2002-08-12 23:31:36

by Marcelo Tosatti

[permalink] [raw]
Subject: Linux 2.4.20-pre2



Hi,

Here goes -pre2.


Summary of changes from v2.4.20-pre1 to v2.4.20-pre2
============================================

<[email protected]>:
o Add pci id to tulip net driver

<[email protected]>:
o VLAN: Fix gcc-3.1 warnings

<[email protected]>:
o Add pci id to tulip net driver

<[email protected]>:
o 1/18 s390 architecture core updates

<[email protected]>:
o Pass 'unsigned long' not 'long' as argument to save_flags, in several old net drivers

<[email protected]>:
o drivers/net/bonding.c: Handle non-ETHTOOL devices more correctly

<[email protected]>:
o x.patch

<[email protected]>:
o include/net/tcp.h: Kill redundant declaration

<[email protected]>:
o Fix spelling in natsemi net driver

<[email protected]>:
o Fix MODULE_DESCRIPTION of olympic and pss drivers

<[email protected]>:
o drivers/sbus/char/Config.in: Avoid using ARCH
o drivers/sbus/char/Config.in: Avoid using ARCH
o Update net driver Config.in texts to indicate their dependency
o Mark drivers/net/Config.in entries that depends on CONFIG_OBSOLETE with "(OBSOLETE)" text.

<[email protected]>:
o Many files
o reiserfs_fs.h, namei.c, bitmap.c
o Configure.help, Config.in
o tail_conversion.c, namei.c
o inode.c
o reiserfs_fs.h, namei.c, journal.c
o Many files

<[email protected]>:
o dump_stack()
o backport yield() and conditional reschedule changes from
o small VM updates from -aa (1/5)
o small VM updates from -aa (2/5)
o small VM updates from -aa (4/5)
o small VM updates from -aa (5/5)
o use slab for kiobufs and allocate it's bhs dynamically
o add missing prototype to arch/i386/kernel/setup.c
o i386 stackoverflow checker
o Re: [PATCH] small VM updates from -aa (3/5)
o conditionally re-enable per-disk stats, convert to seq_file
o for_each_pgdat/for_each_zone
o implement kmem_cache_size()
o advanced f00f bug detection & workaround
o cure the leftovers of the CONFIG_ISA / X86_64 patch
o use for_each_pgdat in try_to_free_pages_nozone
o __set_64bit needs lock prefix
o minor VM changes from -aa (2/3)
o minor VM changes from -aa (3/3)

<[email protected]>:
o net/ipv4/ipconfig.c: [TRIVIAL] fix a typo
o include/linux/netdevice.h: [TRIVIAL] Use ___cacheline_aligned
o include/net/tcp.h: [TRIVIAL] Use ___cacheline_aligned

<[email protected]>:
o Add pci id to orinoco wireless net driver

<[email protected]>:
o Fix typos in 8139cp net driver RxProto{TCP,UDP} constants

<[email protected]>:
o net/ipv4/tcp.c: Fix comment typo
o arch/sparc64/kernel/time.c: Fix comment typo
o net/sched/sch_ingress.c: [TRIVIAL] Fix debugging printk typo

<[email protected]>:
o e1000 net driver cleanups

<[email protected]>:
o Get hamachi net driver RX working again

<[email protected]>:
o Fix epic100 net driver

<[email protected]>:
o Fix use of pointer after kfree(), in au1000_eth net driver

<[email protected]>:
o Update old eepro net driver

<[email protected]>:
o Yet another new tulip pci id

<[email protected]>:
o Add Conexant LANfinity support to tulip net driver

<[email protected]>:
o [DECNET]: Fix route device refcounting
o DECnet bug fix

<[email protected]>:
o Natsemi ethernet driver fixes
o Natsemi ethernet fixes
o Lindent drivers/char/nvram.c in anticipation of more patching
o clean up 'return (x);' style stuff into 'return x' in nvram.c

<[email protected]>:
o Fix a typo, so people can have clean logs. Sheesh :)

<[email protected]>:
o Remove inappropriate use of set_bit in dl2k gige net driver

<[email protected]>:
o APM fix for 2.4.20pre1

<[email protected]>:
o Add two pci ids to 8139too net driver

Alan Cox <[email protected]>:
o fix compile problem with multiquad
o add befs to the list of fs docs
o parisc doc stuff
o janitorials on copy_user
o fix cyclades warning
o fix up the formatting in Config.in add HP stuff
o update Makefile for drivers/char for HP bits
o add author tags to tosh driver
o fix warnings
o pcxx janitorials
o fix wrong bracketing
o fix formatting the x86_64 people borked
o ; yet another missing sign check
o update drivers/Makefile for hp
o remove unneeded parisc special case
o undo formatting mess from x86_64, correct list a bit
o fixup depca, fix a missing sign check
o fixup e100,e1000
o switch ne2100 to static not extern inline
o add new ids to hp driver, plus alignment stuff
o fix compile warning
o arch/parisc
o remove bogus x86_64 junk
o update lasi driver
o fix warnings
o remove bogus x86_64isms, fix warnings
o fix warning
o remove unused bits
o update HP lasi scsi driver
o fix pas16 command option parsing
o typo
o HP fixes for sym2 - approved by maintainer
o HP Zalon scsi driver
o fix warning in bin2hex
o missing semaphore drop on error path
o update framebuffer fbcon logic for HP
o fbmem for hp
o video Makefile for HP
o fix warning in pm3
o befs uses NLS
o ATM warning fixes

Andrew Morton <[email protected]>:
o Fix set_page_dirty race

Ben Collins <[email protected]>:
o Ignore Subversion RCS files

Brad Hards <[email protected]>:
o Remove unneeded #includes from 3c359, sbni, and sdla_ft1 net drivers

Christoph Hellwig <[email protected]>:
o Clean up eepro100 net driver update from David M-T

David S. Miller <[email protected]>:
o IPv4: Fix MSG_DONTWAIT behavior on output fragmentation
o VLAN dev: Fix hard_start_xmit return values
o [NETFILTER]: Add some new iptables modules. (from [email protected])
o include/linux/netdevice.h: Define HAVE_NETDEV_POLL

David Woodhouse <[email protected]>:
o linux-2.4.19-pre10-shared-zlib
o Trivial JFFS2 oops fix

Eric Sandeen <[email protected]>:
o Fix printk, remove dead prototype in rcpci45 net driver
o Fix warning in ppp_generic
o Remove unused var and unused func from ali-ircc IrDA driver

Gerd Knorr <[email protected]>:
o btaudio driver update
o gemtek radio driver fix
o video4linux i2c audio modules update
o bttv documentation update
o video4linux tuner update
o bttv driver update

Greg Kroah-Hartman <[email protected]>:
o USB pl2303 driver
o USB: usb.h cleanups, typedef removed for iso packets, and whitespace changes
o USB: removed urb_t typedef
o USB: removed the devrequest typedef
o USB: added TI edgeport usb to serial driver
o USB: added new host controller driver for HC_SL811 devices
o USB: added aiptek driver
o USB: added tiglusb driver
o USB: added usb-midi driver
o USB: added new drivers to the build
o USB: bluetooth fixes for usb typedef cleanups
o ACPI PCI Hotplug driver update
o IBM PCI Hotplug driver update

Harald Welte <[email protected]>:
o net/ipv4/netfilter/ip_nat_core.c: Fix memory leak on unload
o [IP_{CONNTRACK,NAT}_{IRC,FTP}] Handle helper registration failure properly
o [NETFILTER]: Backport newnat infrastructure to 2.4.x
o [NETFILTER]: REJECT packet should not inherit nfmark of original packet
o [NETFILTER]: Two functions which should be static in ipt_ah.c are not
o [NETFILTER]: Allow owner match module match process names
o include/linux/kernel.h: Define HIPQUAD correctly on little-endian

Jean Tourrilhes <[email protected]>:
o New Wireless Extension API - part2
o wavelan_cs update (v23)
o Fix dev->trans_start in wavelan
o ir240_trivial_fixes-3.diff
o ir240_sys_max_tx-2.diff
o ir240_irnet_disc_ind_again.diff
o ir240_discovery_fixes.diff
o IrDA NSC driver add new chip
o IrDA irtty bugfixes
o IrDA: Make discovery expiry work properly for non default period
o Add new IrDA dongle drivers

Jeff Garzik <[email protected]>:
o Add e100 and e1000 net driver docs
o Merge spelling fixes from Rusty's "trivial" patch collection
o Update 3c509 net driver to move MODULE_LICENSE outside all ifdefs
o Include linux/bitops in e100 net driver, it uses ffs() (Noticed by DaveM)
o Update 8139too net driver to make new rx-reset method the default
o Fix mistake in 8139too net driver Config.in entry
o Proper support for RTL8139 rev K in 8139too net driver
o Release 8139too net driver version 0.9.26
o Fix TX checksumming in 8139cp net driver (the feature is still ifdef'd out by default, however)
o Add 64-bit DMA support to 8139cp net driver

Jens Axboe <[email protected]>:
o Add block IO directly from highmem support
o ext3 __FUNCTION__ usage in 2.4
o sr scatter oops

J?rgen E. Fischer <[email protected]>:
o Fix AHA152X problem

Marcelo Tosatti <[email protected]>:
o Import patch revert-x8664-config-change.patch
o Remove duplicate MAINTAINERS entry
o Import patch revert-wrong-kconfig-syncbanner
o Changed EXTRAVERSION to pre2
o Add missing bracket missing in hch's __free_pages_ok() patch

Neil Brown <[email protected]>:
o knfsd - 1/19 - Tidy up code in nfsd_lookup
o knfsd - 2/19 - Tidyup init/exit of nfsd module
o knfsd - 3/19 - Support fsid=<number> export option to be
o 1 of 11 - Claim semaphore for ->lookup call
o 2 of 11 - Change sunrpc to use more list.h lists
o 3 of 11 - Get sunrpc to use module_init properly
o 4 of 11 - Tidy up SMP locking for svc_sock
o 5 of 11 - Detect and close tcp connections that we cannot
o 6 of 11 - Close idle rpc/tcp sockets
o 7 of 11 - Cope with short read when reading length of
o 8 of 11 - Make sure there is alway adequate sndbuf space
o 9 of 11 - Limit number of active tcp connections to an RPC
o 10 of 11 - Allow SO_REUSEADDR for NFS sockets

Olaf Hering <[email protected]>:
o drivers/macintosh only on ppc32

Paul Mackerras <[email protected]>:
o PPC32: Start moving files to their new locations
o PPC32: adjustments to correspond with the new locations of files
o PPC32: more include and Makefile fixes

Pavel Machek <[email protected]>:
o Remove unnecessary prototypes in eepro100 net driver

[email protected] <[email protected]>:
o Probe port 0x240 too, in eexpress net driver

Petr Vandrovec <[email protected]>:
o ncpfs reports ESTALE in 2.4.19

Rob Radez <[email protected]>:
o SPARC: Get sun4c working again
o Documentation/Configure.help: CONFIG_SPARCAUDIO_DBRI applies to LX systems too

Russell King <[email protected]>:
o if_ether.h: Use packed attribute where necessary
o ip6_tables.c: Uncomment debugging printf

Rusty Russell <[email protected]>:
o warning cleanup for drivers/video/sstfb.c
o TRIVIAL EPERM -> EACCESS
o documentation typos in
o header cleanup - arch_ppc64_kernel_htab.c
o cure gcc3 warning in arch_i386_kernel_apm.c
o Fix type of compute_loop_size()
o Re: header cleanup - drivers_net_wan_sdla_ft1.c
o header cleanup - drivers_net_wan_sbni.c
o make awe_wave use struct isapnp_device_id
o explicit signed char cast in i386 spin_is_locked
o 40) request_region check, 31-40
o 2.5: kconfig synchronise banners (6_16)
o Typo in linux_arch_i386_kernel_apic.c
o cure compiler warnings in arch_i386_kernel_setup.c
o [PATCH][trivial] silence disable_ide_dma warning in
o header cleanup - drivers_bluetooth_hci_ldisc.c
o fix "inline" placement in serial.c
o warning cleanup for drivers_media_video_zr36067.c
o silence APIC errors a bit
o warning cleanup for drivers_atm_atmtcp.c
o warning cleanup for drivers_message_i2o_i2o_pci.c
o Typo in arch_mips_dec_wbflush.c
o Typo in linux_fs_partitions_msdos.c
o Trivial Patch to sched.h for
o Maxium inline patch is 40 kilobytes, not kilobits
o 2.5: kconfig use of $ARCH (1_12)
o Fix conflicting md_cpu_has_mmx definitions
o Fix typo in mm_slab.c
o Typo in linux_arch_i386_kernel_smp.c
o Typo in linux_include_asm-cris_pgtable.h
o 2.5: kconfig spurious bool default value (1_3)
o Typo in linux_arch_mips64_kernel_irq.c
o redundant declarations (#10_15)
o Typo in linux_kernel_pm.c
o 2.5: kconfig synchronise banners 3
o Typos in linux_arch_i386_kernel_io_apic.c
o 2.5: kconfig missing EXPERIMENTAL (1_14)
o 2.5: kconfig use of $ARCH (2_12)
o 2.5: kconfig synchronise banners (4_16)
o 25) request_region check, 21-30
o Typo in linux_arch_i386_kernel_mpparse.c
o redundant declarations (#11_15)
o 2.5: kconfig missing EXPERIMENTAL (3_14)
o 2.5: kconfig use of $ARCH (6_12)
o Typo in linux_include_asm-m68k_mac_via.h
o Typo in linux_net_sunrpc_xprt.c
o 2.5: kconfig synchronise banners 2 (2_3)
o 2.5: kconfig missing EXPERIMENTAL 2 (5_7)
o Typo in linux_include_linux_raid_md_k.h
o [patch, 2.4] cs4232.c doesn't kfree on error path
o 2.5: kconfig synchronise banners (8_16)
o Typos in linux_drivers_mtd_devices_blkmtd.c
o Typos in Documentation_video4linux_meye.txt (2.4.19-rc1)
o ipc_ statics
o correct inaccurate comment regarding zone_table's usage
o 2.5: kconfig synchronise banners (16_16)
o Typo in linux_include_asm-sh_pgtable-2level.h
o Typo in linux_include_linux_brlock.h
o drm_mga bitops -> long fix
o Typo in linux_arch_mips64_math-emu_ieee754.c
o Typo in linux_net_unix_af_unix.c
o 2.5: kconfig use of $ARCH (7_12)
o Typo in linux_kernel_fork.c
o 2.5: kconfig missing EXPERIMENTAL 2 (6_7)
o Typo in linux_arch_ppc64_kernel_pSeries_pci.c
o redundant declarations (#8_15)
o 2.5: kconfig synchronise banners (11_16)
o 2.5: kconfig use of $ARCH (9_12)
o Typo in linux_net_sched_sch_ingress.c
o 2.5: kconfig use of $ARCH (8_12)
o 2.5: kconfig missing EXPERIMENTAL (9_14)
o 2.5: kconfig use of $ARCH (12_12)
o Typo in linux_net_ipv4_tcp.c
o Typos in linux_mm_highmem.c
o ia64 incorrect field name in message
o 2.5: kconfig missing EXPERIMENTAL (6_14)
o Typo in linux_arch_i386_kernel_setup.c
o 2.5: kconfig synchronise banners (14_16)
o Typo in linux_drivers_media_video_pms.c
o 2.5: kconfig EXPERIMENTAL misformed
o 2.5: kconfig use of $ARCH (4_12)
o Fix typo in net_ipv4_ipconfig.c
o 2.5: kconfig missing EXPERIMENTAL (12_14)
o Typo in linux_arch_ia64_sn_fakeprom_README
o Typo in linux_fs_pipe.c
o Typo in linux_drivers_isdn_isdn_ppp.c
o Typo in linux_drivers_sound_cs4232.c
o Typo in linux_drivers_ide_ide-geometry.c
o kerneldoc: In kernel-hacking describe designated
o 2.5: kconfig sychronise banners (2_16)
o Use proper ____cacheline_aligned define in
o 2.5: kconfig use of $ARCH (11_12)
o 2.5: kconfig use of $ARCH (10_12)
o 2.5: kconfig synchronise banners (5_16)
o 2.5: kconfig missing EXPERIMENTAL 2 (2_7)
o 2.5: kconfig synchronise banners (1_16)
o Typo in linux_arch_sparc64_kernel_time.c
o 2.5: kconfig synchronise banners (3_16)
o 2.5: kconfig missing EXPERIMENTAL 2 (1_7)

Tom Rini <[email protected]>:
o PPC32: Remove arch/ppc/kernel/local_irq.h as well as move some headers to include/asm-ppc
o PPC32: Change pcibios_assign_all_busses into a marco
o PPC32: Don't report the TAU feature on any of the MPC745x CPUs, as it's not supported
o PPC32: Move arch/ppc/kernel/ppc_asm.h to include/asm-ppc/ppc_asm.h
o PPC32: Have arch/ppc/kernel/ptrace.c point to 'COPYING' instead of 'README.legal'




2002-08-13 02:13:26

by Mel Gorman

[permalink] [raw]
Subject: Re: Linux 2.4.20-pre2

On Mon, 12 Aug 2002, Marcelo Tosatti wrote:

> Here goes -pre2.
>

I am sending you five patches, all documentation and comments related.
I've posted them to the lkml list already so I won't repost. For
interested parties, they are at http://www.csn.ul.ie/~mel/projects/vm/ and
all against 2.4.20pre2

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

2002-08-13 05:48:25

by Ivan G.

[permalink] [raw]
Subject: Re: Linux 2.4.20-pre2

make[3]: Entering directory `/usr/src/linux-2.4.20-pre2/fs/partitions'
gcc -D__KERNEL__ -I/usr/src/linux-2.4.20-pre2/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2
-fno-strict-aliasing -fno-common -fomit-frame-pointer -pipe -
mpreferred-stack-boundary=2 -march=athlon -nostdinc -I
/usr/lib/gcc-lib/athlon-redhat-linux/3.2/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 `/usr/src/linux-2.4.20-pre2/fs/partitions'
make[2]: *** [first_rule] Error 2
make[2]: Leaving directory `/usr/src/linux-2.4.20-pre2/fs/partitions'
make[1]: *** [_subdir_partitions] Error 2
make[1]: Leaving directory `/usr/src/linux-2.4.20-pre2/fs'
make: *** [_dir_fs] Error 2

2002-08-13 08:24:06

by Adrian Bunk

[permalink] [raw]
Subject: Re: Linux 2.4.20-pre2

On Mon, 12 Aug 2002, Ivan Gyurdiev wrote:

> make[3]: Entering directory `/usr/src/linux-2.4.20-pre2/fs/partitions'
> gcc -D__KERNEL__ -I/usr/src/linux-2.4.20-pre2/include -Wall
> -Wstrict-prototypes -Wno-trigraphs -O2
> -fno-strict-aliasing -fno-common -fomit-frame-pointer -pipe -
> mpreferred-stack-boundary=2 -march=athlon -nostdinc -I
> /usr/lib/gcc-lib/athlon-redhat-linux/3.2/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'
>...


The following patch made by Christoph Hellwig fixes it:


--- 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;


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


2002-08-13 09:15:52

by Adrian Bunk

[permalink] [raw]
Subject: Re: Linux 2.4.20-pre2

On Mon, 12 Aug 2002, Marcelo Tosatti wrote:

>...
> Greg Kroah-Hartman <[email protected]>:
>...
> o USB: removed the devrequest typedef
>...

This broke the compilation of drivers/isdn/hisax/st5481_usb.c:

<-- snip -->

...
gcc -D__KERNEL__ -I/home/bunk/linux/kernel-2.4/linux-2.4.19-full/include
-Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
-pipe -mpreferred-stack-boundary=2 -march=k6 -DHISAX_MAX_CARDS=8 -nostdinc -I
/usr/lib/gcc-lib/i386-linux/2.95.4/include -DKBUILD_BASENAME=st5481_usb -c -o
st5481_usb.o st5481_usb.c
st5481_usb.c: In function `usb_next_ctrl_msg':
st5481_usb.c:43: structure has no member named `request'
st5481_usb.c:44: structure has no member named `value'
st5481_usb.c:45: structure has no member named `index'
st5481_usb.c: In function `usb_ctrl_msg':
st5481_usb.c:72: structure has no member named `requesttype'
st5481_usb.c:73: structure has no member named `request'
st5481_usb.c:74: structure has no member named `value'
st5481_usb.c:75: structure has no member named `index'
st5481_usb.c:76: structure has no member named `length'
st5481_usb.c: In function `usb_ctrl_complete':
st5481_usb.c:143: structure has no member named `request'
st5481_usb.c:147: structure has no member named `index'
st5481_usb.c:148: structure has no member named `index'
st5481_usb.c:152: structure has no member named `index'
st5481_usb.c:152: structure has no member named `index'
st5481_usb.c:153: structure has no member named `index'
st5481_usb.c:154: structure has no member named `index'
make[4]: *** [st5481_usb.o] Error 1
make[4]: Leaving directory
`/home/bunk/linux/kernel-2.4/linux-2.4.19-full/drivers/isdn/hisax'

<-- 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


2002-08-13 10:48:30

by Alan

[permalink] [raw]
Subject: Re: Linux 2.4.20-pre2

On Tue, 2002-08-13 at 10:19, Adrian Bunk wrote:
> On Mon, 12 Aug 2002, Marcelo Tosatti wrote:
>
> >...
> > Greg Kroah-Hartman <[email protected]>:
> >...
> > o USB: removed the devrequest typedef
> >...
>
> This broke the compilation of drivers/isdn/hisax/st5481_usb.c:

Fixes can be found for this and a couple of other usb breakages in the
2.4.20-pre1-ac tree. I'll send them on to Marcelo today

2002-08-13 15:00:28

by Bhavesh P. Davda

[permalink] [raw]
Subject: Re: Linux 2.4.20-pre2


For the Nth time, here again is the patch for fixing the scheduler for
correct SCHED_FIFO and SCHED_RR behaviour.

I waited to see it in 2.4.20-pre1, not there, 2.4.20-pre2, not there...

Please apply it in 2.4.20-pre3

Thanks

- Bhavesh
--
Bhavesh P. Davda
Avaya Inc.
[email protected]


diff -aur linux-2.4.19/kernel/sched.c linux-2.4.19-sched/kernel/sched.c
--- linux-2.4.19/kernel/sched.c Tue Aug 6 11:37:49 2002
+++ linux-2.4.19-sched/kernel/sched.c Tue Aug 6 13:42:20 2002
@@ -318,13 +318,17 @@
/*
* Careful!
*
- * This has to add the process to the _beginning_ of the
- * run-queue, not the end. See the comment about "This is
- * subtle" in the scheduler proper..
+ * This has to add the process to the _end_ of the
+ * run-queue, not the beginning. The goodness value will
+ * determine whether this process will run next. This is
+ * important to get SCHED_FIFO and SCHED_RR right, where
+ * a process that is either pre-empted or its time slice
+ * has expired, should be moved to the tail of the run
+ * queue for its priority - Bhavesh Davda
*/
static inline void add_to_runqueue(struct task_struct * p)
{
- list_add(&p->run_list, &runqueue_head);
+ list_add_tail(&p->run_list, &runqueue_head);
nr_running++;
}

@@ -334,12 +338,6 @@
list_add_tail(&p->run_list, &runqueue_head);
}

-static inline void move_first_runqueue(struct task_struct * p)
-{
- list_del(&p->run_list);
- list_add(&p->run_list, &runqueue_head);
-}
-
/*
* Wake up a process. Put it on the run-queue if it's not
* already there. The "current" process is always on the
@@ -955,8 +953,6 @@
retval = 0;
p->policy = policy;
p->rt_priority = lp.sched_priority;
- if (task_on_runqueue(p))
- move_first_runqueue(p);

current->need_resched = 1;

diff -aur linux-2.4.19/kernel/timer.c linux-2.4.19-sched/kernel/timer.c
--- linux-2.4.19/kernel/timer.c Tue Aug 6 11:37:49 2002
+++ linux-2.4.19-sched/kernel/timer.c Tue Aug 6 13:42:20 2002
@@ -601,7 +601,14 @@
if (p->pid) {
if (--p->counter <= 0) {
p->counter = 0;
- p->need_resched = 1;
+ /*
+ * SCHED_FIFO is priority preemption, so this is
+ * not the place to decide whether to reschedule a
+ * SCHED_FIFO task or not - Bhavesh Davda
+ */
+ if (p->policy != SCHED_FIFO) {
+ p->need_resched = 1;
+ }
}
if (p->nice > 0)
kstat.per_cpu_nice[cpu] += user_tick;

2002-09-07 20:06:31

by Adrian Bunk

[permalink] [raw]
Subject: Re: Linux 2.4.20-pre2

On Mon, 12 Aug 2002, Marcelo Tosatti wrote:

>...
> Harald Welte <[email protected]>:
>...
> o [NETFILTER]: Backport newnat infrastructure to 2.4.x
>...

Hi Harald,

this patch adds a #include <linux/netfilter_ipv4/ip_conntrack_helper.h> to
ip_conntrack.h and since ip_conntrack_helper.h already has a
#include <linux/netfilter_ipv4/ip_conntrack.h> this results in a circular
dependency. Could you please fix this?

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