFIG_VT
  o Fix a warning
  o Make sure xmon doesn't try to tap a hash table when none exist
  o Add asm byteswapped 64 bits accessors
  o Rework inline syscall macros, fix clobbers & gcc3.3 (From Franz Sirl)
  o Remove old gross hack that did nothing good

<bergner@vnet.ibm.com>:
  o Remove kdb from PowerPC-64
  o ppc64 updates to 2.4.21-pre4

<bjorn_helgaas@hp.com[helgaas]>:
  o ia64: Add local_irq_set() and save_and_sti()
  o ia64: Use IA64_PSR_I rather than (1UL << 14)
  o ia64: Reverse SGI scatterlist changes so SGI update will apply
  o ia64: Simple ndelay implementation
  o ia64: Add some default configs
  o ia64: whitespace fixes
  o ia64: add infrastructure for multiple IO port spaces
  o ia64: add support for MMIO and IO port spaces from ACPI _CRS
  o ia64: add iomem_resource and ioport_resource allocation
  o ia64: update defconfigs
  o Rename configs

<cniehaus@handhelds.org>:
  o spelling fix for drivers_usb_usbnet.c

<d3august@dtek.chalmers.se>:
  o USB: small uhci bug

<dave@thedillows.org>:
  o The initial release of the driver for the 3Com 3cr990 family

<davidm@tiger.hpl.hp.com[helgaas]>:
  o ia64: For ia32 emulation, do not turn on O_LARGEFILE automatically
  o ia64: Don't risk running past the end of the unwind-table.  Based on a patch by Suresh Siddha.

<davidm@wailua.hpl.hp.com[helgaas]>:
  o ia64: Fix ia64_fls() so it works for all possible 64-bit values

<eranian@frankl.hpl.hp.com[helgaas]>:
  o ia64: new perfmon patch for 2.4.20
  o ia64: perfmon update

<green@linuxhacker.ru>:
  o radio-cadet compile fix

<henning@meier-geinitz.de>:
  o USB scanner.c: Adjust syslog output

<ionut@badula.org>:
  o VLAN support, 64-bit support, bugfixes

<jbarnes@sgi.com>:
  o MAINTAINERS update for 2.4 SN support

<jgarzik@pobox.com>:
  o Fix undefined references for smp + apm

<jh@sgi.com[helgaas]>:
  o ia64: Update SGI SN files

<jochen@scram.de>:
  o [tokenring smctr] fix MAC address input
  o [tokenring madgemc] fix memory leak, add proper refcounting

<kare.sars@lmf.ericsson.se>:
  o [atm nicstar] fix incorrect traffic class assumption

<m.c.p@wolk-project.de>:
  o Speedup 'make dep'

<meissner@suse.de>:
  o [netdrvr pcnet32] fix multicast on big endian

<mikal@stillhq.com>:
  o Handle scsi_register() failure

<p.guehring@futureware.at>:
  o USB: FTDI driver, new id added

<peter@bergner.org>:
  o PPC64 update

<raul@pleyades.net>:
  o mmap.c corner case fix

<sprite@sprite.fr.eu.org>:
  o [SPARC64]: Avoid use of -e option with echo

<stelian@popies.net>:
  o sonypi and input subsystem integration
  o CREDITS update
  o use correct gcc flags when compiling for
  o sonypi driver update
  o make mousedev accept the jogdial
  o meye suspend/resume capabilities

Adrian Bunk <bunk@fs.tum.de>:
  o fix compile error with two IrDA drivers

Alan Cox <alan@lxorguk.ukuu.org.uk>:
  o ACPI apparently wasnt bios
  o fix wrong date in microcode comment
  o add another legitimate P4 type
  o must disallow write combine on 450NX
  o add framework for ndelay (nanoseconds)
  o first block of parisc resend
  o second block of parisc merge
  o third block of parisc merge
  o Ian Nelson moved
  o update videobook docs to avoid check_region
  o docs for IPMI
  o remove dead init call
  o add AMD hammer rng
  o IPMI driver updates
  o keyboard changes
  o fix wrong test in raw driver
  o fix paths for ide
  o clarify hpt37x config
  o fix more ide paths
  o Paul's fix to do ide_cs handling in task context
  o more ide paths
  o fix use of check_region in umc driver
  o more ide comment/doc info updates
  o promise printk cleanups
  o another wrong path
  o IDE printk/cleanup bits
  o fix serverworks paths/docs
  o clean up the siimage driver
  o update sis driver comments/docs/notes
  o update PIIX driver to know about more errata
  o fix winbond driver for new ide
  o more ide doc/comment updates
  o fix ppc ide paths
  o Ide raid updates
  o fix sbp2 compile failure
  o fix unsafe signed wrap check in pcilynx
  o use kbd_refresh_leds to keep USB/base keyboad lights right
  o clean up radio-cadet locking
  o use skb_padto to fix 3c527 padding
  o fix typo in 3c523 fixups
  o fix ethernet padding on 82596
  o fix ethernet padding on ariadne
  o fix ethernet padding on a2065
  o fix ethernet padding on atarilance
  o fix ethernet padding on am79c961a
  o fix ethernet padding on bagetlance
  o fix ethernet padding on declance
  o fix padding on depca
  o fix padding on eepro driver
  o fix padding on eexpress driver
  o fix ethernet padding on fmv18x
  o fix e2100 crash
  o fix ethernet padding on eth16i
  o fix ethernet padding on lasi
  o fix padding on epic100 driver
  o fix ethernet padding on lp486e
  o fix ethernet padding on lancr
  o fix padding on fmvj18x_cs
  o fix ethernet padding on hp100
  o fix ethernet padding on pcmcia/ray_cs
  o fix ethernet padding on xircom
  o fix ethernet padding on r8169
  o fix ethernet padding on seeq8005
  o fix padding on smc9194
  o fix padding on via_rhine
  o fix padding on yellowfin
  o fix padding on znet
  o fix padding on wavelan
  o update pci.ids for syskonnect
  o add 450NX streaming quirk, add via northbridge detect
  o fix dpt_i2o out of memory check
  o fix eata_generic jiffies check
  o document an ICH errata we have to deal with
  o fix sb_mixer handling
  o dont fail on 5451 reset
  o ide.h changes
  o add prototypes for kbdrefresh_leds
  o add skb_padto operation
  o fix ipc/msg race by dropping optimisation out
  o add skb_pad operation
  o copy OUTBSYNC operation too
  o fix the ide irq masking bug Ross found
  o fix confusing extra DMA off messages
  o add but dont yet use ide_execute_comman
  o sk98 driver vendor update

Alan Stern <stern@rowland.harvard.edu>:
  o USB: Patches for the ECONNRESET error (2.4)

Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>:
  o [TCP]: Do not forget data copy while collapsing retransmission queue

Andi Kleen <ak@muc.de>:
  o [IPV4]: Better behavior for NETDEV_CHANGENAME requests
  o x86-64 update
  o Workaround for AMD 8131 bug
  o Fix get_vm_area

Andrea Arcangeli <andrea@suse.de>:
  o xdr nfs highmem deadlock fix

Andrew Morton <akpm@digeo.com>:
  o ia32 syscall compatibility stubs

Andrey Panin <pazke@orbita1.ru>:
  o [netdrvr eepro100] add config option for PIO register read/write

Benjamin Herrenschmidt <benh@kernel.crashing.org>:
  o PPC32: Implement workarounds for errata on recent G3 and G4 cpus

Bjorn Helgaas <bjorn_helgaas@hp.com>:
  o ia64: Delete all SGI SN defconfig files
  o ia64: Dont execute srlz.d needlessly (reported by Chris Ruemmler)
  o ia64: smp_threads_ready: make non-volatile
  o don't swapon mounted devices
  o ia64: Use has_8259 rather than initdata
  o ia64: Really remove ACPI SPCR parsing
  o Cset exclude: eranian@frankl.hpl.hp.com[helgaas]|ChangeSet|20030103231109|26349
  o ia64: fix perfmon typo (PFM_CPU_SYST_WIDE should be PFM_CPUINFO_SYST_WIDE)

Christoph Hellwig <hch@sgi.com>:
  o handle too large vmallocs gracefully

Dave Jones <davej@codemonkey.org.uk>:
  o [netdrvr sunqe] remove incorrect kfree()

Dave Kleikamp <shaggy@shaggy.austin.ibm.com>:
  o JFS: replace ugly JFS debug macros with simpler ones
  o JFS: Minor update in Documentation/filesystems/jfs.txt
  o JFS: implement get_index_page to replace some uses of read_index_page
  o JFS: Add debug code to help catch elusive bug
  o JFS: simplify jfs_err() to avoid parsing bug in gcc-2.95
  o JFS: Fix jfs_sync_fs

David Brownell <david-b@pacbell.net>:
  o USB: ehci-hcd, more hangs gone

David Gibson <david@gibson.dropbear.id.au>:
  o PPC32: Add work-around for erratum #77 on IBM 405 processors
  o Update orinoco driver to 0.13b

David S. Miller <davem@nuts.ninka.net>:
  o [TG3]: Let chip do pseudo-header csum on rx
  o [TG3]: Add device IDs for 5704S/5702a3/5703a3
  o [TG3]: Prevent dropped frames when flow-control is enabled
  o [TG3]: Correct MIN_DMA and ONE_DMA settings in dma_rwctrl
  o [TG3]: Workaround 5701 back-to-back register write bug
  o [TG3]: Add workaround for third-party phy issues
  o [TG3]: Remove anal grc_misc_cfg board IDs check
  o [TG3]: Fix typos in previous changes
  o [TCP]: In tcp_check_req, handle ACKless packets properly
  o [SPARC]: Add ndelay
  o [SPARC]: Add ndelay ksyms export

David Woodhouse <dwmw2@infradead.org>:
  o Export skb_pad() in 2.4.21-pre4

Gerd Knorr <kraxel@bytesex.org>:
  o bttv documentation update
  o tuner module update
  o video4linux i2c modules update
  o bttv update

Gerd Knorr <kraxel@suse.de>:
  o bttv config fix

Greg Kroah-Hartman <greg@kroah.com>:
  o USB: hid blacklist update
  o USB: more hid blacklist items
  o USB: added tripp device id's to pl2303 driver

Ivan Kokshaysky <ink@jurassic.park.msu.ru>:
  o alpha dma fix
  o alpha update

Jay Vosburgh <fubar@us.ibm.com>:
  o [netdrvr 3c59x] move netif_carrier_off() call outside vortex_debug test

Jeff Garzik <jgarzik@mandrakesoft.com>:
  o arch/i386/Makefile: fix Via C3 build flags with gcc 3.<recent>

Jeff Garzik <jgarzik@redhat.com>:
  o [netdrvr tg3] bump version, tidy comments
  o [netdrvr amd8111e] remove stray ';', fixing register dump [#311]
  o [netdrvr tg3] DMA MRM bit only exists on 5700, 5701
  o [netdrvr fc/iphase] correct PCI probe loop-end test logic [#323]
  o [tokenring smctr] remove stray ';' that prevented a loop from working [#312]
  o [ARM] CREDITS, MAINTAINERS, Documentation/arm/* updates
  o [ARM] misc janitorial cleanups for arch/arm/kernel
  o [ARM] misc janitorial cleanups for arch/arm/mach*, arch/arm/mm
  o [ARM] misc janitorial cleanups for include/asm-arm
  o [netdrvr 8390] if ARM, only redefine EI_SHIFT, not I/O macros
  o [netdrvr] add new ARM net drivers cirrus, ether00
  o [netdrvr bmac] Remove unneeded memset()
  o [netdrvr 8139too] add some boards to the list of tested boards
  o [netdrvr tg3] disable 5701 h/w bug workaround during core clock reset
  o [netdrvr tg3] fix NAPI deadlock
  o [netdrvr tg3] bump version to 1.4c / Feb 18
  o [netdrvr tg3] properly synchronize with TX, in tg3_netif_stop
  o [netdrvr tg3] fix TX race in previous code, and another buglet
  o [netdrvr] Update Doc/networking/netdevices.txt with more locking rules

Jens Axboe <axboe@suse.de>:
  o Remove unused node from ide-probe.c
  o Andrea's elevator backmerge patch]

Johannes Erdfelt <johannes@erdfelt.com>:
  o usb_get_driver_np() gives wrong driver name (usb_mouse)
  o USB: OHCI trivial remove unused field
  o USB: 2.4 OHCI trivial comment cleanup

John Stultz <johnstul@us.ibm.com>:
  o Fix target_cpus()

Kurt Garloff <garloff@suse.de>:
  o Handle SCSI recovered errors

Maksim Krasnyanskiy <maxk@qualcomm.com>:
  o [Bluetooth] Add support for vendor specific commands

Marcelo Tosatti <marcelo@freak.distro.conectiva>:
  o Cset exclude: rusty@rustcorp.com.au|ChangeSet|20030224224251|29662
  o Changed EXTRAVERSION to -pre5
  o Define kmap_nonblock() for non highmem

Mark A. Greer <mgreer@mvista.com>:
  o PPC32: Fix our L2 / L3 cache updates for the bootloader

Martin Devera <devik@cdi.cz>:
  o [NET_SCHED]: HTB scheduler updates from Devik

Martin Schwidefsky <schwidefsky@de.ibm.com>:
  o s390 base architecture update
  o xpram driver fix for 64-bit
  o s390 idals.h update

Matthew Wilcox <willy@debian.org>:
  o [wireless airo] call pci_enable_device, pci_set_master where needed

Olaf Hering <olh@suse.de>:
  o ide_fix_driveid unresolved in usb-storage

Oleg Drokin <green@angband.namesys.com>:
  o reiserfs: Fix DIRECT IO interference with tail packing

Oliver Neukum <oliver@neukum.name>:
  o USB: 2.4 ehci uses SLAB_KERNEL in interrupt
  o USB: kaweth length calculation fix
  o USB: new device id for kaweth

Paul Mackerras <paulus@samba.org>:
  o PPC32: Fix the clone syscall, and make exec clear fp and vr registers
  o PPC32: Clean up exception and oops handling
  o PPC32: Tighten up the stack expansion code
  o PPC32: Fix handling of alignment traps on some PPC processors
  o PPC32: Actually use the FP exception mode requested with prctl()
  o PPC32: use the standard __stringify instead of a local version
  o PPC32: Further fixes for the stack expansion code
  o PPC32: add ndelay(), update udelay() to be more accurate and robust
  o PPC32: Minor cleanups in the CHRP platform code
  o PPC32: Allow for RAM not starting at 0, for APUS (and potentially others)
  o PPC32: PReP platform fixes from Hollis Blanchard, Tom Rini, Leigh Brown and others
  o PPC32: Fixes for byte-swapping macros, from Franz Sirl
  o PPC32: PCI fixes.  We can now restrict I/O windows to 16MB or so because this code lets us move the I/O windows of PCI-PCI bridges if necessary.
  o PPC32: Fix copy_from_user to copy as much as possible even when it gets a fault
  o PPC32: Provide a default implementation of ide_init_hwif_ports in asm-ppc/ide.h and use it if there is no platform-specific version.
  o PPC32: fix compilation error in arch/ppc/platforms/pmac_setup.c
  o PPC32: Move some variable declarations related to the MMU hash table to <asm/mmu.h>

Pete Zaitcev <zaitcev@redhat.com>:
  o [SPARC32]: Backport fixes from 2.5.x

Petko Manolov <petkan@users.sourceforge.net>:
  o USB: pegasus update (2.4)

Roger Luethi <rl@hellgate.ch>:
  o [netdrvr via-rhine] trivial bits
  o [netdrvr via-rhine] fix broken tx-underrun handling
  o [netdrvr via-rhine] various duplex-related fixes
  o [netdrvr via-rhine] reset function rewrite
  o [netdrvr via-rhine] bump version, use constant instead of magic number

Rusty Russell <rusty@rustcorp.com.au>:
  o namespace pollution in procfs
  o arch_ia64_sn_io_sn1_pcibr.c, typo: the the
  o misc register audit fix on qtronix
  o duplicate header in drivers_bluetooth_hci_h4.c
  o write with buffer>2GB returns broken errno
  o misc register audit fix on ppc64's nvram.c
  o USB: Clean up some USB macros
  o available spell fixes
  o correct description of Griffin Powermate
  o namespace pollution in eth bridge driver
  o drivers_net_wan_sdla_x25.c, typo: the the
  o es1372.c doesn't free resources correctly
  o Typos in drivers_s390_net_iucv.c
  o i2c ID addition
  o NCR5380 unbalanced curly brace
  o Fix floppy.h's CROSS_64KB()

Scott Feldman <scott.feldman@intel.com>:
  o [netdrvr e100] math fixes and a cleanup

Stephen C. Tweedie <sct@redhat.com>:
  o Fix signed use of i_blocks in ext3 truncate

Tom Rini <trini@kernel.crashing.org>:
  o PPC32: Change the MPC8xx IRQ code so that things are arranged like other systems.
  o PPC32: Enable PCMCIA and a tested wifi card on some MPC8xx targets
  o PPC32: Change the MontaVista copyright / GPL boilerplate to a condensed version.
  o PPC32: Fix an oops on hardware without an RTC in timer_interrupt()
  o PPC32: Fix building of the IBM Spruce platform and !CONFIG_SERIAL
  o PPC32: Fix some gcc-3.x warnings on the IBM Spruce
  o PPC32: Cleanup the boot code to better deal with no console
  o PPC32: Minor KGDB warning fixes
  o PPC32: Add CONFIG_KGDB_CONSOLE to MPC 8xx systems
  o PPC32: MPC8xx KGDB fixes, from Dan Malek
  o PPC32: Add KGDB support for the IBM Spruce platform
  o PPC32: Ask about CONFIG_BOOTX_TEXT in the 'Kernel hacking' menu
  o PPC32: Put reading of PReP/PPCBUG nvram into CONFIG_PPCBUG_NVRAM
  o PPC32: Add support for the Motorola LoPEC platform
  o PPC32: Remove the 'BK Id' tags from files
  o PPC32: Fix SysRq on IBM Spruce

Trond Myklebust <trond.myklebust@fys.uio.no>:
  o Fix XID allocation race in 2.4.21-pre4

Wolfgang Muees <wolfgang@iksw-muees.de>:
  o USB: updated Auerswald driver

Re: Linux 2.4.21-pre5

[Christoph Hellwig]

> <bergner@vnet.ibm.com>:
>   o Remove kdb from PowerPC-64
>   o ppc64 updates to 2.4.21-pre4

This removes the *xattr and tkill syscalls from ppc64 which is bogus.
It would be really nice if arch maintainers could actually follow mainline
instead of some vendor tree and do blind "fixups"..

Re: Linux 2.4.21-pre5

[Adrian Bunk]

On Thu, Feb 27, 2003 at 03:14:44AM -0300, Marcelo Tosatti wrote:
> 
> So here goes -pre5.
> 
> 
> Summary of changes from v2.4.21-pre4 to v2.4.21-pre5
> ============================================
> 
> <ajoshi@kernel.crashing.org>:
>   o rivafb 0.9.4 update
>...

WTF is this???

Besides the rivafb update it reverts parts of the IEEE 1394 patches that 
were included in -pre4.


Please revert these bogus changes.


cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: Linux 2.4.21-pre5

[Christoph Hellwig]

> <jh@sgi.com[helgaas]>:
>   o ia64: Update SGI SN files

This one adds a strtok_r implementation instead of just using strsep,
strange, strange..

Re: Linux 2.4.21-pre5

[Ben Lau]

Hi,

   I have tried to compile the -pre5 with IEEE1394
support and i got the following error:

 gcc -D__KERNEL__ -I/usr/src/2.4.21pre5/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
-fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=i386
-DMODULE  -nostdinc -iwithprefix include -DKBUILD_BASENAME=raw1394  -c
-o raw1394.o raw1394.c
 In file included from raw1394.c:50:
 raw1394.h:167: field `tq' has incomplete type
 raw1394.c: In function `__alloc_pending_request':
 raw1394.c:110: warning: implicit declaration of function `HPSB_INIT_WORK'
 raw1394.c:118: confused by earlier errors, bailing out
 make[2]: *** [raw1394.o] Error 1
 make[2]: Leaving directory `/usr/src/2.4.21pre5/drivers/ieee1394'
 make[1]: *** [_modsubdir_ieee1394] Error 2
 make[1]: Leaving directory `/usr/src/2.4.21pre5/drivers'
 make: *** [_mod_drivers] Error 2

The definition of hpsb_queue_struct was missing
in the -pre5. I found that it did exist on -pre4

/usr/src/2.4.21pre4/drivers/ieee1394/ieee1394_types.h:45:#define
hpsb_queue_struct tq_struct

Re: Linux 2.4.21-pre5

[Simon Oosthoek]

[-- Attachment #1: Type: text/plain, Size: 1262 bytes --]

On Thu, Feb 27, 2003 at 05:39:51PM +0800, Ben Lau wrote:
> Hi,
> 
>    I have tried to compile the -pre5 with IEEE1394
> support and i got the following error:
> 
>  gcc -D__KERNEL__ -I/usr/src/2.4.21pre5/include -Wall
> -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
> -fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=i386
> -DMODULE  -nostdinc -iwithprefix include -DKBUILD_BASENAME=raw1394  -c
> -o raw1394.o raw1394.c
>  In file included from raw1394.c:50:
>  raw1394.h:167: field `tq' has incomplete type
>  raw1394.c: In function `__alloc_pending_request':
>  raw1394.c:110: warning: implicit declaration of function `HPSB_INIT_WORK'
>  raw1394.c:118: confused by earlier errors, bailing out
>  make[2]: *** [raw1394.o] Error 1
>  make[2]: Leaving directory `/usr/src/2.4.21pre5/drivers/ieee1394'
>  make[1]: *** [_modsubdir_ieee1394] Error 2
>  make[1]: Leaving directory `/usr/src/2.4.21pre5/drivers'
>  make: *** [_mod_drivers] Error 2
> 
> The definition of hpsb_queue_struct was missing
> in the -pre5. I found that it did exist on -pre4
> 
> /usr/src/2.4.21pre4/drivers/ieee1394/ieee1394_types.h:45:#define
> hpsb_queue_struct tq_struct

Same here...

here's a patch that should work to fix this.

Cheers

Simon

[-- Attachment #2: ieee1394.patch-2.4.21-pre5 --]
[-- Type: text/plain, Size: 1102 bytes --]

--- linux-2.4.21-pre5/drivers/ieee1394/ieee1394_types.h	2003-02-27 11:04:08.000000000 +0100
+++ linux-2.4.21-pre4-ac4/drivers/ieee1394/ieee1394_types.h	2003-02-17 20:37:10.000000000 +0100
@@ -10,6 +10,27 @@
 #include <asm/byteorder.h>
 
 
+/* The great kdev_t changeover in 2.5.x */
+#include <linux/kdev_t.h>
+#ifndef minor
+#define minor(dev) MINOR(dev)
+#endif
+
+#ifndef __devexit_p
+#define __devexit_p(x) x
+#endif
+
+#include <linux/spinlock.h>
+
+#ifndef list_for_each_safe
+#define list_for_each_safe(pos, n, head) \
+	for (pos = (head)->next, n = pos->next; pos != (head); \
+		pos = n, n = pos->next)
+
+#endif
+
+#define pte_offset_kernel pte_offset
+
 #ifndef MIN
 #define MIN(a,b) ((a) < (b) ? (a) : (b))
 #endif
@@ -19,6 +40,15 @@
 #endif
 
 
+/* Use task queue */
+#include <linux/tqueue.h>
+#define hpsb_queue_struct tq_struct
+#define hpsb_queue_list list
+#define hpsb_schedule_work(x) schedule_task(x)
+#define HPSB_INIT_WORK(x,y,z) INIT_TQUEUE(x,y,z)
+#define HPSB_PREPARE_WORK(x,y,z) PREPARE_TQUEUE(x,y,z)
+
+
 typedef u32 quadlet_t;
 typedef u64 octlet_t;
 typedef u16 nodeid_t;

Re: Linux 2.4.21-pre5

[Benjamin Herrenschmidt]

On Thu, 2003-02-27 at 11:19, Simon Oosthoek wrote:

> here's a patch that should work to fix this.

I think the proper fix for now is to bring back all of
ieee1394 from pre4 to pre5, it was an incorrect merge
that reverted part of it.

Ben.

Re: Linux 2.4.21-pre5

[Simon Oosthoek]

On Thu, Feb 27, 2003 at 11:48:44AM +0100, Benjamin Herrenschmidt wrote:
> On Thu, 2003-02-27 at 11:19, Simon Oosthoek wrote:
> 
> > here's a patch that should work to fix this.
> 
> I think the proper fix for now is to bring back all of
> ieee1394 from pre4 to pre5, it was an incorrect merge
> that reverted part of it.

where do you think I got the patch from ;-)
I diffed pre5 with pre4(-ac4) to get this one...

I'm sure a proper fix is forthcoming by someone more knowledgable than me, I
just posted this to help people get through a compile.

Cheers

Simon

Re: Linux 2.4.21-pre5

[Ben Collins]

On Thu, Feb 27, 2003 at 09:32:39AM +0100, Adrian Bunk wrote:
> On Thu, Feb 27, 2003 at 03:14:44AM -0300, Marcelo Tosatti wrote:
> > 
> > So here goes -pre5.
> > 
> > 
> > Summary of changes from v2.4.21-pre4 to v2.4.21-pre5
> > ============================================
> > 
> > <ajoshi@kernel.crashing.org>:
> >   o rivafb 0.9.4 update
> >...
> 
> WTF is this???
> 
> Besides the rivafb update it reverts parts of the IEEE 1394 patches that 
> were included in -pre4.

Yes, please revert this, and then make sure _my_ patch gets into -pre6?

-- 
Debian     - http://www.debian.org/
Linux 1394 - http://www.linux1394.org/
Subversion - http://subversion.tigris.org/
Deqo       - http://www.deqo.com/

Re: Linux 2.4.21-pre5

[ajoshi]

Just for the record, the patch I sent to Marcelo did _NOT_ include any
1394 hunks in it.  Marcelo must have accidently mixed mine up with some
old 1394 tree.


ani

On Thu, 27 Feb 2003, Ben Collins wrote:

> On Thu, Feb 27, 2003 at 09:32:39AM +0100, Adrian Bunk wrote:
> > On Thu, Feb 27, 2003 at 03:14:44AM -0300, Marcelo Tosatti wrote:
> > > 
> > > So here goes -pre5.
> > > 
> > > 
> > > Summary of changes from v2.4.21-pre4 to v2.4.21-pre5
> > > ============================================
> > > 
> > > <ajoshi@kernel.crashing.org>:
> > >   o rivafb 0.9.4 update
> > >...
> > 
> > WTF is this???
> > 
> > Besides the rivafb update it reverts parts of the IEEE 1394 patches that 
> > were included in -pre4.
> 
> Yes, please revert this, and then make sure _my_ patch gets into -pre6?
> 
> -- 
> Debian     - http://www.debian.org/
> Linux 1394 - http://www.linux1394.org/
> Subversion - http://subversion.tigris.org/
> Deqo       - http://www.deqo.com/
> 

Re: Linux 2.4.21-pre5

[Ben Collins]

On Thu, Feb 27, 2003 at 01:22:25PM -0600, ajoshi@kernel.crashing.org wrote:
> 
> 
> Just for the record, the patch I sent to Marcelo did _NOT_ include any
> 1394 hunks in it.  Marcelo must have accidently mixed mine up with some
> old 1394 tree.

That's sort of what I figured.

Marcelo, do you want me to send you a complete patch against -pre5
reverting all this junk, or do you want to revert it and then accept a
new patch?

-- 
Debian     - http://www.debian.org/
Linux 1394 - http://www.linux1394.org/
Subversion - http://subversion.tigris.org/
Deqo       - http://www.deqo.com/

[patch] 2.4.21-pre5: fix Auerswald compile

[Adrian Bunk]

On Thu, Feb 27, 2003 at 03:14:44AM -0300, Marcelo Tosatti wrote:
>...
> Summary of changes from v2.4.21-pre4 to v2.4.21-pre5
> ============================================
>...
> Wolfgang Muees <wolfgang@iksw-muees.de>:
>   o USB: updated Auerswald driver

I got the following error at the final linking:

<--  snip  -->

...
        --end-group \
        -o vmlinux
...
drivers/usb/usbdrv.o(.text+0x65061): In function `auerchar_open':
: undefined reference to `auerdev_table_mutex'
drivers/usb/usbdrv.o(.text+0x6506a): In function `auerchar_open':
: undefined reference to `auerdev_table_mutex'
drivers/usb/usbdrv.o(.text+0x65087): In function `auerchar_open':
: undefined reference to `auerdev_table'
drivers/usb/usbdrv.o(.text+0x65094): In function `auerchar_open':
: undefined reference to `auerdev_table_mutex'
drivers/usb/usbdrv.o(.text+0x650c1): In function `auerchar_open':
: undefined reference to `auerdev_table_mutex'
drivers/usb/usbdrv.o(.text+0x650da): In function `auerchar_open':
: undefined reference to `auerdev_table_mutex'
make: *** [vmlinux] Error 1

<--  snip  -->

auerdev_table and auerdev_table_mutex are static in auermain.c but used
from auerchar.c. The following patch makes them non-static:

--- linux-2.4.21-pre5-full/drivers/usb/auermain.c.old	2003-02-28 00:15:45.000000000 +0100
+++ linux-2.4.21-pre5-full/drivers/usb/auermain.c	2003-02-28 00:20:49.000000000 +0100
@@ -66,10 +66,10 @@
 extern devfs_handle_t usb_devfs_handle;
 
 /* array of pointers to our devices that are currently connected */
-static struct auerswald *auerdev_table[AUER_MAX_DEVICES];
+struct auerswald *auerdev_table[AUER_MAX_DEVICES];
 
 /* lock to protect the auerdev_table structure */
-static struct semaphore auerdev_table_mutex;
+struct semaphore auerdev_table_mutex;
 
 /*-------------------------------------------------------------------*/
 /* Forwards */


cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: Linux 2.4.21-pre5

[Ben Lau]

Hi,

    Is there any simple method to trace which patch removed
the code?

I used the following command to check it..

grep define\ *hpsb_queue_struct\ *tq_struct patch-2.4.21-pre4-ac*

However, i have downloaded the patch from alan cox only...
it dont contain the code. any site have contained all the
patches that is -pre5 based on?

Benjamin Herrenschmidt wrote:
> On Thu, 2003-02-27 at 11:19, Simon Oosthoek wrote:
> 
> 
>>here's a patch that should work to fix this.
> 
> 
> I think the proper fix for now is to bring back all of
> ieee1394 from pre4 to pre5, it was an incorrect merge
> that reverted part of it.
> 
> Ben.
> 
> -
> 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/

Re: [patch] 2.4.21-pre5: fix Auerswald compile

[Wolfgang Mües]

Hello Adrian,

On Friday 28 February 2003 00:26, Adrian Bunk wrote:

> auerdev_table and auerdev_table_mutex are static in auermain.c but used
> from auerchar.c. The following patch makes them non-static:

You are right. This is a leftover from the old days where auermain and
auerchar function were both in auermain.c.

I wonder why I never get this error? Are you using a newer toolchain?
My gcc is version 2.95.3.

Greg, would you apply this patch?

best regards
Wolfgang Mües

>
> --- linux-2.4.21-pre5-full/drivers/usb/auermain.c.old	2003-02-28
> 00:15:45.000000000 +0100 +++
> linux-2.4.21-pre5-full/drivers/usb/auermain.c	2003-02-28 00:20:49.000000000
> +0100 @@ -66,10 +66,10 @@
>  extern devfs_handle_t usb_devfs_handle;
>
>  /* array of pointers to our devices that are currently connected */
> -static struct auerswald *auerdev_table[AUER_MAX_DEVICES];
> +struct auerswald *auerdev_table[AUER_MAX_DEVICES];
>
>  /* lock to protect the auerdev_table structure */
> -static struct semaphore auerdev_table_mutex;
> +struct semaphore auerdev_table_mutex;
>
>  /*-------------------------------------------------------------------*/
>  /* Forwards */
>
>
> cu
> Adrian

Re: [patch] 2.4.21-pre5: fix Auerswald compile

[Greg KH]

On Fri, Feb 28, 2003 at 12:26:12AM +0100, Adrian Bunk wrote:
> 
> auerdev_table and auerdev_table_mutex are static in auermain.c but used
> from auerchar.c. The following patch makes them non-static:

Applied to my trees, thanks.

greg k-h

Re: Linux 2.4.21-pre5

[J.A. Magallon]

On 02.27 07:14, Marcelo Tosatti wrote:
> 
> So here goes -pre5.
> 

define_mbool ?

werewolf:/usr/src/linux# make xconfig
rm -f include/asm
( cd include ; ln -sf asm-i386 asm)
make -C scripts kconfig.tk
make[1]: Entering directory `/usr/src/linux-2.4.21-pre5-jam1/scripts'
cat header.tk >> ./kconfig.tk
./tkparse < ../arch/i386/config.in >> kconfig.tk
drivers/net/Config.in: 188: unknown command
make[1]: *** [kconfig.tk] Error 1
make[1]: Leaving directory `/usr/src/linux-2.4.21-pre5-jam1/scripts'
make: *** [xconfig] Error 2

Line is:

      if [ "$CONFIG_VISWS" = "y" ]; then
         define_mbool CONFIG_EEPRO100_PIO y
      else

Changed to define_bool, and it works.

-- 
J.A. Magallon <jamagallon@able.es>      \                 Software is like sex:
werewolf.able.es                         \           It's better when it's free
Mandrake Linux release 9.1 (Cooker) for i586
Linux 2.4.21-pre4-jam1 (gcc 3.2.2 (Mandrake Linux 9.1 3.2.2-1mdk))

Re: Linux 2.4.21-pre5

[Marcelo Tosatti]

On Thu, 27 Feb 2003, Ben Collins wrote:

> On Thu, Feb 27, 2003 at 01:22:25PM -0600, ajoshi@kernel.crashing.org wrote:
> >
> >
> > Just for the record, the patch I sent to Marcelo did _NOT_ include any
> > 1394 hunks in it.  Marcelo must have accidently mixed mine up with some
> > old 1394 tree.
>
> That's sort of what I figured.
>
> Marcelo, do you want me to send you a complete patch against -pre5
> reverting all this junk, or do you want to revert it and then accept a
> new patch?

-BK is fixed now by your latest patches. Thanks.