]> git.hungrycats.org Git - linux/log
linux
23 years ago[PATCH] sys_nanosleep() fix
Andrew Morton [Thu, 20 Mar 2003 13:47:19 +0000 (05:47 -0800)]
[PATCH] sys_nanosleep() fix

The current nanosleep implementation has a signedness problem which can cause
it to sleep for 0x7ffffffe jiffies if a clock interrupt happens at the wrong
time.

The patch fixes that up, and also fixes an wrapping-unsafe 64-bit time
comparison.  Also uninline tstojiffie(), which has three call sites.

23 years agoMerge bk://kernel.bkbits.net/davem/net-2.5
Linus Torvalds [Thu, 20 Mar 2003 13:45:25 +0000 (05:45 -0800)]
Merge bk://kernel.bkbits.net/davem/net-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

23 years agoMerge bk://kernel.bkbits.net/davem/sparc-2.5
Linus Torvalds [Thu, 20 Mar 2003 13:44:34 +0000 (05:44 -0800)]
Merge bk://kernel.bkbits.net/davem/sparc-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

23 years agoFix a rather theoretical race if an NMI happens when a debug fault
Linus Torvalds [Thu, 20 Mar 2003 13:43:59 +0000 (05:43 -0800)]
Fix a rather theoretical race if an NMI happens when a debug fault
happens exactly on the sysenter entry path before the kernel stacks
have been switched to the proper ones.

23 years ago[netdrvr tg3] fix memleak in DMA test
Jeff Garzik [Thu, 20 Mar 2003 19:01:37 +0000 (14:01 -0500)]
[netdrvr tg3] fix memleak in DMA test

Also, bump version to 1.5.

Leak fix contributed by Don Fry @ IBM

23 years agoMerge redhat.com:/garz/repo/e1000-2.5
Jeff Garzik [Thu, 20 Mar 2003 18:49:05 +0000 (13:49 -0500)]
Merge redhat.com:/garz/repo/e1000-2.5
into redhat.com:/garz/repo/net-drivers-2.5

23 years ago[E1000] NAPI re-insertion w/ changes
Jeb J. Cramer [Thu, 20 Mar 2003 18:46:10 +0000 (13:46 -0500)]
[E1000] NAPI re-insertion w/ changes

* Previous patch wiped NAPI support, adding it back here.  But,
  with a twist: this one doesn't disable/enable interrupts each
  time we enter/leave polling.  (It's EXPERIMENTAL).

23 years ago[E1000] whitespace fix from previous patches
Jeb J. Cramer [Thu, 20 Mar 2003 18:45:43 +0000 (13:45 -0500)]
[E1000] whitespace fix from previous patches

* Corrected indentation from previous patches

23 years ago[E1000] Controller wake-up thru ASF fix
Jeb J. Cramer [Thu, 20 Mar 2003 18:45:19 +0000 (13:45 -0500)]
[E1000] Controller wake-up thru ASF fix

* Fixed controller wake-up through ASF

23 years ago[E1000] Added Interrupt Throttle Rate tuning support
Jeb J. Cramer [Thu, 20 Mar 2003 18:44:54 +0000 (13:44 -0500)]
[E1000] Added Interrupt Throttle Rate tuning support

* Added Interrupt Throttle Rate tuning support

23 years ago[E1000] Added Tx FIFO flush routine
Jeb J. Cramer [Thu, 20 Mar 2003 18:44:29 +0000 (13:44 -0500)]
[E1000] Added Tx FIFO flush routine

* Added method to flush Tx FIFO after link disconnect; the hardware
  hangs on to Tx skb's that were in flight prior to link loss

23 years ago[E1000] Whitespace changes
Jeb J. Cramer [Thu, 20 Mar 2003 18:44:04 +0000 (13:44 -0500)]
[E1000] Whitespace changes

* Miscellaneous whitespace changes

23 years ago[E1000] Compaq to HP branding change
Jeb J. Cramer [Thu, 20 Mar 2003 18:43:38 +0000 (13:43 -0500)]
[E1000] Compaq to HP branding change

* Changed "Compaq" branding to "HP"

23 years ago[E1000] Read/Write register macro optimizations
Jeb J. Cramer [Thu, 20 Mar 2003 18:43:13 +0000 (13:43 -0500)]
[E1000] Read/Write register macro optimizations

* Optimized E1000_*_REG macros

23 years ago[E1000] Tx Descriptor cleanup
Jeb J. Cramer [Thu, 20 Mar 2003 18:42:48 +0000 (13:42 -0500)]
[E1000] Tx Descriptor cleanup

* Completely clean Tx descriptor to avoid potential dirty descriptor
  fetching (rare, but possible)

23 years ago[E1000] Perform single PCI read per interrupt
Jeb J. Cramer [Thu, 20 Mar 2003 18:42:24 +0000 (13:42 -0500)]
[E1000] Perform single PCI read per interrupt

* ISR cleanup; performing single PCI read

23 years ago[E1000] Modulus math removed
Jeb J. Cramer [Thu, 20 Mar 2003 18:41:59 +0000 (13:41 -0500)]
[E1000] Modulus math removed

* Removed modulus math; decreases CPU utilization, especially on PPC64
  [anton@samba.org]

23 years ago[E1000] Added MII support
Jeb J. Cramer [Thu, 20 Mar 2003 18:41:34 +0000 (13:41 -0500)]
[E1000] Added MII support

* Added MII support

23 years ago[E1000] Added 82541 & 82547 support
Jeb J. Cramer [Thu, 20 Mar 2003 18:41:10 +0000 (13:41 -0500)]
[E1000] Added 82541 & 82547 support

* Added support for 82541 and 82547 gigabit ethernet adapters

23 years ago[E1000] IRQ registration fix
Jeb J. Cramer [Thu, 20 Mar 2003 18:40:37 +0000 (13:40 -0500)]
[E1000] IRQ registration fix

* Fixed IRQ registration bug; IRQ now registered after resources are
  acquired

23 years ago[E1000] Spd/dplx abstraction; eeprom size changes
Jeb J. Cramer [Thu, 20 Mar 2003 18:40:12 +0000 (13:40 -0500)]
[E1000] Spd/dplx abstraction; eeprom size changes

* Setting speed/duplex is now it's own routine
* Update ETHTOOL_GEEPROM routine to use new eeprom size variable

23 years ago[E1000] Version, copyright, changelog and MAINTAINERS
Jeb J. Cramer [Thu, 20 Mar 2003 18:39:47 +0000 (13:39 -0500)]
[E1000] Version, copyright, changelog and MAINTAINERS

* Version, copyright, changelog and MAINTAINERS updates

23 years ago[E1000] Documentation/networking/e1000.txt updates
Jeb J. Cramer [Thu, 20 Mar 2003 18:39:20 +0000 (13:39 -0500)]
[E1000] Documentation/networking/e1000.txt updates

* Documentation/networking/e1000.txt updates

23 years ago[PATCH] Support PC-9800 subarchitecture (9/14) NIC
Osamu Tomita [Thu, 20 Mar 2003 18:08:10 +0000 (13:08 -0500)]
[PATCH] Support PC-9800 subarchitecture (9/14) NIC

This is the patch to support NEC PC-9800 subarchitecture
against 2.5.65-ac1. (9/14)

C-bus(PC98's legacy bus like ISA) network cards support.
Change IO port and IRQ assign.
Add NE2000 compatible driver for PC-9800.
  PCI netwwork card works fine without patch.

Regards,
Osamu Tomita

23 years ago[E100] ASF wakeup enabled, but only if set in EEPROM
Scott Feldman [Thu, 20 Mar 2003 17:10:09 +0000 (12:10 -0500)]
[E100] ASF wakeup enabled, but only if set in EEPROM

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Check if ASF is enabled in EEPROM, and if so, enable
  PME wakeup when suspending.

23 years ago[E100] ethtool EEPROM and GSTRING fixes
Scott Feldman [Thu, 20 Mar 2003 17:09:50 +0000 (12:09 -0500)]
[E100] ethtool EEPROM and GSTRING fixes

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Bug fix: read wrong byte in EEPROM when offset is odd number
* Bug fix: memory leak in ETHTOOL_GSTRINGS
  [Oleg Drokin <green@linuxhacker.ru]

23 years ago[E100] Validate updates to MAC address
Scott Feldman [Thu, 20 Mar 2003 17:09:31 +0000 (12:09 -0500)]
[E100] Validate updates to MAC address

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Validate updates to MAC address as valid ethernet address.

23 years ago[E100] interrupt handler free fix
Scott Feldman [Thu, 20 Mar 2003 17:09:13 +0000 (12:09 -0500)]
[E100] interrupt handler free fix

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Bug fix on e100_close when repeating hot remove/hot add
  from team.  Basically need to disable interurpts and
  unregister handler before shutting h/w down.
* Need to mask only the relevant bits in the interrupt
  status register

23 years ago[E100] Honor WOL settings in EEPROM
Scott Feldman [Thu, 20 Mar 2003 17:08:55 +0000 (12:08 -0500)]
[E100] Honor WOL settings in EEPROM

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Honor WOL settings in EEPROM: only advertise WOL magic
  packet if in EEPROM.

23 years ago[E100] ICH5 support added
Scott Feldman [Thu, 20 Mar 2003 17:08:35 +0000 (12:08 -0500)]
[E100] ICH5 support added

On Thu, 20 Mar 2003, Scott Feldman wrote:

* ICH5 support: chipset integrated LAN (8255x)
* PHY loopback diags is broken on all ICHs

23 years ago[E100] forced speed/duplex link recover
Scott Feldman [Thu, 20 Mar 2003 17:08:16 +0000 (12:08 -0500)]
[E100] forced speed/duplex link recover

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Bug fix when changing to non-autoneg, device may lose
  link with some switches, so try to recover link by
  forcing PHY.

23 years ago[E100] Banish strong branding marketing strings
Scott Feldman [Thu, 20 Mar 2003 17:07:58 +0000 (12:07 -0500)]
[E100] Banish strong branding marketing strings

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Get rid of all of the strong marketing brand strings
  and replace with simple pci_device_id table.  pci.ids
  should be the master list for device ID/strings.

23 years ago[E100] Bug fix on setting up Tx csum
Scott Feldman [Thu, 20 Mar 2003 17:07:34 +0000 (12:07 -0500)]
[E100] Bug fix on setting up Tx csum

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Bug fix on setting up Tx csum

23 years ago[E100] Clean up #include order
Scott Feldman [Thu, 20 Mar 2003 17:07:15 +0000 (12:07 -0500)]
[E100] Clean up #include order

On Thu, 20 Mar 2003, Scott Feldman wrote:

* clean up #includes

23 years ago[E100] Add support for VLAN hw offload
Scott Feldman [Thu, 20 Mar 2003 17:06:55 +0000 (12:06 -0500)]
[E100] Add support for VLAN hw offload

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Add support for VLAN hw offload

23 years ago[E100] Spelling mistakes
Scott Feldman [Thu, 20 Mar 2003 17:06:35 +0000 (12:06 -0500)]
[E100] Spelling mistakes

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Spelling mistakes

23 years ago[E100] update version, copyright year, changelog
Scott Feldman [Thu, 20 Mar 2003 17:06:16 +0000 (12:06 -0500)]
[E100] update version, copyright year, changelog

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Update version, copyright year, changelog

23 years ago[E100] Update Documentation/networking/e100.txt
Scott Feldman [Thu, 20 Mar 2003 17:05:58 +0000 (12:05 -0500)]
[E100] Update Documentation/networking/e100.txt

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Update Documentation/networking/e100.txt

23 years ago[E100] back out memleak patch cuz it messed up following
Scott Feldman [Thu, 20 Mar 2003 17:05:39 +0000 (12:05 -0500)]
[E100] back out memleak patch cuz it messed up following

On Thu, 20 Mar 2003, Scott Feldman wrote:

* Back this patch out - we'll add it later.  I was working against
  2.5.64 when this was checked into 2.5.65, so it messed up
  my patches.

23 years ago[NET]: Optimize handling of CONFIG_NET=n.
Matthew Wilcox [Wed, 19 Mar 2003 16:55:41 +0000 (08:55 -0800)]
[NET]: Optimize handling of CONFIG_NET=n.

23 years ago[NF/IPV6]: Remove all ipv6_ext_hdrs from ip6tables.
Adrian Bunk [Wed, 19 Mar 2003 16:49:27 +0000 (08:49 -0800)]
[NF/IPV6]: Remove all ipv6_ext_hdrs from ip6tables.

23 years ago[ATM]: Fix idt77252/sch_atm/pppoatm compilation.
Chas Williams [Wed, 19 Mar 2003 16:45:22 +0000 (08:45 -0800)]
[ATM]: Fix idt77252/sch_atm/pppoatm compilation.

23 years ago[PATCH] Add C99 initializers for net/ipv6/netfilter code.
Art Haas [Wed, 19 Mar 2003 16:43:52 +0000 (08:43 -0800)]
[PATCH] Add C99 initializers for net/ipv6/netfilter code.

23 years ago[PATCH] Add C99 initializers to net/ipv4/netfilter.
Art Haas [Wed, 19 Mar 2003 16:42:38 +0000 (08:42 -0800)]
[PATCH] Add C99 initializers to net/ipv4/netfilter.

23 years ago[IPSEC]: return error when no dst in ah & esp output.
James Morris [Wed, 19 Mar 2003 16:39:12 +0000 (08:39 -0800)]
[IPSEC]: return error when no dst in ah & esp output.

23 years ago[IPSEC]: fix skb leak in ah and esp.
James Morris [Wed, 19 Mar 2003 16:36:04 +0000 (08:36 -0800)]
[IPSEC]: fix skb leak in ah and esp.

23 years ago[IPSEC]: Fix null authentication/encryption.
Thomas Walpuski [Wed, 19 Mar 2003 16:32:09 +0000 (08:32 -0800)]
[IPSEC]: Fix null authentication/encryption.

23 years agoMerge http://linux-lksctp.bkbits.net/lksctp-2.5
David S. Miller [Wed, 19 Mar 2003 16:12:23 +0000 (08:12 -0800)]
Merge http://linux-lksctp.bkbits.net/lksctp-2.5
into nuts.ninka.net:/home/davem/src/BK/net-2.5

23 years ago[ebtables] bugfix in ebt_ip.c
Bart De Schuymer [Wed, 19 Mar 2003 16:00:25 +0000 (08:00 -0800)]
[ebtables] bugfix in ebt_ip.c

23 years agoMerge nuts.ninka.net:/home/davem/src/BK/network-2.5
David S. Miller [Wed, 19 Mar 2003 15:53:10 +0000 (07:53 -0800)]
Merge nuts.ninka.net:/home/davem/src/BK/network-2.5
into nuts.ninka.net:/home/davem/src/BK/net-2.5

23 years ago[SPARC64]: Fix thread_info offsets to match restart_block layout changes.
David S. Miller [Wed, 19 Mar 2003 15:44:59 +0000 (07:44 -0800)]
[SPARC64]: Fix thread_info offsets to match restart_block layout changes.

23 years ago[SPARC]: Add die_counter changes.
David S. Miller [Wed, 19 Mar 2003 14:56:57 +0000 (06:56 -0800)]
[SPARC]: Add die_counter changes.

23 years agoMerge nuts.ninka.net:/home/davem/src/BK/sparcwork-2.5
David S. Miller [Wed, 19 Mar 2003 14:51:48 +0000 (06:51 -0800)]
Merge nuts.ninka.net:/home/davem/src/BK/sparcwork-2.5
into nuts.ninka.net:/home/davem/src/BK/sparc-2.5

23 years ago[NET]: Kill NETIF_F_DYNALLOC, based upon ideas from Adam J. Richter
David S. Miller [Wed, 19 Mar 2003 13:17:20 +0000 (05:17 -0800)]
[NET]: Kill NETIF_F_DYNALLOC, based upon ideas from Adam J. Richter

23 years ago[IPSEC]: fragmentation & tcp mss calculation.
Alexey Kuznetsov [Wed, 19 Mar 2003 11:54:09 +0000 (03:54 -0800)]
[IPSEC]: fragmentation & tcp mss calculation.
1. Add local_df field to struct sk_buff to mark packets which
   are to be fragmented locally despite of their IPv6ness of IP DF flag
2. Add ext2_header_len to tcp_opt to keep memory of part of header length
   depending on route
3. Add trailer_len to struct dst_entry and xfrm_state to know how
   much of space should be reserved at tail of frame for subsequent
   transformations.
4. [BUG] icv_trun_len must be used while mss claculation, not
   icv_full_length.

23 years ago[NET]: miscellaneous fixes.
Alexey Kuznetsov [Wed, 19 Mar 2003 11:43:44 +0000 (03:43 -0800)]
[NET]: miscellaneous fixes.
1. Fix illegal dereference of potentially freed memory in xfrm_policy.c
2. Complete wildcard flow addresses to real ones in xfrm_lookup().
3. Respect optional flag when chacking for input policy.
4. Delete orphaned comments in ip.h.
5. Fix mistakedly freed route in tcp connect.

23 years ago[NET]: hard_header reservation.
Alexey Kuznetsov [Wed, 19 Mar 2003 11:34:55 +0000 (03:34 -0800)]
[NET]: hard_header reservation.
1. Fix bad reservation in xfrm_state_check_space()
2. Macroize formula for reservation, use the macro over all the places
   in IP.

23 years ago[IPV6]: Undo __constant_{n,h}to{n,h}l from anycast patch.
David S. Miller [Wed, 19 Mar 2003 11:28:07 +0000 (03:28 -0800)]
[IPV6]: Undo __constant_{n,h}to{n,h}l from anycast patch.

23 years ago[IPV6]: ndisc_recv_ns returns void.
David S. Miller [Wed, 19 Mar 2003 10:13:02 +0000 (02:13 -0800)]
[IPV6]: ndisc_recv_ns returns void.

23 years ago[IPV6]: Add anycast support.
David Stevens [Wed, 19 Mar 2003 09:26:22 +0000 (01:26 -0800)]
[IPV6]: Add anycast support.

23 years ago[SPARC64]: Clear tick_cmpr ints properly in bootup assembly.
David S. Miller [Tue, 18 Mar 2003 10:26:42 +0000 (02:26 -0800)]
[SPARC64]: Clear tick_cmpr ints properly in bootup assembly.

23 years ago[SPARC64]: Sanitize all TICK privileged bit handling in tick drivers.
David S. Miller [Tue, 18 Mar 2003 10:04:41 +0000 (02:04 -0800)]
[SPARC64]: Sanitize all TICK privileged bit handling in tick drivers.

23 years ago[FB ATY]: CONFIG_FB_ATY needs cfbcopyarea.o
David S. Miller [Tue, 18 Mar 2003 09:46:05 +0000 (01:46 -0800)]
[FB ATY]: CONFIG_FB_ATY needs cfbcopyarea.o

23 years ago[SPARC64]: Do a dummy write to STICK in hbird_init_tick.
David S. Miller [Tue, 18 Mar 2003 09:17:00 +0000 (01:17 -0800)]
[SPARC64]: Do a dummy write to STICK in hbird_init_tick.

23 years ago[SPARC64]: Ignore bit 63 of Hummingbird STICK when computing COMPARE register values.
David S. Miller [Tue, 18 Mar 2003 09:14:19 +0000 (01:14 -0800)]
[SPARC64]: Ignore bit 63 of Hummingbird STICK when computing COMPARE register values.

23 years ago[SPARC64]: Make TICK comparisons wrap-around safe by using jiffies macros.
David S. Miller [Tue, 18 Mar 2003 09:11:48 +0000 (01:11 -0800)]
[SPARC64]: Make TICK comparisons wrap-around safe by using jiffies macros.

23 years ago[SPARC64]: In __hbird_write_compare, write high then low part.
David S. Miller [Tue, 18 Mar 2003 09:09:28 +0000 (01:09 -0800)]
[SPARC64]: In __hbird_write_compare, write high then low part.

23 years ago[PATCH] Add error checking get_disk()
Andrew Morton [Tue, 18 Mar 2003 05:20:37 +0000 (21:20 -0800)]
[PATCH] Add error checking get_disk()

Patch from Bob Miller <rem@osdl.org>

The get_disk() function should check the return value from kobject_get()
before passing it to to_disk().  This patch fixes this error.

(Acked by Pat)

23 years ago[PATCH] Fix nfsd_symlink() failure path
Andrew Morton [Tue, 18 Mar 2003 05:20:29 +0000 (21:20 -0800)]
[PATCH] Fix nfsd_symlink() failure path

Patch from Andreas Gruenbacher <agruen@suse.de>

In both 2.5 and 2.4, the fs/nfsd/vfs.c:nfsd_symlink() function calls down to
notify_change().  If notify_change fails for some reason, the error code is
not converted to an nfs no-the-wire error code as is should.  The attached
patches fix that (one for 2.4, the other for 2.5).

23 years ago[PATCH] Tighten CONFIG_NUMA preconditions
Andrew Morton [Tue, 18 Mar 2003 05:20:20 +0000 (21:20 -0800)]
[PATCH] Tighten CONFIG_NUMA preconditions

Patch from Martin J. Bligh and Dave Hansen

People with ordinary PCs are accidentally turning on NUMA support, and people
with NUMA machines are finding the NUMA option mysteriously disappearing.
This patch sets the defaults to sane things for everyone, and only allows you
to turn on NUMA with both SMP and 64Gb support on (it's useful for the
distros on non-Summit boxes, but not on their UP kernels ;-)).

I've also moved it below the highmem options, as it logically depends on
them, so this makes more sense.  For those searching for NUMA support on
*real* NUMA machine, Dave has provided some guiding comments to show them
what they messed up (it's totally non-obvious).  Hopefully this will stop
people's recent unfortunate foot-wounds (I think UP machines were defaulting
to NUMA on ...  oops).

23 years ago[PATCH] /proc/sysrq-trigger: trigger sysrq functions via
Andrew Morton [Tue, 18 Mar 2003 05:20:10 +0000 (21:20 -0800)]
[PATCH] /proc/sysrq-trigger: trigger sysrq functions via

This makes sysrq facilities available to remote users.

Writing a 'C' to /proc/sysrq-trigger receives the same treatment as typing
sysrq-C on the local keyboard.

23 years ago[PATCH] pgd_index/pmd_index/pte_index commentary
Andrew Morton [Tue, 18 Mar 2003 05:20:00 +0000 (21:20 -0800)]
[PATCH] pgd_index/pmd_index/pte_index commentary

Patch from Dave Hansen <haveblue@us.ibm.com>

Adds some commentary to these newly-introduced macros.

23 years ago[PATCH] fix oprofile timer race
Andrew Morton [Tue, 18 Mar 2003 05:19:50 +0000 (21:19 -0800)]
[PATCH] fix oprofile timer race

Patch from John Levon <levon@movementarian.org>

wli got an oops from this. The callbacks call mod_timer so the timer had
better be setup by then

23 years ago[PATCH] io-apic.c: DO_ACTION cleanup
Andrew Morton [Tue, 18 Mar 2003 05:19:40 +0000 (21:19 -0800)]
[PATCH] io-apic.c: DO_ACTION cleanup

Patch from "Martin J. Bligh" <mbligh@aracnet.com>

This removes the DO_ACTION stuff.  The downside is that we add some boring
and repetive code.  The upside is that it's simple, and mere mortals can read
it without screwing their brains into a small piece of silly putty and
bouncing it off the wall.  I think that's more important than pure source
code size.

23 years ago[PATCH] OOPS instance counters
Andrew Morton [Tue, 18 Mar 2003 05:19:27 +0000 (21:19 -0800)]
[PATCH] OOPS instance counters

Patch from "Randy.Dunlap" <rddunlap@osdl.org>

Adds an oops counter to the oops messages, such as:

Oops: 0002 [#2]

So we can tell whether oops reports refer to the first oops, or to some
less-interesting followon oops.

23 years ago[PATCH] posix timers update
Andrew Morton [Tue, 18 Mar 2003 05:19:04 +0000 (21:19 -0800)]
[PATCH] posix timers update

Patch from george anzinger <george@mvista.com>

Fix the "large sleep returns EINVAL" problem, and clean a few things up.

23 years ago[PATCH] Memleak in fs/ufs/util.c
Andrew Morton [Tue, 18 Mar 2003 05:18:44 +0000 (21:18 -0800)]
[PATCH] Memleak in fs/ufs/util.c

Patch from Oleg Drokin <green@namesys.com>

    There is trivial memleak on error exit path in
    fs/ufs/util.c::_ubh_bread_()

23 years ago[PATCH] memleak in fs/nfs/inode.c::nfs_get_sb()
Andrew Morton [Tue, 18 Mar 2003 05:18:24 +0000 (21:18 -0800)]
[PATCH] memleak in fs/nfs/inode.c::nfs_get_sb()

Patch from Oleg Drokin <green@namesys.com>

   There is trivial memleak on error exit path in nfs get_sb function.

23 years ago[PATCH] slab changes for !CONFIG_MMU
Andrew Morton [Tue, 18 Mar 2003 05:18:04 +0000 (21:18 -0800)]
[PATCH] slab changes for !CONFIG_MMU

Patch from Christoph Hellwig <hch@lst.de>

It extends the maximum amount of memory which may be kmalloced on nommu
machines.  This is needed because these machines cannot perform vmalloc().

We couldn't really find a way of doing this which avoided the ifdef tangle.

23 years ago[PATCH] a few missing stubs for !CONFIG_MMU
Andrew Morton [Tue, 18 Mar 2003 05:17:46 +0000 (21:17 -0800)]
[PATCH] a few missing stubs for !CONFIG_MMU

Patch from Christoph Hellwig <hch@lst.de>

This is from the uClinux patches - there are a few more stubs needed
in nommu.c to get the mmuless plattforms working.

23 years ago[PATCH] stack reduction in drivers/char/vt_ioctl.c
Andrew Morton [Tue, 18 Mar 2003 05:17:22 +0000 (21:17 -0800)]
[PATCH] stack reduction in drivers/char/vt_ioctl.c

Patch from "Randy.Dunlap" <randy.dunlap@verizon.net>

This patch (to 2.5.64) reduces the stack usage in vt_ioctl()
from 0x334 bytes to 0xec bytes (P4, UP, gcc 2.96).

23 years ago[PATCH] file_list: less locking
Andrew Morton [Tue, 18 Mar 2003 05:16:57 +0000 (21:16 -0800)]
[PATCH] file_list: less locking

- optimise file_kill() to not take the global lock if the file is not on a
  list.

- Use optimised file_kill() in a few places rather than open-coding the
  list removal.

23 years ago[PATCH] file_table: remove the private freelist
Andrew Morton [Tue, 18 Mar 2003 05:16:31 +0000 (21:16 -0800)]
[PATCH] file_table: remove the private freelist

- Remove the private freelist.  There's no point in special-casing file
  structure allocations in this way.

- Hence the freeing of files can be moved outside file_list_lock()

- Replace euid test with capable(CAP_SYS_ADMIN).

- Tidy various other things up.

23 years ago[PATCH] file_list cleanup
Andrew Morton [Tue, 18 Mar 2003 05:15:48 +0000 (21:15 -0800)]
[PATCH] file_list cleanup

Replace the odd handling of f_list.next = NULL with list_emptiness.

23 years ago[PATCH] file->f_list locking in tty_io.c
Andrew Morton [Tue, 18 Mar 2003 05:15:26 +0000 (21:15 -0800)]
[PATCH] file->f_list locking in tty_io.c

release_mem() is altering the file->f_list lists without taking the
appropriate spinlock.

23 years ago[PATCH] file_list_lock contention fixes
Andrew Morton [Tue, 18 Mar 2003 05:15:03 +0000 (21:15 -0800)]
[PATCH] file_list_lock contention fixes

Patch from Manfred Spraul <manfred@colorfullife.com>

Fixes the lock contention over file_list_lock by simply removing the unneeded
anon_list.  So filp allocation does not need to take a global lock any more.

The use of a spinlock to protect files_stat.nr_files is a bit awkward, but
the alternative is a custom sysctl handler, and isn't much more efficient
anyway.

23 years ago[PATCH] Fix memory leak in copy_thread
Andrew Morton [Tue, 18 Mar 2003 05:14:44 +0000 (21:14 -0800)]
[PATCH] Fix memory leak in copy_thread

Patch from Andi Kleen <ak@muc.de>

copy_thread could leak memory if you had a io bitmap and passed wrong
arguments to the new clone flags.

23 years ago[PATCH] use set_current_state in mm
Andrew Morton [Tue, 18 Mar 2003 05:14:27 +0000 (21:14 -0800)]
[PATCH] use set_current_state in mm

Patch from Robert Love <rml@tech9.net>

There are a couple uses of 'p->state=foo' in mm/ which are open coded.
This patch converts them to the proper [__]set_current_state() function.

23 years ago[PATCH] use set_current_state in fs
Andrew Morton [Tue, 18 Mar 2003 05:14:09 +0000 (21:14 -0800)]
[PATCH] use set_current_state in fs

Patch from Robert Love <rml@tech9.net>

This patch is by Inaky Perez-Gonzalez.

There are a couple uses of 'p->state=foo' in fs/ which are open coded.
This patch converts them to the proper [__]set_current_state() function.

23 years ago[PATCH] timer re-addition lockup fix
Andrew Morton [Tue, 18 Mar 2003 05:13:53 +0000 (21:13 -0800)]
[PATCH] timer re-addition lockup fix

This is a forward-port of Andrea's fix in 2.4.

If a timer handler re-adds a timer to go off right now, __run_timers() will
never terminate.  (I wrote a test.  It happens.)

Fix that up by teaching internal_add_timer() to detect when it is being
called from within the context of __run_timers() and to park newly-added
timers onto a temp list instead.  These timers are then added for real by
__run_timers(), after it has finished processing all pending timers.

23 years ago[PATCH] timer code cleanup
Andrew Morton [Tue, 18 Mar 2003 05:13:40 +0000 (21:13 -0800)]
[PATCH] timer code cleanup

- Use list_head functions rather than open-coding them

- Use time comparison macros rather than open-coding them

- Hide some ifdefs

- uninline internal_add_timer().  Saves half a kilobyte of text.

23 years ago[PATCH] remove unused block congestion code
Andrew Morton [Tue, 18 Mar 2003 05:13:31 +0000 (21:13 -0800)]
[PATCH] remove unused block congestion code

Patch from: Hugh Dickins <hugh@veritas.com>

Removes a ton of dead code from ll_rw_blk.c.  I don't expect we'll be using
this now.

23 years ago[PATCH] Pass the load address into ELF_PLAT_INIT()
Andrew Morton [Tue, 18 Mar 2003 05:13:21 +0000 (21:13 -0800)]
[PATCH] Pass the load address into ELF_PLAT_INIT()

Patch from Anton Blanchard <anton@samba.org>

With ppc64 64bit dynamic executables we have to relocate the contents of the
function descriptor.  Passing in the load address to ELF_PLAT_INIT allows us
to do this.

The patch allows ppc64 to run 64-bit executables and is a no-op for other
architectures.

23 years ago[PATCH] Fix noirqbalance
Andrew Morton [Tue, 18 Mar 2003 05:13:12 +0000 (21:13 -0800)]
[PATCH] Fix noirqbalance

Patch from Zwane Mwaikambo <zwane@linuxpower.ca>

This patch fixes what seems to have been a longstanding bug. Ever since we
moved cpu bringup later into the boot process, we end up programming the
ioapics before we have any of our possible cpus in the cpu_online_map.
Therefore leading to the following current situation;

For walmart-smp, bigsmp and summit we set the logical destination for cpu
to TARGET_CPUS which can depend on the cpu_online_map, so what you would
normally see with noirqbalance would be all interrupts handled on cpu0
since at that stage no other cpu apart from the BSP is online.

You can check for this by looking at the ioredtbls at boottime for a two
way system;

.... IRQ redirection table:
 NR Log Phy Mask Trig IRR Pol Stat Dest Deli Vect:
 00 000 00  1    0    0   0   0    0    0    00
 01 001 01  0    0    0   0   0    1    1    39
 02 001 01  0    0    0   0   0    1    1    31
 03 001 01  0    0    0   0   0    1    1    41
 04 001 01  0    0    0   0   0    1    1    49
 05 001 01  0    0    0   0   0    1    1    51
 06 001 01  0    0    0   0   0    1    1    59

Notice that 'Log' is set to 1 instead of 3.

This patch will simply reprogram all the ioredtbls to handle the other
online cpus.

Patch tested on my 2way P2-400 and a 16way NUMAQ both with noirqbalance.
It will not affect the irqbalance case because we are simply setting
TARGET_CPUS which is done anyway.

before:
  CPU0       CPU1
  0:    1495632          0    IO-APIC-edge  timer
  1:       4270          0    IO-APIC-edge  i8042
  2:          0          0          XT-PIC  cascade
  8:          1          0    IO-APIC-edge  rtc
 12:      83592          0    IO-APIC-edge  i8042
 14:      93791          0    IO-APIC-edge  ide0
 15:     103167          0    IO-APIC-edge  ide1
 17:    1396088          0   IO-APIC-level  EMU10K1, eth0
 18:      56125          0   IO-APIC-level  aic7xxx, aic7xxx
 19:       2258          0   IO-APIC-level  uhci-hcd, eth1, serial
NMI:          0          0
LOC:    1495566    1497133

after:
           CPU0       CPU1
  0:    1046157    1015670    IO-APIC-edge  timer
  1:       4923       4173    IO-APIC-edge  i8042
  2:          0          0          XT-PIC  cascade
  8:          1          0    IO-APIC-edge  rtc
 12:      48596      48968    IO-APIC-edge  i8042
 14:       4238       3416    IO-APIC-edge  ide0
 15:      25362      31525    IO-APIC-edge  ide1
 17:       3757       4014   IO-APIC-level  EMU10K1, eth0
 18:        335        366   IO-APIC-level  aic7xxx, aic7xxx
 19:       1052        908   IO-APIC-level  uhci-hcd, eth1
NMI:          0          0
LOC:    2061856    2061893

23 years ago[PATCH] PARISC update
Matthew Wilcox [Tue, 18 Mar 2003 04:49:12 +0000 (20:49 -0800)]
[PATCH] PARISC update

PA-RISC patches for 2.5.65:

 - Only remove palo.conf on a `make mrproper'
 - Add a \ continuation to kernel/Makefile
 - Reindent cache.c
 - Always call schedule_tail
 - Fix some typos in the hardware database
 - Signal handling changes
 - RAID, Device Mapper & BLKSZGET ioctl32 translations
 - Clean up pci host bridge handling a bit.
 - Make IDE link again
 - Fix up compat changes
 - Improve stack dumping code

23 years ago[SPARC64]: Fix __hbird_read_stick signedness, also hbird softint_mask.
David S. Miller [Mon, 17 Mar 2003 15:59:26 +0000 (07:59 -0800)]
[SPARC64]: Fix __hbird_read_stick signedness, also hbird softint_mask.

23 years agoMerge nuts.ninka.net:/home/davem/src/BK/linus-2.5
David S. Miller [Mon, 17 Mar 2003 08:31:02 +0000 (00:31 -0800)]
Merge nuts.ninka.net:/home/davem/src/BK/linus-2.5
into nuts.ninka.net:/home/davem/src/BK/sparc-2.5

23 years agoMerge nuts.ninka.net:/home/davem/src/BK/sparcwork-2.5
David S. Miller [Mon, 17 Mar 2003 05:58:21 +0000 (21:58 -0800)]
Merge nuts.ninka.net:/home/davem/src/BK/sparcwork-2.5
into nuts.ninka.net:/home/davem/src/BK/sparc-2.5

23 years agoLinux 2.5.65 v2.5.65
Linus Torvalds [Mon, 17 Mar 2003 05:36:50 +0000 (21:36 -0800)]
Linux 2.5.65

23 years agoMerge kroah.com:/home/greg/linux/BK/bleed-2.5
Greg Kroah-Hartman [Mon, 17 Mar 2003 02:43:39 +0000 (18:43 -0800)]
Merge kroah.com:/home/greg/linux/BK/bleed-2.5
into kroah.com:/home/greg/linux/BK/gregkh-2.5