Menu
▾
▴
linux-vax-k25-commit — Notifications of CVS commits to kernel 2.5 tree
You can subscribe to this list here.
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(11) |
Jun
(66) |
Jul
(16) |
Aug
(2) |
Sep
(7) |
Oct
(17) |
Nov
(1) |
Dec
(220) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2003 |
Jan
(154) |
Feb
(167) |
Mar
(159) |
Apr
(172) |
May
(35) |
Jun
(58) |
Jul
(97) |
Aug
(285) |
Sep
(139) |
Oct
(252) |
Nov
(8) |
Dec
(3) |
| 2004 |
Jan
(13) |
Feb
(159) |
Mar
(136) |
Apr
(33) |
May
(50) |
Jun
(42) |
Jul
(140) |
Aug
(42) |
Sep
(199) |
Oct
(31) |
Nov
(55) |
Dec
|
| 2005 |
Jan
|
Feb
(12) |
Mar
(214) |
Apr
(119) |
May
(21) |
Jun
(2) |
Jul
(127) |
Aug
(10) |
Sep
(3) |
Oct
(24) |
Nov
(1) |
Dec
|
| 2006 |
Jan
|
Feb
|
Mar
|
Apr
(45) |
May
(13) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(5) |
Nov
(26) |
Dec
|
| 2007 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(35) |
Sep
(1) |
Oct
|
Nov
|
Dec
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-11 08:19:58
|
arch/vax/kernel/cpu_ka62.c | 8 ++++----
include/asm-vax/vaxcpu.h | 16 ++++++++--------
2 files changed, 12 insertions(+), 12 deletions(-)
New commits:
commit d8b81c7f32310ae0ee593c7231de54c1198a7cee
Merge: ac17098... df5fc5d...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 10:19:05 2006 +0200
Merge branch 'vax4linus'
commit df5fc5d29b0e820d48df582bd37b6fb0b000da26
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 10:17:47 2006 +0200
[VAX] Whitespace fixes
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-11 08:19:58
|
arch/vax/kernel/cpu_ka62.c | 8 ++++----
include/asm-vax/vaxcpu.h | 16 ++++++++--------
2 files changed, 12 insertions(+), 12 deletions(-)
New commits:
commit 83f0f97552cb487382231dda0991cf33aedb0284
Merge: e06593f... df5fc5d...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 10:18:08 2006 +0200
Merge branch 'vax4linus'
commit df5fc5d29b0e820d48df582bd37b6fb0b000da26
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 10:17:47 2006 +0200
[VAX] Whitespace fixes
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-11 08:19:56
|
arch/vax/kernel/cpu_ka62.c | 8 ++++----
include/asm-vax/vaxcpu.h | 16 ++++++++--------
2 files changed, 12 insertions(+), 12 deletions(-)
New commits:
commit df5fc5d29b0e820d48df582bd37b6fb0b000da26
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 10:17:47 2006 +0200
[VAX] Whitespace fixes
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-11 08:13:44
|
include/asm-vax/mv.h | 25 +++++++++++++++++++++++++
1 files changed, 25 insertions(+)
New commits:
commit ac17098b0c3970f4e7c215c272db37539ec82004
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 10:13:09 2006 +0200
[VAX] <asm/mv.h> updates.
These were forgotten in the last commit...
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-11 08:02:46
|
arch/vax/kernel/cpu_ka410.c | 8 ++++++--
arch/vax/kernel/cpu_ka55.c | 1 -
arch/vax/kernel/cpu_ka62.c | 11 +++++++----
arch/vax/kernel/cpu_ka630.c | 6 +++++-
arch/vax/kernel/cpu_ka640.c | 6 ++++--
arch/vax/kernel/cpu_ka650.c | 3 +--
arch/vax/kernel/cpu_ka660.c | 3 +--
arch/vax/kernel/cpu_vxt.c | 3 +--
include/asm-vax/clock.h | 2 +-
9 files changed, 26 insertions(+), 17 deletions(-)
New commits:
commit 6c3e757cff44adb137e77a7ac3b741d548e7d1a9
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 10:02:31 2006 +0200
[VAX] Further fixes for the RTC rework.
This includes the addition of additional is_kaXXX() functions.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
commit 72b71a8c1ea8db0f288116383bf4f2b104a84b5f
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 09:38:31 2006 +0200
[VAX] Fix path in comment in <asm/clock.h>
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
commit b58ab36dcb5181359c0177c9c9bed3fcc26e34f0
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 09:29:18 2006 +0200
[VAX] Zap old mv->clock_init from cpu_ka55 -- for real.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-11 07:27:45
|
arch/vax/kernel/time.c | 13 -------------
1 files changed, 13 deletions(-)
New commits:
commit 4d9bb0abb696474c1cf712a206153ca8fb4f6116
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 09:27:24 2006 +0200
[VAX] Don't fiddle with RTC early at kernel startup-time.
Instead of setting some time early at kernel's start up, do it later on
guarded by the RTC lib.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-11 07:08:46
|
arch/vax/kernel/asm-offsets.c | 2 --
1 files changed, 2 deletions(-)
New commits:
commit b6d4ca56d1446516684953e7807ae53f48e6679b
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Tue Apr 11 09:08:21 2006 +0200
[VAX] Remove old mv->clock helpers from asm-offsets.c
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-10 21:02:20
|
Documentation/DocBook/libata.tmpl | 2
Documentation/scsi/scsi_eh.txt | 14
Documentation/scsi/scsi_mid_low_api.txt | 19 -
arch/sparc64/Kconfig | 3
arch/sparc64/defconfig | 10
arch/sparc64/kernel/ptrace.c | 23 +
arch/sparc64/kernel/smp.c | 35 --
arch/sparc64/kernel/traps.c | 4
drivers/char/random.c | 1
drivers/net/irda/irda-usb.c | 363 +++++++++++++++++++--
drivers/net/irda/irda-usb.h | 43 +-
drivers/net/irda/smsc-ircc2.c | 311 +++++++++++++++++
drivers/net/netconsole.c | 1
drivers/net/tg3.c | 100 +++--
drivers/net/tg3.h | 3
drivers/scsi/ahci.c | 1
drivers/scsi/ata_piix.c | 1
drivers/scsi/hosts.c | 12
drivers/scsi/libata-core.c | 1
drivers/scsi/libata-scsi.c | 8
drivers/scsi/libata.h | 1
drivers/scsi/pdc_adma.c | 1
drivers/scsi/sata_mv.c | 1
drivers/scsi/sata_nv.c | 1
drivers/scsi/sata_promise.c | 1
drivers/scsi/sata_qstor.c | 1
drivers/scsi/sata_sil.c | 1
drivers/scsi/sata_sil24.c | 1
drivers/scsi/sata_sis.c | 1
drivers/scsi/sata_svw.c | 1
drivers/scsi/sata_sx4.c | 1
drivers/scsi/sata_uli.c | 1
drivers/scsi/sata_via.c | 1
drivers/scsi/sata_vsc.c | 1
drivers/scsi/scsi_error.c | 4
include/linux/libata.h | 1
include/linux/mm.h | 5
include/linux/netfilter.h | 42 ++
include/linux/netfilter_ipv4.h | 2
include/linux/netfilter_ipv4/ip_conntrack_h323.h | 52 +++
include/linux/netfilter_ipv6.h | 3
include/linux/page-flags.h | 8
include/net/inet6_hashtables.h | 70 ----
include/net/ip.h | 1
include/net/x25device.h | 1
include/scsi/scsi_host.h | 1
include/scsi/scsi_transport.h | 5
mm/page_alloc.c | 31 +
net/bluetooth/sco.c | 2
net/bridge/br_netfilter.c | 13
net/core/dev.c | 5
net/core/dv.c | 5
net/core/flow.c | 4
net/core/gen_estimator.c | 3
net/core/neighbour.c | 14
net/core/net-sysfs.c | 2
net/core/request_sock.c | 4
net/ipv4/fib_trie.c | 14
net/ipv4/ip_fragment.c | 15
net/ipv4/ip_gre.c | 2
net/ipv4/ip_output.c | 6
net/ipv4/ipip.c | 4
net/ipv4/netfilter.c | 50 ++
net/ipv4/netfilter/Kconfig | 1
net/ipv4/netfilter/arptable_filter.c | 19 -
net/ipv4/netfilter/ip_conntrack_helper_h323.c | 66 +--
net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c | 2
net/ipv4/netfilter/ip_conntrack_proto_icmp.c | 23 -
net/ipv4/netfilter/ip_conntrack_proto_tcp.c | 7
net/ipv4/netfilter/ip_conntrack_proto_udp.c | 7
net/ipv4/netfilter/ip_conntrack_standalone.c | 268 ++++++---------
net/ipv4/netfilter/ip_nat_helper_h323.c | 71 ----
net/ipv4/netfilter/ip_nat_rule.c | 2
net/ipv4/netfilter/ip_nat_standalone.c | 159 +++------
net/ipv4/netfilter/ip_queue.c | 31 -
net/ipv4/netfilter/ipt_CLUSTERIP.c | 40 --
net/ipv4/netfilter/ipt_REJECT.c | 9
net/ipv4/netfilter/iptable_filter.c | 21 -
net/ipv4/netfilter/iptable_mangle.c | 33 -
net/ipv4/netfilter/iptable_raw.c | 35 --
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 223 ++++--------
net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 20 -
net/ipv4/tunnel4.c | 8
net/ipv4/xfrm4_input.c | 4
net/ipv6/inet6_hashtables.c | 80 ++++
net/ipv6/ip6_tunnel.c | 12
net/ipv6/netfilter.c | 52 ++-
net/ipv6/netfilter/ip6_queue.c | 31 -
net/ipv6/netfilter/ip6table_filter.c | 21 -
net/ipv6/netfilter/ip6table_mangle.c | 33 -
net/ipv6/netfilter/ip6table_raw.c | 15
net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 179 +++-------
net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 12
net/ipv6/sit.c | 2
net/ipv6/tunnel6.c | 8
net/netfilter/core.c | 51 ++
net/netfilter/nf_conntrack_proto_tcp.c | 50 --
net/netfilter/nf_conntrack_proto_udp.c | 50 --
net/netfilter/nf_conntrack_standalone.c | 115 +++---
net/netfilter/nf_queue.c | 49 --
net/netfilter/nfnetlink_log.c | 25 -
net/netfilter/nfnetlink_queue.c | 27 -
net/sched/act_police.c | 8
security/keys/key.c | 4
security/keys/keyring.c | 1
105 files changed, 1739 insertions(+), 1473 deletions(-)
New commits:
commit 58b298c69358c4c8e22cb64ed2e4346724823d27
Merge: ebe9c64... 1b72373...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Mon Apr 10 22:56:52 2006 +0200
Merge branch 'master'
commit 1b72373491a061be6d456d219a4e2d054ac2aaad
Author: Jeff Garzik <je...@ga...>
Date: Mon Apr 10 14:56:39 2006 -0400
[libata] sata_mv: fix can_queue line accidentally removed in scsi-eh patch
commit 9227c33de80ac01f269ed33624990ce84358e419
Author: Christoph Hellwig <hc...@ls...>
Date: Sat Apr 1 19:21:04 2006 +0200
[PATCH] move ->eh_strategy_handler to the transport class
Overriding the whole EH code is a per-transport, not per-host thing.
Move ->eh_strategy_handler to the transport class, same as
->eh_timed_out.
Downside is that scsi_host_alloc can't check for the total lack of EH
anymore, but the transition period from old EH where we needed it is
long gone already.
Signed-off-by: Christoph Hellwig <hc...@ls...>
Signed-off-by: Jeff Garzik <je...@ga...>
commit 676165a8af7167f488abdcce6851a9bc36e83254
Author: Nick Piggin <pi...@cy...>
Date: Mon Apr 10 11:21:48 2006 +1000
[PATCH] Fix buddy list race that could lead to page lru list corruptions
Rohit found an obscure bug causing buddy list corruption.
page_is_buddy is using a non-atomic test (PagePrivate && page_count == 0)
to determine whether or not a free page's buddy is itself free and in the
buddy lists.
Each of the conjuncts may be true at different times due to unrelated
conditions, so the non-atomic page_is_buddy test may find each conjunct to
be true even if they were not both true at the same time (ie. the page was
not on the buddy lists).
Signed-off-by: Martin Bligh <mb...@go...>
Signed-off-by: Rohit Seth <roh...@go...>
Signed-off-by: Nick Piggin <np...@su...>
Signed-off-by: KAMEZAWA Hiroyuki <kam...@jp...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit c3a9d6541f84ac3ff566982d08389b87c1c36b4e
Author: David Howells <dho...@re...>
Date: Mon Apr 10 15:15:21 2006 +0100
[Security] Keys: Fix oops when adding key to non-keyring
This fixes the problem of an oops occuring when a user attempts to add a
key to a non-keyring key [CVE-2006-1522].
The problem is that __keyring_search_one() doesn't check that the
keyring it's been given is actually a keyring.
I've fixed this problem by:
(1) declaring that caller of __keyring_search_one() must guarantee that
the keyring is a keyring; and
(2) making key_create_or_update() check that the keyring is a keyring,
and return -ENOTDIR if it isn't.
This can be tested by:
keyctl add user b b `keyctl add user a a @s`
Signed-off-by: David Howells <dho...@re...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 460fbf82c0842cad3f3c744c4dcb81978b7829f3
Merge: 68491d5... b1a7ffc...
Author: Linus Torvalds <tor...@g5...>
Date: Mon Apr 10 08:46:00 2006 -0700
Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: (35 commits)
[IPV6]: Deinline few large functions in inet6 code
[IPV4] ip_fragment: Always compute hash with ipfrag_lock held.
[NETFILTER]: Fix DNAT in LOCAL_OUT
[X25]: Restore skb->dev setting in x25_type_trans().
[NET]: Fix hotplug race during device registration.
[IPV6]: Unexport secure_ipv6_port_ephemeral
[NETFILTER]: Fix build with CONFIG_NETFILTER=y/m on IA64
[NET]: More kzalloc conversions.
[NET] kzalloc: use in alloc_netdev
[PKT_SCHED] act_police: Rename methods.
[TG3]: Speed up SRAM access (2nd version)
[TG3]: Kill some less useful flags
[NETFILTER]: H.323 helper: remove changelog
[NETFILTER]: Convert conntrack/ipt_REJECT to new checksumming functions
[NETFILTER]: Add address family specific checksum helpers
[NETFILTER]: Introduce infrastructure for address family specific operations
[NETFILTER]: Fix IP_NF_CONNTRACK_NETLINK dependency
[NETFILTER]: H.323 helper: add parameter 'default_rrq_ttl'
[NETFILTER]: H.323 helper: make get_h245_addr() static
[NETFILTER]: H.323 helper: change EXPORT_SYMBOL to EXPORT_SYMBOL_GPL
...
commit 68491d5892defca59ad4f604cace2b1e30889c14
Author: David S. Miller <da...@su...>
Date: Thu Apr 6 20:28:11 2006 -0700
[SPARC64]: Set ARCH_SELECT_MEMORY_MODEL
Otherwise the build breaks with EXPERIMENTAL disabled
because SPARSEMEM will not get selected properly. See
mm/Kconfig for how that works.
Signed-off-by: David S. Miller <da...@da...>
commit aa1d1a0af6022f02fb601508d3feaabafd405299
Author: David S. Miller <da...@su...>
Date: Thu Apr 6 16:54:33 2006 -0700
[SPARC64]: smp_call_function() fixups...
1) Take doc-book function comment from i386 implementation.
2) cacheline align call_lock, taken from powerpc
3) Need memory barrier after setting call_data
4) Remove timeout
Signed-off-by: David S. Miller <da...@da...>
commit 731bbe431f7dbbcbdc5293cfb187a916c375e83b
Author: David S. Miller <da...@su...>
Date: Tue Apr 4 16:54:40 2006 -0700
[SPARC64]: Translate PTRACE_GETEVENTMSG for 32-bit tasks.
Signed-off-by: David S. Miller <da...@da...>
commit 1608a96e7998bffd10fcb2440e8175cfb703fd95
Author: David S. Miller <da...@su...>
Date: Sun Apr 2 19:31:30 2006 -0700
[SPARC64]: Update defconfig.
Signed-off-by: David S. Miller <da...@da...>
commit 955c054f7905e0a9ee4483b03f866f61e6929bec
Author: David S. Miller <da...@su...>
Date: Sat Apr 1 23:29:56 2006 -0800
[SPARC64]: Print out return PC in cheetah_log_errors().
This makes debugging things a little bit easier.
Signed-off-by: David S. Miller <da...@da...>
commit 1759e58ed2684b7a01cbb96864f23d18884f42ba
Author: David S. Miller <da...@su...>
Date: Sat Apr 1 23:28:10 2006 -0800
[SPARC64]: Add dummy PTRACE_PEEKUSR for gdb.
GDB uses a PTRACE_PEEKUSR call with offset 0 to see
if a thread is alive, so provide a success return for
this particular special case.
Signed-off-by: David S. Miller <da...@da...>
commit b1a7ffcb7a047e99ab02424e651e0492f36095f7
Author: Denis Vlasenko <vd...@il...>
Date: Sun Apr 9 22:48:59 2006 -0700
[IPV6]: Deinline few large functions in inet6 code
Deinline a few functions which produce 200+ bytes of code.
Size Uses Wasted Name and definition
===== ==== ====== ================================================
429 3 818 __inet6_lookup include/net/inet6_hashtables.h
404 2 384 __inet6_lookup_established include/net/inet6_hashtables.h
206 3 372 __inet6_hash include/net/inet6_hashtables.h
Signed-off-by: Denis Vlasenko <vd...@il...>
Signed-off-by: David S. Miller <da...@da...>
commit 55c0022e53452360064ea264c41410c70565d9f8
Author: David S. Miller <da...@su...>
Date: Sun Apr 9 22:43:55 2006 -0700
[IPV4] ip_fragment: Always compute hash with ipfrag_lock held.
Otherwise we could compute an inaccurate hash due to the
random seed changing.
Noticed by Zach Brown and patch is based upon some feedback
from Herbert Xu.
Signed-off-by: David S. Miller <da...@da...>
commit 19910d1aec0b2b96ff4b4a93e2aed0d32643e850
Author: Patrick McHardy <ka...@tr...>
Date: Sun Apr 9 22:38:29 2006 -0700
[NETFILTER]: Fix DNAT in LOCAL_OUT
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 9b591cbd4e0fc2911d105d88d354124467b2cc08
Author: David S. Miller <da...@su...>
Date: Sun Apr 9 22:37:18 2006 -0700
[X25]: Restore skb->dev setting in x25_type_trans().
Noticed by Pascal Schlafer.
Signed-off-by: David S. Miller <da...@da...>
commit 9469d458b90bfb9117cbb488cfa645d94c3921b1
Author: Sergey Vlasov <vs...@al...>
Date: Sun Apr 9 22:32:48 2006 -0700
[NET]: Fix hotplug race during device registration.
From: Thomas de Grenier de Latour <deg...@ea...>
On Sun, 9 Apr 2006 21:56:59 +0400,
Sergey Vlasov <vs...@al...> wrote:
> However, show_address() does not output anything unless
> dev->reg_state == NETREG_REGISTERED - and this state is set by
> netdev_run_todo() only after netdev_register_sysfs() returns, so in
> the meantime (while netdev_register_sysfs() is busy adding the
> "statistics" attribute group) some process may see an empty "address"
> attribute.
I've tried the attached patch, suggested by Sergey Vlasov on
hotplug-devel@, and as far as i can test it works just fine.
Signed-off-by: David S. Miller <da...@da...>
commit 30aaa154fc21ad1ee4400e28009732a04a80862f
Author: Adrian Bunk <bu...@st...>
Date: Sun Apr 9 22:29:17 2006 -0700
[IPV6]: Unexport secure_ipv6_port_ephemeral
This patch removes the unused EXPORT_SYMBOL(secure_ipv6_port_ephemeral).
Signed-off-by: Adrian Bunk <bu...@st...>
Signed-off-by: David S. Miller <da...@da...>
commit 503e4faad18baa62bb818537b920ad939749823e
Author: Brian Haley <bri...@hp...>
Date: Fri Apr 7 15:00:06 2006 -0700
[NETFILTER]: Fix build with CONFIG_NETFILTER=y/m on IA64
Can't build with CONFIG_NETFILTER=y/m on IA64, there's a missing
#include in net/ipv6/netfilter.c
net/ipv6/netfilter.c: In function `nf_ip6_checksum':
net/ipv6/netfilter.c:92: warning: implicit declaration of function
`csum_ipv6_magic'
Signed-off-by: Brian Haley <bri...@hp...>
Signed-off-by: David S. Miller <da...@da...>
commit 77d04bd957ddca9d48a664e28b40f33993f4550e
Author: Andrew Morton <ak...@os...>
Date: Fri Apr 7 14:52:59 2006 -0700
[NET]: More kzalloc conversions.
Signed-off-by: David S. Miller <da...@da...>
commit 31380de95cc3183bbb379339e67f83d69e56fbd6
Author: Paolo 'Blaisorblade' Giarrusso <bla...@ya...>
Date: Thu Apr 6 22:38:28 2006 -0700
[NET] kzalloc: use in alloc_netdev
Noticed this use, fixed it.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <bla...@ya...>
Signed-off-by: David S. Miller <da...@da...>
commit 83b950c89c8cc0dcc1b079c638be25915c9945f1
Author: Jamal Hadi Salim <ha...@cy...>
Date: Thu Apr 6 22:24:22 2006 -0700
[PKT_SCHED] act_police: Rename methods.
Rename policer specific _generic_ methods to be specific to
_act_police_
Signed-off-by: Jamal Hadi Salim <ha...@cy...>
Signed-off-by: David S. Miller <da...@da...>
commit bbadf503d7c7e6efe0a4cd731f8855ba08276215
Author: Michael Chan <mc...@br...>
Date: Thu Apr 6 21:46:34 2006 -0700
[TG3]: Speed up SRAM access (2nd version)
Speed up SRAM read and write functions if possible by using MMIO
instead of config. cycles. With this change, the post reset signature
done at the end of D3 power change must now be moved before the D3
power change.
IBM reported a problem on powerpc blades during ethtool self test that
was caused by the memory test taking excessively long. Config. cycles
are very slow on powerpc and the memory test can take more than 10
seconds to complete using config. cycles.
David Miller informed me that an earlier version of the patch caused
problems on sparc64 systems with built-in tg3 chips. This version
fixes the problem by excluding all SUN built-in tg3 chips from doing
MMIO SRAM access.
TG3_FLAG_EEPROM_WRITE_PROT is also set unconditionally when
TG3_FLG2_SUN_570X is set. This should be sane as all SUN chips are
built-in and do not require Vaux switching.
Signed-off-by: Michael Chan <mc...@br...>
Signed-off-by: David S. Miller <da...@da...>
commit d2d746f83b74022a50d28f7f0f496842c9cde330
Author: Michael Chan <mc...@br...>
Date: Thu Apr 6 21:45:39 2006 -0700
[TG3]: Kill some less useful flags
Kill the TG3_FLAG_NO_{TX|RX}_PSEUDO_CSUM flags because they are not
very useful. This will free up some bits for new flags.
Signed-off-by: Michael Chan <mc...@br...>
Signed-off-by: David S. Miller <da...@da...>
commit 7a43c9955178f1cc88374abe1186c0f2ef21e040
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 16:16:51 2006 -0700
[NETFILTER]: H.323 helper: remove changelog
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 96f6bf82ea3abc77d255d5d554df5f349651f6de
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:19:24 2006 -0700
[NETFILTER]: Convert conntrack/ipt_REJECT to new checksumming functions
Besides removing lots of duplicate code, all converted users benefit
from improved HW checksum error handling. Tested with and without HW
checksums in almost all combinations.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 422c346fad806e2abaeffac686860ebc98dfe33e
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:18:43 2006 -0700
[NETFILTER]: Add address family specific checksum helpers
Add checksum operation which takes care of verifying the checksum and
dealing with HW checksum errors and avoids multiple checksum
operations by setting ip_summed to CHECKSUM_UNNECESSARY after
successful verification.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit bce8032ef3cc58170ab3550e9e271dba7b4c4764
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:18:09 2006 -0700
[NETFILTER]: Introduce infrastructure for address family specific operations
Change the queue rerouter intrastructure to a generic usable
infrastructure for address family specific operations as a base for
some cleanups.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit a0aed49bdb2bbb4234789f241cffb607fd2e213d
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:17:27 2006 -0700
[NETFILTER]: Fix IP_NF_CONNTRACK_NETLINK dependency
When NAT is built as a module, ip_conntrack_netlink can not be linked
statically.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit a0b7db5e86d30f470dc1849f4fa44ff77a813091
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:15:33 2006 -0700
[NETFILTER]: H.323 helper: add parameter 'default_rrq_ttl'
default_rrq_ttl is used when no TTL is included in the RRQ.
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 51d42f5e4ee43fc98aa0c7a2a104808bb602276f
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:14:59 2006 -0700
[NETFILTER]: H.323 helper: make get_h245_addr() static
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 0f249685fde399c01e51a63f2bdfe4a667db083c
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:14:11 2006 -0700
[NETFILTER]: H.323 helper: change EXPORT_SYMBOL to EXPORT_SYMBOL_GPL
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 48bfee5fad0e46f4f18d46285efceba39e897482
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:13:42 2006 -0700
[NETFILTER]: H.323 helper: move some function prototypes to ip_conntrack_h323.h
Move prototypes of NAT callbacks to ip_conntrack_h323.h. Because the
use of typedefs as arguments, some header files need to be moved as
well.
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 32292a7ff1d9306841a8da6ea286847b1070cc6a
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:11:30 2006 -0700
[NETFILTER]: Fix section mismatch warnings
Fix section mismatch warnings caused by netfilter's init_or_cleanup
functions used in many places by splitting the init from the cleanup
parts.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 964ddaa10de8f3aeed12bc2a30726514ff309e64
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:09:49 2006 -0700
[NETFILTER]: Clean up hook registration
Clean up hook registration by makeing use of the new mass registration and
unregistration helpers.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 972d1cb1427946f4980240363aac4e73fb375290
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:09:12 2006 -0700
[NETFILTER]: Add helper functions for mass hook registration/unregistration
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 137dc0233fba0bfa19679bdd96eb104f0e659c5a
Author: Samuel Ortiz <sam...@no...>
Date: Wed Apr 5 22:39:14 2006 -0700
[IRDA]: Support for Sigmatel STIR421x chip
This patch enables support for the Sigmatel's STIR421x IrDA chip.
Once patched with Sigmatel's firmware, this chip "almost" follows the
USB-IrDA spec. Thus this patch is against irda-usb.[ch].
The code has been tested by Nick Fedchik on an STIR4210 chipset based
dongle.
Signed-off-by: Samuel Ortiz <sam...@no...>
Signed-off-by: David S. Miller <da...@da...>
commit c1e14a6ea2ae34993f431d2eb9a0f228ac3574eb
Author: Linus Walleij <tr...@df...>
Date: Wed Apr 5 22:33:59 2006 -0700
[IRDA]: smcinit merged into smsc-ircc driver
This patch integrates the smcinit code into the smsc-ircc driver.
Some laptops have their smsc-ircc chip not properly configured by the
BIOS and needs some preconfiguration. Currently, this can be done from
userspace with smcinit, a utility that comes with the irda-utils
package. It messes with ioports and PCI settings, from userspace. Now
with this patch, if we happen to be on one of the known to be faulty
laptops, we preconfigure the chip from the driver.
Patch from Linus Walleij <tr...@df...>
Signed-off-by: Samuel Ortiz <sam...@no...>
Signed-off-by: David S. Miller <da...@da...>
commit 45af08be6d120690d379cb8421ebaf9d9f86ba52
Author: Herbert Xu <he...@go...>
Date: Wed Apr 5 22:31:19 2006 -0700
[INET]: Use port unreachable instead of proto for tunnels
This patch changes GRE and SIT to generate port unreachable instead of
protocol unreachable errors when we can't find a matching tunnel for a
packet.
This removes the ambiguity as to whether the error is caused by no
tunnel being found or by the lack of support for the given tunnel
type.
Signed-off-by: Herbert Xu <he...@go...>
Signed-off-by: David S. Miller <da...@da...>
commit cdee5751bf91d02616aaf30a5affef56105e3b79
Author: Eric Sesterhenn <sna...@gm...>
Date: Wed Apr 5 22:28:14 2006 -0700
[BLUETOOTH] sco: Possible double free.
this fixes coverity bug id #1068.
hci_send_sco() frees skb if (skb->len > hdev->sco_mtu).
Since it returns a negative error value only in this case, we
can directly return here.
Signed-off-by: Eric Sesterhenn <sna...@gm...>
Signed-off-by: David S. Miller <da...@da...>
commit ad96b485b4224785bcca478cfba02d06c46ee575
Author: Adrian Bunk <bu...@st...>
Date: Wed Apr 5 22:21:04 2006 -0700
[TG3]: Fix a memory leak.
This patch fixes a memory leak (buf wasn't freed) spotted by the
Coverity checker.
Signed-off-by: Adrian Bunk <bu...@st...>
Signed-off-by: David S. Miller <da...@da...>
commit e3a5cd9edff9a7a20de3c88c9d479704da98fb85
Author: Adrian Bunk <bu...@st...>
Date: Wed Apr 5 22:19:47 2006 -0700
[NET]: Fix an off-by-21-or-49 error.
This patch fixes an off-by-21-or-49 error ;-) spotted by the Coverity
checker.
Signed-off-by: Adrian Bunk <bu...@st...>
Signed-off-by: David S. Miller <da...@da...>
commit d938ab44c0c5418bb74a97b422a070e2cdccce22
Author: Randy Dunlap <rd...@xe...>
Date: Tue Apr 4 20:11:56 2006 -0700
[NET] netconsole: set .name in struct console
Set .name in netconsole's struct console to identify the
struct's owner.
Signed-off-by: Randy Dunlap <rd...@xe...>
Acked-by: Matt Mackall <mp...@se...>
Signed-off-by: David S. Miller <da...@da...>
commit 50fba2aa7cefa6b0e1768cb350c9e69042320c03
Author: Herbert Xu <he...@go...>
Date: Tue Apr 4 13:50:45 2006 -0700
[INET]: Move no-tunnel ICMP error to tunnel4/tunnel6
This patch moves the sending of ICMP messages when there are no IPv4/IPv6
tunnels present to tunnel4/tunnel6 respectively. Please note that for now
if xfrm4_tunnel/xfrm6_tunnel is loaded then no ICMP messages will ever be
sent. This is similar to how we handle AH/ESP/IPCOMP.
This move fixes the bug where we always send an ICMP message when there is
no ip6_tunnel device present for a given packet even if it is later handled
by IPsec. It also causes ICMP messages to be sent when no IPIP tunnel is
present.
I've decided to use the "port unreachable" ICMP message over the current
value of "address unreachable" (and "protocol unreachable" by GRE) because
it is not ambiguous unlike the other ones which can be triggered by other
conditions. There seems to be no standard specifying what value must be
used so this change should be OK. In fact we should change GRE to use
this value as well.
Incidentally, this patch also fixes a fairly serious bug in xfrm6_tunnel
where we don't check whether the embedded IPv6 header is present before
dereferencing it for the inside source address.
This patch is inspired by a previous patch by Hugo Santos <hs...@av...>.
Signed-off-by: Herbert Xu <he...@go...>
Signed-off-by: David S. Miller <da...@da...>
commit 2e2f7aefa8a8ba4adb6ecee8cbb43fbe9ca4cc89
Author: Patrick McHardy <ka...@tr...>
Date: Tue Apr 4 13:42:35 2006 -0700
[NETFILTER]: Fix fragmentation issues with bridge netfilter
The conntrack code doesn't do re-fragmentation of defragmented packets
anymore but relies on fragmentation in the IP layer. Purely bridged
packets don't pass through the IP layer, so the bridge netfilter code
needs to take care of fragmentation itself.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 550e29bc96e6f1ced2bca82dace197b009434367
Author: Robert Olsson <rob...@it...>
Date: Tue Apr 4 12:53:35 2006 -0700
[FIB_TRIE]: Fix leaf freeing.
Seems like leaf (end-nodes) has been freed by __tnode_free_rcu and not
by __leaf_free_rcu. This fixes the problem. Only tnode_free is now
used which checks for appropriate node type. free_leaf can be removed.
Signed-off-by: Robert Olsson <rob...@it...>
Signed-off-by: David S. Miller <da...@da...>
commit 8bf4b8a1083694d5aac292f92705ddd3aec29be6
Author: Herbert Xu <he...@go...>
Date: Tue Apr 4 12:51:05 2006 -0700
[IPSEC]: Check x->encap before dereferencing it
We need to dereference x->encap before dereferencing it for encap_type.
If it's absent then the encap_type is zero.
Signed-off-by: Herbert Xu <he...@go...>
Signed-off-by: David S. Miller <da...@da...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-10 21:02:19
|
Documentation/DocBook/libata.tmpl | 2
Documentation/scsi/scsi_eh.txt | 14
Documentation/scsi/scsi_mid_low_api.txt | 19 -
arch/sparc64/Kconfig | 3
arch/sparc64/defconfig | 10
arch/sparc64/kernel/ptrace.c | 23 +
arch/sparc64/kernel/smp.c | 35 --
arch/sparc64/kernel/traps.c | 4
drivers/char/random.c | 1
drivers/net/irda/irda-usb.c | 363 +++++++++++++++++++--
drivers/net/irda/irda-usb.h | 43 +-
drivers/net/irda/smsc-ircc2.c | 311 +++++++++++++++++
drivers/net/netconsole.c | 1
drivers/net/tg3.c | 100 +++--
drivers/net/tg3.h | 3
drivers/scsi/ahci.c | 1
drivers/scsi/ata_piix.c | 1
drivers/scsi/hosts.c | 12
drivers/scsi/libata-core.c | 1
drivers/scsi/libata-scsi.c | 8
drivers/scsi/libata.h | 1
drivers/scsi/pdc_adma.c | 1
drivers/scsi/sata_mv.c | 1
drivers/scsi/sata_nv.c | 1
drivers/scsi/sata_promise.c | 1
drivers/scsi/sata_qstor.c | 1
drivers/scsi/sata_sil.c | 1
drivers/scsi/sata_sil24.c | 1
drivers/scsi/sata_sis.c | 1
drivers/scsi/sata_svw.c | 1
drivers/scsi/sata_sx4.c | 1
drivers/scsi/sata_uli.c | 1
drivers/scsi/sata_via.c | 1
drivers/scsi/sata_vsc.c | 1
drivers/scsi/scsi_error.c | 4
include/linux/libata.h | 1
include/linux/mm.h | 5
include/linux/netfilter.h | 42 ++
include/linux/netfilter_ipv4.h | 2
include/linux/netfilter_ipv4/ip_conntrack_h323.h | 52 +++
include/linux/netfilter_ipv6.h | 3
include/linux/page-flags.h | 8
include/net/inet6_hashtables.h | 70 ----
include/net/ip.h | 1
include/net/x25device.h | 1
include/scsi/scsi_host.h | 1
include/scsi/scsi_transport.h | 5
mm/page_alloc.c | 31 +
net/bluetooth/sco.c | 2
net/bridge/br_netfilter.c | 13
net/core/dev.c | 5
net/core/dv.c | 5
net/core/flow.c | 4
net/core/gen_estimator.c | 3
net/core/neighbour.c | 14
net/core/net-sysfs.c | 2
net/core/request_sock.c | 4
net/ipv4/fib_trie.c | 14
net/ipv4/ip_fragment.c | 15
net/ipv4/ip_gre.c | 2
net/ipv4/ip_output.c | 6
net/ipv4/ipip.c | 4
net/ipv4/netfilter.c | 50 ++
net/ipv4/netfilter/Kconfig | 1
net/ipv4/netfilter/arptable_filter.c | 19 -
net/ipv4/netfilter/ip_conntrack_helper_h323.c | 66 +--
net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c | 2
net/ipv4/netfilter/ip_conntrack_proto_icmp.c | 23 -
net/ipv4/netfilter/ip_conntrack_proto_tcp.c | 7
net/ipv4/netfilter/ip_conntrack_proto_udp.c | 7
net/ipv4/netfilter/ip_conntrack_standalone.c | 268 ++++++---------
net/ipv4/netfilter/ip_nat_helper_h323.c | 71 ----
net/ipv4/netfilter/ip_nat_rule.c | 2
net/ipv4/netfilter/ip_nat_standalone.c | 159 +++------
net/ipv4/netfilter/ip_queue.c | 31 -
net/ipv4/netfilter/ipt_CLUSTERIP.c | 40 --
net/ipv4/netfilter/ipt_REJECT.c | 9
net/ipv4/netfilter/iptable_filter.c | 21 -
net/ipv4/netfilter/iptable_mangle.c | 33 -
net/ipv4/netfilter/iptable_raw.c | 35 --
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 223 ++++--------
net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 20 -
net/ipv4/tunnel4.c | 8
net/ipv4/xfrm4_input.c | 4
net/ipv6/inet6_hashtables.c | 80 ++++
net/ipv6/ip6_tunnel.c | 12
net/ipv6/netfilter.c | 52 ++-
net/ipv6/netfilter/ip6_queue.c | 31 -
net/ipv6/netfilter/ip6table_filter.c | 21 -
net/ipv6/netfilter/ip6table_mangle.c | 33 -
net/ipv6/netfilter/ip6table_raw.c | 15
net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 179 +++-------
net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 12
net/ipv6/sit.c | 2
net/ipv6/tunnel6.c | 8
net/netfilter/core.c | 51 ++
net/netfilter/nf_conntrack_proto_tcp.c | 50 --
net/netfilter/nf_conntrack_proto_udp.c | 50 --
net/netfilter/nf_conntrack_standalone.c | 115 +++---
net/netfilter/nf_queue.c | 49 --
net/netfilter/nfnetlink_log.c | 25 -
net/netfilter/nfnetlink_queue.c | 27 -
net/sched/act_police.c | 8
security/keys/key.c | 4
security/keys/keyring.c | 1
105 files changed, 1739 insertions(+), 1473 deletions(-)
New commits:
commit bcf11adc5da3d790ded7bfd22bfe121633569fd2
Merge: 0660a65... 58b298c...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Mon Apr 10 22:58:14 2006 +0200
Merge branch 'vax4linus'
commit 58b298c69358c4c8e22cb64ed2e4346724823d27
Merge: ebe9c64... 1b72373...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Mon Apr 10 22:56:52 2006 +0200
Merge branch 'master'
commit 1b72373491a061be6d456d219a4e2d054ac2aaad
Author: Jeff Garzik <je...@ga...>
Date: Mon Apr 10 14:56:39 2006 -0400
[libata] sata_mv: fix can_queue line accidentally removed in scsi-eh patch
commit 9227c33de80ac01f269ed33624990ce84358e419
Author: Christoph Hellwig <hc...@ls...>
Date: Sat Apr 1 19:21:04 2006 +0200
[PATCH] move ->eh_strategy_handler to the transport class
Overriding the whole EH code is a per-transport, not per-host thing.
Move ->eh_strategy_handler to the transport class, same as
->eh_timed_out.
Downside is that scsi_host_alloc can't check for the total lack of EH
anymore, but the transition period from old EH where we needed it is
long gone already.
Signed-off-by: Christoph Hellwig <hc...@ls...>
Signed-off-by: Jeff Garzik <je...@ga...>
commit 676165a8af7167f488abdcce6851a9bc36e83254
Author: Nick Piggin <pi...@cy...>
Date: Mon Apr 10 11:21:48 2006 +1000
[PATCH] Fix buddy list race that could lead to page lru list corruptions
Rohit found an obscure bug causing buddy list corruption.
page_is_buddy is using a non-atomic test (PagePrivate && page_count == 0)
to determine whether or not a free page's buddy is itself free and in the
buddy lists.
Each of the conjuncts may be true at different times due to unrelated
conditions, so the non-atomic page_is_buddy test may find each conjunct to
be true even if they were not both true at the same time (ie. the page was
not on the buddy lists).
Signed-off-by: Martin Bligh <mb...@go...>
Signed-off-by: Rohit Seth <roh...@go...>
Signed-off-by: Nick Piggin <np...@su...>
Signed-off-by: KAMEZAWA Hiroyuki <kam...@jp...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit c3a9d6541f84ac3ff566982d08389b87c1c36b4e
Author: David Howells <dho...@re...>
Date: Mon Apr 10 15:15:21 2006 +0100
[Security] Keys: Fix oops when adding key to non-keyring
This fixes the problem of an oops occuring when a user attempts to add a
key to a non-keyring key [CVE-2006-1522].
The problem is that __keyring_search_one() doesn't check that the
keyring it's been given is actually a keyring.
I've fixed this problem by:
(1) declaring that caller of __keyring_search_one() must guarantee that
the keyring is a keyring; and
(2) making key_create_or_update() check that the keyring is a keyring,
and return -ENOTDIR if it isn't.
This can be tested by:
keyctl add user b b `keyctl add user a a @s`
Signed-off-by: David Howells <dho...@re...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 460fbf82c0842cad3f3c744c4dcb81978b7829f3
Merge: 68491d5... b1a7ffc...
Author: Linus Torvalds <tor...@g5...>
Date: Mon Apr 10 08:46:00 2006 -0700
Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: (35 commits)
[IPV6]: Deinline few large functions in inet6 code
[IPV4] ip_fragment: Always compute hash with ipfrag_lock held.
[NETFILTER]: Fix DNAT in LOCAL_OUT
[X25]: Restore skb->dev setting in x25_type_trans().
[NET]: Fix hotplug race during device registration.
[IPV6]: Unexport secure_ipv6_port_ephemeral
[NETFILTER]: Fix build with CONFIG_NETFILTER=y/m on IA64
[NET]: More kzalloc conversions.
[NET] kzalloc: use in alloc_netdev
[PKT_SCHED] act_police: Rename methods.
[TG3]: Speed up SRAM access (2nd version)
[TG3]: Kill some less useful flags
[NETFILTER]: H.323 helper: remove changelog
[NETFILTER]: Convert conntrack/ipt_REJECT to new checksumming functions
[NETFILTER]: Add address family specific checksum helpers
[NETFILTER]: Introduce infrastructure for address family specific operations
[NETFILTER]: Fix IP_NF_CONNTRACK_NETLINK dependency
[NETFILTER]: H.323 helper: add parameter 'default_rrq_ttl'
[NETFILTER]: H.323 helper: make get_h245_addr() static
[NETFILTER]: H.323 helper: change EXPORT_SYMBOL to EXPORT_SYMBOL_GPL
...
commit 68491d5892defca59ad4f604cace2b1e30889c14
Author: David S. Miller <da...@su...>
Date: Thu Apr 6 20:28:11 2006 -0700
[SPARC64]: Set ARCH_SELECT_MEMORY_MODEL
Otherwise the build breaks with EXPERIMENTAL disabled
because SPARSEMEM will not get selected properly. See
mm/Kconfig for how that works.
Signed-off-by: David S. Miller <da...@da...>
commit aa1d1a0af6022f02fb601508d3feaabafd405299
Author: David S. Miller <da...@su...>
Date: Thu Apr 6 16:54:33 2006 -0700
[SPARC64]: smp_call_function() fixups...
1) Take doc-book function comment from i386 implementation.
2) cacheline align call_lock, taken from powerpc
3) Need memory barrier after setting call_data
4) Remove timeout
Signed-off-by: David S. Miller <da...@da...>
commit 731bbe431f7dbbcbdc5293cfb187a916c375e83b
Author: David S. Miller <da...@su...>
Date: Tue Apr 4 16:54:40 2006 -0700
[SPARC64]: Translate PTRACE_GETEVENTMSG for 32-bit tasks.
Signed-off-by: David S. Miller <da...@da...>
commit 1608a96e7998bffd10fcb2440e8175cfb703fd95
Author: David S. Miller <da...@su...>
Date: Sun Apr 2 19:31:30 2006 -0700
[SPARC64]: Update defconfig.
Signed-off-by: David S. Miller <da...@da...>
commit 955c054f7905e0a9ee4483b03f866f61e6929bec
Author: David S. Miller <da...@su...>
Date: Sat Apr 1 23:29:56 2006 -0800
[SPARC64]: Print out return PC in cheetah_log_errors().
This makes debugging things a little bit easier.
Signed-off-by: David S. Miller <da...@da...>
commit 1759e58ed2684b7a01cbb96864f23d18884f42ba
Author: David S. Miller <da...@su...>
Date: Sat Apr 1 23:28:10 2006 -0800
[SPARC64]: Add dummy PTRACE_PEEKUSR for gdb.
GDB uses a PTRACE_PEEKUSR call with offset 0 to see
if a thread is alive, so provide a success return for
this particular special case.
Signed-off-by: David S. Miller <da...@da...>
commit b1a7ffcb7a047e99ab02424e651e0492f36095f7
Author: Denis Vlasenko <vd...@il...>
Date: Sun Apr 9 22:48:59 2006 -0700
[IPV6]: Deinline few large functions in inet6 code
Deinline a few functions which produce 200+ bytes of code.
Size Uses Wasted Name and definition
===== ==== ====== ================================================
429 3 818 __inet6_lookup include/net/inet6_hashtables.h
404 2 384 __inet6_lookup_established include/net/inet6_hashtables.h
206 3 372 __inet6_hash include/net/inet6_hashtables.h
Signed-off-by: Denis Vlasenko <vd...@il...>
Signed-off-by: David S. Miller <da...@da...>
commit 55c0022e53452360064ea264c41410c70565d9f8
Author: David S. Miller <da...@su...>
Date: Sun Apr 9 22:43:55 2006 -0700
[IPV4] ip_fragment: Always compute hash with ipfrag_lock held.
Otherwise we could compute an inaccurate hash due to the
random seed changing.
Noticed by Zach Brown and patch is based upon some feedback
from Herbert Xu.
Signed-off-by: David S. Miller <da...@da...>
commit 19910d1aec0b2b96ff4b4a93e2aed0d32643e850
Author: Patrick McHardy <ka...@tr...>
Date: Sun Apr 9 22:38:29 2006 -0700
[NETFILTER]: Fix DNAT in LOCAL_OUT
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 9b591cbd4e0fc2911d105d88d354124467b2cc08
Author: David S. Miller <da...@su...>
Date: Sun Apr 9 22:37:18 2006 -0700
[X25]: Restore skb->dev setting in x25_type_trans().
Noticed by Pascal Schlafer.
Signed-off-by: David S. Miller <da...@da...>
commit 9469d458b90bfb9117cbb488cfa645d94c3921b1
Author: Sergey Vlasov <vs...@al...>
Date: Sun Apr 9 22:32:48 2006 -0700
[NET]: Fix hotplug race during device registration.
From: Thomas de Grenier de Latour <deg...@ea...>
On Sun, 9 Apr 2006 21:56:59 +0400,
Sergey Vlasov <vs...@al...> wrote:
> However, show_address() does not output anything unless
> dev->reg_state == NETREG_REGISTERED - and this state is set by
> netdev_run_todo() only after netdev_register_sysfs() returns, so in
> the meantime (while netdev_register_sysfs() is busy adding the
> "statistics" attribute group) some process may see an empty "address"
> attribute.
I've tried the attached patch, suggested by Sergey Vlasov on
hotplug-devel@, and as far as i can test it works just fine.
Signed-off-by: David S. Miller <da...@da...>
commit 30aaa154fc21ad1ee4400e28009732a04a80862f
Author: Adrian Bunk <bu...@st...>
Date: Sun Apr 9 22:29:17 2006 -0700
[IPV6]: Unexport secure_ipv6_port_ephemeral
This patch removes the unused EXPORT_SYMBOL(secure_ipv6_port_ephemeral).
Signed-off-by: Adrian Bunk <bu...@st...>
Signed-off-by: David S. Miller <da...@da...>
commit 503e4faad18baa62bb818537b920ad939749823e
Author: Brian Haley <bri...@hp...>
Date: Fri Apr 7 15:00:06 2006 -0700
[NETFILTER]: Fix build with CONFIG_NETFILTER=y/m on IA64
Can't build with CONFIG_NETFILTER=y/m on IA64, there's a missing
#include in net/ipv6/netfilter.c
net/ipv6/netfilter.c: In function `nf_ip6_checksum':
net/ipv6/netfilter.c:92: warning: implicit declaration of function
`csum_ipv6_magic'
Signed-off-by: Brian Haley <bri...@hp...>
Signed-off-by: David S. Miller <da...@da...>
commit 77d04bd957ddca9d48a664e28b40f33993f4550e
Author: Andrew Morton <ak...@os...>
Date: Fri Apr 7 14:52:59 2006 -0700
[NET]: More kzalloc conversions.
Signed-off-by: David S. Miller <da...@da...>
commit 31380de95cc3183bbb379339e67f83d69e56fbd6
Author: Paolo 'Blaisorblade' Giarrusso <bla...@ya...>
Date: Thu Apr 6 22:38:28 2006 -0700
[NET] kzalloc: use in alloc_netdev
Noticed this use, fixed it.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <bla...@ya...>
Signed-off-by: David S. Miller <da...@da...>
commit 83b950c89c8cc0dcc1b079c638be25915c9945f1
Author: Jamal Hadi Salim <ha...@cy...>
Date: Thu Apr 6 22:24:22 2006 -0700
[PKT_SCHED] act_police: Rename methods.
Rename policer specific _generic_ methods to be specific to
_act_police_
Signed-off-by: Jamal Hadi Salim <ha...@cy...>
Signed-off-by: David S. Miller <da...@da...>
commit bbadf503d7c7e6efe0a4cd731f8855ba08276215
Author: Michael Chan <mc...@br...>
Date: Thu Apr 6 21:46:34 2006 -0700
[TG3]: Speed up SRAM access (2nd version)
Speed up SRAM read and write functions if possible by using MMIO
instead of config. cycles. With this change, the post reset signature
done at the end of D3 power change must now be moved before the D3
power change.
IBM reported a problem on powerpc blades during ethtool self test that
was caused by the memory test taking excessively long. Config. cycles
are very slow on powerpc and the memory test can take more than 10
seconds to complete using config. cycles.
David Miller informed me that an earlier version of the patch caused
problems on sparc64 systems with built-in tg3 chips. This version
fixes the problem by excluding all SUN built-in tg3 chips from doing
MMIO SRAM access.
TG3_FLAG_EEPROM_WRITE_PROT is also set unconditionally when
TG3_FLG2_SUN_570X is set. This should be sane as all SUN chips are
built-in and do not require Vaux switching.
Signed-off-by: Michael Chan <mc...@br...>
Signed-off-by: David S. Miller <da...@da...>
commit d2d746f83b74022a50d28f7f0f496842c9cde330
Author: Michael Chan <mc...@br...>
Date: Thu Apr 6 21:45:39 2006 -0700
[TG3]: Kill some less useful flags
Kill the TG3_FLAG_NO_{TX|RX}_PSEUDO_CSUM flags because they are not
very useful. This will free up some bits for new flags.
Signed-off-by: Michael Chan <mc...@br...>
Signed-off-by: David S. Miller <da...@da...>
commit 7a43c9955178f1cc88374abe1186c0f2ef21e040
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 16:16:51 2006 -0700
[NETFILTER]: H.323 helper: remove changelog
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 96f6bf82ea3abc77d255d5d554df5f349651f6de
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:19:24 2006 -0700
[NETFILTER]: Convert conntrack/ipt_REJECT to new checksumming functions
Besides removing lots of duplicate code, all converted users benefit
from improved HW checksum error handling. Tested with and without HW
checksums in almost all combinations.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 422c346fad806e2abaeffac686860ebc98dfe33e
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:18:43 2006 -0700
[NETFILTER]: Add address family specific checksum helpers
Add checksum operation which takes care of verifying the checksum and
dealing with HW checksum errors and avoids multiple checksum
operations by setting ip_summed to CHECKSUM_UNNECESSARY after
successful verification.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit bce8032ef3cc58170ab3550e9e271dba7b4c4764
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:18:09 2006 -0700
[NETFILTER]: Introduce infrastructure for address family specific operations
Change the queue rerouter intrastructure to a generic usable
infrastructure for address family specific operations as a base for
some cleanups.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit a0aed49bdb2bbb4234789f241cffb607fd2e213d
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:17:27 2006 -0700
[NETFILTER]: Fix IP_NF_CONNTRACK_NETLINK dependency
When NAT is built as a module, ip_conntrack_netlink can not be linked
statically.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit a0b7db5e86d30f470dc1849f4fa44ff77a813091
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:15:33 2006 -0700
[NETFILTER]: H.323 helper: add parameter 'default_rrq_ttl'
default_rrq_ttl is used when no TTL is included in the RRQ.
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 51d42f5e4ee43fc98aa0c7a2a104808bb602276f
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:14:59 2006 -0700
[NETFILTER]: H.323 helper: make get_h245_addr() static
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 0f249685fde399c01e51a63f2bdfe4a667db083c
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:14:11 2006 -0700
[NETFILTER]: H.323 helper: change EXPORT_SYMBOL to EXPORT_SYMBOL_GPL
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 48bfee5fad0e46f4f18d46285efceba39e897482
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:13:42 2006 -0700
[NETFILTER]: H.323 helper: move some function prototypes to ip_conntrack_h323.h
Move prototypes of NAT callbacks to ip_conntrack_h323.h. Because the
use of typedefs as arguments, some header files need to be moved as
well.
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 32292a7ff1d9306841a8da6ea286847b1070cc6a
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:11:30 2006 -0700
[NETFILTER]: Fix section mismatch warnings
Fix section mismatch warnings caused by netfilter's init_or_cleanup
functions used in many places by splitting the init from the cleanup
parts.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 964ddaa10de8f3aeed12bc2a30726514ff309e64
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:09:49 2006 -0700
[NETFILTER]: Clean up hook registration
Clean up hook registration by makeing use of the new mass registration and
unregistration helpers.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 972d1cb1427946f4980240363aac4e73fb375290
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:09:12 2006 -0700
[NETFILTER]: Add helper functions for mass hook registration/unregistration
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 137dc0233fba0bfa19679bdd96eb104f0e659c5a
Author: Samuel Ortiz <sam...@no...>
Date: Wed Apr 5 22:39:14 2006 -0700
[IRDA]: Support for Sigmatel STIR421x chip
This patch enables support for the Sigmatel's STIR421x IrDA chip.
Once patched with Sigmatel's firmware, this chip "almost" follows the
USB-IrDA spec. Thus this patch is against irda-usb.[ch].
The code has been tested by Nick Fedchik on an STIR4210 chipset based
dongle.
Signed-off-by: Samuel Ortiz <sam...@no...>
Signed-off-by: David S. Miller <da...@da...>
commit c1e14a6ea2ae34993f431d2eb9a0f228ac3574eb
Author: Linus Walleij <tr...@df...>
Date: Wed Apr 5 22:33:59 2006 -0700
[IRDA]: smcinit merged into smsc-ircc driver
This patch integrates the smcinit code into the smsc-ircc driver.
Some laptops have their smsc-ircc chip not properly configured by the
BIOS and needs some preconfiguration. Currently, this can be done from
userspace with smcinit, a utility that comes with the irda-utils
package. It messes with ioports and PCI settings, from userspace. Now
with this patch, if we happen to be on one of the known to be faulty
laptops, we preconfigure the chip from the driver.
Patch from Linus Walleij <tr...@df...>
Signed-off-by: Samuel Ortiz <sam...@no...>
Signed-off-by: David S. Miller <da...@da...>
commit 45af08be6d120690d379cb8421ebaf9d9f86ba52
Author: Herbert Xu <he...@go...>
Date: Wed Apr 5 22:31:19 2006 -0700
[INET]: Use port unreachable instead of proto for tunnels
This patch changes GRE and SIT to generate port unreachable instead of
protocol unreachable errors when we can't find a matching tunnel for a
packet.
This removes the ambiguity as to whether the error is caused by no
tunnel being found or by the lack of support for the given tunnel
type.
Signed-off-by: Herbert Xu <he...@go...>
Signed-off-by: David S. Miller <da...@da...>
commit cdee5751bf91d02616aaf30a5affef56105e3b79
Author: Eric Sesterhenn <sna...@gm...>
Date: Wed Apr 5 22:28:14 2006 -0700
[BLUETOOTH] sco: Possible double free.
this fixes coverity bug id #1068.
hci_send_sco() frees skb if (skb->len > hdev->sco_mtu).
Since it returns a negative error value only in this case, we
can directly return here.
Signed-off-by: Eric Sesterhenn <sna...@gm...>
Signed-off-by: David S. Miller <da...@da...>
commit ad96b485b4224785bcca478cfba02d06c46ee575
Author: Adrian Bunk <bu...@st...>
Date: Wed Apr 5 22:21:04 2006 -0700
[TG3]: Fix a memory leak.
This patch fixes a memory leak (buf wasn't freed) spotted by the
Coverity checker.
Signed-off-by: Adrian Bunk <bu...@st...>
Signed-off-by: David S. Miller <da...@da...>
commit e3a5cd9edff9a7a20de3c88c9d479704da98fb85
Author: Adrian Bunk <bu...@st...>
Date: Wed Apr 5 22:19:47 2006 -0700
[NET]: Fix an off-by-21-or-49 error.
This patch fixes an off-by-21-or-49 error ;-) spotted by the Coverity
checker.
Signed-off-by: Adrian Bunk <bu...@st...>
Signed-off-by: David S. Miller <da...@da...>
commit d938ab44c0c5418bb74a97b422a070e2cdccce22
Author: Randy Dunlap <rd...@xe...>
Date: Tue Apr 4 20:11:56 2006 -0700
[NET] netconsole: set .name in struct console
Set .name in netconsole's struct console to identify the
struct's owner.
Signed-off-by: Randy Dunlap <rd...@xe...>
Acked-by: Matt Mackall <mp...@se...>
Signed-off-by: David S. Miller <da...@da...>
commit 50fba2aa7cefa6b0e1768cb350c9e69042320c03
Author: Herbert Xu <he...@go...>
Date: Tue Apr 4 13:50:45 2006 -0700
[INET]: Move no-tunnel ICMP error to tunnel4/tunnel6
This patch moves the sending of ICMP messages when there are no IPv4/IPv6
tunnels present to tunnel4/tunnel6 respectively. Please note that for now
if xfrm4_tunnel/xfrm6_tunnel is loaded then no ICMP messages will ever be
sent. This is similar to how we handle AH/ESP/IPCOMP.
This move fixes the bug where we always send an ICMP message when there is
no ip6_tunnel device present for a given packet even if it is later handled
by IPsec. It also causes ICMP messages to be sent when no IPIP tunnel is
present.
I've decided to use the "port unreachable" ICMP message over the current
value of "address unreachable" (and "protocol unreachable" by GRE) because
it is not ambiguous unlike the other ones which can be triggered by other
conditions. There seems to be no standard specifying what value must be
used so this change should be OK. In fact we should change GRE to use
this value as well.
Incidentally, this patch also fixes a fairly serious bug in xfrm6_tunnel
where we don't check whether the embedded IPv6 header is present before
dereferencing it for the inside source address.
This patch is inspired by a previous patch by Hugo Santos <hs...@av...>.
Signed-off-by: Herbert Xu <he...@go...>
Signed-off-by: David S. Miller <da...@da...>
commit 2e2f7aefa8a8ba4adb6ecee8cbb43fbe9ca4cc89
Author: Patrick McHardy <ka...@tr...>
Date: Tue Apr 4 13:42:35 2006 -0700
[NETFILTER]: Fix fragmentation issues with bridge netfilter
The conntrack code doesn't do re-fragmentation of defragmented packets
anymore but relies on fragmentation in the IP layer. Purely bridged
packets don't pass through the IP layer, so the bridge netfilter code
needs to take care of fragmentation itself.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 550e29bc96e6f1ced2bca82dace197b009434367
Author: Robert Olsson <rob...@it...>
Date: Tue Apr 4 12:53:35 2006 -0700
[FIB_TRIE]: Fix leaf freeing.
Seems like leaf (end-nodes) has been freed by __tnode_free_rcu and not
by __leaf_free_rcu. This fixes the problem. Only tnode_free is now
used which checks for appropriate node type. free_leaf can be removed.
Signed-off-by: Robert Olsson <rob...@it...>
Signed-off-by: David S. Miller <da...@da...>
commit 8bf4b8a1083694d5aac292f92705ddd3aec29be6
Author: Herbert Xu <he...@go...>
Date: Tue Apr 4 12:51:05 2006 -0700
[IPSEC]: Check x->encap before dereferencing it
We need to dereference x->encap before dereferencing it for encap_type.
If it's absent then the encap_type is zero.
Signed-off-by: Herbert Xu <he...@go...>
Signed-off-by: David S. Miller <da...@da...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-10 21:02:19
|
Documentation/DocBook/libata.tmpl | 2
Documentation/scsi/scsi_eh.txt | 14
Documentation/scsi/scsi_mid_low_api.txt | 19 -
arch/sparc64/Kconfig | 3
arch/sparc64/defconfig | 10
arch/sparc64/kernel/ptrace.c | 23 +
arch/sparc64/kernel/smp.c | 35 --
arch/sparc64/kernel/traps.c | 4
drivers/char/random.c | 1
drivers/net/irda/irda-usb.c | 363 +++++++++++++++++++--
drivers/net/irda/irda-usb.h | 43 +-
drivers/net/irda/smsc-ircc2.c | 311 +++++++++++++++++
drivers/net/netconsole.c | 1
drivers/net/tg3.c | 100 +++--
drivers/net/tg3.h | 3
drivers/scsi/ahci.c | 1
drivers/scsi/ata_piix.c | 1
drivers/scsi/hosts.c | 12
drivers/scsi/libata-core.c | 1
drivers/scsi/libata-scsi.c | 8
drivers/scsi/libata.h | 1
drivers/scsi/pdc_adma.c | 1
drivers/scsi/sata_mv.c | 1
drivers/scsi/sata_nv.c | 1
drivers/scsi/sata_promise.c | 1
drivers/scsi/sata_qstor.c | 1
drivers/scsi/sata_sil.c | 1
drivers/scsi/sata_sil24.c | 1
drivers/scsi/sata_sis.c | 1
drivers/scsi/sata_svw.c | 1
drivers/scsi/sata_sx4.c | 1
drivers/scsi/sata_uli.c | 1
drivers/scsi/sata_via.c | 1
drivers/scsi/sata_vsc.c | 1
drivers/scsi/scsi_error.c | 4
include/linux/libata.h | 1
include/linux/mm.h | 5
include/linux/netfilter.h | 42 ++
include/linux/netfilter_ipv4.h | 2
include/linux/netfilter_ipv4/ip_conntrack_h323.h | 52 +++
include/linux/netfilter_ipv6.h | 3
include/linux/page-flags.h | 8
include/net/inet6_hashtables.h | 70 ----
include/net/ip.h | 1
include/net/x25device.h | 1
include/scsi/scsi_host.h | 1
include/scsi/scsi_transport.h | 5
mm/page_alloc.c | 31 +
net/bluetooth/sco.c | 2
net/bridge/br_netfilter.c | 13
net/core/dev.c | 5
net/core/dv.c | 5
net/core/flow.c | 4
net/core/gen_estimator.c | 3
net/core/neighbour.c | 14
net/core/net-sysfs.c | 2
net/core/request_sock.c | 4
net/ipv4/fib_trie.c | 14
net/ipv4/ip_fragment.c | 15
net/ipv4/ip_gre.c | 2
net/ipv4/ip_output.c | 6
net/ipv4/ipip.c | 4
net/ipv4/netfilter.c | 50 ++
net/ipv4/netfilter/Kconfig | 1
net/ipv4/netfilter/arptable_filter.c | 19 -
net/ipv4/netfilter/ip_conntrack_helper_h323.c | 66 +--
net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c | 2
net/ipv4/netfilter/ip_conntrack_proto_icmp.c | 23 -
net/ipv4/netfilter/ip_conntrack_proto_tcp.c | 7
net/ipv4/netfilter/ip_conntrack_proto_udp.c | 7
net/ipv4/netfilter/ip_conntrack_standalone.c | 268 ++++++---------
net/ipv4/netfilter/ip_nat_helper_h323.c | 71 ----
net/ipv4/netfilter/ip_nat_rule.c | 2
net/ipv4/netfilter/ip_nat_standalone.c | 159 +++------
net/ipv4/netfilter/ip_queue.c | 31 -
net/ipv4/netfilter/ipt_CLUSTERIP.c | 40 --
net/ipv4/netfilter/ipt_REJECT.c | 9
net/ipv4/netfilter/iptable_filter.c | 21 -
net/ipv4/netfilter/iptable_mangle.c | 33 -
net/ipv4/netfilter/iptable_raw.c | 35 --
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 223 ++++--------
net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 20 -
net/ipv4/tunnel4.c | 8
net/ipv4/xfrm4_input.c | 4
net/ipv6/inet6_hashtables.c | 80 ++++
net/ipv6/ip6_tunnel.c | 12
net/ipv6/netfilter.c | 52 ++-
net/ipv6/netfilter/ip6_queue.c | 31 -
net/ipv6/netfilter/ip6table_filter.c | 21 -
net/ipv6/netfilter/ip6table_mangle.c | 33 -
net/ipv6/netfilter/ip6table_raw.c | 15
net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 179 +++-------
net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 12
net/ipv6/sit.c | 2
net/ipv6/tunnel6.c | 8
net/netfilter/core.c | 51 ++
net/netfilter/nf_conntrack_proto_tcp.c | 50 --
net/netfilter/nf_conntrack_proto_udp.c | 50 --
net/netfilter/nf_conntrack_standalone.c | 115 +++---
net/netfilter/nf_queue.c | 49 --
net/netfilter/nfnetlink_log.c | 25 -
net/netfilter/nfnetlink_queue.c | 27 -
net/sched/act_police.c | 8
security/keys/key.c | 4
security/keys/keyring.c | 1
105 files changed, 1739 insertions(+), 1473 deletions(-)
New commits:
commit e06593fc3d6f73622ebe6a4ec75d6b48cf118341
Merge: dce6b2b... 58b298c...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Mon Apr 10 22:57:27 2006 +0200
Merge branch 'vax4linus'
commit 58b298c69358c4c8e22cb64ed2e4346724823d27
Merge: ebe9c64... 1b72373...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Mon Apr 10 22:56:52 2006 +0200
Merge branch 'master'
commit 1b72373491a061be6d456d219a4e2d054ac2aaad
Author: Jeff Garzik <je...@ga...>
Date: Mon Apr 10 14:56:39 2006 -0400
[libata] sata_mv: fix can_queue line accidentally removed in scsi-eh patch
commit 9227c33de80ac01f269ed33624990ce84358e419
Author: Christoph Hellwig <hc...@ls...>
Date: Sat Apr 1 19:21:04 2006 +0200
[PATCH] move ->eh_strategy_handler to the transport class
Overriding the whole EH code is a per-transport, not per-host thing.
Move ->eh_strategy_handler to the transport class, same as
->eh_timed_out.
Downside is that scsi_host_alloc can't check for the total lack of EH
anymore, but the transition period from old EH where we needed it is
long gone already.
Signed-off-by: Christoph Hellwig <hc...@ls...>
Signed-off-by: Jeff Garzik <je...@ga...>
commit 676165a8af7167f488abdcce6851a9bc36e83254
Author: Nick Piggin <pi...@cy...>
Date: Mon Apr 10 11:21:48 2006 +1000
[PATCH] Fix buddy list race that could lead to page lru list corruptions
Rohit found an obscure bug causing buddy list corruption.
page_is_buddy is using a non-atomic test (PagePrivate && page_count == 0)
to determine whether or not a free page's buddy is itself free and in the
buddy lists.
Each of the conjuncts may be true at different times due to unrelated
conditions, so the non-atomic page_is_buddy test may find each conjunct to
be true even if they were not both true at the same time (ie. the page was
not on the buddy lists).
Signed-off-by: Martin Bligh <mb...@go...>
Signed-off-by: Rohit Seth <roh...@go...>
Signed-off-by: Nick Piggin <np...@su...>
Signed-off-by: KAMEZAWA Hiroyuki <kam...@jp...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit c3a9d6541f84ac3ff566982d08389b87c1c36b4e
Author: David Howells <dho...@re...>
Date: Mon Apr 10 15:15:21 2006 +0100
[Security] Keys: Fix oops when adding key to non-keyring
This fixes the problem of an oops occuring when a user attempts to add a
key to a non-keyring key [CVE-2006-1522].
The problem is that __keyring_search_one() doesn't check that the
keyring it's been given is actually a keyring.
I've fixed this problem by:
(1) declaring that caller of __keyring_search_one() must guarantee that
the keyring is a keyring; and
(2) making key_create_or_update() check that the keyring is a keyring,
and return -ENOTDIR if it isn't.
This can be tested by:
keyctl add user b b `keyctl add user a a @s`
Signed-off-by: David Howells <dho...@re...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 460fbf82c0842cad3f3c744c4dcb81978b7829f3
Merge: 68491d5... b1a7ffc...
Author: Linus Torvalds <tor...@g5...>
Date: Mon Apr 10 08:46:00 2006 -0700
Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: (35 commits)
[IPV6]: Deinline few large functions in inet6 code
[IPV4] ip_fragment: Always compute hash with ipfrag_lock held.
[NETFILTER]: Fix DNAT in LOCAL_OUT
[X25]: Restore skb->dev setting in x25_type_trans().
[NET]: Fix hotplug race during device registration.
[IPV6]: Unexport secure_ipv6_port_ephemeral
[NETFILTER]: Fix build with CONFIG_NETFILTER=y/m on IA64
[NET]: More kzalloc conversions.
[NET] kzalloc: use in alloc_netdev
[PKT_SCHED] act_police: Rename methods.
[TG3]: Speed up SRAM access (2nd version)
[TG3]: Kill some less useful flags
[NETFILTER]: H.323 helper: remove changelog
[NETFILTER]: Convert conntrack/ipt_REJECT to new checksumming functions
[NETFILTER]: Add address family specific checksum helpers
[NETFILTER]: Introduce infrastructure for address family specific operations
[NETFILTER]: Fix IP_NF_CONNTRACK_NETLINK dependency
[NETFILTER]: H.323 helper: add parameter 'default_rrq_ttl'
[NETFILTER]: H.323 helper: make get_h245_addr() static
[NETFILTER]: H.323 helper: change EXPORT_SYMBOL to EXPORT_SYMBOL_GPL
...
commit 68491d5892defca59ad4f604cace2b1e30889c14
Author: David S. Miller <da...@su...>
Date: Thu Apr 6 20:28:11 2006 -0700
[SPARC64]: Set ARCH_SELECT_MEMORY_MODEL
Otherwise the build breaks with EXPERIMENTAL disabled
because SPARSEMEM will not get selected properly. See
mm/Kconfig for how that works.
Signed-off-by: David S. Miller <da...@da...>
commit aa1d1a0af6022f02fb601508d3feaabafd405299
Author: David S. Miller <da...@su...>
Date: Thu Apr 6 16:54:33 2006 -0700
[SPARC64]: smp_call_function() fixups...
1) Take doc-book function comment from i386 implementation.
2) cacheline align call_lock, taken from powerpc
3) Need memory barrier after setting call_data
4) Remove timeout
Signed-off-by: David S. Miller <da...@da...>
commit 731bbe431f7dbbcbdc5293cfb187a916c375e83b
Author: David S. Miller <da...@su...>
Date: Tue Apr 4 16:54:40 2006 -0700
[SPARC64]: Translate PTRACE_GETEVENTMSG for 32-bit tasks.
Signed-off-by: David S. Miller <da...@da...>
commit 1608a96e7998bffd10fcb2440e8175cfb703fd95
Author: David S. Miller <da...@su...>
Date: Sun Apr 2 19:31:30 2006 -0700
[SPARC64]: Update defconfig.
Signed-off-by: David S. Miller <da...@da...>
commit 955c054f7905e0a9ee4483b03f866f61e6929bec
Author: David S. Miller <da...@su...>
Date: Sat Apr 1 23:29:56 2006 -0800
[SPARC64]: Print out return PC in cheetah_log_errors().
This makes debugging things a little bit easier.
Signed-off-by: David S. Miller <da...@da...>
commit 1759e58ed2684b7a01cbb96864f23d18884f42ba
Author: David S. Miller <da...@su...>
Date: Sat Apr 1 23:28:10 2006 -0800
[SPARC64]: Add dummy PTRACE_PEEKUSR for gdb.
GDB uses a PTRACE_PEEKUSR call with offset 0 to see
if a thread is alive, so provide a success return for
this particular special case.
Signed-off-by: David S. Miller <da...@da...>
commit b1a7ffcb7a047e99ab02424e651e0492f36095f7
Author: Denis Vlasenko <vd...@il...>
Date: Sun Apr 9 22:48:59 2006 -0700
[IPV6]: Deinline few large functions in inet6 code
Deinline a few functions which produce 200+ bytes of code.
Size Uses Wasted Name and definition
===== ==== ====== ================================================
429 3 818 __inet6_lookup include/net/inet6_hashtables.h
404 2 384 __inet6_lookup_established include/net/inet6_hashtables.h
206 3 372 __inet6_hash include/net/inet6_hashtables.h
Signed-off-by: Denis Vlasenko <vd...@il...>
Signed-off-by: David S. Miller <da...@da...>
commit 55c0022e53452360064ea264c41410c70565d9f8
Author: David S. Miller <da...@su...>
Date: Sun Apr 9 22:43:55 2006 -0700
[IPV4] ip_fragment: Always compute hash with ipfrag_lock held.
Otherwise we could compute an inaccurate hash due to the
random seed changing.
Noticed by Zach Brown and patch is based upon some feedback
from Herbert Xu.
Signed-off-by: David S. Miller <da...@da...>
commit 19910d1aec0b2b96ff4b4a93e2aed0d32643e850
Author: Patrick McHardy <ka...@tr...>
Date: Sun Apr 9 22:38:29 2006 -0700
[NETFILTER]: Fix DNAT in LOCAL_OUT
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 9b591cbd4e0fc2911d105d88d354124467b2cc08
Author: David S. Miller <da...@su...>
Date: Sun Apr 9 22:37:18 2006 -0700
[X25]: Restore skb->dev setting in x25_type_trans().
Noticed by Pascal Schlafer.
Signed-off-by: David S. Miller <da...@da...>
commit 9469d458b90bfb9117cbb488cfa645d94c3921b1
Author: Sergey Vlasov <vs...@al...>
Date: Sun Apr 9 22:32:48 2006 -0700
[NET]: Fix hotplug race during device registration.
From: Thomas de Grenier de Latour <deg...@ea...>
On Sun, 9 Apr 2006 21:56:59 +0400,
Sergey Vlasov <vs...@al...> wrote:
> However, show_address() does not output anything unless
> dev->reg_state == NETREG_REGISTERED - and this state is set by
> netdev_run_todo() only after netdev_register_sysfs() returns, so in
> the meantime (while netdev_register_sysfs() is busy adding the
> "statistics" attribute group) some process may see an empty "address"
> attribute.
I've tried the attached patch, suggested by Sergey Vlasov on
hotplug-devel@, and as far as i can test it works just fine.
Signed-off-by: David S. Miller <da...@da...>
commit 30aaa154fc21ad1ee4400e28009732a04a80862f
Author: Adrian Bunk <bu...@st...>
Date: Sun Apr 9 22:29:17 2006 -0700
[IPV6]: Unexport secure_ipv6_port_ephemeral
This patch removes the unused EXPORT_SYMBOL(secure_ipv6_port_ephemeral).
Signed-off-by: Adrian Bunk <bu...@st...>
Signed-off-by: David S. Miller <da...@da...>
commit 503e4faad18baa62bb818537b920ad939749823e
Author: Brian Haley <bri...@hp...>
Date: Fri Apr 7 15:00:06 2006 -0700
[NETFILTER]: Fix build with CONFIG_NETFILTER=y/m on IA64
Can't build with CONFIG_NETFILTER=y/m on IA64, there's a missing
#include in net/ipv6/netfilter.c
net/ipv6/netfilter.c: In function `nf_ip6_checksum':
net/ipv6/netfilter.c:92: warning: implicit declaration of function
`csum_ipv6_magic'
Signed-off-by: Brian Haley <bri...@hp...>
Signed-off-by: David S. Miller <da...@da...>
commit 77d04bd957ddca9d48a664e28b40f33993f4550e
Author: Andrew Morton <ak...@os...>
Date: Fri Apr 7 14:52:59 2006 -0700
[NET]: More kzalloc conversions.
Signed-off-by: David S. Miller <da...@da...>
commit 31380de95cc3183bbb379339e67f83d69e56fbd6
Author: Paolo 'Blaisorblade' Giarrusso <bla...@ya...>
Date: Thu Apr 6 22:38:28 2006 -0700
[NET] kzalloc: use in alloc_netdev
Noticed this use, fixed it.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <bla...@ya...>
Signed-off-by: David S. Miller <da...@da...>
commit 83b950c89c8cc0dcc1b079c638be25915c9945f1
Author: Jamal Hadi Salim <ha...@cy...>
Date: Thu Apr 6 22:24:22 2006 -0700
[PKT_SCHED] act_police: Rename methods.
Rename policer specific _generic_ methods to be specific to
_act_police_
Signed-off-by: Jamal Hadi Salim <ha...@cy...>
Signed-off-by: David S. Miller <da...@da...>
commit bbadf503d7c7e6efe0a4cd731f8855ba08276215
Author: Michael Chan <mc...@br...>
Date: Thu Apr 6 21:46:34 2006 -0700
[TG3]: Speed up SRAM access (2nd version)
Speed up SRAM read and write functions if possible by using MMIO
instead of config. cycles. With this change, the post reset signature
done at the end of D3 power change must now be moved before the D3
power change.
IBM reported a problem on powerpc blades during ethtool self test that
was caused by the memory test taking excessively long. Config. cycles
are very slow on powerpc and the memory test can take more than 10
seconds to complete using config. cycles.
David Miller informed me that an earlier version of the patch caused
problems on sparc64 systems with built-in tg3 chips. This version
fixes the problem by excluding all SUN built-in tg3 chips from doing
MMIO SRAM access.
TG3_FLAG_EEPROM_WRITE_PROT is also set unconditionally when
TG3_FLG2_SUN_570X is set. This should be sane as all SUN chips are
built-in and do not require Vaux switching.
Signed-off-by: Michael Chan <mc...@br...>
Signed-off-by: David S. Miller <da...@da...>
commit d2d746f83b74022a50d28f7f0f496842c9cde330
Author: Michael Chan <mc...@br...>
Date: Thu Apr 6 21:45:39 2006 -0700
[TG3]: Kill some less useful flags
Kill the TG3_FLAG_NO_{TX|RX}_PSEUDO_CSUM flags because they are not
very useful. This will free up some bits for new flags.
Signed-off-by: Michael Chan <mc...@br...>
Signed-off-by: David S. Miller <da...@da...>
commit 7a43c9955178f1cc88374abe1186c0f2ef21e040
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 16:16:51 2006 -0700
[NETFILTER]: H.323 helper: remove changelog
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 96f6bf82ea3abc77d255d5d554df5f349651f6de
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:19:24 2006 -0700
[NETFILTER]: Convert conntrack/ipt_REJECT to new checksumming functions
Besides removing lots of duplicate code, all converted users benefit
from improved HW checksum error handling. Tested with and without HW
checksums in almost all combinations.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 422c346fad806e2abaeffac686860ebc98dfe33e
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:18:43 2006 -0700
[NETFILTER]: Add address family specific checksum helpers
Add checksum operation which takes care of verifying the checksum and
dealing with HW checksum errors and avoids multiple checksum
operations by setting ip_summed to CHECKSUM_UNNECESSARY after
successful verification.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit bce8032ef3cc58170ab3550e9e271dba7b4c4764
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:18:09 2006 -0700
[NETFILTER]: Introduce infrastructure for address family specific operations
Change the queue rerouter intrastructure to a generic usable
infrastructure for address family specific operations as a base for
some cleanups.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit a0aed49bdb2bbb4234789f241cffb607fd2e213d
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:17:27 2006 -0700
[NETFILTER]: Fix IP_NF_CONNTRACK_NETLINK dependency
When NAT is built as a module, ip_conntrack_netlink can not be linked
statically.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit a0b7db5e86d30f470dc1849f4fa44ff77a813091
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:15:33 2006 -0700
[NETFILTER]: H.323 helper: add parameter 'default_rrq_ttl'
default_rrq_ttl is used when no TTL is included in the RRQ.
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 51d42f5e4ee43fc98aa0c7a2a104808bb602276f
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:14:59 2006 -0700
[NETFILTER]: H.323 helper: make get_h245_addr() static
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 0f249685fde399c01e51a63f2bdfe4a667db083c
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:14:11 2006 -0700
[NETFILTER]: H.323 helper: change EXPORT_SYMBOL to EXPORT_SYMBOL_GPL
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 48bfee5fad0e46f4f18d46285efceba39e897482
Author: Jing Min Zhao <zha...@us...>
Date: Thu Apr 6 14:13:42 2006 -0700
[NETFILTER]: H.323 helper: move some function prototypes to ip_conntrack_h323.h
Move prototypes of NAT callbacks to ip_conntrack_h323.h. Because the
use of typedefs as arguments, some header files need to be moved as
well.
Signed-off-by: Jing Min Zhao <zha...@us...>
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 32292a7ff1d9306841a8da6ea286847b1070cc6a
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:11:30 2006 -0700
[NETFILTER]: Fix section mismatch warnings
Fix section mismatch warnings caused by netfilter's init_or_cleanup
functions used in many places by splitting the init from the cleanup
parts.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 964ddaa10de8f3aeed12bc2a30726514ff309e64
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:09:49 2006 -0700
[NETFILTER]: Clean up hook registration
Clean up hook registration by makeing use of the new mass registration and
unregistration helpers.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 972d1cb1427946f4980240363aac4e73fb375290
Author: Patrick McHardy <ka...@tr...>
Date: Thu Apr 6 14:09:12 2006 -0700
[NETFILTER]: Add helper functions for mass hook registration/unregistration
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 137dc0233fba0bfa19679bdd96eb104f0e659c5a
Author: Samuel Ortiz <sam...@no...>
Date: Wed Apr 5 22:39:14 2006 -0700
[IRDA]: Support for Sigmatel STIR421x chip
This patch enables support for the Sigmatel's STIR421x IrDA chip.
Once patched with Sigmatel's firmware, this chip "almost" follows the
USB-IrDA spec. Thus this patch is against irda-usb.[ch].
The code has been tested by Nick Fedchik on an STIR4210 chipset based
dongle.
Signed-off-by: Samuel Ortiz <sam...@no...>
Signed-off-by: David S. Miller <da...@da...>
commit c1e14a6ea2ae34993f431d2eb9a0f228ac3574eb
Author: Linus Walleij <tr...@df...>
Date: Wed Apr 5 22:33:59 2006 -0700
[IRDA]: smcinit merged into smsc-ircc driver
This patch integrates the smcinit code into the smsc-ircc driver.
Some laptops have their smsc-ircc chip not properly configured by the
BIOS and needs some preconfiguration. Currently, this can be done from
userspace with smcinit, a utility that comes with the irda-utils
package. It messes with ioports and PCI settings, from userspace. Now
with this patch, if we happen to be on one of the known to be faulty
laptops, we preconfigure the chip from the driver.
Patch from Linus Walleij <tr...@df...>
Signed-off-by: Samuel Ortiz <sam...@no...>
Signed-off-by: David S. Miller <da...@da...>
commit 45af08be6d120690d379cb8421ebaf9d9f86ba52
Author: Herbert Xu <he...@go...>
Date: Wed Apr 5 22:31:19 2006 -0700
[INET]: Use port unreachable instead of proto for tunnels
This patch changes GRE and SIT to generate port unreachable instead of
protocol unreachable errors when we can't find a matching tunnel for a
packet.
This removes the ambiguity as to whether the error is caused by no
tunnel being found or by the lack of support for the given tunnel
type.
Signed-off-by: Herbert Xu <he...@go...>
Signed-off-by: David S. Miller <da...@da...>
commit cdee5751bf91d02616aaf30a5affef56105e3b79
Author: Eric Sesterhenn <sna...@gm...>
Date: Wed Apr 5 22:28:14 2006 -0700
[BLUETOOTH] sco: Possible double free.
this fixes coverity bug id #1068.
hci_send_sco() frees skb if (skb->len > hdev->sco_mtu).
Since it returns a negative error value only in this case, we
can directly return here.
Signed-off-by: Eric Sesterhenn <sna...@gm...>
Signed-off-by: David S. Miller <da...@da...>
commit ad96b485b4224785bcca478cfba02d06c46ee575
Author: Adrian Bunk <bu...@st...>
Date: Wed Apr 5 22:21:04 2006 -0700
[TG3]: Fix a memory leak.
This patch fixes a memory leak (buf wasn't freed) spotted by the
Coverity checker.
Signed-off-by: Adrian Bunk <bu...@st...>
Signed-off-by: David S. Miller <da...@da...>
commit e3a5cd9edff9a7a20de3c88c9d479704da98fb85
Author: Adrian Bunk <bu...@st...>
Date: Wed Apr 5 22:19:47 2006 -0700
[NET]: Fix an off-by-21-or-49 error.
This patch fixes an off-by-21-or-49 error ;-) spotted by the Coverity
checker.
Signed-off-by: Adrian Bunk <bu...@st...>
Signed-off-by: David S. Miller <da...@da...>
commit d938ab44c0c5418bb74a97b422a070e2cdccce22
Author: Randy Dunlap <rd...@xe...>
Date: Tue Apr 4 20:11:56 2006 -0700
[NET] netconsole: set .name in struct console
Set .name in netconsole's struct console to identify the
struct's owner.
Signed-off-by: Randy Dunlap <rd...@xe...>
Acked-by: Matt Mackall <mp...@se...>
Signed-off-by: David S. Miller <da...@da...>
commit 50fba2aa7cefa6b0e1768cb350c9e69042320c03
Author: Herbert Xu <he...@go...>
Date: Tue Apr 4 13:50:45 2006 -0700
[INET]: Move no-tunnel ICMP error to tunnel4/tunnel6
This patch moves the sending of ICMP messages when there are no IPv4/IPv6
tunnels present to tunnel4/tunnel6 respectively. Please note that for now
if xfrm4_tunnel/xfrm6_tunnel is loaded then no ICMP messages will ever be
sent. This is similar to how we handle AH/ESP/IPCOMP.
This move fixes the bug where we always send an ICMP message when there is
no ip6_tunnel device present for a given packet even if it is later handled
by IPsec. It also causes ICMP messages to be sent when no IPIP tunnel is
present.
I've decided to use the "port unreachable" ICMP message over the current
value of "address unreachable" (and "protocol unreachable" by GRE) because
it is not ambiguous unlike the other ones which can be triggered by other
conditions. There seems to be no standard specifying what value must be
used so this change should be OK. In fact we should change GRE to use
this value as well.
Incidentally, this patch also fixes a fairly serious bug in xfrm6_tunnel
where we don't check whether the embedded IPv6 header is present before
dereferencing it for the inside source address.
This patch is inspired by a previous patch by Hugo Santos <hs...@av...>.
Signed-off-by: Herbert Xu <he...@go...>
Signed-off-by: David S. Miller <da...@da...>
commit 2e2f7aefa8a8ba4adb6ecee8cbb43fbe9ca4cc89
Author: Patrick McHardy <ka...@tr...>
Date: Tue Apr 4 13:42:35 2006 -0700
[NETFILTER]: Fix fragmentation issues with bridge netfilter
The conntrack code doesn't do re-fragmentation of defragmented packets
anymore but relies on fragmentation in the IP layer. Purely bridged
packets don't pass through the IP layer, so the bridge netfilter code
needs to take care of fragmentation itself.
Signed-off-by: Patrick McHardy <ka...@tr...>
Signed-off-by: David S. Miller <da...@da...>
commit 550e29bc96e6f1ced2bca82dace197b009434367
Author: Robert Olsson <rob...@it...>
Date: Tue Apr 4 12:53:35 2006 -0700
[FIB_TRIE]: Fix leaf freeing.
Seems like leaf (end-nodes) has been freed by __tnode_free_rcu and not
by __leaf_free_rcu. This fixes the problem. Only tnode_free is now
used which checks for appropriate node type. free_leaf can be removed.
Signed-off-by: Robert Olsson <rob...@it...>
Signed-off-by: David S. Miller <da...@da...>
commit 8bf4b8a1083694d5aac292f92705ddd3aec29be6
Author: Herbert Xu <he...@go...>
Date: Tue Apr 4 12:51:05 2006 -0700
[IPSEC]: Check x->encap before dereferencing it
We need to dereference x->encap before dereferencing it for encap_type.
If it's absent then the encap_type is zero.
Signed-off-by: Herbert Xu <he...@go...>
Signed-off-by: David S. Miller <da...@da...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-09 20:15:58
|
Documentation/x86_64/boot-options.txt | 5
arch/i386/kernel/acpi/boot.c | 5
arch/i386/kernel/apic.c | 23 ++-
arch/i386/kernel/cpu/cpufreq/powernow-k8.c | 18 ++-
arch/i386/kernel/mpparse.c | 21 ---
arch/i386/kernel/reboot_fixups.c | 2
arch/i386/kernel/setup.c | 36 +++++-
arch/i386/mm/init.c | 2
arch/i386/pci/direct.c | 9 +
arch/i386/pci/mmconfig.c | 67 +++++++----
arch/x86_64/Kconfig | 5
arch/x86_64/Makefile | 24 ++--
arch/x86_64/defconfig | 42 ++++---
arch/x86_64/ia32/ia32entry.S | 23 ++-
arch/x86_64/kernel/aperture.c | 2
arch/x86_64/kernel/e820.c | 36 +++++-
arch/x86_64/kernel/entry.S | 28 +---
arch/x86_64/kernel/mce.c | 8 +
arch/x86_64/kernel/nmi.c | 7 +
arch/x86_64/kernel/pci-dma.c | 2
arch/x86_64/kernel/process.c | 10 +
arch/x86_64/kernel/setup.c | 4
arch/x86_64/kernel/time.c | 4
arch/x86_64/kernel/vmlinux.lds.S | 2
arch/x86_64/kernel/x8664_ksyms.c | 2
arch/x86_64/mm/init.c | 37 +++++-
arch/x86_64/mm/numa.c | 46 ++++++-
arch/x86_64/mm/srat.c | 170 +++++++++++++++++++++++++++--
arch/x86_64/pci/mmconfig.c | 53 ++++++---
drivers/acpi/Kconfig | 2
include/asm-i386/apic.h | 2
include/asm-i386/e820.h | 4
include/asm-i386/hpet.h | 1
include/asm-x86_64/e820.h | 3
include/asm-x86_64/hpet.h | 2
include/asm-x86_64/ia32_unistd.h | 2
include/asm-x86_64/mce.h | 7 +
include/asm-x86_64/numa.h | 2
include/asm-x86_64/numnodes.h | 2
include/linux/bootmem.h | 1
include/linux/init.h | 3
include/linux/jiffies.h | 6 +
include/linux/memory_hotplug.h | 14 +-
kernel/timer.c | 2
mm/bootmem.c | 9 +
security/selinux/xfrm.c | 4
46 files changed, 575 insertions(+), 184 deletions(-)
New commits:
commit dce6b2b7b90c60ec4a37aed710cc36a7906bfb0d
Merge: 686e8d1... ebe9c64...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Sun Apr 9 22:13:52 2006 +0200
Merge branch 'vax4linus'
commit ebe9c645f0738c1254e7b6208e3dec8a1b9be2ba
Merge: 6b61940... 6764472...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Sun Apr 9 22:12:03 2006 +0200
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6
commit 67644726317a8274be4a3d0ef85b9ccebaa90304
Author: Dave Jones <da...@re...>
Date: Sun Apr 2 23:34:19 2006 -0700
[SELINUX] Fix build after ipsec decap state changes.
security/selinux/xfrm.c: In function 'selinux_socket_getpeer_dgram':
security/selinux/xfrm.c:284: error: 'struct sec_path' has no member named 'x'
security/selinux/xfrm.c: In function 'selinux_xfrm_sock_rcv_skb':
security/selinux/xfrm.c:317: error: 'struct sec_path' has no member named 'x'
Signed-off-by: Dave Jones <da...@re...>
Signed-off-by: David S. Miller <da...@da...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 66004a6ca23f2a2408b32cbe27fda0389fb8f9dc
Author: Linus Torvalds <tor...@g5...>
Date: Sun Apr 9 12:14:02 2006 -0700
Move request_standard_resources() back to before PCI probing
This effectively undoes the PCI resource allocation changes done in
commit b408cbc704352eccee301e1103b23203ba1c3a0e, but leaves the cleanups
of that commit in place.
We're going back to marking the resources reported by e820 busy _before_
doing PCI probing, so that any PCI resource that clashes with the BIOS-
reported memory map will be reloacted to a non-clashing area.
The reason? Larry Finger reports that his laptop has the cardbus
controller set up by the BIOS so that it conflicts with the e820 memory
map, and needs to be relocated. See
http://bugzilla.kernel.org/show_bug.cgi?id=6337
for more details.
We'll have to work out how to handle the fbcon problem that caused that
commit in the first place in some other way.
Cc: Ivan Kokshaysky <in...@ju...>
Cc: Greg Kroah-Hartman <gr...@su...>
Cc: Antonino A. Daplas <ad...@po...>
Cc: <bj...@lu...>
Tested-by: Larry Finger <Lar...@lw...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit b8feb47f992d314c956add15c1118430120635bb
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:34 2006 +0200
[PATCH] x86_64: Update 32-bit system call table
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 67d53ea5a3d42aadeb1584e757ca4660c0e8a810
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:31 2006 +0200
[PATCH] x86_64: Eliminate IA32_NR_syscalls define
Or rather compute it based on the table length automatically.
This also has the intended side effect of not warning for new system calls
anymore.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit bbd3aff89d4b34ef17a748e4c001ecc5b43e3e55
Author: Sam Ravnborg <sa...@ra...>
Date: Fri Apr 7 19:50:28 2006 +0200
[PATCH] x86_64: fix CONFIG_REORDER
Fix CONFIG_REORDER.
The value of cflags-y was assined to CFLAGS before cflags-y was assigned
the value used for CONFIG_REORDER.
Use cflags-y for all CFLAGS options in the Makefile to avoid this
happening again.
Signed-off-by: Sam Ravnborg <sa...@ra...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 97c2803c9c694cafbd9f5e43a25903e0abf25188
Author: John Blackwood <joh...@cc...>
Date: Fri Apr 7 19:50:25 2006 +0200
[PATCH] x86_64: Plug GS leak in arch_prctl()
In linux-2.6.16, we have noticed a problem where the gs base value
returned from an arch_prtcl(ARCH_GET_GS, ...) call will be incorrect if:
- the current/calling task has NOT set its own gs base yet to a
non-zero value,
- some other task that ran on the same processor previously set their
own gs base to a non-zero value.
In this situation, the ARCH_GET_GS code will read and return the
MSR_KERNEL_GS_BASE msr register.
However, since the __switch_to() code does NOT load/zero the
MSR_KERNEL_GS_BASE register when the task that is switched IN has a zero
next->gs value, the caller of arch_prctl(ARCH_GET_GS, ...) will get back
the value of some previous tasks's gs base value instead of 0.
Change the arch_prctl() ARCH_GET_GS code to only read and return
the MSR_KERNEL_GS_BASE msr register if the 'gs' register of the calling
task is non-zero.
Side note: Since in addition to using arch_prctl(ARCH_SET_GS, ...),
a task can also setup a gs base value by using modify_ldt() and write
an index value into 'gs' from user space, the patch below reads
'gs' instead of using thread.gs, since in the modify_ldt() case,
the thread.gs value will be 0, and incorrect value would be returned
(the task->thread.gs value).
When the user has not set its own gs base value and the 'gs'
register is zero, then the MSR_KERNEL_GS_BASE register will not be
read and a value of zero will be returned by reading and returning
'task->thread.gs'.
The first patch shown below is an attempt at implementing this
approach.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit e48c4729d23a026f3711d5e36add5cce894b4913
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:21 2006 +0200
[PATCH] i386: Remove printk about reboot fixups at reboot
Printk doesn't have any value
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit b20367a6c2a0cd937cb1f0a8cf848f1402fef99c
Author: Jordan Hargrave <jor...@de...>
Date: Fri Apr 7 19:50:18 2006 +0200
[PATCH] x86_64: Fix drift with HPET timer enabled
If the HPET timer is enabled, the clock can drift by ~3 seconds a day.
This is due to the HPET timer not being initialized with the correct
setting (still using PIT count).
If HZ changes, this drift can become even more pronounced.
HPET patch initializes tick_nsec with correct tick_nsec settings for
HPET timer.
Vojtech comments:
"It's not entirely correct (it assumes the HPET ticks totally
exactly), but it's significantly better than assuming the PIT error
there."
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 49c93e84d8b2d602a07c302c7e3cd4fa09095fbb
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:15 2006 +0200
[PATCH] i386/x86-64: Return defined error value for bad PCI config space accesses
Mostly to get better handling when a extended config space
access has to fallback to Type1.
Cc: gr...@su...
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 8c30b1a74aed4041f183e183a149b7dfbdc6c20e
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:12 2006 +0200
[PATCH] i386/x86_64: Check if MCFG works for the first 16 busses
Previously only the first bus would be checked against Type 1.
Why 16? Checking all would need too much memory and we
can assume that systems with more than 16 busses have better than
average quality BIOS.
This is an additional defense against bad MCFG tables.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit e405d067298b2b960bf20318e91ed842157c65bc
Author: Ravikiran G Thirumalai <ki...@sc...>
Date: Fri Apr 7 19:50:09 2006 +0200
[PATCH] x86_64: Fixup read_mostly section on internode cache line size for vSMP
Fixup the read mostly section to start at internode cacheline boundary.
Signed-off-by: Ravikiran Thirumalai <ki...@sc...>
Signed-off-by: Shai Fultheim <sh...@sc...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 3d34ee6891e274dfb6a22930546d37738cdbe9c4
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:06 2006 +0200
[PATCH] x86_64: Don't return error for HPET initialization in initcall
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit ac04dcaf6f567307fbeef9c3c1fff35280e53f02
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:03 2006 +0200
[PATCH] x86_64: Don't export strlen twice
Fix
WARNING: vmlinux: 'strlen' exported twice. Previous export was in vmlinux
Reported by Mats Johannesson
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 7bf36bbc5e0c09271f9efe22162f8cc3f8ebd3d2
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:00 2006 +0200
[PATCH] x86_64: When user could have changed RIP always force IRET
Intel EM64T CPUs handle uncanonical return addresses differently
from AMD CPUs.
The exception is reported in the SYSRET, not the next instruction.
This leads to the kernel exception handler running on the user stack
with the wrong GS because the kernel didn't expect exceptions
on this instruction.
This version of the patch has the teething problems that plagued an earlier
version fixed.
This is CVE-2006-0744
Thanks to Ernie Petrides and Asit B. Mallick for analysis and initial
patches.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 553f265fe883a23502ee351845f09334790f18b8
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:57 2006 +0200
[PATCH] x86_64: Don't run NMI watchdog during machine checks
Machine checks can stall the machine for a long time and
it's not good to trigger the nmi watchdog during that.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit be56db6186999a8571ae480cf2b929578f6dfd68
Author: Dave Hansen <hav...@us...>
Date: Fri Apr 7 19:49:54 2006 +0200
[PATCH] x86_64: extra NODES_SHIFT definition
The generic linux/numa.h file defines NODES_SHIFT to 0 in case
the architecture did not.
Every architecture which has a NUMA config option defines
NODES_SHIFT in its asm-$ARCH headers, but only if NUMA is
enabled, except for x86_64.
This should make it like all the rest.
Signed-off-by: Dave Hansen <hav...@us...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 4211a30349e8d2b724cfb4ce2584604f5e59c299
Author: Jacob Shin <jac...@am...>
Date: Fri Apr 7 19:49:51 2006 +0200
[PATCH] x86_64: Proper null pointer check in powernow_k8_get
This prevents crashes on dual core system when enough ticks are lost.
Replaces earlier patch by me.
Cc: Dave Jones <da...@re...>
Signed-off-by: Thomas Renninger <tr...@su...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit d7fa706ce2c29cb751c15ca00f3aa7b223e3c9f0
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:48 2006 +0200
[PATCH] x86_64: Revert earlier powernow-k8 change
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 95d769aaf47abfc77b600631403ff5af6c990cff
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:45 2006 +0200
[PATCH] i386: Consolidate modern APIC handling
AMD systems have a modern APIC that supports 8 bit IDs, but
don't have a XAPIC version number. Add a new "modern_apic"
subfunction that handles this correctly and use it (nearly)
everywhere where XAPIC is tested for.
I removed one wart: the code specified that external APICs
would use an 8bit APIC ID. But I checked a real 82093 data sheet
and it says clearly that they only use 4bit. So I removed
this special case since it would a bit awkward to implement now.
I removed the valid APIC tests in mptable parsing completely. On any modern
system they only check against the full field width (8bit) anyways
and are no-ops. This also fixes them doing the wrong thing
on >8 core Opterons.
This makes i386 boot again on 16 core Opterons.
Cc: Ingo Molnar <mi...@el...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit d1530d82e02fd96d4634a6d6f6538c8b778c43af
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:42 2006 +0200
[PATCH] x86_64: Clear APIC feature bit when local APIC is disabled
Needed for other checks later in ACPI.
Pointed out by Len Brown
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit d3b6a349d233aecf2c52f7f4c150ca09f684f2d8
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:39 2006 +0200
[PATCH] x86-64/i386: Don't process APICs/IO-APICs in ACPI when APIC is disabled.
When nolapic was passed or the local APIC was disabled
for another reason ACPI would still parse the IO-APICs
until these were explicitely disabled with noapic.
Usually this resulted in a non booting configuration unless
"nolapic noapic" was used.
I also disabled the local APIC parsing in this case, although
that's only cosmetic (suppresses a few printks)
This hopefully makes nolapic work in all cases.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit ec0f08eeea6ac1d8c925f47e3677e4c985fd8f63
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:36 2006 +0200
[PATCH] x86_64: Don't sanity check Type 1 PCI bus access on newer systems
Horus systems don't have anything on bus 0 which makes
the Type 1 sanity checks fail. Use the DMI BIOS year to
check for newer systems and always assume Type 1 works on them.
I used 2001 as an pretty arbitary cutoff year.
Cc: gr...@su...
Cc: Navin Boppuri <nav...@ne...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit fa47dd0ba303599f8adf8d8336ed2fb74efc47c5
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:33 2006 +0200
[PATCH] x86_64: Fix compilation with CONFIG_PCI=n / allnoconfig
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 946f2ee5c7312e8acac4f3ab6629e7e2d36a3646
Author: Arjan van de Ven <ar...@li...>
Date: Fri Apr 7 19:49:30 2006 +0200
[PATCH] i386/x86-64: Check that MCFG points to an e820 reserved area
This patch introduces a user for the e820_all_mapped function:
There have been several machines that don't have a working MMCONFIG,
often because of a buggy MCFG table in the ACPI bios. This patch adds a
simple sanity check that detects a whole bunch of these cases, and when
it detects it, linux now boots rather than crash-and-burns.
The accuracy of this detection can in principle be improved if there was
a "is this entire range in e820 with THIS attribute", but no such
function exist and the complexity needed for this is not really worth
it; this simple check already catches most cases anyway.
Signed-off-by: Arjan van de Ven <ar...@li...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 952223683ec989e86328c24808fdb962c4dbeb0a
Author: Arjan van de Ven <ar...@li...>
Date: Fri Apr 7 19:49:27 2006 +0200
[PATCH] x86_64: Introduce e820_all_mapped
Introduce a e820_all_mapped() function which checks if the entire range
<start,end> is mapped with type.
This is done by moving the local start variable to the end of each
known-good region; if at the end of the function the start address is
still before end, there must be a part that's not of the correct type;
otherwise it's a good region.
Signed-off-by: Arjan van de Ven <ar...@li...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit eee5a9fa63c97366cdea6ab3aa2ed9e3601812d0
Author: Arjan van de Ven <ar...@li...>
Date: Fri Apr 7 19:49:24 2006 +0200
[PATCH] x86_64: Rename e820_mapped to e820_any_mapped
Rename e820_mapped to e820_any_mapped since it tests if any part of the
range is mapped according to the type.
Later steps will introduce e820_all_mapped which will check if the
entire range is mapped with the type. Both have their merit.
Signed-off-by: Arjan van de Ven <ar...@li...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit a8062231d80239cf3405982858c02aea21a6066a
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:21 2006 +0200
[PATCH] x86_64: Handle empty PXMs that only contain hotplug memory
The node setup code would try to allocate the node metadata in the node
itself, but that fails if there is no memory in there.
This can happen with memory hotplug when the hotplug area defines an so
far empty node.
Now use bootmem to try to allocate the mem_map in other nodes.
And if it fails don't panic, but just ignore the node.
To make this work I added a new __alloc_bootmem_nopanic function that
does what its name implies.
TBD should try to use nearby nodes here. Currently we just use any.
It's hard to do it better because bootmem doesn't have proper fallback
lists yet.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 68a3a7feb08f960095072f28ec20f7900793c506
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:18 2006 +0200
[PATCH] x86_64: Reserve SRAT hotadd memory on x86-64
From: Keith Mannthey, Andi Kleen
Implement memory hotadd without sparsemem. The memory in the SRAT
hotadd area is just preserved instead and can be activated later.
There are a few restrictions:
- Only one continuous hotadd area allowed per node
The main problem is dealing with the many buggy SRAT tables
that are out there. The strategy here is to reject anything
suspicious.
Originally from Keith Mannthey, with several hacks and changes by AK
and also contributions from Andrew Morton
[ TBD: Problems pointed out by KAMEZAWA Hiroyuki <kam...@jp...>:
1) Goto's rebuild_zonelist patch will not work if CONFIG_MEMORY_HOTPLUG=n.
Rebuilding zonelist is necessary when the system has just memory <
4G at boot, and hot add memory > 4G. because x86_64 has DMA32,
ZONE_NORAML is not included into zonelist at boot time if system
doesn't have memory >4G at boot.
[AK: should just force the higher zones at boot time when SRAT tells us]
2) zone and node's spanned_pages and present_pages are not incremented.
They should be.
For example, our server (ia64/Fujitsu PrimeQuest) can equip memory
from 4G to 1T(maybe 2T in future), and SRAT will *always* say we have
possible 1T +memory. (Microsoft requires "write all possible memory
in SRAT") When we reserve memmap for possible 1T memory, Linux will
not work well in +minimum 4G configuraion ;)
[AK: needs limiting to 5-10% of max memory]
]
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 9d99aaa31f5994d1923c3713ce9144c4c42332e1
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:15 2006 +0200
[PATCH] x86_64: Support memory hotadd without sparsemem
Memory hotadd doesn't need SPARSEMEM, but can be handled by just preallocating
mem_maps. This only needs some untangling of ifdefs to enable the necessary
code even without SPARSEMEM.
Originally from Keith Mannthey, hacked by AK.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 805e8c03c9ea9bdb402a36341e02ec24825d5417
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:12 2006 +0200
[PATCH] x86_64: Clean up execve path
Just call IRET always, no need for any special cases.
Needed for the next bug fix.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 903fcc608e9f531749024172277dc2fd15d5a587
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:09 2006 +0200
[PATCH] x86_64: Update defconfig
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt
index 1921353..f2cd6ef 100644
--- a/Documentation/x86_64/boot-options.txt
+++ b/Documentation/x86_64/boot-options.txt
@@ -151,6 +151,11 @@ NUMA
numa=fake=X Fake X nodes and ignore NUMA setup of the actual machine.
+ numa=hotadd=percent
+ Only allow hotadd memory to preallocate page structures upto
+ percent of already available memory.
+ numa=hotadd=0 will disable hotadd memory.
+
ACPI
acpi=off Don't enable ACPI
diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 0330661..8dab352 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -215,7 +215,7 @@ static int __init acpi_parse_madt(unsign
{
struct acpi_table_madt *madt = NULL;
- if (!phys_addr || !size)
+ if (!phys_addr || !size || !cpu_has_apic)
return -EINVAL;
madt = (struct acpi_table_madt *)__acpi_map_table(phys_addr, size);
@@ -751,6 +751,9 @@ static int __init acpi_parse_madt_ioapic
return -ENODEV;
}
+ if (!cpu_has_apic)
+ return -ENODEV;
+
/*
* if "noapic" boot option, don't look for IO-APICs
*/
diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 6273bf7..254cee9 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -62,6 +62,18 @@ int apic_verbosity;
static void apic_pm_activate(void);
+int modern_apic(void)
+{
+ unsigned int lvr, version;
+ /* AMD systems use old APIC versions, so check the CPU */
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
+ boot_cpu_data.x86 >= 0xf)
+ return 1;
+ lvr = apic_read(APIC_LVR);
+ version = GET_APIC_VERSION(lvr);
+ return version >= 0x14;
+}
+
/*
* 'what should we do if we get a hw irq event on an illegal vector'.
* each architecture has to answer this themselves.
@@ -119,10 +131,7 @@ void enable_NMI_through_LVT0 (void * dum
int get_physical_broadcast(void)
{
- unsigned int lvr, version;
- lvr = apic_read(APIC_LVR);
- version = GET_APIC_VERSION(lvr);
- if (!APIC_INTEGRATED(version) || version >= 0x14)
+ if (modern_apic())
return 0xff;
else
return 0xf;
@@ -349,9 +358,9 @@ int __init verify_local_APIC(void)
void __init sync_Arb_IDs(void)
{
- /* Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 */
- unsigned int ver = GET_APIC_VERSION(apic_read(APIC_LVR));
- if (ver >= 0x14) /* P4 or higher */
+ /* Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1
+ And not needed on AMD */
+ if (modern_apic())
return;
/*
* Wait for idle.
diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
index 712a26b..7c0e160 100644
--- a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
@@ -46,7 +46,7 @@
#define PFX "powernow-k8: "
#define BFX PFX "BIOS error: "
-#define VERSION "version 1.60.1"
+#define VERSION "version 1.60.2"
#include "powernow-k8.h"
/* serialize freq changes */
@@ -55,7 +55,7 @@ static DEFINE_MUTEX(fidvid_mutex);
static struct powernow_k8_data *powernow_data[NR_CPUS];
#ifndef CONFIG_SMP
-static cpumask_t cpu_core_map[1] = { CPU_MASK_ALL };
+static cpumask_t cpu_core_map[1];
#endif
/* Return a frequency in MHz, given an input fid */
@@ -910,6 +910,9 @@ static int powernowk8_target(struct cpuf
unsigned int newstate;
int ret = -EIO;
+ if (!data)
+ return -EINVAL;
+
/* only run on specific CPU from here on */
oldmask = current->cpus_allowed;
set_cpus_allowed(current, cpumask_of_cpu(pol->cpu));
@@ -969,6 +972,9 @@ static int powernowk8_verify(struct cpuf
{
struct powernow_k8_data *data = powernow_data[pol->cpu];
+ if (!data)
+ return -EINVAL;
+
return cpufreq_frequency_table_verify(pol, data->powernow_table);
}
@@ -977,7 +983,7 @@ static int __cpuinit powernowk8_cpu_init
{
struct powernow_k8_data *data;
cpumask_t oldmask = CPU_MASK_ALL;
- int rc, i;
+ int rc;
if (!cpu_online(pol->cpu))
return -ENODEV;
@@ -1063,8 +1069,7 @@ static int __cpuinit powernowk8_cpu_init
printk("cpu_init done, current fid 0x%x, vid 0x%x\n",
data->currfid, data->currvid);
- for_each_cpu_mask(i, cpu_core_map[pol->cpu])
- powernow_data[i] = data;
+ powernow_data[pol->cpu] = data;
return 0;
@@ -1104,6 +1109,9 @@ static unsigned int powernowk8_get (unsi
if (!data)
return -EINVAL;
+ if (!data)
+ return -EINVAL;
+
set_cpus_allowed(current, cpumask_of_cpu(cpu));
if (smp_processor_id() != cpu) {
printk(KERN_ERR PFX "limiting to CPU %d failed in powernowk8_get\n", cpu);
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index 8d8aa9d..db12017 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -110,21 +110,6 @@ static int __init mpf_checksum(unsigned
static int mpc_record;
static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __initdata;
-#ifdef CONFIG_X86_NUMAQ
-static int MP_valid_apicid(int apicid, int version)
-{
- return hweight_long(apicid & 0xf) == 1 && (apicid >> 4) != 0xf;
-}
-#else
-static int MP_valid_apicid(int apicid, int version)
-{
- if (version >= 0x14)
- return apicid < 0xff;
- else
- return apicid < 0xf;
-}
-#endif
-
static void __devinit MP_processor_info (struct mpc_config_processor *m)
{
int ver, apicid;
@@ -190,12 +175,6 @@ static void __devinit MP_processor_info
ver = m->mpc_apicver;
- if (!MP_valid_apicid(apicid, ver)) {
- printk(KERN_WARNING "Processor #%d INVALID. (Max ID: %d).\n",
- m->mpc_apicid, MAX_APICS);
- return;
- }
-
/*
* Validate version
*/
diff --git a/arch/i386/kernel/reboot_fixups.c b/arch/i386/kernel/reboot_fixups.c
index 10e21a4..99aab41 100644
--- a/arch/i386/kernel/reboot_fixups.c
+++ b/arch/i386/kernel/reboot_fixups.c
@@ -51,7 +51,5 @@ void mach_reboot_fixups(void)
cur->reboot_fixup(dev);
}
-
- printk(KERN_WARNING "No reboot fixup found for your hardware\n");
}
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index eacc3f0..80cb3b2 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -963,6 +963,36 @@ efi_memory_present_wrapper(unsigned long
return 0;
}
+ /*
+ * This function checks if the entire range <start,end> is mapped with type.
+ *
+ * Note: this function only works correct if the e820 table is sorted and
+ * not-overlapping, which is the case
+ */
+int __init
+e820_all_mapped(unsigned long start, unsigned long end, unsigned type)
+{
+ int i;
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+ if (type && ei->type != type)
+ continue;
+ /* is the region (part) in overlap with the current region ?*/
+ if (ei->addr >= end || ei->addr + ei->size <= start)
+ continue;
+ /* if the region is at the beginning of <start,end> we move
+ * start to the end of the region since it's ok until there
+ */
+ if (ei->addr <= start)
+ start = ei->addr + ei->size;
+ /* if start is now at or beyond end, we're done, full
+ * coverage */
+ if (start >= end)
+ return 1; /* we're done */
+ }
+ return 0;
+}
+
/*
* Find the highest page frame number we have available
*/
@@ -1317,8 +1347,8 @@ legacy_init_iomem_resources(struct resou
/*
* Request address space for all standard resources
*
- * This is called just before pcibios_assign_resources(), which is also
- * an fs_initcall, but is linked in later (in arch/i386/pci/i386.c).
+ * This is called just before pcibios_init(), which is also a
+ * subsys_initcall, but is linked in later (in arch/i386/pci/common.c).
*/
static int __init request_standard_resources(void)
{
@@ -1339,7 +1369,7 @@ static int __init request_standard_resou
return 0;
}
-fs_initcall(request_standard_resources);
+subsys_initcall(request_standard_resources);
static void __init register_memory(void)
{
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 9f66ac5..ae6534a 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -651,6 +651,7 @@ void __init mem_init(void)
* Specifically, in the case of x86, we will always add
* memory to the highmem for now.
*/
+#ifdef CONFIG_HOTPLUG_MEMORY
#ifndef CONFIG_NEED_MULTIPLE_NODES
int add_memory(u64 start, u64 size)
{
@@ -667,6 +668,7 @@ int remove_memory(u64 start, u64 size)
return -EINVAL;
}
#endif
+#endif
kmem_cache_t *pgd_cache;
kmem_cache_t *pmd_cache;
diff --git a/arch/i386/pci/direct.c b/arch/i386/pci/direct.c
index 99012b9..0659ced 100644
--- a/arch/i386/pci/direct.c
+++ b/arch/i386/pci/direct.c
@@ -4,6 +4,7 @@
#include <linux/pci.h>
#include <linux/init.h>
+#include <linux/dmi.h>
#include "pci.h"
/*
@@ -18,8 +19,10 @@ int pci_conf1_read(unsigned int seg, uns
{
unsigned long flags;
- if (!value || (bus > 255) || (devfn > 255) || (reg > 255))
+ if (!value || (bus > 255) || (devfn > 255) || (reg > 255)) {
+ *value = -1;
return -EINVAL;
+ }
spin_lock_irqsave(&pci_config_lock, flags);
@@ -188,6 +191,10 @@ static int __init pci_sanity_check(struc
if (pci_probe & PCI_NO_CHECKS)
return 1;
+ /* Assume Type 1 works for newer systems.
+ This handles machines that don't have anything on PCI Bus 0. */
+ if (dmi_get_year(DMI_BIOS_DATE) >= 2001)
+ return 1;
for (devfn = 0; devfn < 0x100; devfn++) {
if (o->read(0, 0, devfn, PCI_CLASS_DEVICE, 2, &x))
diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c
index 6137890..f77d7f8 100644
--- a/arch/i386/pci/mmconfig.c
+++ b/arch/i386/pci/mmconfig.c
@@ -12,14 +12,20 @@
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/acpi.h>
+#include <asm/e820.h>
#include "pci.h"
+#define MMCONFIG_APER_SIZE (256*1024*1024)
+
+/* Assume systems with more busses have correct MCFG */
+#define MAX_CHECK_BUS 16
+
#define mmcfg_virt_addr ((void __iomem *) fix_to_virt(FIX_PCIE_MCFG))
/* The base address of the last MMCONFIG device accessed */
static u32 mmcfg_last_accessed_device;
-static DECLARE_BITMAP(fallback_slots, 32);
+static DECLARE_BITMAP(fallback_slots, MAX_CHECK_BUS*32);
/*
* Functions for accessing PCI configuration space with MMCONFIG accesses
@@ -29,8 +35,8 @@ static u32 get_base_addr(unsigned int se
int cfg_num = -1;
struct acpi_table_mcfg_config *cfg;
- if (seg == 0 && bus == 0 &&
- test_bit(PCI_SLOT(devfn), fallback_slots))
+ if (seg == 0 && bus < MAX_CHECK_BUS &&
+ test_bit(PCI_SLOT(devfn) + 32*bus, fallback_slots))
return 0;
while (1) {
@@ -74,8 +80,10 @@ static int pci_mmcfg_read(unsigned int s
unsigned long flags;
u32 base;
- if (!value || (bus > 255) || (devfn > 255) || (reg > 4095))
+ if (!value || (bus > 255) || (devfn > 255) || (reg > 4095)) {
+ *value = -1;
return -EINVAL;
+ }
base = get_base_addr(seg, bus, devfn);
if (!base)
@@ -146,29 +154,34 @@ static struct pci_raw_ops pci_mmcfg = {
Normally this can be expressed in the MCFG by not listing them
and assigning suitable _SEGs, but this isn't implemented in some BIOS.
Instead try to discover all devices on bus 0 that are unreachable using MM
- and fallback for them.
- We only do this for bus 0/seg 0 */
+ and fallback for them. */
static __init void unreachable_devices(void)
{
- int i;
+ int i, k;
unsigned long flags;
- for (i = 0; i < 32; i++) {
- u32 val1;
- u32 addr;
-
- pci_conf1_read(0, 0, PCI_DEVFN(i, 0), 0, 4, &val1);
- if (val1 == 0xffffffff)
- continue;
-
- /* Locking probably not needed, but safer */
- spin_lock_irqsave(&pci_config_lock, flags);
- addr = get_base_addr(0, 0, PCI_DEVFN(i, 0));
- if (addr != 0)
- pci_exp_set_dev_base(addr, 0, PCI_DEVFN(i, 0));
- if (addr == 0 || readl((u32 __iomem *)mmcfg_virt_addr) != val1)
- set_bit(i, fallback_slots);
- spin_unlock_irqrestore(&pci_config_lock, flags);
+ for (k = 0; k < MAX_CHECK_BUS; k++) {
+ for (i = 0; i < 32; i++) {
+ u32 val1;
+ u32 addr;
+
+ pci_conf1_read(0, k, PCI_DEVFN(i, 0), 0, 4, &val1);
+ if (val1 == 0xffffffff)
+ continue;
+
+ /* Locking probably not needed, but safer */
+ spin_lock_irqsave(&pci_config_lock, flags);
+ addr = get_base_addr(0, k, PCI_DEVFN(i, 0));
+ if (addr != 0)
+ pci_exp_set_dev_base(addr, k, PCI_DEVFN(i, 0));
+ if (addr == 0 ||
+ readl((u32 __iomem *)mmcfg_virt_addr) != val1) {
+ set_bit(i, fallback_slots);
+ printk(KERN_NOTICE
+ "PCI: No mmconfig possible on %x:%x\n", k, i);
+ }
+ spin_unlock_irqrestore(&pci_config_lock, flags);
+ }
}
}
@@ -183,6 +196,14 @@ void __init pci_mmcfg_init(void)
(pci_mmcfg_config[0].base_address == 0))
return;
+ if (!e820_all_mapped(pci_mmcfg_config[0].base_address,
+ pci_mmcfg_config[0].base_address + MMCONFIG_APER_SIZE,
+ E820_RESERVED)) {
+ printk(KERN_ERR "PCI: BIOS Bug: MCFG area is not E820-reserved\n");
+ printk(KERN_ERR "PCI: Not using MMCONFIG.\n");
+ return;
+ }
+
printk(KERN_INFO "PCI: Using MMCONFIG\n");
raw_pci_ops = &pci_mmcfg;
pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF;
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 4310b4a..7df2fe1 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -136,6 +136,11 @@ config X86_L1_CACHE_SHIFT
default "7" if GENERIC_CPU || MPSC
default "6" if MK8
+config X86_INTERNODE_CACHE_BYTES
+ int
+ default "4096" if X86_VSMP
+ default X86_L1_CACHE_BYTES if !X86_VSMP
+
config X86_TSC
bool
default y
diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile
index 585fd4a..e573e2a 100644
--- a/arch/x86_64/Makefile
+++ b/arch/x86_64/Makefile
@@ -24,37 +24,37 @@
LDFLAGS := -m elf_x86_64
OBJCOPYFLAGS := -O binary -R .note -R .comment -S
LDFLAGS_vmlinux :=
-
CHECKFLAGS += -D__x86_64__ -m64
+cflags-y :=
cflags-$(CONFIG_MK8) += $(call cc-option,-march=k8)
cflags-$(CONFIG_MPSC) += $(call cc-option,-march=nocona)
cflags-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mtune=generic)
-CFLAGS += $(cflags-y)
-CFLAGS += -m64
-CFLAGS += -mno-red-zone
-CFLAGS += -mcmodel=kernel
-CFLAGS += -pipe
+cflags-y += -m64
+cflags-y += -mno-red-zone
+cflags-y += -mcmodel=kernel
+cflags-y += -pipe
cflags-$(CONFIG_REORDER) += -ffunction-sections
# this makes reading assembly source easier, but produces worse code
# actually it makes the kernel smaller too.
-CFLAGS += -fno-reorder-blocks
-CFLAGS += -Wno-sign-compare
+cflags-y += -fno-reorder-blocks
+cflags-y += -Wno-sign-compare
ifneq ($(CONFIG_UNWIND_INFO),y)
-CFLAGS += -fno-asynchronous-unwind-tables
+cflags-y += -fno-asynchronous-unwind-tables
endif
ifneq ($(CONFIG_DEBUG_INFO),y)
# -fweb shrinks the kernel a bit, but the difference is very small
# it also messes up debugging, so don't use it for now.
-#CFLAGS += $(call cc-option,-fweb)
+#cflags-y += $(call cc-option,-fweb)
endif
# -funit-at-a-time shrinks the kernel .text considerably
# unfortunately it makes reading oopses harder.
-CFLAGS += $(call cc-option,-funit-at-a-time)
+cflags-y += $(call cc-option,-funit-at-a-time)
# prevent gcc from generating any FP code by mistake
-CFLAGS += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
+cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
+CFLAGS += $(cflags-y)
AFLAGS += -m64
head-y := arch/x86_64/kernel/head.o arch/x86_64/kernel/head64.o arch/x86_64/kernel/init_task.o
diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig
index 566ecc9..3c45ec2 100644
--- a/arch/x86_64/defconfig
+++ b/arch/x86_64/defconfig
@@ -1,7 +1,7 @@
#
# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.16-git9
-# Sat Mar 25 15:18:40 2006
+# Linux kernel version: 2.6.17-rc1
+# Mon Apr 3 16:11:14 2006
#
CONFIG_X86_64=y
CONFIG_64BIT=y
@@ -9,6 +9,7 @@ CONFIG_X86=y
CONFIG_SEMAPHORE_SLEEPERS=y
CONFIG_MMU=y
CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HWEIGHT=y
CONFIG_GENERIC_CALIBRATE_DELAY=y
CONFIG_X86_CMPXCHG=y
CONFIG_EARLY_PRINTK=y
@@ -55,10 +56,6 @@ CONFIG_BASE_FULL=y
CONFIG_FUTEX=y
CONFIG_EPOLL=y
CONFIG_SHMEM=y
-CONFIG_CC_ALIGN_FUNCTIONS=0
-CONFIG_CC_ALIGN_LABELS=0
-CONFIG_CC_ALIGN_LOOPS=0
-CONFIG_CC_ALIGN_JUMPS=0
CONFIG_SLAB=y
# CONFIG_TINY_SHMEM is not set
CONFIG_BASE_SMALL=0
@@ -70,7 +67,6 @@ CONFIG_BASE_SMALL=0
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
CONFIG_MODULE_FORCE_UNLOAD=y
-CONFIG_OBSOLETE_MODPARM=y
# CONFIG_MODVERSIONS is not set
# CONFIG_MODULE_SRCVERSION_ALL is not set
# CONFIG_KMOD is not set
@@ -81,6 +77,7 @@ CONFIG_STOP_MACHINE=y
#
CONFIG_LBD=y
# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
#
# IO Schedulers
@@ -105,6 +102,7 @@ CONFIG_X86_PC=y
CONFIG_GENERIC_CPU=y
CONFIG_X86_L1_CACHE_BYTES=128
CONFIG_X86_L1_CACHE_SHIFT=7
+CONFIG_X86_INTERNODE_CACHE_BYTES=128
CONFIG_X86_TSC=y
CONFIG_X86_GOOD_APIC=y
# CONFIG_MICROCODE is not set
@@ -116,6 +114,7 @@ CONFIG_X86_LOCAL_APIC=y
CONFIG_MTRR=y
CONFIG_SMP=y
CONFIG_SCHED_SMT=y
+CONFIG_SCHED_MC=y
# CONFIG_PREEMPT_NONE is not set
CONFIG_PREEMPT_VOLUNTARY=y
# CONFIG_PREEMPT is not set
@@ -138,6 +137,7 @@ CONFIG_NEED_MULTIPLE_NODES=y
CONFIG_SPLIT_PTLOCK_CPUS=4
CONFIG_MIGRATION=y
CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID=y
+CONFIG_OUT_OF_LINE_PFN_TO_PAGE=y
CONFIG_NR_CPUS=32
CONFIG_HOTPLUG_CPU=y
CONFIG_HPET_TIMER=y
@@ -289,6 +289,7 @@ CONFIG_IP_PNP_DHCP=y
# CONFIG_INET_AH is not set
# CONFIG_INET_ESP is not set
# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
# CONFIG_INET_TUNNEL is not set
CONFIG_INET_DIAG=y
CONFIG_INET_TCP_DIAG=y
@@ -300,6 +301,7 @@ CONFIG_IPV6=y
# CONFIG_INET6_AH is not set
# CONFIG_INET6_ESP is not set
# CONFIG_INET6_IPCOMP is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
# CONFIG_INET6_TUNNEL is not set
# CONFIG_IPV6_TUNNEL is not set
# CONFIG_NETFILTER is not set
@@ -704,7 +706,6 @@ CONFIG_S2IO=m
# Wireless LAN (non-hamradio)
#
# CONFIG_NET_RADIO is not set
-# CONFIG_NET_WIRELESS_RTNETLINK is not set
#
# Wan interfaces
@@ -791,7 +792,7 @@ CONFIG_HW_CONSOLE=y
#
CONFIG_SERIAL_8250=y
CONFIG_SERIAL_8250_CONSOLE=y
-# CONFIG_SERIAL_8250_ACPI is not set
+CONFIG_SERIAL_8250_PCI=y
CONFIG_SERIAL_8250_NR_UARTS=4
CONFIG_SERIAL_8250_RUNTIME_UARTS=4
# CONFIG_SERIAL_8250_EXTENDED is not set
@@ -921,6 +922,7 @@ CONFIG_HWMON=y
# Digital Video Broadcasting Devices
#
# CONFIG_DVB is not set
+# CONFIG_USB_DABUSB is not set
#
# Graphics support
@@ -932,6 +934,8 @@ CONFIG_VIDEO_SELECT=y
# Console display driver support
#
CONFIG_VGA_CONSOLE=y
+CONFIG_VGACON_SOFT_SCROLLBACK=y
+CONFIG_VGACON_SOFT_SCROLLBACK_SIZE=256
CONFIG_DUMMY_CONSOLE=y
#
@@ -1058,15 +1062,6 @@ CONFIG_USB_HIDINPUT=y
# CONFIG_USB_MICROTEK is not set
#
-# USB Multimedia devices
-#
-# CONFIG_USB_DABUSB is not set
-
-#
-# Video4Linux support is needed for USB Multimedia device support
-#
-
-#
# USB Network Adapters
#
# CONFIG_USB_CATC is not set
@@ -1118,9 +1113,15 @@ CONFIG_USB_MON=y
# CONFIG_MMC is not set
#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
# InfiniBand support
#
# CONFIG_INFINIBAND is not set
+# CONFIG_IPATH_CORE is not set
#
# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
@@ -1128,6 +1129,11 @@ CONFIG_USB_MON=y
# CONFIG_EDAC is not set
#
+# Real Time Clock
+#
+# CONFIG_RTC_CLASS is not set
+
+#
# Firmware Drivers
#
# CONFIG_EDD is not set
diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 35b2fac..5a98026 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -15,6 +15,8 @@
#include <asm/vsyscall32.h>
#include <linux/linkage.h>
+#define IA32_NR_syscalls ((ia32_syscall_end - ia32_sys_call_table)/8)
+
.macro IA32_ARG_FIXUP noebp=0
movl %edi,%r8d
.if \noebp
@@ -109,8 +111,8 @@ ENTRY(ia32_sysenter_target)
CFI_REMEMBER_STATE
jnz sysenter_tracesys
sysenter_do_call:
- cmpl $(IA32_NR_syscalls),%eax
- jae ia32_badsys
+ cmpl $(IA32_NR_syscalls-1),%eax
+ ja ia32_badsys
IA32_ARG_FIXUP 1
call *ia32_sys_call_table(,%rax,8)
movq %rax,RAX-ARGOFFSET(%rsp)
@@ -210,8 +212,8 @@ ENTRY(ia32_cstar_target)
CFI_REMEMBER_STATE
jnz cstar_tracesys
cstar_do_call:
- cmpl $IA32_NR_syscalls,%eax
- jae ia32_badsys
+ cmpl $IA32_NR_syscalls-1,%eax
+ ja ia32_badsys
IA32_ARG_FIXUP 1
call *ia32_sys_call_table(,%rax,8)
movq %rax,RAX-ARGOFFSET(%rsp)
@@ -296,8 +298,8 @@ ENTRY(ia32_syscall)
testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
jnz ia32_tracesys
ia32_do_syscall:
- cmpl $(IA32_NR_syscalls),%eax
- jae ia32_badsys
+ cmpl $(IA32_NR_syscalls-1),%eax
+ ja ia32_badsys
IA32_ARG_FIXUP
call *ia32_sys_call_table(,%rax,8) # xxx: rip relative
ia32_sysret:
@@ -685,12 +687,11 @@ ia32_sys_call_table:
.quad sys_readlinkat /* 305 */
.quad sys_fchmodat
.quad sys_faccessat
- .quad sys_ni_syscall /* pselect6 for now */
- .quad sys_ni_syscall /* ppoll for now */
+ .quad quiet_ni_syscall /* pselect6 for now */
+ .quad quiet_ni_syscall /* ppoll for now */
.quad sys_unshare /* 310 */
.quad compat_sys_set_robust_list
.quad compat_sys_get_robust_list
+ .quad sys_splice
+ .quad sys_sync_file_range
ia32_syscall_end:
- .rept IA32_NR_syscalls-(ia32_syscall_end-ia32_sys_call_table)/8
- .quad ni_syscall
- .endr
diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c
index fffd6b0..70b9d21 100644
--- a/arch/x86_64/kernel/aperture.c
+++ b/arch/x86_64/kernel/aperture.c
@@ -80,7 +80,7 @@ static int __init aperture_valid(char *n
printk("Aperture from %s beyond 4GB. Ignoring.\n",name);
return 0;
}
- if (e820_mapped(aper_base, aper_base + aper_size, E820_RAM)) {
+ if (e820_any_mapped(aper_base, aper_base + aper_size, E820_RAM)) {
printk("Aperture from %s pointing to e820 RAM. Ignoring.\n",name);
return 0;
}
diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index 293cd71..62776c0 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -80,7 +80,12 @@ static inline int bad_addr(unsigned long
return 0;
}
-int __init e820_mapped(unsigned long start, unsigned long end, unsigned type)
+/*
+ * This function checks if any part of the range <start,end> is mapped
+ * with type.
+ */
+int __meminit
+e820_any_mapped(unsigned long start, unsigned long end, unsigned type)
{
int i;
for (i = 0; i < e820.nr_map; i++) {
@@ -94,6 +99,35 @@ int __init e820_mapped(unsigned long sta
return 0;
}
+/*
+ * This function checks if the entire range <start,end> is mapped with type.
+ *
+ * Note: this function only works correct if the e820 table is sorted and
+ * not-overlapping, which is the case
+ */
+int __init e820_all_mapped(unsigned long start, unsigned long end, unsigned type)
+{
+ int i;
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+ if (type && ei->type != type)
+ continue;
+ /* is the region (part) in overlap with the current region ?*/
+ if (ei->addr >= end || ei->addr + ei->size <= start)
+ continue;
+
+ /* if the region is at the beginning of <start,end> we move
+ * start to the end of the region since it's ok until there
+ */
+ if (ei->addr <= start)
+ start = ei->addr + ei->size;
+ /* if start is now at or beyond end, we're done, full coverage */
+ if (start >= end)
+ return 1; /* we're done */
+ }
+ return 0;
+}
+
/*
* Find a free area in a specific range.
*/
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 8538bfe..c946e4f 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -180,6 +180,10 @@ rff_trace:
*
* XXX if we had a free scratch register we could save the RSP into the stack frame
* and report it properly in ps. Unfortunately we haven't.
+ *
+ * When user can change the frames always force IRET. That is because
+ * it deals with uncanonical addresses better. SYSRET has trouble
+ * with them due to bugs in both AMD and Intel CPUs.
*/
ENTRY(system_call)
@@ -254,7 +258,10 @@ sysret_signal:
xorl %esi,%esi # oldset -> arg2
call ptregscall_common
1: movl $_TIF_NEED_RESCHED,%edi
- jmp sysret_check
+ /* Use IRET because user could have changed frame. This
+ works because ptregscall_common has called FIXUP_TOP_OF_STACK. */
+ cli
+ jmp int_with_check
badsys:
movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
@@ -280,7 +287,8 @@ tracesys:
call syscall_trace_leave
RESTORE_TOP_OF_STACK %rbx
RESTORE_REST
- jmp ret_from_sys_call
+ /* Use IRET because user could have changed frame */
+ jmp int_ret_from_sys_call
CFI_ENDPROC
/*
@@ -408,25 +416,9 @@ ENTRY(stub_execve)
CFI_ADJUST_CFA_OFFSET -8
CFI_REGISTER rip, r11
SAVE_REST
- movq %r11, %r15
- CFI_REGISTER rip, r15
FIXUP_TOP_OF_STACK %r11
call sys_execve
- GET_THREAD_INFO(%rcx)
- bt $TIF_IA32,threadinfo_flags(%rcx)
- CFI_REMEMBER_STATE
- jc exec_32bit
RESTORE_TOP_OF_STACK %r11
- movq %r15, %r11
- CFI_REGISTER rip, r11
- RESTORE_REST
- pushq %r11
- CFI_ADJUST_CFA_OFFSET 8
- CFI_REL_OFFSET rip, 0
- ret
-
-exec_32bit:
- CFI_RESTORE_STATE
movq %rax,RAX(%rsp)
RESTORE_REST
jmp int_ret_from_sys_call
diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 10b3e34..6f0790e 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -29,6 +29,8 @@
#define MISC_MCELOG_MINOR 227
#define NR_BANKS 6
+atomic_t mce_entry;
+
static int mce_dont_init;
/* 0: always panic, 1: panic if deadlock possible, 2: try to avoid panic,
@@ -172,10 +174,12 @@ void do_machine_check(struct pt_regs * r
int i;
int panicm_found = 0;
+ atomic_inc(&mce_entry);
+
if (regs)
notify_die(DIE_NMI, "machine check", regs, error_code, 18, SIGKILL);
if (!banks)
- return;
+ goto out2;
memset(&m, 0, sizeof(struct mce));
m.cpu = safe_smp_processor_id();
@@ -266,6 +270,8 @@ void do_machine_check(struct pt_regs * r
out:
/* Last thing done in the machine check exception to clear state. */
wrmsrl(MSR_IA32_MCG_STATUS, 0);
+ out2:
+ atomic_dec(&mce_entry);
}
/*
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index d9e4067..4e6357f 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -34,6 +34,7 @@
#include <asm/proto.h>
#include <asm/kdebug.h>
#include <asm/local.h>
+#include <asm/mce.h>
/*
* lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
@@ -480,6 +481,12 @@ void __kprobes nmi_watchdog_tick(struct
__get_cpu_var(nmi_touch) = 0;
touched = 1;
}
+#ifdef CONFIG_X86_MCE
+ /* Could check oops_in_progress here too, but it's safer
+ not too */
+ if (atomic_read(&mce_entry) > 0)
+ touched = 1;
+#endif
if (!touched && __get_cpu_var(last_irq_sum) == sum) {
/*
* Ayiee, looks like this CPU is stuck ...
diff --git a/arch/x86_64/kernel/pci-dma.c b/arch/x86_64/kernel/pci-dma.c
index 03c9eee..af035ed 100644
--- a/arch/x86_64/kernel/pci-dma.c
+++ b/arch/x86_64/kernel/pci-dma.c
@@ -48,9 +48,11 @@ dma_alloc_pages(struct device *dev, gfp_
{
struct page *page;
int node;
+#ifdef CONFIG_PCI
if (dev->bus == &pci_bus_type)
node = pcibus_to_node(to_pci_dev(dev)->bus);
else
+#endif
node = numa_node_id();
page = alloc_pages_node(node, gfp, order);
return page ? page_address(page) : NULL;
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 70dd8e5..1c44b53 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -781,10 +781,16 @@ long do_arch_prctl(struct task_struct *t
}
case ARCH_GET_GS: {
unsigned long base;
+ unsigned gsindex;
if (task->thread.gsindex == GS_TLS_SEL)
base = read_32bit_tls(task, GS_TLS);
- else if (doit)
- rdmsrl(MSR_KERNEL_GS_BASE, base);
+ else if (doit) {
+ asm("movl %%gs,%0" : "=r" (gsindex));
+ if (gsindex)
+ rdmsrl(MSR_KERNEL_GS_BASE, base);
+ else
+ base = task->thread.gs;
+ }
else
base = task->thread.gs;
ret = put_user(base, (unsigned long __user *)addr);
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 0856ad4..c50b067 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -353,8 +353,10 @@ static __init void parse_cmdline_early (
if (fullarg(from, "enable_timer_pin_1"))
disable_timer_pin_1 = -1;
- if (fullarg(from, "nolapic") || fullarg(from, "disableapic"))
+ if (fullarg(from, "nolapic") || fullarg(from, "disableapic")) {
+ clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
disable_apic = 1;
+ }
if (fullarg(from, "noapic"))
skip_ioapic_setup = 1;
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index ef8bc46..7392570 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -726,7 +726,7 @@ static __init int late_hpet_init(void)
unsigned int ntimer;
if (!vxtime.hpet_address)
- return -1;
+ return 0;
memset(&hd, 0, sizeof (hd));
@@ -917,6 +917,8 @@ void __init time_init(void)
vxtime.hpet_address = 0;
if (hpet_use_timer) {
+ /* set tick_nsec to use the proper rate for HPET */
+ tick_nsec = TICK_NSEC_HPET;
cpu_khz = hpet_calibrate_tsc();
timename = "HPET";
#ifdef CONFIG_X86_PM_TIMER
diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index 39ff070..b81f473 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/arch/x86_64/kernel/vmlinux.lds.S
@@ -65,7 +65,7 @@ SECTIONS
.data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) {
*(.data.cacheline_aligned)
}
- . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
+ . = ALIGN(CONFIG_X86_INTERNODE_CACHE_BYTES);
.data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET) {
*(.data.read_mostly)
}
diff --git a/arch/x86_64/kernel/x8664_ksyms.c b/arch/x86_64/kernel/x8664_ksyms.c
index d78f460..fec4e52 100644
--- a/arch/x86_64/kernel/x8664_ksyms.c
+++ b/arch/x86_64/kernel/x8664_ksyms.c
@@ -112,7 +112,6 @@ EXPORT_SYMBOL_GPL(unset_nmi_callback);
#undef memcpy
#undef memset
#undef memmove
-#undef strlen
extern void * memset(void *,int,__kernel_size_t);
extern size_t strlen(const char *);
@@ -121,7 +120,6 @@ extern void * memcpy(void *,const void *
extern void * __memcpy(void *,const void *,__kernel_size_t);
EXPORT_SYMBOL(memset);
-EXPORT_SYMBOL(strlen);
EXPORT_SYMBOL(strpbrk);
EXPORT_SYMBOL(memmove);
EXPORT_SYMBOL(memcpy);
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index e5f7f1c..4ba34e9 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -305,7 +305,7 @@ static void __meminit phys_pud_init(pud_
if (paddr >= end)
break;
- if (!after_bootmem && !e820_mapped(paddr, paddr+PUD_SIZE, 0)) {
+ if (!after_bootmem && !e820_any_mapped(paddr, paddr+PUD_SIZE, 0)) {
set_pud(pud, __pud(0));
continue;
}
@@ -507,9 +507,8 @@ void __init clear_kernel_mapping(unsigne
/*
* Memory hotplug specific functions
- * These are only for non-NUMA machines right now.
*/
-#ifdef CONFIG_MEMORY_HOTPLUG
+#if defined(CONFIG_ACPI_HOTPLUG_MEMORY) || defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)
void online_page(struct page *page)
{
@@ -520,6 +519,38 @@ void online_page(struct page *page)
num_physpages++;
}
+#ifndef CONFIG_MEMORY_HOTPLUG
+/*
+ * Memory Hotadd without sparsemem. The mem_maps have been allocated in advance,
+ * just online the pages.
+ */
+int __add_pages(struct zone *z, unsigned long start_pfn, unsigned long nr_pages)
+{
+ int err = -EIO;
+ unsigned long pfn;
+ unsigned long total = 0, mem = 0;
+ for (pfn = start_pfn; pfn < start_pfn + nr_pages; pfn++) {
+ if (pfn_valid(pfn)) {
+ online_page(pfn_to_page(pfn));
+ err = 0;
+ mem++;
+ }
+ total++;
+ }
+ if (!err) {
+ z->spanned_pages += total;
+ z->present_pages += mem;
+ z->zone_pgdat->node_spanned_pages += total;
+ z->zone_pgdat->node_present_pages += mem;
+ }
+ return err;
+}
+#endif
+
+/*
+ * Memory is added always to NORMAL zone. This means you will never get
+ * additional DMA/DMA32 memory.
+ */
int add_memory(u64 start, u64 size)
{
struct pglist_data *pgdat = NODE_DATA(0);
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index 4be82d6..cc02573 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -100,11 +100,30 @@ int early_pfn_to_nid(unsigned long pfn)
}
#endif
+static void * __init
+early_node_mem(int nodeid, unsigned long start, unsigned long end,
+ unsigned long size)
+{
+ unsigned long mem = find_e820_area(start, end, size);
+ void *ptr;
+ if (mem != -1L)
+ return __va(mem);
+ ptr = __alloc_bootmem_nopanic(size,
+ SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS));
+ if (ptr == 0) {
+ printk(KERN_ERR "Cannot find %lu bytes in node %d\n",
+ size, nodeid);
+ return NULL;
+ }
+ return ptr;
+}
+
/* Initialize bootmem allocator for a node */
void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long end)
{
unsigned long start_pfn, end_pfn, bootmap_pages, bootmap_size, bootmap_start;
unsigned long nodedata_phys;
+ void *bootmap;
const int pgdat_size = round_up(sizeof(pg_data_t), PAGE_SIZE);
start = round_up(start, ZONE_ALIGN);
@@ -114,13 +133,11 @@ void __init setup_node_bootmem(int nodei
start_pfn = start >> PAGE_SHIFT;
end_pfn = end >> PAGE_SHIFT;
- nodedata_phys = find_e820_area(start, end, pgdat_size);
- if (nodedata_phys == -1L)
- panic("Cannot find memory pgdat in node %d\n", nodeid);
-
- Dprintk("nodedata_phys %lx\n", nodedata_phys);
+ node_data[nodeid] = early_node_mem(nodeid, start, end, pgdat_size);
+ if (node_data[nodeid] == NULL)
+ return;
+ nodedata_phys = __pa(node_data[nodeid]);
- node_data[nodeid] = phys_to_virt(nodedata_phys);
memset(NODE_DATA(nodeid), 0, sizeof(pg_data_t));
NODE_DATA(nodeid)->bdata = &plat_node_bdata[nodeid];
NODE_DATA(nodeid)->node_start_pfn = start_pfn;
@@ -129,9 +146,15 @@ void __init setup_node_bootmem(int nodei
/* Find a place for the bootmem map */
bootmap_pages = bootmem_bootmap_pages(end_pfn - start_pfn);
bootmap_start = round_up(nodedata_phys + pgdat_size, PAGE_SIZE);
- bootmap_start = find_e820_area(bootmap_start, end, bootmap_pages<<PAGE_SHIFT);
- if (bootmap_start == -1L)
- panic("Not enough continuous space for bootmap on node %d", nodeid);
+ bootmap = early_node_mem(nodeid, bootmap_start, end,
+ bootmap_pages<<PAGE_SHIFT);
+ if (bootmap == NULL) {
+ if (nodedata_phys < start || nodedata_phys >= end)
+ free_bootmem((unsigned long)node_data[nodeid],pgdat_size);
+ node_data[nodeid] = NULL;
+ return;
+ }
+ bootmap_start = __pa(bootmap);
Dprintk("bootmap start %lu pages %lu\n", bootmap_start, bootmap_pages);
bootmap_size = init_bootmem_node(NODE_DATA(nodeid),
@@ -142,6 +165,9 @@ void __init setup_node_bootmem(int nodei
reserve_bootmem_node(NODE_DATA(nodeid), nodedata_phys, pgdat_size);
reserve_bootmem_node(NODE_DATA(nodeid), bootmap_start, bootmap_pages<<PAGE_SHIFT);
+#ifdef CONFIG_ACPI_NUMA
+ srat_reserve_add_area(nodeid);
+#endif
node_set_online(nodeid);
}
@@ -335,6 +361,8 @@ __init int numa_setup(char *opt)
#ifdef CONFIG_ACPI_NUMA
if (!strncmp(opt,"noacpi",6))
acpi_numa = -1;
+ if (!strncmp(opt,"hotadd=", 7))
+ hotadd_percent = simple_strtoul(opt+7, NULL, 10);
#endif
return 1;
}
diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index 2eb8795..15ae9fc 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -15,15 +15,26 @@
#include <linux/bitmap.h>
#include <linux/module.h>
#include <linux/topology.h>
+#include <linux/bootmem.h>
+#include <linux/mm.h>
#include <asm/proto.h>
#include <asm/numa.h>
#include <asm/e820.h>
+#if (defined(CONFIG_ACPI_HOTPLUG_MEMORY) || \
+ defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)) \
+ && !defined(CONFIG_MEMORY_HOTPLUG)
+#define RESERVE_HOTADD 1
+#endif
+
static struct acpi_table_slit *acpi_slit;
static nodemask_t nodes_parsed __initdata;
static nodemask_t nodes_found __initdata;
static struct bootnode nodes[MAX_NUMNODES] __initdata;
+static struct bootnode nodes_add[MAX_NUMNODES] __initdata;
+static int found_add_area __initdata;
+int hotadd_percent __initdata = 10;
static u8 pxm2node[256] = { [0 ... 255] = 0xff };
/* Too small nodes confuse the VM badly. Usually they result
@@ -71,6 +82,10 @@ static __init int conflicting_nodes(unsi
static __init void cutoff_node(int i, unsigned long start, unsigned long end)
{
struct bootnode *nd = &nodes[i];
+
+ if (found_add_area)
+ return;
+
if (nd->start < start) {
nd->start = start;
if (nd->end < nd->start)
@@ -90,6 +105,8 @@ static __init void bad_srat(void)
acpi_numa = -1;
for (i = 0; i < MAX_LOCAL_APIC; i++)
apicid_to_node[i] = NUMA_NO_NODE;
+ for (i = 0; i < MAX_NUMNODES; i++)
+ nodes_add[i].start = nodes[i].end = 0;
}
static __init inline int srat_disabled(void)
@@ -155,11 +172,114 @@ acpi_numa_processor_affinity_init(struct
pxm, pa->apic_id, node);
}
+#ifdef RESERVE_HOTADD
+/*
+ * Protect against too large hotadd areas that would fill up memory.
+ */
+static int hotadd_enough_memory(struct bootnode *nd)
+{
+ static unsigned long allocated;
+ static unsigned long las...
[truncated message content] |
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-09 20:15:58
|
Documentation/x86_64/boot-options.txt | 5
arch/i386/kernel/acpi/boot.c | 5
arch/i386/kernel/apic.c | 23 ++-
arch/i386/kernel/cpu/cpufreq/powernow-k8.c | 18 ++-
arch/i386/kernel/mpparse.c | 21 ---
arch/i386/kernel/reboot_fixups.c | 2
arch/i386/kernel/setup.c | 36 +++++-
arch/i386/mm/init.c | 2
arch/i386/pci/direct.c | 9 +
arch/i386/pci/mmconfig.c | 67 +++++++----
arch/x86_64/Kconfig | 5
arch/x86_64/Makefile | 24 ++--
arch/x86_64/defconfig | 42 ++++---
arch/x86_64/ia32/ia32entry.S | 23 ++-
arch/x86_64/kernel/aperture.c | 2
arch/x86_64/kernel/e820.c | 36 +++++-
arch/x86_64/kernel/entry.S | 28 +---
arch/x86_64/kernel/mce.c | 8 +
arch/x86_64/kernel/nmi.c | 7 +
arch/x86_64/kernel/pci-dma.c | 2
arch/x86_64/kernel/process.c | 10 +
arch/x86_64/kernel/setup.c | 4
arch/x86_64/kernel/time.c | 4
arch/x86_64/kernel/vmlinux.lds.S | 2
arch/x86_64/kernel/x8664_ksyms.c | 2
arch/x86_64/mm/init.c | 37 +++++-
arch/x86_64/mm/numa.c | 46 ++++++-
arch/x86_64/mm/srat.c | 170 +++++++++++++++++++++++++++--
arch/x86_64/pci/mmconfig.c | 53 ++++++---
drivers/acpi/Kconfig | 2
include/asm-i386/apic.h | 2
include/asm-i386/e820.h | 4
include/asm-i386/hpet.h | 1
include/asm-x86_64/e820.h | 3
include/asm-x86_64/hpet.h | 2
include/asm-x86_64/ia32_unistd.h | 2
include/asm-x86_64/mce.h | 7 +
include/asm-x86_64/numa.h | 2
include/asm-x86_64/numnodes.h | 2
include/linux/bootmem.h | 1
include/linux/init.h | 3
include/linux/jiffies.h | 6 +
include/linux/memory_hotplug.h | 14 +-
kernel/timer.c | 2
mm/bootmem.c | 9 +
security/selinux/xfrm.c | 4
46 files changed, 575 insertions(+), 184 deletions(-)
New commits:
commit 0660a65891b42630ae856d73ff19212745565d0a
Merge: 315ecb5... ebe9c64...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Sun Apr 9 22:14:28 2006 +0200
Merge branch 'vax4linus'
commit ebe9c645f0738c1254e7b6208e3dec8a1b9be2ba
Merge: 6b61940... 6764472...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Sun Apr 9 22:12:03 2006 +0200
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6
commit 67644726317a8274be4a3d0ef85b9ccebaa90304
Author: Dave Jones <da...@re...>
Date: Sun Apr 2 23:34:19 2006 -0700
[SELINUX] Fix build after ipsec decap state changes.
security/selinux/xfrm.c: In function 'selinux_socket_getpeer_dgram':
security/selinux/xfrm.c:284: error: 'struct sec_path' has no member named 'x'
security/selinux/xfrm.c: In function 'selinux_xfrm_sock_rcv_skb':
security/selinux/xfrm.c:317: error: 'struct sec_path' has no member named 'x'
Signed-off-by: Dave Jones <da...@re...>
Signed-off-by: David S. Miller <da...@da...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 66004a6ca23f2a2408b32cbe27fda0389fb8f9dc
Author: Linus Torvalds <tor...@g5...>
Date: Sun Apr 9 12:14:02 2006 -0700
Move request_standard_resources() back to before PCI probing
This effectively undoes the PCI resource allocation changes done in
commit b408cbc704352eccee301e1103b23203ba1c3a0e, but leaves the cleanups
of that commit in place.
We're going back to marking the resources reported by e820 busy _before_
doing PCI probing, so that any PCI resource that clashes with the BIOS-
reported memory map will be reloacted to a non-clashing area.
The reason? Larry Finger reports that his laptop has the cardbus
controller set up by the BIOS so that it conflicts with the e820 memory
map, and needs to be relocated. See
http://bugzilla.kernel.org/show_bug.cgi?id=6337
for more details.
We'll have to work out how to handle the fbcon problem that caused that
commit in the first place in some other way.
Cc: Ivan Kokshaysky <in...@ju...>
Cc: Greg Kroah-Hartman <gr...@su...>
Cc: Antonino A. Daplas <ad...@po...>
Cc: <bj...@lu...>
Tested-by: Larry Finger <Lar...@lw...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit b8feb47f992d314c956add15c1118430120635bb
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:34 2006 +0200
[PATCH] x86_64: Update 32-bit system call table
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 67d53ea5a3d42aadeb1584e757ca4660c0e8a810
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:31 2006 +0200
[PATCH] x86_64: Eliminate IA32_NR_syscalls define
Or rather compute it based on the table length automatically.
This also has the intended side effect of not warning for new system calls
anymore.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit bbd3aff89d4b34ef17a748e4c001ecc5b43e3e55
Author: Sam Ravnborg <sa...@ra...>
Date: Fri Apr 7 19:50:28 2006 +0200
[PATCH] x86_64: fix CONFIG_REORDER
Fix CONFIG_REORDER.
The value of cflags-y was assined to CFLAGS before cflags-y was assigned
the value used for CONFIG_REORDER.
Use cflags-y for all CFLAGS options in the Makefile to avoid this
happening again.
Signed-off-by: Sam Ravnborg <sa...@ra...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 97c2803c9c694cafbd9f5e43a25903e0abf25188
Author: John Blackwood <joh...@cc...>
Date: Fri Apr 7 19:50:25 2006 +0200
[PATCH] x86_64: Plug GS leak in arch_prctl()
In linux-2.6.16, we have noticed a problem where the gs base value
returned from an arch_prtcl(ARCH_GET_GS, ...) call will be incorrect if:
- the current/calling task has NOT set its own gs base yet to a
non-zero value,
- some other task that ran on the same processor previously set their
own gs base to a non-zero value.
In this situation, the ARCH_GET_GS code will read and return the
MSR_KERNEL_GS_BASE msr register.
However, since the __switch_to() code does NOT load/zero the
MSR_KERNEL_GS_BASE register when the task that is switched IN has a zero
next->gs value, the caller of arch_prctl(ARCH_GET_GS, ...) will get back
the value of some previous tasks's gs base value instead of 0.
Change the arch_prctl() ARCH_GET_GS code to only read and return
the MSR_KERNEL_GS_BASE msr register if the 'gs' register of the calling
task is non-zero.
Side note: Since in addition to using arch_prctl(ARCH_SET_GS, ...),
a task can also setup a gs base value by using modify_ldt() and write
an index value into 'gs' from user space, the patch below reads
'gs' instead of using thread.gs, since in the modify_ldt() case,
the thread.gs value will be 0, and incorrect value would be returned
(the task->thread.gs value).
When the user has not set its own gs base value and the 'gs'
register is zero, then the MSR_KERNEL_GS_BASE register will not be
read and a value of zero will be returned by reading and returning
'task->thread.gs'.
The first patch shown below is an attempt at implementing this
approach.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit e48c4729d23a026f3711d5e36add5cce894b4913
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:21 2006 +0200
[PATCH] i386: Remove printk about reboot fixups at reboot
Printk doesn't have any value
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit b20367a6c2a0cd937cb1f0a8cf848f1402fef99c
Author: Jordan Hargrave <jor...@de...>
Date: Fri Apr 7 19:50:18 2006 +0200
[PATCH] x86_64: Fix drift with HPET timer enabled
If the HPET timer is enabled, the clock can drift by ~3 seconds a day.
This is due to the HPET timer not being initialized with the correct
setting (still using PIT count).
If HZ changes, this drift can become even more pronounced.
HPET patch initializes tick_nsec with correct tick_nsec settings for
HPET timer.
Vojtech comments:
"It's not entirely correct (it assumes the HPET ticks totally
exactly), but it's significantly better than assuming the PIT error
there."
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 49c93e84d8b2d602a07c302c7e3cd4fa09095fbb
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:15 2006 +0200
[PATCH] i386/x86-64: Return defined error value for bad PCI config space accesses
Mostly to get better handling when a extended config space
access has to fallback to Type1.
Cc: gr...@su...
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 8c30b1a74aed4041f183e183a149b7dfbdc6c20e
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:12 2006 +0200
[PATCH] i386/x86_64: Check if MCFG works for the first 16 busses
Previously only the first bus would be checked against Type 1.
Why 16? Checking all would need too much memory and we
can assume that systems with more than 16 busses have better than
average quality BIOS.
This is an additional defense against bad MCFG tables.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit e405d067298b2b960bf20318e91ed842157c65bc
Author: Ravikiran G Thirumalai <ki...@sc...>
Date: Fri Apr 7 19:50:09 2006 +0200
[PATCH] x86_64: Fixup read_mostly section on internode cache line size for vSMP
Fixup the read mostly section to start at internode cacheline boundary.
Signed-off-by: Ravikiran Thirumalai <ki...@sc...>
Signed-off-by: Shai Fultheim <sh...@sc...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 3d34ee6891e274dfb6a22930546d37738cdbe9c4
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:06 2006 +0200
[PATCH] x86_64: Don't return error for HPET initialization in initcall
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit ac04dcaf6f567307fbeef9c3c1fff35280e53f02
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:03 2006 +0200
[PATCH] x86_64: Don't export strlen twice
Fix
WARNING: vmlinux: 'strlen' exported twice. Previous export was in vmlinux
Reported by Mats Johannesson
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 7bf36bbc5e0c09271f9efe22162f8cc3f8ebd3d2
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:00 2006 +0200
[PATCH] x86_64: When user could have changed RIP always force IRET
Intel EM64T CPUs handle uncanonical return addresses differently
from AMD CPUs.
The exception is reported in the SYSRET, not the next instruction.
This leads to the kernel exception handler running on the user stack
with the wrong GS because the kernel didn't expect exceptions
on this instruction.
This version of the patch has the teething problems that plagued an earlier
version fixed.
This is CVE-2006-0744
Thanks to Ernie Petrides and Asit B. Mallick for analysis and initial
patches.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 553f265fe883a23502ee351845f09334790f18b8
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:57 2006 +0200
[PATCH] x86_64: Don't run NMI watchdog during machine checks
Machine checks can stall the machine for a long time and
it's not good to trigger the nmi watchdog during that.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit be56db6186999a8571ae480cf2b929578f6dfd68
Author: Dave Hansen <hav...@us...>
Date: Fri Apr 7 19:49:54 2006 +0200
[PATCH] x86_64: extra NODES_SHIFT definition
The generic linux/numa.h file defines NODES_SHIFT to 0 in case
the architecture did not.
Every architecture which has a NUMA config option defines
NODES_SHIFT in its asm-$ARCH headers, but only if NUMA is
enabled, except for x86_64.
This should make it like all the rest.
Signed-off-by: Dave Hansen <hav...@us...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 4211a30349e8d2b724cfb4ce2584604f5e59c299
Author: Jacob Shin <jac...@am...>
Date: Fri Apr 7 19:49:51 2006 +0200
[PATCH] x86_64: Proper null pointer check in powernow_k8_get
This prevents crashes on dual core system when enough ticks are lost.
Replaces earlier patch by me.
Cc: Dave Jones <da...@re...>
Signed-off-by: Thomas Renninger <tr...@su...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit d7fa706ce2c29cb751c15ca00f3aa7b223e3c9f0
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:48 2006 +0200
[PATCH] x86_64: Revert earlier powernow-k8 change
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 95d769aaf47abfc77b600631403ff5af6c990cff
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:45 2006 +0200
[PATCH] i386: Consolidate modern APIC handling
AMD systems have a modern APIC that supports 8 bit IDs, but
don't have a XAPIC version number. Add a new "modern_apic"
subfunction that handles this correctly and use it (nearly)
everywhere where XAPIC is tested for.
I removed one wart: the code specified that external APICs
would use an 8bit APIC ID. But I checked a real 82093 data sheet
and it says clearly that they only use 4bit. So I removed
this special case since it would a bit awkward to implement now.
I removed the valid APIC tests in mptable parsing completely. On any modern
system they only check against the full field width (8bit) anyways
and are no-ops. This also fixes them doing the wrong thing
on >8 core Opterons.
This makes i386 boot again on 16 core Opterons.
Cc: Ingo Molnar <mi...@el...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit d1530d82e02fd96d4634a6d6f6538c8b778c43af
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:42 2006 +0200
[PATCH] x86_64: Clear APIC feature bit when local APIC is disabled
Needed for other checks later in ACPI.
Pointed out by Len Brown
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit d3b6a349d233aecf2c52f7f4c150ca09f684f2d8
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:39 2006 +0200
[PATCH] x86-64/i386: Don't process APICs/IO-APICs in ACPI when APIC is disabled.
When nolapic was passed or the local APIC was disabled
for another reason ACPI would still parse the IO-APICs
until these were explicitely disabled with noapic.
Usually this resulted in a non booting configuration unless
"nolapic noapic" was used.
I also disabled the local APIC parsing in this case, although
that's only cosmetic (suppresses a few printks)
This hopefully makes nolapic work in all cases.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit ec0f08eeea6ac1d8c925f47e3677e4c985fd8f63
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:36 2006 +0200
[PATCH] x86_64: Don't sanity check Type 1 PCI bus access on newer systems
Horus systems don't have anything on bus 0 which makes
the Type 1 sanity checks fail. Use the DMI BIOS year to
check for newer systems and always assume Type 1 works on them.
I used 2001 as an pretty arbitary cutoff year.
Cc: gr...@su...
Cc: Navin Boppuri <nav...@ne...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit fa47dd0ba303599f8adf8d8336ed2fb74efc47c5
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:33 2006 +0200
[PATCH] x86_64: Fix compilation with CONFIG_PCI=n / allnoconfig
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 946f2ee5c7312e8acac4f3ab6629e7e2d36a3646
Author: Arjan van de Ven <ar...@li...>
Date: Fri Apr 7 19:49:30 2006 +0200
[PATCH] i386/x86-64: Check that MCFG points to an e820 reserved area
This patch introduces a user for the e820_all_mapped function:
There have been several machines that don't have a working MMCONFIG,
often because of a buggy MCFG table in the ACPI bios. This patch adds a
simple sanity check that detects a whole bunch of these cases, and when
it detects it, linux now boots rather than crash-and-burns.
The accuracy of this detection can in principle be improved if there was
a "is this entire range in e820 with THIS attribute", but no such
function exist and the complexity needed for this is not really worth
it; this simple check already catches most cases anyway.
Signed-off-by: Arjan van de Ven <ar...@li...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 952223683ec989e86328c24808fdb962c4dbeb0a
Author: Arjan van de Ven <ar...@li...>
Date: Fri Apr 7 19:49:27 2006 +0200
[PATCH] x86_64: Introduce e820_all_mapped
Introduce a e820_all_mapped() function which checks if the entire range
<start,end> is mapped with type.
This is done by moving the local start variable to the end of each
known-good region; if at the end of the function the start address is
still before end, there must be a part that's not of the correct type;
otherwise it's a good region.
Signed-off-by: Arjan van de Ven <ar...@li...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit eee5a9fa63c97366cdea6ab3aa2ed9e3601812d0
Author: Arjan van de Ven <ar...@li...>
Date: Fri Apr 7 19:49:24 2006 +0200
[PATCH] x86_64: Rename e820_mapped to e820_any_mapped
Rename e820_mapped to e820_any_mapped since it tests if any part of the
range is mapped according to the type.
Later steps will introduce e820_all_mapped which will check if the
entire range is mapped with the type. Both have their merit.
Signed-off-by: Arjan van de Ven <ar...@li...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit a8062231d80239cf3405982858c02aea21a6066a
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:21 2006 +0200
[PATCH] x86_64: Handle empty PXMs that only contain hotplug memory
The node setup code would try to allocate the node metadata in the node
itself, but that fails if there is no memory in there.
This can happen with memory hotplug when the hotplug area defines an so
far empty node.
Now use bootmem to try to allocate the mem_map in other nodes.
And if it fails don't panic, but just ignore the node.
To make this work I added a new __alloc_bootmem_nopanic function that
does what its name implies.
TBD should try to use nearby nodes here. Currently we just use any.
It's hard to do it better because bootmem doesn't have proper fallback
lists yet.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 68a3a7feb08f960095072f28ec20f7900793c506
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:18 2006 +0200
[PATCH] x86_64: Reserve SRAT hotadd memory on x86-64
From: Keith Mannthey, Andi Kleen
Implement memory hotadd without sparsemem. The memory in the SRAT
hotadd area is just preserved instead and can be activated later.
There are a few restrictions:
- Only one continuous hotadd area allowed per node
The main problem is dealing with the many buggy SRAT tables
that are out there. The strategy here is to reject anything
suspicious.
Originally from Keith Mannthey, with several hacks and changes by AK
and also contributions from Andrew Morton
[ TBD: Problems pointed out by KAMEZAWA Hiroyuki <kam...@jp...>:
1) Goto's rebuild_zonelist patch will not work if CONFIG_MEMORY_HOTPLUG=n.
Rebuilding zonelist is necessary when the system has just memory <
4G at boot, and hot add memory > 4G. because x86_64 has DMA32,
ZONE_NORAML is not included into zonelist at boot time if system
doesn't have memory >4G at boot.
[AK: should just force the higher zones at boot time when SRAT tells us]
2) zone and node's spanned_pages and present_pages are not incremented.
They should be.
For example, our server (ia64/Fujitsu PrimeQuest) can equip memory
from 4G to 1T(maybe 2T in future), and SRAT will *always* say we have
possible 1T +memory. (Microsoft requires "write all possible memory
in SRAT") When we reserve memmap for possible 1T memory, Linux will
not work well in +minimum 4G configuraion ;)
[AK: needs limiting to 5-10% of max memory]
]
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 9d99aaa31f5994d1923c3713ce9144c4c42332e1
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:15 2006 +0200
[PATCH] x86_64: Support memory hotadd without sparsemem
Memory hotadd doesn't need SPARSEMEM, but can be handled by just preallocating
mem_maps. This only needs some untangling of ifdefs to enable the necessary
code even without SPARSEMEM.
Originally from Keith Mannthey, hacked by AK.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 805e8c03c9ea9bdb402a36341e02ec24825d5417
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:12 2006 +0200
[PATCH] x86_64: Clean up execve path
Just call IRET always, no need for any special cases.
Needed for the next bug fix.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 903fcc608e9f531749024172277dc2fd15d5a587
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:09 2006 +0200
[PATCH] x86_64: Update defconfig
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt
index 1921353..f2cd6ef 100644
--- a/Documentation/x86_64/boot-options.txt
+++ b/Documentation/x86_64/boot-options.txt
@@ -151,6 +151,11 @@ NUMA
numa=fake=X Fake X nodes and ignore NUMA setup of the actual machine.
+ numa=hotadd=percent
+ Only allow hotadd memory to preallocate page structures upto
+ percent of already available memory.
+ numa=hotadd=0 will disable hotadd memory.
+
ACPI
acpi=off Don't enable ACPI
diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 0330661..8dab352 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -215,7 +215,7 @@ static int __init acpi_parse_madt(unsign
{
struct acpi_table_madt *madt = NULL;
- if (!phys_addr || !size)
+ if (!phys_addr || !size || !cpu_has_apic)
return -EINVAL;
madt = (struct acpi_table_madt *)__acpi_map_table(phys_addr, size);
@@ -751,6 +751,9 @@ static int __init acpi_parse_madt_ioapic
return -ENODEV;
}
+ if (!cpu_has_apic)
+ return -ENODEV;
+
/*
* if "noapic" boot option, don't look for IO-APICs
*/
diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 6273bf7..254cee9 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -62,6 +62,18 @@ int apic_verbosity;
static void apic_pm_activate(void);
+int modern_apic(void)
+{
+ unsigned int lvr, version;
+ /* AMD systems use old APIC versions, so check the CPU */
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
+ boot_cpu_data.x86 >= 0xf)
+ return 1;
+ lvr = apic_read(APIC_LVR);
+ version = GET_APIC_VERSION(lvr);
+ return version >= 0x14;
+}
+
/*
* 'what should we do if we get a hw irq event on an illegal vector'.
* each architecture has to answer this themselves.
@@ -119,10 +131,7 @@ void enable_NMI_through_LVT0 (void * dum
int get_physical_broadcast(void)
{
- unsigned int lvr, version;
- lvr = apic_read(APIC_LVR);
- version = GET_APIC_VERSION(lvr);
- if (!APIC_INTEGRATED(version) || version >= 0x14)
+ if (modern_apic())
return 0xff;
else
return 0xf;
@@ -349,9 +358,9 @@ int __init verify_local_APIC(void)
void __init sync_Arb_IDs(void)
{
- /* Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 */
- unsigned int ver = GET_APIC_VERSION(apic_read(APIC_LVR));
- if (ver >= 0x14) /* P4 or higher */
+ /* Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1
+ And not needed on AMD */
+ if (modern_apic())
return;
/*
* Wait for idle.
diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
index 712a26b..7c0e160 100644
--- a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
@@ -46,7 +46,7 @@
#define PFX "powernow-k8: "
#define BFX PFX "BIOS error: "
-#define VERSION "version 1.60.1"
+#define VERSION "version 1.60.2"
#include "powernow-k8.h"
/* serialize freq changes */
@@ -55,7 +55,7 @@ static DEFINE_MUTEX(fidvid_mutex);
static struct powernow_k8_data *powernow_data[NR_CPUS];
#ifndef CONFIG_SMP
-static cpumask_t cpu_core_map[1] = { CPU_MASK_ALL };
+static cpumask_t cpu_core_map[1];
#endif
/* Return a frequency in MHz, given an input fid */
@@ -910,6 +910,9 @@ static int powernowk8_target(struct cpuf
unsigned int newstate;
int ret = -EIO;
+ if (!data)
+ return -EINVAL;
+
/* only run on specific CPU from here on */
oldmask = current->cpus_allowed;
set_cpus_allowed(current, cpumask_of_cpu(pol->cpu));
@@ -969,6 +972,9 @@ static int powernowk8_verify(struct cpuf
{
struct powernow_k8_data *data = powernow_data[pol->cpu];
+ if (!data)
+ return -EINVAL;
+
return cpufreq_frequency_table_verify(pol, data->powernow_table);
}
@@ -977,7 +983,7 @@ static int __cpuinit powernowk8_cpu_init
{
struct powernow_k8_data *data;
cpumask_t oldmask = CPU_MASK_ALL;
- int rc, i;
+ int rc;
if (!cpu_online(pol->cpu))
return -ENODEV;
@@ -1063,8 +1069,7 @@ static int __cpuinit powernowk8_cpu_init
printk("cpu_init done, current fid 0x%x, vid 0x%x\n",
data->currfid, data->currvid);
- for_each_cpu_mask(i, cpu_core_map[pol->cpu])
- powernow_data[i] = data;
+ powernow_data[pol->cpu] = data;
return 0;
@@ -1104,6 +1109,9 @@ static unsigned int powernowk8_get (unsi
if (!data)
return -EINVAL;
+ if (!data)
+ return -EINVAL;
+
set_cpus_allowed(current, cpumask_of_cpu(cpu));
if (smp_processor_id() != cpu) {
printk(KERN_ERR PFX "limiting to CPU %d failed in powernowk8_get\n", cpu);
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index 8d8aa9d..db12017 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -110,21 +110,6 @@ static int __init mpf_checksum(unsigned
static int mpc_record;
static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __initdata;
-#ifdef CONFIG_X86_NUMAQ
-static int MP_valid_apicid(int apicid, int version)
-{
- return hweight_long(apicid & 0xf) == 1 && (apicid >> 4) != 0xf;
-}
-#else
-static int MP_valid_apicid(int apicid, int version)
-{
- if (version >= 0x14)
- return apicid < 0xff;
- else
- return apicid < 0xf;
-}
-#endif
-
static void __devinit MP_processor_info (struct mpc_config_processor *m)
{
int ver, apicid;
@@ -190,12 +175,6 @@ static void __devinit MP_processor_info
ver = m->mpc_apicver;
- if (!MP_valid_apicid(apicid, ver)) {
- printk(KERN_WARNING "Processor #%d INVALID. (Max ID: %d).\n",
- m->mpc_apicid, MAX_APICS);
- return;
- }
-
/*
* Validate version
*/
diff --git a/arch/i386/kernel/reboot_fixups.c b/arch/i386/kernel/reboot_fixups.c
index 10e21a4..99aab41 100644
--- a/arch/i386/kernel/reboot_fixups.c
+++ b/arch/i386/kernel/reboot_fixups.c
@@ -51,7 +51,5 @@ void mach_reboot_fixups(void)
cur->reboot_fixup(dev);
}
-
- printk(KERN_WARNING "No reboot fixup found for your hardware\n");
}
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index eacc3f0..80cb3b2 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -963,6 +963,36 @@ efi_memory_present_wrapper(unsigned long
return 0;
}
+ /*
+ * This function checks if the entire range <start,end> is mapped with type.
+ *
+ * Note: this function only works correct if the e820 table is sorted and
+ * not-overlapping, which is the case
+ */
+int __init
+e820_all_mapped(unsigned long start, unsigned long end, unsigned type)
+{
+ int i;
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+ if (type && ei->type != type)
+ continue;
+ /* is the region (part) in overlap with the current region ?*/
+ if (ei->addr >= end || ei->addr + ei->size <= start)
+ continue;
+ /* if the region is at the beginning of <start,end> we move
+ * start to the end of the region since it's ok until there
+ */
+ if (ei->addr <= start)
+ start = ei->addr + ei->size;
+ /* if start is now at or beyond end, we're done, full
+ * coverage */
+ if (start >= end)
+ return 1; /* we're done */
+ }
+ return 0;
+}
+
/*
* Find the highest page frame number we have available
*/
@@ -1317,8 +1347,8 @@ legacy_init_iomem_resources(struct resou
/*
* Request address space for all standard resources
*
- * This is called just before pcibios_assign_resources(), which is also
- * an fs_initcall, but is linked in later (in arch/i386/pci/i386.c).
+ * This is called just before pcibios_init(), which is also a
+ * subsys_initcall, but is linked in later (in arch/i386/pci/common.c).
*/
static int __init request_standard_resources(void)
{
@@ -1339,7 +1369,7 @@ static int __init request_standard_resou
return 0;
}
-fs_initcall(request_standard_resources);
+subsys_initcall(request_standard_resources);
static void __init register_memory(void)
{
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 9f66ac5..ae6534a 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -651,6 +651,7 @@ void __init mem_init(void)
* Specifically, in the case of x86, we will always add
* memory to the highmem for now.
*/
+#ifdef CONFIG_HOTPLUG_MEMORY
#ifndef CONFIG_NEED_MULTIPLE_NODES
int add_memory(u64 start, u64 size)
{
@@ -667,6 +668,7 @@ int remove_memory(u64 start, u64 size)
return -EINVAL;
}
#endif
+#endif
kmem_cache_t *pgd_cache;
kmem_cache_t *pmd_cache;
diff --git a/arch/i386/pci/direct.c b/arch/i386/pci/direct.c
index 99012b9..0659ced 100644
--- a/arch/i386/pci/direct.c
+++ b/arch/i386/pci/direct.c
@@ -4,6 +4,7 @@
#include <linux/pci.h>
#include <linux/init.h>
+#include <linux/dmi.h>
#include "pci.h"
/*
@@ -18,8 +19,10 @@ int pci_conf1_read(unsigned int seg, uns
{
unsigned long flags;
- if (!value || (bus > 255) || (devfn > 255) || (reg > 255))
+ if (!value || (bus > 255) || (devfn > 255) || (reg > 255)) {
+ *value = -1;
return -EINVAL;
+ }
spin_lock_irqsave(&pci_config_lock, flags);
@@ -188,6 +191,10 @@ static int __init pci_sanity_check(struc
if (pci_probe & PCI_NO_CHECKS)
return 1;
+ /* Assume Type 1 works for newer systems.
+ This handles machines that don't have anything on PCI Bus 0. */
+ if (dmi_get_year(DMI_BIOS_DATE) >= 2001)
+ return 1;
for (devfn = 0; devfn < 0x100; devfn++) {
if (o->read(0, 0, devfn, PCI_CLASS_DEVICE, 2, &x))
diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c
index 6137890..f77d7f8 100644
--- a/arch/i386/pci/mmconfig.c
+++ b/arch/i386/pci/mmconfig.c
@@ -12,14 +12,20 @@
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/acpi.h>
+#include <asm/e820.h>
#include "pci.h"
+#define MMCONFIG_APER_SIZE (256*1024*1024)
+
+/* Assume systems with more busses have correct MCFG */
+#define MAX_CHECK_BUS 16
+
#define mmcfg_virt_addr ((void __iomem *) fix_to_virt(FIX_PCIE_MCFG))
/* The base address of the last MMCONFIG device accessed */
static u32 mmcfg_last_accessed_device;
-static DECLARE_BITMAP(fallback_slots, 32);
+static DECLARE_BITMAP(fallback_slots, MAX_CHECK_BUS*32);
/*
* Functions for accessing PCI configuration space with MMCONFIG accesses
@@ -29,8 +35,8 @@ static u32 get_base_addr(unsigned int se
int cfg_num = -1;
struct acpi_table_mcfg_config *cfg;
- if (seg == 0 && bus == 0 &&
- test_bit(PCI_SLOT(devfn), fallback_slots))
+ if (seg == 0 && bus < MAX_CHECK_BUS &&
+ test_bit(PCI_SLOT(devfn) + 32*bus, fallback_slots))
return 0;
while (1) {
@@ -74,8 +80,10 @@ static int pci_mmcfg_read(unsigned int s
unsigned long flags;
u32 base;
- if (!value || (bus > 255) || (devfn > 255) || (reg > 4095))
+ if (!value || (bus > 255) || (devfn > 255) || (reg > 4095)) {
+ *value = -1;
return -EINVAL;
+ }
base = get_base_addr(seg, bus, devfn);
if (!base)
@@ -146,29 +154,34 @@ static struct pci_raw_ops pci_mmcfg = {
Normally this can be expressed in the MCFG by not listing them
and assigning suitable _SEGs, but this isn't implemented in some BIOS.
Instead try to discover all devices on bus 0 that are unreachable using MM
- and fallback for them.
- We only do this for bus 0/seg 0 */
+ and fallback for them. */
static __init void unreachable_devices(void)
{
- int i;
+ int i, k;
unsigned long flags;
- for (i = 0; i < 32; i++) {
- u32 val1;
- u32 addr;
-
- pci_conf1_read(0, 0, PCI_DEVFN(i, 0), 0, 4, &val1);
- if (val1 == 0xffffffff)
- continue;
-
- /* Locking probably not needed, but safer */
- spin_lock_irqsave(&pci_config_lock, flags);
- addr = get_base_addr(0, 0, PCI_DEVFN(i, 0));
- if (addr != 0)
- pci_exp_set_dev_base(addr, 0, PCI_DEVFN(i, 0));
- if (addr == 0 || readl((u32 __iomem *)mmcfg_virt_addr) != val1)
- set_bit(i, fallback_slots);
- spin_unlock_irqrestore(&pci_config_lock, flags);
+ for (k = 0; k < MAX_CHECK_BUS; k++) {
+ for (i = 0; i < 32; i++) {
+ u32 val1;
+ u32 addr;
+
+ pci_conf1_read(0, k, PCI_DEVFN(i, 0), 0, 4, &val1);
+ if (val1 == 0xffffffff)
+ continue;
+
+ /* Locking probably not needed, but safer */
+ spin_lock_irqsave(&pci_config_lock, flags);
+ addr = get_base_addr(0, k, PCI_DEVFN(i, 0));
+ if (addr != 0)
+ pci_exp_set_dev_base(addr, k, PCI_DEVFN(i, 0));
+ if (addr == 0 ||
+ readl((u32 __iomem *)mmcfg_virt_addr) != val1) {
+ set_bit(i, fallback_slots);
+ printk(KERN_NOTICE
+ "PCI: No mmconfig possible on %x:%x\n", k, i);
+ }
+ spin_unlock_irqrestore(&pci_config_lock, flags);
+ }
}
}
@@ -183,6 +196,14 @@ void __init pci_mmcfg_init(void)
(pci_mmcfg_config[0].base_address == 0))
return;
+ if (!e820_all_mapped(pci_mmcfg_config[0].base_address,
+ pci_mmcfg_config[0].base_address + MMCONFIG_APER_SIZE,
+ E820_RESERVED)) {
+ printk(KERN_ERR "PCI: BIOS Bug: MCFG area is not E820-reserved\n");
+ printk(KERN_ERR "PCI: Not using MMCONFIG.\n");
+ return;
+ }
+
printk(KERN_INFO "PCI: Using MMCONFIG\n");
raw_pci_ops = &pci_mmcfg;
pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF;
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 4310b4a..7df2fe1 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -136,6 +136,11 @@ config X86_L1_CACHE_SHIFT
default "7" if GENERIC_CPU || MPSC
default "6" if MK8
+config X86_INTERNODE_CACHE_BYTES
+ int
+ default "4096" if X86_VSMP
+ default X86_L1_CACHE_BYTES if !X86_VSMP
+
config X86_TSC
bool
default y
diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile
index 585fd4a..e573e2a 100644
--- a/arch/x86_64/Makefile
+++ b/arch/x86_64/Makefile
@@ -24,37 +24,37 @@
LDFLAGS := -m elf_x86_64
OBJCOPYFLAGS := -O binary -R .note -R .comment -S
LDFLAGS_vmlinux :=
-
CHECKFLAGS += -D__x86_64__ -m64
+cflags-y :=
cflags-$(CONFIG_MK8) += $(call cc-option,-march=k8)
cflags-$(CONFIG_MPSC) += $(call cc-option,-march=nocona)
cflags-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mtune=generic)
-CFLAGS += $(cflags-y)
-CFLAGS += -m64
-CFLAGS += -mno-red-zone
-CFLAGS += -mcmodel=kernel
-CFLAGS += -pipe
+cflags-y += -m64
+cflags-y += -mno-red-zone
+cflags-y += -mcmodel=kernel
+cflags-y += -pipe
cflags-$(CONFIG_REORDER) += -ffunction-sections
# this makes reading assembly source easier, but produces worse code
# actually it makes the kernel smaller too.
-CFLAGS += -fno-reorder-blocks
-CFLAGS += -Wno-sign-compare
+cflags-y += -fno-reorder-blocks
+cflags-y += -Wno-sign-compare
ifneq ($(CONFIG_UNWIND_INFO),y)
-CFLAGS += -fno-asynchronous-unwind-tables
+cflags-y += -fno-asynchronous-unwind-tables
endif
ifneq ($(CONFIG_DEBUG_INFO),y)
# -fweb shrinks the kernel a bit, but the difference is very small
# it also messes up debugging, so don't use it for now.
-#CFLAGS += $(call cc-option,-fweb)
+#cflags-y += $(call cc-option,-fweb)
endif
# -funit-at-a-time shrinks the kernel .text considerably
# unfortunately it makes reading oopses harder.
-CFLAGS += $(call cc-option,-funit-at-a-time)
+cflags-y += $(call cc-option,-funit-at-a-time)
# prevent gcc from generating any FP code by mistake
-CFLAGS += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
+cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
+CFLAGS += $(cflags-y)
AFLAGS += -m64
head-y := arch/x86_64/kernel/head.o arch/x86_64/kernel/head64.o arch/x86_64/kernel/init_task.o
diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig
index 566ecc9..3c45ec2 100644
--- a/arch/x86_64/defconfig
+++ b/arch/x86_64/defconfig
@@ -1,7 +1,7 @@
#
# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.16-git9
-# Sat Mar 25 15:18:40 2006
+# Linux kernel version: 2.6.17-rc1
+# Mon Apr 3 16:11:14 2006
#
CONFIG_X86_64=y
CONFIG_64BIT=y
@@ -9,6 +9,7 @@ CONFIG_X86=y
CONFIG_SEMAPHORE_SLEEPERS=y
CONFIG_MMU=y
CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HWEIGHT=y
CONFIG_GENERIC_CALIBRATE_DELAY=y
CONFIG_X86_CMPXCHG=y
CONFIG_EARLY_PRINTK=y
@@ -55,10 +56,6 @@ CONFIG_BASE_FULL=y
CONFIG_FUTEX=y
CONFIG_EPOLL=y
CONFIG_SHMEM=y
-CONFIG_CC_ALIGN_FUNCTIONS=0
-CONFIG_CC_ALIGN_LABELS=0
-CONFIG_CC_ALIGN_LOOPS=0
-CONFIG_CC_ALIGN_JUMPS=0
CONFIG_SLAB=y
# CONFIG_TINY_SHMEM is not set
CONFIG_BASE_SMALL=0
@@ -70,7 +67,6 @@ CONFIG_BASE_SMALL=0
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
CONFIG_MODULE_FORCE_UNLOAD=y
-CONFIG_OBSOLETE_MODPARM=y
# CONFIG_MODVERSIONS is not set
# CONFIG_MODULE_SRCVERSION_ALL is not set
# CONFIG_KMOD is not set
@@ -81,6 +77,7 @@ CONFIG_STOP_MACHINE=y
#
CONFIG_LBD=y
# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
#
# IO Schedulers
@@ -105,6 +102,7 @@ CONFIG_X86_PC=y
CONFIG_GENERIC_CPU=y
CONFIG_X86_L1_CACHE_BYTES=128
CONFIG_X86_L1_CACHE_SHIFT=7
+CONFIG_X86_INTERNODE_CACHE_BYTES=128
CONFIG_X86_TSC=y
CONFIG_X86_GOOD_APIC=y
# CONFIG_MICROCODE is not set
@@ -116,6 +114,7 @@ CONFIG_X86_LOCAL_APIC=y
CONFIG_MTRR=y
CONFIG_SMP=y
CONFIG_SCHED_SMT=y
+CONFIG_SCHED_MC=y
# CONFIG_PREEMPT_NONE is not set
CONFIG_PREEMPT_VOLUNTARY=y
# CONFIG_PREEMPT is not set
@@ -138,6 +137,7 @@ CONFIG_NEED_MULTIPLE_NODES=y
CONFIG_SPLIT_PTLOCK_CPUS=4
CONFIG_MIGRATION=y
CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID=y
+CONFIG_OUT_OF_LINE_PFN_TO_PAGE=y
CONFIG_NR_CPUS=32
CONFIG_HOTPLUG_CPU=y
CONFIG_HPET_TIMER=y
@@ -289,6 +289,7 @@ CONFIG_IP_PNP_DHCP=y
# CONFIG_INET_AH is not set
# CONFIG_INET_ESP is not set
# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
# CONFIG_INET_TUNNEL is not set
CONFIG_INET_DIAG=y
CONFIG_INET_TCP_DIAG=y
@@ -300,6 +301,7 @@ CONFIG_IPV6=y
# CONFIG_INET6_AH is not set
# CONFIG_INET6_ESP is not set
# CONFIG_INET6_IPCOMP is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
# CONFIG_INET6_TUNNEL is not set
# CONFIG_IPV6_TUNNEL is not set
# CONFIG_NETFILTER is not set
@@ -704,7 +706,6 @@ CONFIG_S2IO=m
# Wireless LAN (non-hamradio)
#
# CONFIG_NET_RADIO is not set
-# CONFIG_NET_WIRELESS_RTNETLINK is not set
#
# Wan interfaces
@@ -791,7 +792,7 @@ CONFIG_HW_CONSOLE=y
#
CONFIG_SERIAL_8250=y
CONFIG_SERIAL_8250_CONSOLE=y
-# CONFIG_SERIAL_8250_ACPI is not set
+CONFIG_SERIAL_8250_PCI=y
CONFIG_SERIAL_8250_NR_UARTS=4
CONFIG_SERIAL_8250_RUNTIME_UARTS=4
# CONFIG_SERIAL_8250_EXTENDED is not set
@@ -921,6 +922,7 @@ CONFIG_HWMON=y
# Digital Video Broadcasting Devices
#
# CONFIG_DVB is not set
+# CONFIG_USB_DABUSB is not set
#
# Graphics support
@@ -932,6 +934,8 @@ CONFIG_VIDEO_SELECT=y
# Console display driver support
#
CONFIG_VGA_CONSOLE=y
+CONFIG_VGACON_SOFT_SCROLLBACK=y
+CONFIG_VGACON_SOFT_SCROLLBACK_SIZE=256
CONFIG_DUMMY_CONSOLE=y
#
@@ -1058,15 +1062,6 @@ CONFIG_USB_HIDINPUT=y
# CONFIG_USB_MICROTEK is not set
#
-# USB Multimedia devices
-#
-# CONFIG_USB_DABUSB is not set
-
-#
-# Video4Linux support is needed for USB Multimedia device support
-#
-
-#
# USB Network Adapters
#
# CONFIG_USB_CATC is not set
@@ -1118,9 +1113,15 @@ CONFIG_USB_MON=y
# CONFIG_MMC is not set
#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
# InfiniBand support
#
# CONFIG_INFINIBAND is not set
+# CONFIG_IPATH_CORE is not set
#
# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
@@ -1128,6 +1129,11 @@ CONFIG_USB_MON=y
# CONFIG_EDAC is not set
#
+# Real Time Clock
+#
+# CONFIG_RTC_CLASS is not set
+
+#
# Firmware Drivers
#
# CONFIG_EDD is not set
diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 35b2fac..5a98026 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -15,6 +15,8 @@
#include <asm/vsyscall32.h>
#include <linux/linkage.h>
+#define IA32_NR_syscalls ((ia32_syscall_end - ia32_sys_call_table)/8)
+
.macro IA32_ARG_FIXUP noebp=0
movl %edi,%r8d
.if \noebp
@@ -109,8 +111,8 @@ ENTRY(ia32_sysenter_target)
CFI_REMEMBER_STATE
jnz sysenter_tracesys
sysenter_do_call:
- cmpl $(IA32_NR_syscalls),%eax
- jae ia32_badsys
+ cmpl $(IA32_NR_syscalls-1),%eax
+ ja ia32_badsys
IA32_ARG_FIXUP 1
call *ia32_sys_call_table(,%rax,8)
movq %rax,RAX-ARGOFFSET(%rsp)
@@ -210,8 +212,8 @@ ENTRY(ia32_cstar_target)
CFI_REMEMBER_STATE
jnz cstar_tracesys
cstar_do_call:
- cmpl $IA32_NR_syscalls,%eax
- jae ia32_badsys
+ cmpl $IA32_NR_syscalls-1,%eax
+ ja ia32_badsys
IA32_ARG_FIXUP 1
call *ia32_sys_call_table(,%rax,8)
movq %rax,RAX-ARGOFFSET(%rsp)
@@ -296,8 +298,8 @@ ENTRY(ia32_syscall)
testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
jnz ia32_tracesys
ia32_do_syscall:
- cmpl $(IA32_NR_syscalls),%eax
- jae ia32_badsys
+ cmpl $(IA32_NR_syscalls-1),%eax
+ ja ia32_badsys
IA32_ARG_FIXUP
call *ia32_sys_call_table(,%rax,8) # xxx: rip relative
ia32_sysret:
@@ -685,12 +687,11 @@ ia32_sys_call_table:
.quad sys_readlinkat /* 305 */
.quad sys_fchmodat
.quad sys_faccessat
- .quad sys_ni_syscall /* pselect6 for now */
- .quad sys_ni_syscall /* ppoll for now */
+ .quad quiet_ni_syscall /* pselect6 for now */
+ .quad quiet_ni_syscall /* ppoll for now */
.quad sys_unshare /* 310 */
.quad compat_sys_set_robust_list
.quad compat_sys_get_robust_list
+ .quad sys_splice
+ .quad sys_sync_file_range
ia32_syscall_end:
- .rept IA32_NR_syscalls-(ia32_syscall_end-ia32_sys_call_table)/8
- .quad ni_syscall
- .endr
diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c
index fffd6b0..70b9d21 100644
--- a/arch/x86_64/kernel/aperture.c
+++ b/arch/x86_64/kernel/aperture.c
@@ -80,7 +80,7 @@ static int __init aperture_valid(char *n
printk("Aperture from %s beyond 4GB. Ignoring.\n",name);
return 0;
}
- if (e820_mapped(aper_base, aper_base + aper_size, E820_RAM)) {
+ if (e820_any_mapped(aper_base, aper_base + aper_size, E820_RAM)) {
printk("Aperture from %s pointing to e820 RAM. Ignoring.\n",name);
return 0;
}
diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index 293cd71..62776c0 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -80,7 +80,12 @@ static inline int bad_addr(unsigned long
return 0;
}
-int __init e820_mapped(unsigned long start, unsigned long end, unsigned type)
+/*
+ * This function checks if any part of the range <start,end> is mapped
+ * with type.
+ */
+int __meminit
+e820_any_mapped(unsigned long start, unsigned long end, unsigned type)
{
int i;
for (i = 0; i < e820.nr_map; i++) {
@@ -94,6 +99,35 @@ int __init e820_mapped(unsigned long sta
return 0;
}
+/*
+ * This function checks if the entire range <start,end> is mapped with type.
+ *
+ * Note: this function only works correct if the e820 table is sorted and
+ * not-overlapping, which is the case
+ */
+int __init e820_all_mapped(unsigned long start, unsigned long end, unsigned type)
+{
+ int i;
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+ if (type && ei->type != type)
+ continue;
+ /* is the region (part) in overlap with the current region ?*/
+ if (ei->addr >= end || ei->addr + ei->size <= start)
+ continue;
+
+ /* if the region is at the beginning of <start,end> we move
+ * start to the end of the region since it's ok until there
+ */
+ if (ei->addr <= start)
+ start = ei->addr + ei->size;
+ /* if start is now at or beyond end, we're done, full coverage */
+ if (start >= end)
+ return 1; /* we're done */
+ }
+ return 0;
+}
+
/*
* Find a free area in a specific range.
*/
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 8538bfe..c946e4f 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -180,6 +180,10 @@ rff_trace:
*
* XXX if we had a free scratch register we could save the RSP into the stack frame
* and report it properly in ps. Unfortunately we haven't.
+ *
+ * When user can change the frames always force IRET. That is because
+ * it deals with uncanonical addresses better. SYSRET has trouble
+ * with them due to bugs in both AMD and Intel CPUs.
*/
ENTRY(system_call)
@@ -254,7 +258,10 @@ sysret_signal:
xorl %esi,%esi # oldset -> arg2
call ptregscall_common
1: movl $_TIF_NEED_RESCHED,%edi
- jmp sysret_check
+ /* Use IRET because user could have changed frame. This
+ works because ptregscall_common has called FIXUP_TOP_OF_STACK. */
+ cli
+ jmp int_with_check
badsys:
movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
@@ -280,7 +287,8 @@ tracesys:
call syscall_trace_leave
RESTORE_TOP_OF_STACK %rbx
RESTORE_REST
- jmp ret_from_sys_call
+ /* Use IRET because user could have changed frame */
+ jmp int_ret_from_sys_call
CFI_ENDPROC
/*
@@ -408,25 +416,9 @@ ENTRY(stub_execve)
CFI_ADJUST_CFA_OFFSET -8
CFI_REGISTER rip, r11
SAVE_REST
- movq %r11, %r15
- CFI_REGISTER rip, r15
FIXUP_TOP_OF_STACK %r11
call sys_execve
- GET_THREAD_INFO(%rcx)
- bt $TIF_IA32,threadinfo_flags(%rcx)
- CFI_REMEMBER_STATE
- jc exec_32bit
RESTORE_TOP_OF_STACK %r11
- movq %r15, %r11
- CFI_REGISTER rip, r11
- RESTORE_REST
- pushq %r11
- CFI_ADJUST_CFA_OFFSET 8
- CFI_REL_OFFSET rip, 0
- ret
-
-exec_32bit:
- CFI_RESTORE_STATE
movq %rax,RAX(%rsp)
RESTORE_REST
jmp int_ret_from_sys_call
diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 10b3e34..6f0790e 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -29,6 +29,8 @@
#define MISC_MCELOG_MINOR 227
#define NR_BANKS 6
+atomic_t mce_entry;
+
static int mce_dont_init;
/* 0: always panic, 1: panic if deadlock possible, 2: try to avoid panic,
@@ -172,10 +174,12 @@ void do_machine_check(struct pt_regs * r
int i;
int panicm_found = 0;
+ atomic_inc(&mce_entry);
+
if (regs)
notify_die(DIE_NMI, "machine check", regs, error_code, 18, SIGKILL);
if (!banks)
- return;
+ goto out2;
memset(&m, 0, sizeof(struct mce));
m.cpu = safe_smp_processor_id();
@@ -266,6 +270,8 @@ void do_machine_check(struct pt_regs * r
out:
/* Last thing done in the machine check exception to clear state. */
wrmsrl(MSR_IA32_MCG_STATUS, 0);
+ out2:
+ atomic_dec(&mce_entry);
}
/*
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index d9e4067..4e6357f 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -34,6 +34,7 @@
#include <asm/proto.h>
#include <asm/kdebug.h>
#include <asm/local.h>
+#include <asm/mce.h>
/*
* lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
@@ -480,6 +481,12 @@ void __kprobes nmi_watchdog_tick(struct
__get_cpu_var(nmi_touch) = 0;
touched = 1;
}
+#ifdef CONFIG_X86_MCE
+ /* Could check oops_in_progress here too, but it's safer
+ not too */
+ if (atomic_read(&mce_entry) > 0)
+ touched = 1;
+#endif
if (!touched && __get_cpu_var(last_irq_sum) == sum) {
/*
* Ayiee, looks like this CPU is stuck ...
diff --git a/arch/x86_64/kernel/pci-dma.c b/arch/x86_64/kernel/pci-dma.c
index 03c9eee..af035ed 100644
--- a/arch/x86_64/kernel/pci-dma.c
+++ b/arch/x86_64/kernel/pci-dma.c
@@ -48,9 +48,11 @@ dma_alloc_pages(struct device *dev, gfp_
{
struct page *page;
int node;
+#ifdef CONFIG_PCI
if (dev->bus == &pci_bus_type)
node = pcibus_to_node(to_pci_dev(dev)->bus);
else
+#endif
node = numa_node_id();
page = alloc_pages_node(node, gfp, order);
return page ? page_address(page) : NULL;
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 70dd8e5..1c44b53 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -781,10 +781,16 @@ long do_arch_prctl(struct task_struct *t
}
case ARCH_GET_GS: {
unsigned long base;
+ unsigned gsindex;
if (task->thread.gsindex == GS_TLS_SEL)
base = read_32bit_tls(task, GS_TLS);
- else if (doit)
- rdmsrl(MSR_KERNEL_GS_BASE, base);
+ else if (doit) {
+ asm("movl %%gs,%0" : "=r" (gsindex));
+ if (gsindex)
+ rdmsrl(MSR_KERNEL_GS_BASE, base);
+ else
+ base = task->thread.gs;
+ }
else
base = task->thread.gs;
ret = put_user(base, (unsigned long __user *)addr);
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 0856ad4..c50b067 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -353,8 +353,10 @@ static __init void parse_cmdline_early (
if (fullarg(from, "enable_timer_pin_1"))
disable_timer_pin_1 = -1;
- if (fullarg(from, "nolapic") || fullarg(from, "disableapic"))
+ if (fullarg(from, "nolapic") || fullarg(from, "disableapic")) {
+ clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
disable_apic = 1;
+ }
if (fullarg(from, "noapic"))
skip_ioapic_setup = 1;
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index ef8bc46..7392570 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -726,7 +726,7 @@ static __init int late_hpet_init(void)
unsigned int ntimer;
if (!vxtime.hpet_address)
- return -1;
+ return 0;
memset(&hd, 0, sizeof (hd));
@@ -917,6 +917,8 @@ void __init time_init(void)
vxtime.hpet_address = 0;
if (hpet_use_timer) {
+ /* set tick_nsec to use the proper rate for HPET */
+ tick_nsec = TICK_NSEC_HPET;
cpu_khz = hpet_calibrate_tsc();
timename = "HPET";
#ifdef CONFIG_X86_PM_TIMER
diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index 39ff070..b81f473 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/arch/x86_64/kernel/vmlinux.lds.S
@@ -65,7 +65,7 @@ SECTIONS
.data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) {
*(.data.cacheline_aligned)
}
- . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
+ . = ALIGN(CONFIG_X86_INTERNODE_CACHE_BYTES);
.data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET) {
*(.data.read_mostly)
}
diff --git a/arch/x86_64/kernel/x8664_ksyms.c b/arch/x86_64/kernel/x8664_ksyms.c
index d78f460..fec4e52 100644
--- a/arch/x86_64/kernel/x8664_ksyms.c
+++ b/arch/x86_64/kernel/x8664_ksyms.c
@@ -112,7 +112,6 @@ EXPORT_SYMBOL_GPL(unset_nmi_callback);
#undef memcpy
#undef memset
#undef memmove
-#undef strlen
extern void * memset(void *,int,__kernel_size_t);
extern size_t strlen(const char *);
@@ -121,7 +120,6 @@ extern void * memcpy(void *,const void *
extern void * __memcpy(void *,const void *,__kernel_size_t);
EXPORT_SYMBOL(memset);
-EXPORT_SYMBOL(strlen);
EXPORT_SYMBOL(strpbrk);
EXPORT_SYMBOL(memmove);
EXPORT_SYMBOL(memcpy);
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index e5f7f1c..4ba34e9 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -305,7 +305,7 @@ static void __meminit phys_pud_init(pud_
if (paddr >= end)
break;
- if (!after_bootmem && !e820_mapped(paddr, paddr+PUD_SIZE, 0)) {
+ if (!after_bootmem && !e820_any_mapped(paddr, paddr+PUD_SIZE, 0)) {
set_pud(pud, __pud(0));
continue;
}
@@ -507,9 +507,8 @@ void __init clear_kernel_mapping(unsigne
/*
* Memory hotplug specific functions
- * These are only for non-NUMA machines right now.
*/
-#ifdef CONFIG_MEMORY_HOTPLUG
+#if defined(CONFIG_ACPI_HOTPLUG_MEMORY) || defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)
void online_page(struct page *page)
{
@@ -520,6 +519,38 @@ void online_page(struct page *page)
num_physpages++;
}
+#ifndef CONFIG_MEMORY_HOTPLUG
+/*
+ * Memory Hotadd without sparsemem. The mem_maps have been allocated in advance,
+ * just online the pages.
+ */
+int __add_pages(struct zone *z, unsigned long start_pfn, unsigned long nr_pages)
+{
+ int err = -EIO;
+ unsigned long pfn;
+ unsigned long total = 0, mem = 0;
+ for (pfn = start_pfn; pfn < start_pfn + nr_pages; pfn++) {
+ if (pfn_valid(pfn)) {
+ online_page(pfn_to_page(pfn));
+ err = 0;
+ mem++;
+ }
+ total++;
+ }
+ if (!err) {
+ z->spanned_pages += total;
+ z->present_pages += mem;
+ z->zone_pgdat->node_spanned_pages += total;
+ z->zone_pgdat->node_present_pages += mem;
+ }
+ return err;
+}
+#endif
+
+/*
+ * Memory is added always to NORMAL zone. This means you will never get
+ * additional DMA/DMA32 memory.
+ */
int add_memory(u64 start, u64 size)
{
struct pglist_data *pgdat = NODE_DATA(0);
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index 4be82d6..cc02573 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -100,11 +100,30 @@ int early_pfn_to_nid(unsigned long pfn)
}
#endif
+static void * __init
+early_node_mem(int nodeid, unsigned long start, unsigned long end,
+ unsigned long size)
+{
+ unsigned long mem = find_e820_area(start, end, size);
+ void *ptr;
+ if (mem != -1L)
+ return __va(mem);
+ ptr = __alloc_bootmem_nopanic(size,
+ SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS));
+ if (ptr == 0) {
+ printk(KERN_ERR "Cannot find %lu bytes in node %d\n",
+ size, nodeid);
+ return NULL;
+ }
+ return ptr;
+}
+
/* Initialize bootmem allocator for a node */
void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long end)
{
unsigned long start_pfn, end_pfn, bootmap_pages, bootmap_size, bootmap_start;
unsigned long nodedata_phys;
+ void *bootmap;
const int pgdat_size = round_up(sizeof(pg_data_t), PAGE_SIZE);
start = round_up(start, ZONE_ALIGN);
@@ -114,13 +133,11 @@ void __init setup_node_bootmem(int nodei
start_pfn = start >> PAGE_SHIFT;
end_pfn = end >> PAGE_SHIFT;
- nodedata_phys = find_e820_area(start, end, pgdat_size);
- if (nodedata_phys == -1L)
- panic("Cannot find memory pgdat in node %d\n", nodeid);
-
- Dprintk("nodedata_phys %lx\n", nodedata_phys);
+ node_data[nodeid] = early_node_mem(nodeid, start, end, pgdat_size);
+ if (node_data[nodeid] == NULL)
+ return;
+ nodedata_phys = __pa(node_data[nodeid]);
- node_data[nodeid] = phys_to_virt(nodedata_phys);
memset(NODE_DATA(nodeid), 0, sizeof(pg_data_t));
NODE_DATA(nodeid)->bdata = &plat_node_bdata[nodeid];
NODE_DATA(nodeid)->node_start_pfn = start_pfn;
@@ -129,9 +146,15 @@ void __init setup_node_bootmem(int nodei
/* Find a place for the bootmem map */
bootmap_pages = bootmem_bootmap_pages(end_pfn - start_pfn);
bootmap_start = round_up(nodedata_phys + pgdat_size, PAGE_SIZE);
- bootmap_start = find_e820_area(bootmap_start, end, bootmap_pages<<PAGE_SHIFT);
- if (bootmap_start == -1L)
- panic("Not enough continuous space for bootmap on node %d", nodeid);
+ bootmap = early_node_mem(nodeid, bootmap_start, end,
+ bootmap_pages<<PAGE_SHIFT);
+ if (bootmap == NULL) {
+ if (nodedata_phys < start || nodedata_phys >= end)
+ free_bootmem((unsigned long)node_data[nodeid],pgdat_size);
+ node_data[nodeid] = NULL;
+ return;
+ }
+ bootmap_start = __pa(bootmap);
Dprintk("bootmap start %lu pages %lu\n", bootmap_start, bootmap_pages);
bootmap_size = init_bootmem_node(NODE_DATA(nodeid),
@@ -142,6 +165,9 @@ void __init setup_node_bootmem(int nodei
reserve_bootmem_node(NODE_DATA(nodeid), nodedata_phys, pgdat_size);
reserve_bootmem_node(NODE_DATA(nodeid), bootmap_start, bootmap_pages<<PAGE_SHIFT);
+#ifdef CONFIG_ACPI_NUMA
+ srat_reserve_add_area(nodeid);
+#endif
node_set_online(nodeid);
}
@@ -335,6 +361,8 @@ __init int numa_setup(char *opt)
#ifdef CONFIG_ACPI_NUMA
if (!strncmp(opt,"noacpi",6))
acpi_numa = -1;
+ if (!strncmp(opt,"hotadd=", 7))
+ hotadd_percent = simple_strtoul(opt+7, NULL, 10);
#endif
return 1;
}
diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index 2eb8795..15ae9fc 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -15,15 +15,26 @@
#include <linux/bitmap.h>
#include <linux/module.h>
#include <linux/topology.h>
+#include <linux/bootmem.h>
+#include <linux/mm.h>
#include <asm/proto.h>
#include <asm/numa.h>
#include <asm/e820.h>
+#if (defined(CONFIG_ACPI_HOTPLUG_MEMORY) || \
+ defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)) \
+ && !defined(CONFIG_MEMORY_HOTPLUG)
+#define RESERVE_HOTADD 1
+#endif
+
static struct acpi_table_slit *acpi_slit;
static nodemask_t nodes_parsed __initdata;
static nodemask_t nodes_found __initdata;
static struct bootnode nodes[MAX_NUMNODES] __initdata;
+static struct bootnode nodes_add[MAX_NUMNODES] __initdata;
+static int found_add_area __initdata;
+int hotadd_percent __initdata = 10;
static u8 pxm2node[256] = { [0 ... 255] = 0xff };
/* Too small nodes confuse the VM badly. Usually they result
@@ -71,6 +82,10 @@ static __init int conflicting_nodes(unsi
static __init void cutoff_node(int i, unsigned long start, unsigned long end)
{
struct bootnode *nd = &nodes[i];
+
+ if (found_add_area)
+ return;
+
if (nd->start < start) {
nd->start = start;
if (nd->end < nd->start)
@@ -90,6 +105,8 @@ static __init void bad_srat(void)
acpi_numa = -1;
for (i = 0; i < MAX_LOCAL_APIC; i++)
apicid_to_node[i] = NUMA_NO_NODE;
+ for (i = 0; i < MAX_NUMNODES; i++)
+ nodes_add[i].start = nodes[i].end = 0;
}
static __init inline int srat_disabled(void)
@@ -155,11 +172,114 @@ acpi_numa_processor_affinity_init(struct
pxm, pa->apic_id, node);
}
+#ifdef RESERVE_HOTADD
+/*
+ * Protect against too large hotadd areas that would fill up memory.
+ */
+static int hotadd_enough_memory(struct bootnode *nd)
+{
+ static unsigned long allocated;
+ static unsigned long las...
[truncated message content] |
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-09 20:15:45
|
Documentation/x86_64/boot-options.txt | 5
arch/i386/kernel/acpi/boot.c | 5
arch/i386/kernel/apic.c | 23 ++-
arch/i386/kernel/cpu/cpufreq/powernow-k8.c | 18 ++-
arch/i386/kernel/mpparse.c | 21 ---
arch/i386/kernel/reboot_fixups.c | 2
arch/i386/kernel/setup.c | 36 +++++-
arch/i386/mm/init.c | 2
arch/i386/pci/direct.c | 9 +
arch/i386/pci/mmconfig.c | 67 +++++++----
arch/x86_64/Kconfig | 5
arch/x86_64/Makefile | 24 ++--
arch/x86_64/defconfig | 42 ++++---
arch/x86_64/ia32/ia32entry.S | 23 ++-
arch/x86_64/kernel/aperture.c | 2
arch/x86_64/kernel/e820.c | 36 +++++-
arch/x86_64/kernel/entry.S | 28 +---
arch/x86_64/kernel/mce.c | 8 +
arch/x86_64/kernel/nmi.c | 7 +
arch/x86_64/kernel/pci-dma.c | 2
arch/x86_64/kernel/process.c | 10 +
arch/x86_64/kernel/setup.c | 4
arch/x86_64/kernel/time.c | 4
arch/x86_64/kernel/vmlinux.lds.S | 2
arch/x86_64/kernel/x8664_ksyms.c | 2
arch/x86_64/mm/init.c | 37 +++++-
arch/x86_64/mm/numa.c | 46 ++++++-
arch/x86_64/mm/srat.c | 170 +++++++++++++++++++++++++++--
arch/x86_64/pci/mmconfig.c | 53 ++++++---
drivers/acpi/Kconfig | 2
include/asm-i386/apic.h | 2
include/asm-i386/e820.h | 4
include/asm-i386/hpet.h | 1
include/asm-x86_64/e820.h | 3
include/asm-x86_64/hpet.h | 2
include/asm-x86_64/ia32_unistd.h | 2
include/asm-x86_64/mce.h | 7 +
include/asm-x86_64/numa.h | 2
include/asm-x86_64/numnodes.h | 2
include/linux/bootmem.h | 1
include/linux/init.h | 3
include/linux/jiffies.h | 6 +
include/linux/memory_hotplug.h | 14 +-
kernel/timer.c | 2
mm/bootmem.c | 9 +
security/selinux/xfrm.c | 4
46 files changed, 575 insertions(+), 184 deletions(-)
New commits:
commit ebe9c645f0738c1254e7b6208e3dec8a1b9be2ba
Merge: 6b61940... 6764472...
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Sun Apr 9 22:12:03 2006 +0200
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6
commit 67644726317a8274be4a3d0ef85b9ccebaa90304
Author: Dave Jones <da...@re...>
Date: Sun Apr 2 23:34:19 2006 -0700
[SELINUX] Fix build after ipsec decap state changes.
security/selinux/xfrm.c: In function 'selinux_socket_getpeer_dgram':
security/selinux/xfrm.c:284: error: 'struct sec_path' has no member named 'x'
security/selinux/xfrm.c: In function 'selinux_xfrm_sock_rcv_skb':
security/selinux/xfrm.c:317: error: 'struct sec_path' has no member named 'x'
Signed-off-by: Dave Jones <da...@re...>
Signed-off-by: David S. Miller <da...@da...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 66004a6ca23f2a2408b32cbe27fda0389fb8f9dc
Author: Linus Torvalds <tor...@g5...>
Date: Sun Apr 9 12:14:02 2006 -0700
Move request_standard_resources() back to before PCI probing
This effectively undoes the PCI resource allocation changes done in
commit b408cbc704352eccee301e1103b23203ba1c3a0e, but leaves the cleanups
of that commit in place.
We're going back to marking the resources reported by e820 busy _before_
doing PCI probing, so that any PCI resource that clashes with the BIOS-
reported memory map will be reloacted to a non-clashing area.
The reason? Larry Finger reports that his laptop has the cardbus
controller set up by the BIOS so that it conflicts with the e820 memory
map, and needs to be relocated. See
http://bugzilla.kernel.org/show_bug.cgi?id=6337
for more details.
We'll have to work out how to handle the fbcon problem that caused that
commit in the first place in some other way.
Cc: Ivan Kokshaysky <in...@ju...>
Cc: Greg Kroah-Hartman <gr...@su...>
Cc: Antonino A. Daplas <ad...@po...>
Cc: <bj...@lu...>
Tested-by: Larry Finger <Lar...@lw...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit b8feb47f992d314c956add15c1118430120635bb
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:34 2006 +0200
[PATCH] x86_64: Update 32-bit system call table
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 67d53ea5a3d42aadeb1584e757ca4660c0e8a810
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:31 2006 +0200
[PATCH] x86_64: Eliminate IA32_NR_syscalls define
Or rather compute it based on the table length automatically.
This also has the intended side effect of not warning for new system calls
anymore.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit bbd3aff89d4b34ef17a748e4c001ecc5b43e3e55
Author: Sam Ravnborg <sa...@ra...>
Date: Fri Apr 7 19:50:28 2006 +0200
[PATCH] x86_64: fix CONFIG_REORDER
Fix CONFIG_REORDER.
The value of cflags-y was assined to CFLAGS before cflags-y was assigned
the value used for CONFIG_REORDER.
Use cflags-y for all CFLAGS options in the Makefile to avoid this
happening again.
Signed-off-by: Sam Ravnborg <sa...@ra...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 97c2803c9c694cafbd9f5e43a25903e0abf25188
Author: John Blackwood <joh...@cc...>
Date: Fri Apr 7 19:50:25 2006 +0200
[PATCH] x86_64: Plug GS leak in arch_prctl()
In linux-2.6.16, we have noticed a problem where the gs base value
returned from an arch_prtcl(ARCH_GET_GS, ...) call will be incorrect if:
- the current/calling task has NOT set its own gs base yet to a
non-zero value,
- some other task that ran on the same processor previously set their
own gs base to a non-zero value.
In this situation, the ARCH_GET_GS code will read and return the
MSR_KERNEL_GS_BASE msr register.
However, since the __switch_to() code does NOT load/zero the
MSR_KERNEL_GS_BASE register when the task that is switched IN has a zero
next->gs value, the caller of arch_prctl(ARCH_GET_GS, ...) will get back
the value of some previous tasks's gs base value instead of 0.
Change the arch_prctl() ARCH_GET_GS code to only read and return
the MSR_KERNEL_GS_BASE msr register if the 'gs' register of the calling
task is non-zero.
Side note: Since in addition to using arch_prctl(ARCH_SET_GS, ...),
a task can also setup a gs base value by using modify_ldt() and write
an index value into 'gs' from user space, the patch below reads
'gs' instead of using thread.gs, since in the modify_ldt() case,
the thread.gs value will be 0, and incorrect value would be returned
(the task->thread.gs value).
When the user has not set its own gs base value and the 'gs'
register is zero, then the MSR_KERNEL_GS_BASE register will not be
read and a value of zero will be returned by reading and returning
'task->thread.gs'.
The first patch shown below is an attempt at implementing this
approach.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit e48c4729d23a026f3711d5e36add5cce894b4913
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:21 2006 +0200
[PATCH] i386: Remove printk about reboot fixups at reboot
Printk doesn't have any value
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit b20367a6c2a0cd937cb1f0a8cf848f1402fef99c
Author: Jordan Hargrave <jor...@de...>
Date: Fri Apr 7 19:50:18 2006 +0200
[PATCH] x86_64: Fix drift with HPET timer enabled
If the HPET timer is enabled, the clock can drift by ~3 seconds a day.
This is due to the HPET timer not being initialized with the correct
setting (still using PIT count).
If HZ changes, this drift can become even more pronounced.
HPET patch initializes tick_nsec with correct tick_nsec settings for
HPET timer.
Vojtech comments:
"It's not entirely correct (it assumes the HPET ticks totally
exactly), but it's significantly better than assuming the PIT error
there."
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 49c93e84d8b2d602a07c302c7e3cd4fa09095fbb
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:15 2006 +0200
[PATCH] i386/x86-64: Return defined error value for bad PCI config space accesses
Mostly to get better handling when a extended config space
access has to fallback to Type1.
Cc: gr...@su...
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 8c30b1a74aed4041f183e183a149b7dfbdc6c20e
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:12 2006 +0200
[PATCH] i386/x86_64: Check if MCFG works for the first 16 busses
Previously only the first bus would be checked against Type 1.
Why 16? Checking all would need too much memory and we
can assume that systems with more than 16 busses have better than
average quality BIOS.
This is an additional defense against bad MCFG tables.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit e405d067298b2b960bf20318e91ed842157c65bc
Author: Ravikiran G Thirumalai <ki...@sc...>
Date: Fri Apr 7 19:50:09 2006 +0200
[PATCH] x86_64: Fixup read_mostly section on internode cache line size for vSMP
Fixup the read mostly section to start at internode cacheline boundary.
Signed-off-by: Ravikiran Thirumalai <ki...@sc...>
Signed-off-by: Shai Fultheim <sh...@sc...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 3d34ee6891e274dfb6a22930546d37738cdbe9c4
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:06 2006 +0200
[PATCH] x86_64: Don't return error for HPET initialization in initcall
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit ac04dcaf6f567307fbeef9c3c1fff35280e53f02
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:03 2006 +0200
[PATCH] x86_64: Don't export strlen twice
Fix
WARNING: vmlinux: 'strlen' exported twice. Previous export was in vmlinux
Reported by Mats Johannesson
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 7bf36bbc5e0c09271f9efe22162f8cc3f8ebd3d2
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:50:00 2006 +0200
[PATCH] x86_64: When user could have changed RIP always force IRET
Intel EM64T CPUs handle uncanonical return addresses differently
from AMD CPUs.
The exception is reported in the SYSRET, not the next instruction.
This leads to the kernel exception handler running on the user stack
with the wrong GS because the kernel didn't expect exceptions
on this instruction.
This version of the patch has the teething problems that plagued an earlier
version fixed.
This is CVE-2006-0744
Thanks to Ernie Petrides and Asit B. Mallick for analysis and initial
patches.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 553f265fe883a23502ee351845f09334790f18b8
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:57 2006 +0200
[PATCH] x86_64: Don't run NMI watchdog during machine checks
Machine checks can stall the machine for a long time and
it's not good to trigger the nmi watchdog during that.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit be56db6186999a8571ae480cf2b929578f6dfd68
Author: Dave Hansen <hav...@us...>
Date: Fri Apr 7 19:49:54 2006 +0200
[PATCH] x86_64: extra NODES_SHIFT definition
The generic linux/numa.h file defines NODES_SHIFT to 0 in case
the architecture did not.
Every architecture which has a NUMA config option defines
NODES_SHIFT in its asm-$ARCH headers, but only if NUMA is
enabled, except for x86_64.
This should make it like all the rest.
Signed-off-by: Dave Hansen <hav...@us...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 4211a30349e8d2b724cfb4ce2584604f5e59c299
Author: Jacob Shin <jac...@am...>
Date: Fri Apr 7 19:49:51 2006 +0200
[PATCH] x86_64: Proper null pointer check in powernow_k8_get
This prevents crashes on dual core system when enough ticks are lost.
Replaces earlier patch by me.
Cc: Dave Jones <da...@re...>
Signed-off-by: Thomas Renninger <tr...@su...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit d7fa706ce2c29cb751c15ca00f3aa7b223e3c9f0
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:48 2006 +0200
[PATCH] x86_64: Revert earlier powernow-k8 change
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 95d769aaf47abfc77b600631403ff5af6c990cff
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:45 2006 +0200
[PATCH] i386: Consolidate modern APIC handling
AMD systems have a modern APIC that supports 8 bit IDs, but
don't have a XAPIC version number. Add a new "modern_apic"
subfunction that handles this correctly and use it (nearly)
everywhere where XAPIC is tested for.
I removed one wart: the code specified that external APICs
would use an 8bit APIC ID. But I checked a real 82093 data sheet
and it says clearly that they only use 4bit. So I removed
this special case since it would a bit awkward to implement now.
I removed the valid APIC tests in mptable parsing completely. On any modern
system they only check against the full field width (8bit) anyways
and are no-ops. This also fixes them doing the wrong thing
on >8 core Opterons.
This makes i386 boot again on 16 core Opterons.
Cc: Ingo Molnar <mi...@el...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit d1530d82e02fd96d4634a6d6f6538c8b778c43af
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:42 2006 +0200
[PATCH] x86_64: Clear APIC feature bit when local APIC is disabled
Needed for other checks later in ACPI.
Pointed out by Len Brown
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit d3b6a349d233aecf2c52f7f4c150ca09f684f2d8
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:39 2006 +0200
[PATCH] x86-64/i386: Don't process APICs/IO-APICs in ACPI when APIC is disabled.
When nolapic was passed or the local APIC was disabled
for another reason ACPI would still parse the IO-APICs
until these were explicitely disabled with noapic.
Usually this resulted in a non booting configuration unless
"nolapic noapic" was used.
I also disabled the local APIC parsing in this case, although
that's only cosmetic (suppresses a few printks)
This hopefully makes nolapic work in all cases.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit ec0f08eeea6ac1d8c925f47e3677e4c985fd8f63
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:36 2006 +0200
[PATCH] x86_64: Don't sanity check Type 1 PCI bus access on newer systems
Horus systems don't have anything on bus 0 which makes
the Type 1 sanity checks fail. Use the DMI BIOS year to
check for newer systems and always assume Type 1 works on them.
I used 2001 as an pretty arbitary cutoff year.
Cc: gr...@su...
Cc: Navin Boppuri <nav...@ne...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit fa47dd0ba303599f8adf8d8336ed2fb74efc47c5
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:33 2006 +0200
[PATCH] x86_64: Fix compilation with CONFIG_PCI=n / allnoconfig
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 946f2ee5c7312e8acac4f3ab6629e7e2d36a3646
Author: Arjan van de Ven <ar...@li...>
Date: Fri Apr 7 19:49:30 2006 +0200
[PATCH] i386/x86-64: Check that MCFG points to an e820 reserved area
This patch introduces a user for the e820_all_mapped function:
There have been several machines that don't have a working MMCONFIG,
often because of a buggy MCFG table in the ACPI bios. This patch adds a
simple sanity check that detects a whole bunch of these cases, and when
it detects it, linux now boots rather than crash-and-burns.
The accuracy of this detection can in principle be improved if there was
a "is this entire range in e820 with THIS attribute", but no such
function exist and the complexity needed for this is not really worth
it; this simple check already catches most cases anyway.
Signed-off-by: Arjan van de Ven <ar...@li...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 952223683ec989e86328c24808fdb962c4dbeb0a
Author: Arjan van de Ven <ar...@li...>
Date: Fri Apr 7 19:49:27 2006 +0200
[PATCH] x86_64: Introduce e820_all_mapped
Introduce a e820_all_mapped() function which checks if the entire range
<start,end> is mapped with type.
This is done by moving the local start variable to the end of each
known-good region; if at the end of the function the start address is
still before end, there must be a part that's not of the correct type;
otherwise it's a good region.
Signed-off-by: Arjan van de Ven <ar...@li...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit eee5a9fa63c97366cdea6ab3aa2ed9e3601812d0
Author: Arjan van de Ven <ar...@li...>
Date: Fri Apr 7 19:49:24 2006 +0200
[PATCH] x86_64: Rename e820_mapped to e820_any_mapped
Rename e820_mapped to e820_any_mapped since it tests if any part of the
range is mapped according to the type.
Later steps will introduce e820_all_mapped which will check if the
entire range is mapped with the type. Both have their merit.
Signed-off-by: Arjan van de Ven <ar...@li...>
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit a8062231d80239cf3405982858c02aea21a6066a
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:21 2006 +0200
[PATCH] x86_64: Handle empty PXMs that only contain hotplug memory
The node setup code would try to allocate the node metadata in the node
itself, but that fails if there is no memory in there.
This can happen with memory hotplug when the hotplug area defines an so
far empty node.
Now use bootmem to try to allocate the mem_map in other nodes.
And if it fails don't panic, but just ignore the node.
To make this work I added a new __alloc_bootmem_nopanic function that
does what its name implies.
TBD should try to use nearby nodes here. Currently we just use any.
It's hard to do it better because bootmem doesn't have proper fallback
lists yet.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 68a3a7feb08f960095072f28ec20f7900793c506
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:18 2006 +0200
[PATCH] x86_64: Reserve SRAT hotadd memory on x86-64
From: Keith Mannthey, Andi Kleen
Implement memory hotadd without sparsemem. The memory in the SRAT
hotadd area is just preserved instead and can be activated later.
There are a few restrictions:
- Only one continuous hotadd area allowed per node
The main problem is dealing with the many buggy SRAT tables
that are out there. The strategy here is to reject anything
suspicious.
Originally from Keith Mannthey, with several hacks and changes by AK
and also contributions from Andrew Morton
[ TBD: Problems pointed out by KAMEZAWA Hiroyuki <kam...@jp...>:
1) Goto's rebuild_zonelist patch will not work if CONFIG_MEMORY_HOTPLUG=n.
Rebuilding zonelist is necessary when the system has just memory <
4G at boot, and hot add memory > 4G. because x86_64 has DMA32,
ZONE_NORAML is not included into zonelist at boot time if system
doesn't have memory >4G at boot.
[AK: should just force the higher zones at boot time when SRAT tells us]
2) zone and node's spanned_pages and present_pages are not incremented.
They should be.
For example, our server (ia64/Fujitsu PrimeQuest) can equip memory
from 4G to 1T(maybe 2T in future), and SRAT will *always* say we have
possible 1T +memory. (Microsoft requires "write all possible memory
in SRAT") When we reserve memmap for possible 1T memory, Linux will
not work well in +minimum 4G configuraion ;)
[AK: needs limiting to 5-10% of max memory]
]
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 9d99aaa31f5994d1923c3713ce9144c4c42332e1
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:15 2006 +0200
[PATCH] x86_64: Support memory hotadd without sparsemem
Memory hotadd doesn't need SPARSEMEM, but can be handled by just preallocating
mem_maps. This only needs some untangling of ifdefs to enable the necessary
code even without SPARSEMEM.
Originally from Keith Mannthey, hacked by AK.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 805e8c03c9ea9bdb402a36341e02ec24825d5417
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:12 2006 +0200
[PATCH] x86_64: Clean up execve path
Just call IRET always, no need for any special cases.
Needed for the next bug fix.
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
commit 903fcc608e9f531749024172277dc2fd15d5a587
Author: Andi Kleen <ak...@su...>
Date: Fri Apr 7 19:49:09 2006 +0200
[PATCH] x86_64: Update defconfig
Signed-off-by: Andi Kleen <ak...@su...>
Signed-off-by: Linus Torvalds <tor...@os...>
diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt
index 1921353..f2cd6ef 100644
--- a/Documentation/x86_64/boot-options.txt
+++ b/Documentation/x86_64/boot-options.txt
@@ -151,6 +151,11 @@ NUMA
numa=fake=X Fake X nodes and ignore NUMA setup of the actual machine.
+ numa=hotadd=percent
+ Only allow hotadd memory to preallocate page structures upto
+ percent of already available memory.
+ numa=hotadd=0 will disable hotadd memory.
+
ACPI
acpi=off Don't enable ACPI
diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 0330661..8dab352 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -215,7 +215,7 @@ static int __init acpi_parse_madt(unsign
{
struct acpi_table_madt *madt = NULL;
- if (!phys_addr || !size)
+ if (!phys_addr || !size || !cpu_has_apic)
return -EINVAL;
madt = (struct acpi_table_madt *)__acpi_map_table(phys_addr, size);
@@ -751,6 +751,9 @@ static int __init acpi_parse_madt_ioapic
return -ENODEV;
}
+ if (!cpu_has_apic)
+ return -ENODEV;
+
/*
* if "noapic" boot option, don't look for IO-APICs
*/
diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 6273bf7..254cee9 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -62,6 +62,18 @@ int apic_verbosity;
static void apic_pm_activate(void);
+int modern_apic(void)
+{
+ unsigned int lvr, version;
+ /* AMD systems use old APIC versions, so check the CPU */
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
+ boot_cpu_data.x86 >= 0xf)
+ return 1;
+ lvr = apic_read(APIC_LVR);
+ version = GET_APIC_VERSION(lvr);
+ return version >= 0x14;
+}
+
/*
* 'what should we do if we get a hw irq event on an illegal vector'.
* each architecture has to answer this themselves.
@@ -119,10 +131,7 @@ void enable_NMI_through_LVT0 (void * dum
int get_physical_broadcast(void)
{
- unsigned int lvr, version;
- lvr = apic_read(APIC_LVR);
- version = GET_APIC_VERSION(lvr);
- if (!APIC_INTEGRATED(version) || version >= 0x14)
+ if (modern_apic())
return 0xff;
else
return 0xf;
@@ -349,9 +358,9 @@ int __init verify_local_APIC(void)
void __init sync_Arb_IDs(void)
{
- /* Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 */
- unsigned int ver = GET_APIC_VERSION(apic_read(APIC_LVR));
- if (ver >= 0x14) /* P4 or higher */
+ /* Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1
+ And not needed on AMD */
+ if (modern_apic())
return;
/*
* Wait for idle.
diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
index 712a26b..7c0e160 100644
--- a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
@@ -46,7 +46,7 @@
#define PFX "powernow-k8: "
#define BFX PFX "BIOS error: "
-#define VERSION "version 1.60.1"
+#define VERSION "version 1.60.2"
#include "powernow-k8.h"
/* serialize freq changes */
@@ -55,7 +55,7 @@ static DEFINE_MUTEX(fidvid_mutex);
static struct powernow_k8_data *powernow_data[NR_CPUS];
#ifndef CONFIG_SMP
-static cpumask_t cpu_core_map[1] = { CPU_MASK_ALL };
+static cpumask_t cpu_core_map[1];
#endif
/* Return a frequency in MHz, given an input fid */
@@ -910,6 +910,9 @@ static int powernowk8_target(struct cpuf
unsigned int newstate;
int ret = -EIO;
+ if (!data)
+ return -EINVAL;
+
/* only run on specific CPU from here on */
oldmask = current->cpus_allowed;
set_cpus_allowed(current, cpumask_of_cpu(pol->cpu));
@@ -969,6 +972,9 @@ static int powernowk8_verify(struct cpuf
{
struct powernow_k8_data *data = powernow_data[pol->cpu];
+ if (!data)
+ return -EINVAL;
+
return cpufreq_frequency_table_verify(pol, data->powernow_table);
}
@@ -977,7 +983,7 @@ static int __cpuinit powernowk8_cpu_init
{
struct powernow_k8_data *data;
cpumask_t oldmask = CPU_MASK_ALL;
- int rc, i;
+ int rc;
if (!cpu_online(pol->cpu))
return -ENODEV;
@@ -1063,8 +1069,7 @@ static int __cpuinit powernowk8_cpu_init
printk("cpu_init done, current fid 0x%x, vid 0x%x\n",
data->currfid, data->currvid);
- for_each_cpu_mask(i, cpu_core_map[pol->cpu])
- powernow_data[i] = data;
+ powernow_data[pol->cpu] = data;
return 0;
@@ -1104,6 +1109,9 @@ static unsigned int powernowk8_get (unsi
if (!data)
return -EINVAL;
+ if (!data)
+ return -EINVAL;
+
set_cpus_allowed(current, cpumask_of_cpu(cpu));
if (smp_processor_id() != cpu) {
printk(KERN_ERR PFX "limiting to CPU %d failed in powernowk8_get\n", cpu);
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index 8d8aa9d..db12017 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -110,21 +110,6 @@ static int __init mpf_checksum(unsigned
static int mpc_record;
static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __initdata;
-#ifdef CONFIG_X86_NUMAQ
-static int MP_valid_apicid(int apicid, int version)
-{
- return hweight_long(apicid & 0xf) == 1 && (apicid >> 4) != 0xf;
-}
-#else
-static int MP_valid_apicid(int apicid, int version)
-{
- if (version >= 0x14)
- return apicid < 0xff;
- else
- return apicid < 0xf;
-}
-#endif
-
static void __devinit MP_processor_info (struct mpc_config_processor *m)
{
int ver, apicid;
@@ -190,12 +175,6 @@ static void __devinit MP_processor_info
ver = m->mpc_apicver;
- if (!MP_valid_apicid(apicid, ver)) {
- printk(KERN_WARNING "Processor #%d INVALID. (Max ID: %d).\n",
- m->mpc_apicid, MAX_APICS);
- return;
- }
-
/*
* Validate version
*/
diff --git a/arch/i386/kernel/reboot_fixups.c b/arch/i386/kernel/reboot_fixups.c
index 10e21a4..99aab41 100644
--- a/arch/i386/kernel/reboot_fixups.c
+++ b/arch/i386/kernel/reboot_fixups.c
@@ -51,7 +51,5 @@ void mach_reboot_fixups(void)
cur->reboot_fixup(dev);
}
-
- printk(KERN_WARNING "No reboot fixup found for your hardware\n");
}
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index eacc3f0..80cb3b2 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -963,6 +963,36 @@ efi_memory_present_wrapper(unsigned long
return 0;
}
+ /*
+ * This function checks if the entire range <start,end> is mapped with type.
+ *
+ * Note: this function only works correct if the e820 table is sorted and
+ * not-overlapping, which is the case
+ */
+int __init
+e820_all_mapped(unsigned long start, unsigned long end, unsigned type)
+{
+ int i;
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+ if (type && ei->type != type)
+ continue;
+ /* is the region (part) in overlap with the current region ?*/
+ if (ei->addr >= end || ei->addr + ei->size <= start)
+ continue;
+ /* if the region is at the beginning of <start,end> we move
+ * start to the end of the region since it's ok until there
+ */
+ if (ei->addr <= start)
+ start = ei->addr + ei->size;
+ /* if start is now at or beyond end, we're done, full
+ * coverage */
+ if (start >= end)
+ return 1; /* we're done */
+ }
+ return 0;
+}
+
/*
* Find the highest page frame number we have available
*/
@@ -1317,8 +1347,8 @@ legacy_init_iomem_resources(struct resou
/*
* Request address space for all standard resources
*
- * This is called just before pcibios_assign_resources(), which is also
- * an fs_initcall, but is linked in later (in arch/i386/pci/i386.c).
+ * This is called just before pcibios_init(), which is also a
+ * subsys_initcall, but is linked in later (in arch/i386/pci/common.c).
*/
static int __init request_standard_resources(void)
{
@@ -1339,7 +1369,7 @@ static int __init request_standard_resou
return 0;
}
-fs_initcall(request_standard_resources);
+subsys_initcall(request_standard_resources);
static void __init register_memory(void)
{
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 9f66ac5..ae6534a 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -651,6 +651,7 @@ void __init mem_init(void)
* Specifically, in the case of x86, we will always add
* memory to the highmem for now.
*/
+#ifdef CONFIG_HOTPLUG_MEMORY
#ifndef CONFIG_NEED_MULTIPLE_NODES
int add_memory(u64 start, u64 size)
{
@@ -667,6 +668,7 @@ int remove_memory(u64 start, u64 size)
return -EINVAL;
}
#endif
+#endif
kmem_cache_t *pgd_cache;
kmem_cache_t *pmd_cache;
diff --git a/arch/i386/pci/direct.c b/arch/i386/pci/direct.c
index 99012b9..0659ced 100644
--- a/arch/i386/pci/direct.c
+++ b/arch/i386/pci/direct.c
@@ -4,6 +4,7 @@
#include <linux/pci.h>
#include <linux/init.h>
+#include <linux/dmi.h>
#include "pci.h"
/*
@@ -18,8 +19,10 @@ int pci_conf1_read(unsigned int seg, uns
{
unsigned long flags;
- if (!value || (bus > 255) || (devfn > 255) || (reg > 255))
+ if (!value || (bus > 255) || (devfn > 255) || (reg > 255)) {
+ *value = -1;
return -EINVAL;
+ }
spin_lock_irqsave(&pci_config_lock, flags);
@@ -188,6 +191,10 @@ static int __init pci_sanity_check(struc
if (pci_probe & PCI_NO_CHECKS)
return 1;
+ /* Assume Type 1 works for newer systems.
+ This handles machines that don't have anything on PCI Bus 0. */
+ if (dmi_get_year(DMI_BIOS_DATE) >= 2001)
+ return 1;
for (devfn = 0; devfn < 0x100; devfn++) {
if (o->read(0, 0, devfn, PCI_CLASS_DEVICE, 2, &x))
diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c
index 6137890..f77d7f8 100644
--- a/arch/i386/pci/mmconfig.c
+++ b/arch/i386/pci/mmconfig.c
@@ -12,14 +12,20 @@
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/acpi.h>
+#include <asm/e820.h>
#include "pci.h"
+#define MMCONFIG_APER_SIZE (256*1024*1024)
+
+/* Assume systems with more busses have correct MCFG */
+#define MAX_CHECK_BUS 16
+
#define mmcfg_virt_addr ((void __iomem *) fix_to_virt(FIX_PCIE_MCFG))
/* The base address of the last MMCONFIG device accessed */
static u32 mmcfg_last_accessed_device;
-static DECLARE_BITMAP(fallback_slots, 32);
+static DECLARE_BITMAP(fallback_slots, MAX_CHECK_BUS*32);
/*
* Functions for accessing PCI configuration space with MMCONFIG accesses
@@ -29,8 +35,8 @@ static u32 get_base_addr(unsigned int se
int cfg_num = -1;
struct acpi_table_mcfg_config *cfg;
- if (seg == 0 && bus == 0 &&
- test_bit(PCI_SLOT(devfn), fallback_slots))
+ if (seg == 0 && bus < MAX_CHECK_BUS &&
+ test_bit(PCI_SLOT(devfn) + 32*bus, fallback_slots))
return 0;
while (1) {
@@ -74,8 +80,10 @@ static int pci_mmcfg_read(unsigned int s
unsigned long flags;
u32 base;
- if (!value || (bus > 255) || (devfn > 255) || (reg > 4095))
+ if (!value || (bus > 255) || (devfn > 255) || (reg > 4095)) {
+ *value = -1;
return -EINVAL;
+ }
base = get_base_addr(seg, bus, devfn);
if (!base)
@@ -146,29 +154,34 @@ static struct pci_raw_ops pci_mmcfg = {
Normally this can be expressed in the MCFG by not listing them
and assigning suitable _SEGs, but this isn't implemented in some BIOS.
Instead try to discover all devices on bus 0 that are unreachable using MM
- and fallback for them.
- We only do this for bus 0/seg 0 */
+ and fallback for them. */
static __init void unreachable_devices(void)
{
- int i;
+ int i, k;
unsigned long flags;
- for (i = 0; i < 32; i++) {
- u32 val1;
- u32 addr;
-
- pci_conf1_read(0, 0, PCI_DEVFN(i, 0), 0, 4, &val1);
- if (val1 == 0xffffffff)
- continue;
-
- /* Locking probably not needed, but safer */
- spin_lock_irqsave(&pci_config_lock, flags);
- addr = get_base_addr(0, 0, PCI_DEVFN(i, 0));
- if (addr != 0)
- pci_exp_set_dev_base(addr, 0, PCI_DEVFN(i, 0));
- if (addr == 0 || readl((u32 __iomem *)mmcfg_virt_addr) != val1)
- set_bit(i, fallback_slots);
- spin_unlock_irqrestore(&pci_config_lock, flags);
+ for (k = 0; k < MAX_CHECK_BUS; k++) {
+ for (i = 0; i < 32; i++) {
+ u32 val1;
+ u32 addr;
+
+ pci_conf1_read(0, k, PCI_DEVFN(i, 0), 0, 4, &val1);
+ if (val1 == 0xffffffff)
+ continue;
+
+ /* Locking probably not needed, but safer */
+ spin_lock_irqsave(&pci_config_lock, flags);
+ addr = get_base_addr(0, k, PCI_DEVFN(i, 0));
+ if (addr != 0)
+ pci_exp_set_dev_base(addr, k, PCI_DEVFN(i, 0));
+ if (addr == 0 ||
+ readl((u32 __iomem *)mmcfg_virt_addr) != val1) {
+ set_bit(i, fallback_slots);
+ printk(KERN_NOTICE
+ "PCI: No mmconfig possible on %x:%x\n", k, i);
+ }
+ spin_unlock_irqrestore(&pci_config_lock, flags);
+ }
}
}
@@ -183,6 +196,14 @@ void __init pci_mmcfg_init(void)
(pci_mmcfg_config[0].base_address == 0))
return;
+ if (!e820_all_mapped(pci_mmcfg_config[0].base_address,
+ pci_mmcfg_config[0].base_address + MMCONFIG_APER_SIZE,
+ E820_RESERVED)) {
+ printk(KERN_ERR "PCI: BIOS Bug: MCFG area is not E820-reserved\n");
+ printk(KERN_ERR "PCI: Not using MMCONFIG.\n");
+ return;
+ }
+
printk(KERN_INFO "PCI: Using MMCONFIG\n");
raw_pci_ops = &pci_mmcfg;
pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF;
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 4310b4a..7df2fe1 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -136,6 +136,11 @@ config X86_L1_CACHE_SHIFT
default "7" if GENERIC_CPU || MPSC
default "6" if MK8
+config X86_INTERNODE_CACHE_BYTES
+ int
+ default "4096" if X86_VSMP
+ default X86_L1_CACHE_BYTES if !X86_VSMP
+
config X86_TSC
bool
default y
diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile
index 585fd4a..e573e2a 100644
--- a/arch/x86_64/Makefile
+++ b/arch/x86_64/Makefile
@@ -24,37 +24,37 @@
LDFLAGS := -m elf_x86_64
OBJCOPYFLAGS := -O binary -R .note -R .comment -S
LDFLAGS_vmlinux :=
-
CHECKFLAGS += -D__x86_64__ -m64
+cflags-y :=
cflags-$(CONFIG_MK8) += $(call cc-option,-march=k8)
cflags-$(CONFIG_MPSC) += $(call cc-option,-march=nocona)
cflags-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mtune=generic)
-CFLAGS += $(cflags-y)
-CFLAGS += -m64
-CFLAGS += -mno-red-zone
-CFLAGS += -mcmodel=kernel
-CFLAGS += -pipe
+cflags-y += -m64
+cflags-y += -mno-red-zone
+cflags-y += -mcmodel=kernel
+cflags-y += -pipe
cflags-$(CONFIG_REORDER) += -ffunction-sections
# this makes reading assembly source easier, but produces worse code
# actually it makes the kernel smaller too.
-CFLAGS += -fno-reorder-blocks
-CFLAGS += -Wno-sign-compare
+cflags-y += -fno-reorder-blocks
+cflags-y += -Wno-sign-compare
ifneq ($(CONFIG_UNWIND_INFO),y)
-CFLAGS += -fno-asynchronous-unwind-tables
+cflags-y += -fno-asynchronous-unwind-tables
endif
ifneq ($(CONFIG_DEBUG_INFO),y)
# -fweb shrinks the kernel a bit, but the difference is very small
# it also messes up debugging, so don't use it for now.
-#CFLAGS += $(call cc-option,-fweb)
+#cflags-y += $(call cc-option,-fweb)
endif
# -funit-at-a-time shrinks the kernel .text considerably
# unfortunately it makes reading oopses harder.
-CFLAGS += $(call cc-option,-funit-at-a-time)
+cflags-y += $(call cc-option,-funit-at-a-time)
# prevent gcc from generating any FP code by mistake
-CFLAGS += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
+cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
+CFLAGS += $(cflags-y)
AFLAGS += -m64
head-y := arch/x86_64/kernel/head.o arch/x86_64/kernel/head64.o arch/x86_64/kernel/init_task.o
diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig
index 566ecc9..3c45ec2 100644
--- a/arch/x86_64/defconfig
+++ b/arch/x86_64/defconfig
@@ -1,7 +1,7 @@
#
# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.16-git9
-# Sat Mar 25 15:18:40 2006
+# Linux kernel version: 2.6.17-rc1
+# Mon Apr 3 16:11:14 2006
#
CONFIG_X86_64=y
CONFIG_64BIT=y
@@ -9,6 +9,7 @@ CONFIG_X86=y
CONFIG_SEMAPHORE_SLEEPERS=y
CONFIG_MMU=y
CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HWEIGHT=y
CONFIG_GENERIC_CALIBRATE_DELAY=y
CONFIG_X86_CMPXCHG=y
CONFIG_EARLY_PRINTK=y
@@ -55,10 +56,6 @@ CONFIG_BASE_FULL=y
CONFIG_FUTEX=y
CONFIG_EPOLL=y
CONFIG_SHMEM=y
-CONFIG_CC_ALIGN_FUNCTIONS=0
-CONFIG_CC_ALIGN_LABELS=0
-CONFIG_CC_ALIGN_LOOPS=0
-CONFIG_CC_ALIGN_JUMPS=0
CONFIG_SLAB=y
# CONFIG_TINY_SHMEM is not set
CONFIG_BASE_SMALL=0
@@ -70,7 +67,6 @@ CONFIG_BASE_SMALL=0
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
CONFIG_MODULE_FORCE_UNLOAD=y
-CONFIG_OBSOLETE_MODPARM=y
# CONFIG_MODVERSIONS is not set
# CONFIG_MODULE_SRCVERSION_ALL is not set
# CONFIG_KMOD is not set
@@ -81,6 +77,7 @@ CONFIG_STOP_MACHINE=y
#
CONFIG_LBD=y
# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
#
# IO Schedulers
@@ -105,6 +102,7 @@ CONFIG_X86_PC=y
CONFIG_GENERIC_CPU=y
CONFIG_X86_L1_CACHE_BYTES=128
CONFIG_X86_L1_CACHE_SHIFT=7
+CONFIG_X86_INTERNODE_CACHE_BYTES=128
CONFIG_X86_TSC=y
CONFIG_X86_GOOD_APIC=y
# CONFIG_MICROCODE is not set
@@ -116,6 +114,7 @@ CONFIG_X86_LOCAL_APIC=y
CONFIG_MTRR=y
CONFIG_SMP=y
CONFIG_SCHED_SMT=y
+CONFIG_SCHED_MC=y
# CONFIG_PREEMPT_NONE is not set
CONFIG_PREEMPT_VOLUNTARY=y
# CONFIG_PREEMPT is not set
@@ -138,6 +137,7 @@ CONFIG_NEED_MULTIPLE_NODES=y
CONFIG_SPLIT_PTLOCK_CPUS=4
CONFIG_MIGRATION=y
CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID=y
+CONFIG_OUT_OF_LINE_PFN_TO_PAGE=y
CONFIG_NR_CPUS=32
CONFIG_HOTPLUG_CPU=y
CONFIG_HPET_TIMER=y
@@ -289,6 +289,7 @@ CONFIG_IP_PNP_DHCP=y
# CONFIG_INET_AH is not set
# CONFIG_INET_ESP is not set
# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
# CONFIG_INET_TUNNEL is not set
CONFIG_INET_DIAG=y
CONFIG_INET_TCP_DIAG=y
@@ -300,6 +301,7 @@ CONFIG_IPV6=y
# CONFIG_INET6_AH is not set
# CONFIG_INET6_ESP is not set
# CONFIG_INET6_IPCOMP is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
# CONFIG_INET6_TUNNEL is not set
# CONFIG_IPV6_TUNNEL is not set
# CONFIG_NETFILTER is not set
@@ -704,7 +706,6 @@ CONFIG_S2IO=m
# Wireless LAN (non-hamradio)
#
# CONFIG_NET_RADIO is not set
-# CONFIG_NET_WIRELESS_RTNETLINK is not set
#
# Wan interfaces
@@ -791,7 +792,7 @@ CONFIG_HW_CONSOLE=y
#
CONFIG_SERIAL_8250=y
CONFIG_SERIAL_8250_CONSOLE=y
-# CONFIG_SERIAL_8250_ACPI is not set
+CONFIG_SERIAL_8250_PCI=y
CONFIG_SERIAL_8250_NR_UARTS=4
CONFIG_SERIAL_8250_RUNTIME_UARTS=4
# CONFIG_SERIAL_8250_EXTENDED is not set
@@ -921,6 +922,7 @@ CONFIG_HWMON=y
# Digital Video Broadcasting Devices
#
# CONFIG_DVB is not set
+# CONFIG_USB_DABUSB is not set
#
# Graphics support
@@ -932,6 +934,8 @@ CONFIG_VIDEO_SELECT=y
# Console display driver support
#
CONFIG_VGA_CONSOLE=y
+CONFIG_VGACON_SOFT_SCROLLBACK=y
+CONFIG_VGACON_SOFT_SCROLLBACK_SIZE=256
CONFIG_DUMMY_CONSOLE=y
#
@@ -1058,15 +1062,6 @@ CONFIG_USB_HIDINPUT=y
# CONFIG_USB_MICROTEK is not set
#
-# USB Multimedia devices
-#
-# CONFIG_USB_DABUSB is not set
-
-#
-# Video4Linux support is needed for USB Multimedia device support
-#
-
-#
# USB Network Adapters
#
# CONFIG_USB_CATC is not set
@@ -1118,9 +1113,15 @@ CONFIG_USB_MON=y
# CONFIG_MMC is not set
#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
# InfiniBand support
#
# CONFIG_INFINIBAND is not set
+# CONFIG_IPATH_CORE is not set
#
# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
@@ -1128,6 +1129,11 @@ CONFIG_USB_MON=y
# CONFIG_EDAC is not set
#
+# Real Time Clock
+#
+# CONFIG_RTC_CLASS is not set
+
+#
# Firmware Drivers
#
# CONFIG_EDD is not set
diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 35b2fac..5a98026 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -15,6 +15,8 @@
#include <asm/vsyscall32.h>
#include <linux/linkage.h>
+#define IA32_NR_syscalls ((ia32_syscall_end - ia32_sys_call_table)/8)
+
.macro IA32_ARG_FIXUP noebp=0
movl %edi,%r8d
.if \noebp
@@ -109,8 +111,8 @@ ENTRY(ia32_sysenter_target)
CFI_REMEMBER_STATE
jnz sysenter_tracesys
sysenter_do_call:
- cmpl $(IA32_NR_syscalls),%eax
- jae ia32_badsys
+ cmpl $(IA32_NR_syscalls-1),%eax
+ ja ia32_badsys
IA32_ARG_FIXUP 1
call *ia32_sys_call_table(,%rax,8)
movq %rax,RAX-ARGOFFSET(%rsp)
@@ -210,8 +212,8 @@ ENTRY(ia32_cstar_target)
CFI_REMEMBER_STATE
jnz cstar_tracesys
cstar_do_call:
- cmpl $IA32_NR_syscalls,%eax
- jae ia32_badsys
+ cmpl $IA32_NR_syscalls-1,%eax
+ ja ia32_badsys
IA32_ARG_FIXUP 1
call *ia32_sys_call_table(,%rax,8)
movq %rax,RAX-ARGOFFSET(%rsp)
@@ -296,8 +298,8 @@ ENTRY(ia32_syscall)
testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
jnz ia32_tracesys
ia32_do_syscall:
- cmpl $(IA32_NR_syscalls),%eax
- jae ia32_badsys
+ cmpl $(IA32_NR_syscalls-1),%eax
+ ja ia32_badsys
IA32_ARG_FIXUP
call *ia32_sys_call_table(,%rax,8) # xxx: rip relative
ia32_sysret:
@@ -685,12 +687,11 @@ ia32_sys_call_table:
.quad sys_readlinkat /* 305 */
.quad sys_fchmodat
.quad sys_faccessat
- .quad sys_ni_syscall /* pselect6 for now */
- .quad sys_ni_syscall /* ppoll for now */
+ .quad quiet_ni_syscall /* pselect6 for now */
+ .quad quiet_ni_syscall /* ppoll for now */
.quad sys_unshare /* 310 */
.quad compat_sys_set_robust_list
.quad compat_sys_get_robust_list
+ .quad sys_splice
+ .quad sys_sync_file_range
ia32_syscall_end:
- .rept IA32_NR_syscalls-(ia32_syscall_end-ia32_sys_call_table)/8
- .quad ni_syscall
- .endr
diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c
index fffd6b0..70b9d21 100644
--- a/arch/x86_64/kernel/aperture.c
+++ b/arch/x86_64/kernel/aperture.c
@@ -80,7 +80,7 @@ static int __init aperture_valid(char *n
printk("Aperture from %s beyond 4GB. Ignoring.\n",name);
return 0;
}
- if (e820_mapped(aper_base, aper_base + aper_size, E820_RAM)) {
+ if (e820_any_mapped(aper_base, aper_base + aper_size, E820_RAM)) {
printk("Aperture from %s pointing to e820 RAM. Ignoring.\n",name);
return 0;
}
diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index 293cd71..62776c0 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -80,7 +80,12 @@ static inline int bad_addr(unsigned long
return 0;
}
-int __init e820_mapped(unsigned long start, unsigned long end, unsigned type)
+/*
+ * This function checks if any part of the range <start,end> is mapped
+ * with type.
+ */
+int __meminit
+e820_any_mapped(unsigned long start, unsigned long end, unsigned type)
{
int i;
for (i = 0; i < e820.nr_map; i++) {
@@ -94,6 +99,35 @@ int __init e820_mapped(unsigned long sta
return 0;
}
+/*
+ * This function checks if the entire range <start,end> is mapped with type.
+ *
+ * Note: this function only works correct if the e820 table is sorted and
+ * not-overlapping, which is the case
+ */
+int __init e820_all_mapped(unsigned long start, unsigned long end, unsigned type)
+{
+ int i;
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+ if (type && ei->type != type)
+ continue;
+ /* is the region (part) in overlap with the current region ?*/
+ if (ei->addr >= end || ei->addr + ei->size <= start)
+ continue;
+
+ /* if the region is at the beginning of <start,end> we move
+ * start to the end of the region since it's ok until there
+ */
+ if (ei->addr <= start)
+ start = ei->addr + ei->size;
+ /* if start is now at or beyond end, we're done, full coverage */
+ if (start >= end)
+ return 1; /* we're done */
+ }
+ return 0;
+}
+
/*
* Find a free area in a specific range.
*/
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 8538bfe..c946e4f 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -180,6 +180,10 @@ rff_trace:
*
* XXX if we had a free scratch register we could save the RSP into the stack frame
* and report it properly in ps. Unfortunately we haven't.
+ *
+ * When user can change the frames always force IRET. That is because
+ * it deals with uncanonical addresses better. SYSRET has trouble
+ * with them due to bugs in both AMD and Intel CPUs.
*/
ENTRY(system_call)
@@ -254,7 +258,10 @@ sysret_signal:
xorl %esi,%esi # oldset -> arg2
call ptregscall_common
1: movl $_TIF_NEED_RESCHED,%edi
- jmp sysret_check
+ /* Use IRET because user could have changed frame. This
+ works because ptregscall_common has called FIXUP_TOP_OF_STACK. */
+ cli
+ jmp int_with_check
badsys:
movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
@@ -280,7 +287,8 @@ tracesys:
call syscall_trace_leave
RESTORE_TOP_OF_STACK %rbx
RESTORE_REST
- jmp ret_from_sys_call
+ /* Use IRET because user could have changed frame */
+ jmp int_ret_from_sys_call
CFI_ENDPROC
/*
@@ -408,25 +416,9 @@ ENTRY(stub_execve)
CFI_ADJUST_CFA_OFFSET -8
CFI_REGISTER rip, r11
SAVE_REST
- movq %r11, %r15
- CFI_REGISTER rip, r15
FIXUP_TOP_OF_STACK %r11
call sys_execve
- GET_THREAD_INFO(%rcx)
- bt $TIF_IA32,threadinfo_flags(%rcx)
- CFI_REMEMBER_STATE
- jc exec_32bit
RESTORE_TOP_OF_STACK %r11
- movq %r15, %r11
- CFI_REGISTER rip, r11
- RESTORE_REST
- pushq %r11
- CFI_ADJUST_CFA_OFFSET 8
- CFI_REL_OFFSET rip, 0
- ret
-
-exec_32bit:
- CFI_RESTORE_STATE
movq %rax,RAX(%rsp)
RESTORE_REST
jmp int_ret_from_sys_call
diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 10b3e34..6f0790e 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -29,6 +29,8 @@
#define MISC_MCELOG_MINOR 227
#define NR_BANKS 6
+atomic_t mce_entry;
+
static int mce_dont_init;
/* 0: always panic, 1: panic if deadlock possible, 2: try to avoid panic,
@@ -172,10 +174,12 @@ void do_machine_check(struct pt_regs * r
int i;
int panicm_found = 0;
+ atomic_inc(&mce_entry);
+
if (regs)
notify_die(DIE_NMI, "machine check", regs, error_code, 18, SIGKILL);
if (!banks)
- return;
+ goto out2;
memset(&m, 0, sizeof(struct mce));
m.cpu = safe_smp_processor_id();
@@ -266,6 +270,8 @@ void do_machine_check(struct pt_regs * r
out:
/* Last thing done in the machine check exception to clear state. */
wrmsrl(MSR_IA32_MCG_STATUS, 0);
+ out2:
+ atomic_dec(&mce_entry);
}
/*
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index d9e4067..4e6357f 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -34,6 +34,7 @@
#include <asm/proto.h>
#include <asm/kdebug.h>
#include <asm/local.h>
+#include <asm/mce.h>
/*
* lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
@@ -480,6 +481,12 @@ void __kprobes nmi_watchdog_tick(struct
__get_cpu_var(nmi_touch) = 0;
touched = 1;
}
+#ifdef CONFIG_X86_MCE
+ /* Could check oops_in_progress here too, but it's safer
+ not too */
+ if (atomic_read(&mce_entry) > 0)
+ touched = 1;
+#endif
if (!touched && __get_cpu_var(last_irq_sum) == sum) {
/*
* Ayiee, looks like this CPU is stuck ...
diff --git a/arch/x86_64/kernel/pci-dma.c b/arch/x86_64/kernel/pci-dma.c
index 03c9eee..af035ed 100644
--- a/arch/x86_64/kernel/pci-dma.c
+++ b/arch/x86_64/kernel/pci-dma.c
@@ -48,9 +48,11 @@ dma_alloc_pages(struct device *dev, gfp_
{
struct page *page;
int node;
+#ifdef CONFIG_PCI
if (dev->bus == &pci_bus_type)
node = pcibus_to_node(to_pci_dev(dev)->bus);
else
+#endif
node = numa_node_id();
page = alloc_pages_node(node, gfp, order);
return page ? page_address(page) : NULL;
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 70dd8e5..1c44b53 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -781,10 +781,16 @@ long do_arch_prctl(struct task_struct *t
}
case ARCH_GET_GS: {
unsigned long base;
+ unsigned gsindex;
if (task->thread.gsindex == GS_TLS_SEL)
base = read_32bit_tls(task, GS_TLS);
- else if (doit)
- rdmsrl(MSR_KERNEL_GS_BASE, base);
+ else if (doit) {
+ asm("movl %%gs,%0" : "=r" (gsindex));
+ if (gsindex)
+ rdmsrl(MSR_KERNEL_GS_BASE, base);
+ else
+ base = task->thread.gs;
+ }
else
base = task->thread.gs;
ret = put_user(base, (unsigned long __user *)addr);
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 0856ad4..c50b067 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -353,8 +353,10 @@ static __init void parse_cmdline_early (
if (fullarg(from, "enable_timer_pin_1"))
disable_timer_pin_1 = -1;
- if (fullarg(from, "nolapic") || fullarg(from, "disableapic"))
+ if (fullarg(from, "nolapic") || fullarg(from, "disableapic")) {
+ clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
disable_apic = 1;
+ }
if (fullarg(from, "noapic"))
skip_ioapic_setup = 1;
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index ef8bc46..7392570 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -726,7 +726,7 @@ static __init int late_hpet_init(void)
unsigned int ntimer;
if (!vxtime.hpet_address)
- return -1;
+ return 0;
memset(&hd, 0, sizeof (hd));
@@ -917,6 +917,8 @@ void __init time_init(void)
vxtime.hpet_address = 0;
if (hpet_use_timer) {
+ /* set tick_nsec to use the proper rate for HPET */
+ tick_nsec = TICK_NSEC_HPET;
cpu_khz = hpet_calibrate_tsc();
timename = "HPET";
#ifdef CONFIG_X86_PM_TIMER
diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index 39ff070..b81f473 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/arch/x86_64/kernel/vmlinux.lds.S
@@ -65,7 +65,7 @@ SECTIONS
.data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) {
*(.data.cacheline_aligned)
}
- . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
+ . = ALIGN(CONFIG_X86_INTERNODE_CACHE_BYTES);
.data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET) {
*(.data.read_mostly)
}
diff --git a/arch/x86_64/kernel/x8664_ksyms.c b/arch/x86_64/kernel/x8664_ksyms.c
index d78f460..fec4e52 100644
--- a/arch/x86_64/kernel/x8664_ksyms.c
+++ b/arch/x86_64/kernel/x8664_ksyms.c
@@ -112,7 +112,6 @@ EXPORT_SYMBOL_GPL(unset_nmi_callback);
#undef memcpy
#undef memset
#undef memmove
-#undef strlen
extern void * memset(void *,int,__kernel_size_t);
extern size_t strlen(const char *);
@@ -121,7 +120,6 @@ extern void * memcpy(void *,const void *
extern void * __memcpy(void *,const void *,__kernel_size_t);
EXPORT_SYMBOL(memset);
-EXPORT_SYMBOL(strlen);
EXPORT_SYMBOL(strpbrk);
EXPORT_SYMBOL(memmove);
EXPORT_SYMBOL(memcpy);
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index e5f7f1c..4ba34e9 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -305,7 +305,7 @@ static void __meminit phys_pud_init(pud_
if (paddr >= end)
break;
- if (!after_bootmem && !e820_mapped(paddr, paddr+PUD_SIZE, 0)) {
+ if (!after_bootmem && !e820_any_mapped(paddr, paddr+PUD_SIZE, 0)) {
set_pud(pud, __pud(0));
continue;
}
@@ -507,9 +507,8 @@ void __init clear_kernel_mapping(unsigne
/*
* Memory hotplug specific functions
- * These are only for non-NUMA machines right now.
*/
-#ifdef CONFIG_MEMORY_HOTPLUG
+#if defined(CONFIG_ACPI_HOTPLUG_MEMORY) || defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)
void online_page(struct page *page)
{
@@ -520,6 +519,38 @@ void online_page(struct page *page)
num_physpages++;
}
+#ifndef CONFIG_MEMORY_HOTPLUG
+/*
+ * Memory Hotadd without sparsemem. The mem_maps have been allocated in advance,
+ * just online the pages.
+ */
+int __add_pages(struct zone *z, unsigned long start_pfn, unsigned long nr_pages)
+{
+ int err = -EIO;
+ unsigned long pfn;
+ unsigned long total = 0, mem = 0;
+ for (pfn = start_pfn; pfn < start_pfn + nr_pages; pfn++) {
+ if (pfn_valid(pfn)) {
+ online_page(pfn_to_page(pfn));
+ err = 0;
+ mem++;
+ }
+ total++;
+ }
+ if (!err) {
+ z->spanned_pages += total;
+ z->present_pages += mem;
+ z->zone_pgdat->node_spanned_pages += total;
+ z->zone_pgdat->node_present_pages += mem;
+ }
+ return err;
+}
+#endif
+
+/*
+ * Memory is added always to NORMAL zone. This means you will never get
+ * additional DMA/DMA32 memory.
+ */
int add_memory(u64 start, u64 size)
{
struct pglist_data *pgdat = NODE_DATA(0);
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index 4be82d6..cc02573 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -100,11 +100,30 @@ int early_pfn_to_nid(unsigned long pfn)
}
#endif
+static void * __init
+early_node_mem(int nodeid, unsigned long start, unsigned long end,
+ unsigned long size)
+{
+ unsigned long mem = find_e820_area(start, end, size);
+ void *ptr;
+ if (mem != -1L)
+ return __va(mem);
+ ptr = __alloc_bootmem_nopanic(size,
+ SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS));
+ if (ptr == 0) {
+ printk(KERN_ERR "Cannot find %lu bytes in node %d\n",
+ size, nodeid);
+ return NULL;
+ }
+ return ptr;
+}
+
/* Initialize bootmem allocator for a node */
void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long end)
{
unsigned long start_pfn, end_pfn, bootmap_pages, bootmap_size, bootmap_start;
unsigned long nodedata_phys;
+ void *bootmap;
const int pgdat_size = round_up(sizeof(pg_data_t), PAGE_SIZE);
start = round_up(start, ZONE_ALIGN);
@@ -114,13 +133,11 @@ void __init setup_node_bootmem(int nodei
start_pfn = start >> PAGE_SHIFT;
end_pfn = end >> PAGE_SHIFT;
- nodedata_phys = find_e820_area(start, end, pgdat_size);
- if (nodedata_phys == -1L)
- panic("Cannot find memory pgdat in node %d\n", nodeid);
-
- Dprintk("nodedata_phys %lx\n", nodedata_phys);
+ node_data[nodeid] = early_node_mem(nodeid, start, end, pgdat_size);
+ if (node_data[nodeid] == NULL)
+ return;
+ nodedata_phys = __pa(node_data[nodeid]);
- node_data[nodeid] = phys_to_virt(nodedata_phys);
memset(NODE_DATA(nodeid), 0, sizeof(pg_data_t));
NODE_DATA(nodeid)->bdata = &plat_node_bdata[nodeid];
NODE_DATA(nodeid)->node_start_pfn = start_pfn;
@@ -129,9 +146,15 @@ void __init setup_node_bootmem(int nodei
/* Find a place for the bootmem map */
bootmap_pages = bootmem_bootmap_pages(end_pfn - start_pfn);
bootmap_start = round_up(nodedata_phys + pgdat_size, PAGE_SIZE);
- bootmap_start = find_e820_area(bootmap_start, end, bootmap_pages<<PAGE_SHIFT);
- if (bootmap_start == -1L)
- panic("Not enough continuous space for bootmap on node %d", nodeid);
+ bootmap = early_node_mem(nodeid, bootmap_start, end,
+ bootmap_pages<<PAGE_SHIFT);
+ if (bootmap == NULL) {
+ if (nodedata_phys < start || nodedata_phys >= end)
+ free_bootmem((unsigned long)node_data[nodeid],pgdat_size);
+ node_data[nodeid] = NULL;
+ return;
+ }
+ bootmap_start = __pa(bootmap);
Dprintk("bootmap start %lu pages %lu\n", bootmap_start, bootmap_pages);
bootmap_size = init_bootmem_node(NODE_DATA(nodeid),
@@ -142,6 +165,9 @@ void __init setup_node_bootmem(int nodei
reserve_bootmem_node(NODE_DATA(nodeid), nodedata_phys, pgdat_size);
reserve_bootmem_node(NODE_DATA(nodeid), bootmap_start, bootmap_pages<<PAGE_SHIFT);
+#ifdef CONFIG_ACPI_NUMA
+ srat_reserve_add_area(nodeid);
+#endif
node_set_online(nodeid);
}
@@ -335,6 +361,8 @@ __init int numa_setup(char *opt)
#ifdef CONFIG_ACPI_NUMA
if (!strncmp(opt,"noacpi",6))
acpi_numa = -1;
+ if (!strncmp(opt,"hotadd=", 7))
+ hotadd_percent = simple_strtoul(opt+7, NULL, 10);
#endif
return 1;
}
diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index 2eb8795..15ae9fc 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -15,15 +15,26 @@
#include <linux/bitmap.h>
#include <linux/module.h>
#include <linux/topology.h>
+#include <linux/bootmem.h>
+#include <linux/mm.h>
#include <asm/proto.h>
#include <asm/numa.h>
#include <asm/e820.h>
+#if (defined(CONFIG_ACPI_HOTPLUG_MEMORY) || \
+ defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)) \
+ && !defined(CONFIG_MEMORY_HOTPLUG)
+#define RESERVE_HOTADD 1
+#endif
+
static struct acpi_table_slit *acpi_slit;
static nodemask_t nodes_parsed __initdata;
static nodemask_t nodes_found __initdata;
static struct bootnode nodes[MAX_NUMNODES] __initdata;
+static struct bootnode nodes_add[MAX_NUMNODES] __initdata;
+static int found_add_area __initdata;
+int hotadd_percent __initdata = 10;
static u8 pxm2node[256] = { [0 ... 255] = 0xff };
/* Too small nodes confuse the VM badly. Usually they result
@@ -71,6 +82,10 @@ static __init int conflicting_nodes(unsi
static __init void cutoff_node(int i, unsigned long start, unsigned long end)
{
struct bootnode *nd = &nodes[i];
+
+ if (found_add_area)
+ return;
+
if (nd->start < start) {
nd->start = start;
if (nd->end < nd->start)
@@ -90,6 +105,8 @@ static __init void bad_srat(void)
acpi_numa = -1;
for (i = 0; i < MAX_LOCAL_APIC; i++)
apicid_to_node[i] = NUMA_NO_NODE;
+ for (i = 0; i < MAX_NUMNODES; i++)
+ nodes_add[i].start = nodes[i].end = 0;
}
static __init inline int srat_disabled(void)
@@ -155,11 +172,114 @@ acpi_numa_processor_affinity_init(struct
pxm, pa->apic_id, node);
}
+#ifdef RESERVE_HOTADD
+/*
+ * Protect against too large hotadd areas that would fill up memory.
+ */
+static int hotadd_enough_memory(struct bootnode *nd)
+{
+ static unsigned long allocated;
+ static unsigned long last_area_end;
+ unsigned long pages = (nd->end - nd->start) >> PAGE_SHIFT;
+ long mem = pages * sizeof(struct page);
+ unsigned long addr;
+ unsigned long allowed;
+ unsigned long oldpages = pages;
+
+ if (mem <...
[truncated message content] |
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-09 19:49:00
|
New branch 'vax-rtc-rework' available with the following commits: |
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-09 19:48:59
|
New branch 'vax--microsecond-resolution' available with the following commits: |
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-09 19:44:54
|
arch/vax/kernel/cpu_ka650.c | 2 +-
arch/vax/kernel/cpu_vxt.c | 22 +++++++++++-----------
include/asm-vax/mv.h | 2 +-
3 files changed, 13 insertions(+), 13 deletions(-)
New commits:
commit 6b619408df7c0a277e327d0937ff34b0c4728ece
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Sun Apr 9 21:40:17 2006 +0200
[VAX] Remove leading spaces.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
commit bd78fb3ebae6d4db80d4f4bbe977cf1f2172f78d
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Sun Apr 9 21:39:34 2006 +0200
[VAX] Remove leading spaces.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
commit 2f3b3024d760cd8c0b76463d11e2f41d210d1b08
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Sun Apr 9 21:38:26 2006 +0200
[VAX] Remove leading spaces.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
diff --git a/arch/vax/kernel/cpu_ka650.c b/arch/vax/kernel/cpu_ka650.c
index 03b871d..636f828 100644
--- a/arch/vax/kernel/cpu_ka650.c
+++ b/arch/vax/kernel/cpu_ka650.c
@@ -118,7 +118,7 @@ static int __init ka650_platform_device_
platform_device_register(&ka650_cqbic_device);
platform_device_register(&ka650_iprcons_device);
- platform_device_register(&ka650_diag_led_device);
+ platform_device_register(&ka650_diag_led_device);
return 0;
}
diff --git a/arch/vax/kernel/cpu_vxt.c b/arch/vax/kernel/cpu_vxt.c
index d4493c2..7f3ef13 100644
--- a/arch/vax/kernel/cpu_vxt.c
+++ b/arch/vax/kernel/cpu_vxt.c
@@ -1,12 +1,12 @@
/*
- * This file contains machine vector handlers for the
- * VXT CPU in the VXT2000 machines.
+ * This file contains machine vector handlers for the
+ * VXT CPU in the VXT2000 machines.
*
- * From mailing list messages the SID is 14000006 but NetBSD uses 14000008.
+ * From mailing list messages the SID is 14000006 but NetBSD uses 14000008.
*
- * This may work for other ka48 based systems,
+ * This may work for other ka48 based systems,
*
- * may 2002. It looks as if the 20040058 address is right for prom output.
+ * may 2002. It looks as if the 20040058 address is right for prom output.
*/
#include <linux/types.h> /* For NULL */
@@ -48,14 +48,14 @@ struct vax_mv mv_vxt = {
};
static struct cpu_match __CPU_MATCH cpu_vxt = {
- .mv = &mv_vxt,
- .sid_mask = VAX_SID_FAMILY_MASK,
- .sid_match = VAX_SOC << VAX_SID_FAMILY_SHIFT,
+ .mv = &mv_vxt,
+ .sid_mask = VAX_SID_FAMILY_MASK,
+ .sid_match = VAX_SOC << VAX_SID_FAMILY_SHIFT,
- .sidex_addr = SOC_SIDEX_ADDR,
+ .sidex_addr = SOC_SIDEX_ADDR,
- .sidex_mask = SOC_SIDEX_TYPE_MASK,
- .sidex_match = SOC_SIDEX_TYPE_VXT << SOC_SIDEX_TYPE_SHIFT
+ .sidex_mask = SOC_SIDEX_TYPE_MASK,
+ .sidex_match = SOC_SIDEX_TYPE_VXT << SOC_SIDEX_TYPE_SHIFT
};
static struct platform_device vxt_diag_led_device = {
diff --git a/include/asm-vax/mv.h b/include/asm-vax/mv.h
index b61ee9e..6bdcd28 100644
--- a/include/asm-vax/mv.h
+++ b/include/asm-vax/mv.h
@@ -34,7 +34,7 @@ struct vax_mv {
void (*reboot)(void); /* If implemented, these are called from */
void (*halt)(void); /* machine_restart/_halt */
- void (*mcheck)(void *); /* Machine check handler */
+ void (*mcheck)(void *); /* Machine check handler */
void (*init_devices)(void); /* Optional */
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-09 19:10:58
|
arch/vax/kernel/Makefile | 2 +-
1 files changed, 1 insertion(+), 1 deletion(-)
New commits:
commit e40d99d6fdc2de331ba1741897c51cf92c9647ea
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Sun Apr 9 21:10:41 2006 +0200
[VAX] Remove trailing backslash in a makefile where it is not needed.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
diff --git a/arch/vax/kernel/Makefile b/arch/vax/kernel/Makefile
index 671aab3..d5cb600 100644
--- a/arch/vax/kernel/Makefile
+++ b/arch/vax/kernel/Makefile
@@ -9,7 +9,7 @@ extra-y := vmlinux.lds
obj-y := ptrace.o process.o setup.o regdump.o interrupt.o entry.o time.o \
ioprobe.o syscall.o signal.o semaphore.o vax_dev_init.o \
- init_task.o reboot.o cpu_generic.o clock.o \
+ init_task.o reboot.o cpu_generic.o clock.o
obj-y += bootcons/
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-08 11:27:04
|
arch/vax/boot/head.S | 3 ++-
1 files changed, 2 insertions(+), 1 deletion(-)
New commits:
commit 474f6e13d495485df80ca7f19085932d37b12ec3
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Sat Apr 8 13:26:34 2006 +0200
[VAX] Fix a thinko in a comment.
This patch fixes a wrong comment about the content of `mv'.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
diff --git a/arch/vax/boot/head.S b/arch/vax/boot/head.S
index bf59290..528967a 100644
--- a/arch/vax/boot/head.S
+++ b/arch/vax/boot/head.S
@@ -79,7 +79,8 @@ codestart:
* we haven't yet enabled mapping, but want to use the function
* pointers (eg. to print something.)
*/
- movl mv, %r10 /* R10 = &mv */
+ movl mv, %r10 /* R10 = mv (which holds a pointer to
+ the actual struct mv in use. */
moval start, %r8 /* R8 = &start */
/*
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-07 20:12:55
|
arch/vax/boot/head.S | 141 ++++++++++++++++++++++++------------------
arch/vax/kernel/asm-offsets.c | 4 -
2 files changed, 86 insertions(+), 59 deletions(-)
New commits:
commit b36dd710bc4699877e8222f92d59394dc62fceb6
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Fri Apr 7 22:12:17 2006 +0200
[VAX] Zap another magic number from head.S
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
commit 3e87c0b8b1492463b60d9646a57767fb3db6530c
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Fri Apr 7 21:55:02 2006 +0200
[VAX] Updates for head.S
This patch adds some more comments into head.S and zaps one magic constant.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
diff --git a/arch/vax/boot/head.S b/arch/vax/boot/head.S
index 3861e3a..bf59290 100644
--- a/arch/vax/boot/head.S
+++ b/arch/vax/boot/head.S
@@ -1,12 +1,12 @@
-# Copyright atp Nov 1998.
-# Changes for aligning IOMAP/VMALLOC - Copyright ai...@li... - June 2001
-# start of boot. entry point
-# this assumes vmb has does most of the hard work (ie uvax rom vmb)
-# save useful registers. jump to c in boot.c
-#
-# TBD: Some of this stuff could do with being rewritten in C
-# Some of this stuff could be in .init sections and thrown away.
-#
+/*
+ * Copyright atp Nov 1998.
+ * Changes for aligning IOMAP/VMALLOC - Copyright ai...@li... - June 2001
+ * start of boot. entry point
+ * this assumes vmb has does most of the hard work (ie uvax rom vmb)
+ * save useful registers. jump to c in boot.c
+ * TBD: Some of this stuff could do with being rewritten in C
+ * Some of this stuff could be in .init sections and thrown away.
+ */
#include <linux/init.h> /* For __INITDATA */
#include <asm/mtpr.h> /* Processor register definitions */
@@ -16,86 +16,108 @@
#include "boot_sections.h"
+/*
+ * __BOOT is the starting point of the kernel image, put at virtual address
+ * 0x80100000 (cf. arch/vax/kernel/vmlinux.lds.S) and physicalls the address
+ * the firmware jumps to.
+ */
__BOOT
.globl _stext
_stext:
.globl start
start:
- jmp codestart # Word displacement.
+ /*
+ * Entry point, jumped to by the firmware. We jump further down
+ * into the image and keep some space to put the command line
+ * below this.
+ */
+ jmp codestart
.globl bootparam
bootparam:
- .fill 256,1,0 # The boot parameter block. Presently just the
- # command line.
-
+ /*
+ * 256 bytes to keep the boot command line. This can be modified
+ * by `setcmdline'.
+ */
+ .fill 256, 1, 0
codestart:
- # Disable memory mapping
+ /* Disable memory mapping, though it should be disabled already... */
mtpr $0, $PR_MAPEN
mtpr $31, $PR_IPL
- # Save r11, AP, SCBB and location of command line
- movl %ap, boot_ap
- movl %r11, boot_r11
- mfpr $PR_SCBB, boot_scb
- movab bootparam, %r5
- addl2 $PAGE_OFFSET, %r5 # we will only access this when MAPEN=1
- movl %r5, kernel_cmd_line
-
- # Put the sp somewhere safe, over our bootblock in fact
- moval start, %r5
- subl2 $0x200, %r5
- movl %r5,%sp
+ /* Save some important values */
+ movl %ap, boot_ap /* Save AP */
+ movl %r11, boot_r11 /* Save R11 */
+ mfpr $PR_SCBB, boot_scb /* Save System Control Block
+ from its CPU register */
+ movab bootparam, %r5 /* R5 = physical address of command line */
+ addl2 $PAGE_OFFSET, %r5 /* R5 += PAGE_OFFSET (now it's a virtual address) */
+ movl %r5, kernel_cmd_line /* Now kernel_cmd_line points to the right thing. */
+
+ /* Put the sp somewhere safe, over our bootblock in fact */
+ moval start, %r5 /* R5 = &start */
+ subl2 $0x200, %r5 /* R5 -= 0x200 */
+ movl %r5, %sp /* SP = R5 */
+#if 0
# Debug code:
# movzbl $0x42, %r2
# jsb 0x20040058
+#endif
+ /* Identify the CPU we're running on. */
pushal start
- calls $1, idcpu # Identify this CPU and...
- movl %r0, mv # ...put the returned mv ptr into mv.
-
- # Now fix up the machine vector entries. (They currently contain
- # pointers to virtual addresses in S0 space. We need to change
- # the pointers to the functions we use before VM init to point
- # into the newly-loaded kernel image.)
- movl mv, %r10
- moval start, %r8
+ calls $1, idcpu
+ movl %r0, mv /* mv = idcpu(&start) */
+ /*
+ * Prepare to fix some of the Machine Vector pointers to
+ * point to physical memory (instead of virtual addresses) because
+ * we haven't yet enabled mapping, but want to use the function
+ * pointers (eg. to print something.)
+ */
+ movl mv, %r10 /* R10 = &mv */
+ moval start, %r8 /* R8 = &start */
+
+ /*
+ * Fix machine vector addresses to point to physical instead of
+ * virtual addresses because we haven't yet enabled VM.
+ */
subl2 $PAGE_OFFSET+KERNEL_START_PHYS, %r8
addl2 %r8, MV_PRE_VM_PUTCHAR(%r10)
addl2 %r8, MV_PRE_VM_GETCHAR(%r10)
addl2 %r8, MV_CPU_TYPE_STR(%r10)
- calls $0, boot_crlf
- calls $0, boot_crlf
+ pushab msg_welcome /* Welcome to Linux on VAX */
+ calls $1, boot_printstr
- # print the cpu type
+ /* Print the cpu type */
calls $0, boot_print_cpu_id
- # print first line of debug diagnostics
- pushab msg_loaded # ascii string
+ /* Print first line of debug diagnostics */
+ pushab msg_loaded /* "head.s loaded at..." */
calls $1, boot_printstr
- pushal start # where we were loaded
+ pushal start /* Where we were loaded */
calls $1, boot_printint
calls $0, boot_crlf
- pushab msg_registers # ascii string
+ pushab msg_registers /* Print "rpb/r5/ap/sp" */
calls $1, boot_printstr
calls $0, boot_printspace
- movl boot_r11, %r11
- pushl %r11 # r11 (holds the rpb base address, usually 0x0)
+ movl boot_r11, %r11 /* Prepare access to RPB */
+ pushl %r11 /* RPB base address */
calls $1, boot_printint
calls $0, boot_printspace
- # FIXME: magic offset -> asmoffsets.h
- pushl 48(%r11) # saved r5 in RPB (argument to BOOT command)
+
+ pushl RPB_BOOT_R5_OFFSET(%r11)) /* Saved R5 in RPB */
calls $1, boot_printint
calls $0, boot_printspace
- pushl %ap # argument pointer (struct arglist)
+ pushl %ap /* Argument pointer (struct arglist) */
calls $1, boot_printint
calls $0, boot_printspace
- pushl %sp # stack pointer
+ pushl %sp /* Stack pointer */
calls $1, boot_printint
calls $0, boot_crlf
@@ -107,8 +129,9 @@ codestart:
# like the entire .bss section.
movl mv, boot_mv
- # copy the loaded image higher up in physical RAM
-
+ /*
+ * Copy the loaded image higher up in physical RAM
+ */
movl $__bss_start, %r6
subl2 $start, %r6 # byte count to r6
pushl %r6
@@ -156,10 +179,9 @@ reloc:
addl2 %r2, MV_PRE_VM_GETCHAR(%r10)
addl2 %r2, MV_CPU_TYPE_STR(%r10)
- # Print 'relocated at phys addr xxxxx'
- pushab msg_relocated
+ pushab msg_relocated /* Print "relocated at phys addr "... */
calls $1, boot_printstr
- pushal reloc
+ pushal reloc /* ...and the address. */
calls $1, boot_printint
calls $0, boot_crlf
calls $0, boot_crlf
@@ -208,7 +230,7 @@ sysfill:
bisl3 $_PAGE_VALID + _PAGE_UW, %r6, (%r5)+
# set PFN, VALID bit and protection UW in PTE
incl %r6 # next PFN
- cmpl %r6, %r7 # one page of PTE Table -> 128 Pages of PTES
+ cmpl %r6, %r7 # one page of PTE Table -> 128 Pages of PTEs
blssu sysfill
# We need to align the IOMAP/VMALLOC tables (well at least the VMALLOC
@@ -225,7 +247,7 @@ sysfill:
zerofl:
movl $0x00000000, (%r5)+
incl %r6 # next PFN
- cmpl %r6, %r7 # one page of PTE Table -> 128 Pages of PTES
+ cmpl %r6, %r7 # one page of PTE Table -> 128 Pages of PTEs
blssu zerofl
nozerofl:
# Zero out the spare part of the SPT (the entries that will be used
@@ -235,7 +257,7 @@ nozerofl:
sparef1:
movl $0x00000000, (%r5)+
incl %r6 # next PFN
- cmpl %r6, %r7 # one page of PTE Table -> 128 Pages of PTES
+ cmpl %r6, %r7 # one page of PTE Table -> 128 Pages of PTEs
blssu sparef1
movl %r5, vmallocmap_base
@@ -295,8 +317,8 @@ sparefill2:
# and faking a saved PC/PSL on the interrupt stack which we then
# 'return' to
moval init_thread_union, %r0
- addl2 $8192, %r0 # FIXME: taken from <linux/sched.h>
- mtpr %r0,$PR_KSP
+ addl2 $THREAD_SIZE, %r0
+ mtpr %r0, $PR_KSP
pushl $0x001f0000 # IS=0, accmode=prevmode=K, IPL=31
pushab now_on_kstack
@@ -307,6 +329,9 @@ now_on_kstack:
calls $0, vax_start_kernel # should never return
halt
+msg_welcome:
+ .ascii "\r\n\r\n* ** Welcome to Linux on VAX ** *\r\n\r\n\0"
+ .align 1
msg_loaded:
.ascii "Boot Head.S loaded at address \0"
.align 1
diff --git a/arch/vax/kernel/asm-offsets.c b/arch/vax/kernel/asm-offsets.c
index 1801560..0d673f6 100644
--- a/arch/vax/kernel/asm-offsets.c
+++ b/arch/vax/kernel/asm-offsets.c
@@ -45,9 +45,11 @@ void foo (void)
BLANK ();
/*
- * PAGE_OFFSET
+ * Define some other important values. Getting these the direct way
+ * would mean to hack the header files quite hard...
*/
DEFINE (PAGE_OFFSET, PAGE_OFFSET);
+ DEFINE (THREAD_SIZE, THREAD_SIZE);
BLANK ();
/*
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-07 19:02:01
|
arch/vax/Kconfig | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
New commits:
commit 2652469aebd922de19d8d443dab0cbad331ad599
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Fri Apr 7 21:01:48 2006 +0200
[VAX] Call it SCSI instead of Scsi.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
diff --git a/arch/vax/Kconfig b/arch/vax/Kconfig
index 00ff1f5..d62323f 100644
--- a/arch/vax/Kconfig
+++ b/arch/vax/Kconfig
@@ -199,12 +199,12 @@ if SCSI
comment "VAX SCSI low-level drivers"
config SCSI_VAX_5380
- bool "NCR53C80 Scsi Driver (used in VAXstation/MicroVAX 3100 family)"
+ bool "NCR53C80 SCSI Driver (used in VAXstation/MicroVAX 3100 family)"
select SCSI_SPI_ATTRS
default n
config SCSI_VAX_53C94
- bool "NCR53C94 Scsi Driver (used in VAXstation 4000 family) (NOT WORKING)"
+ bool "NCR53C94 SCSI Driver (used in VAXstation 4000 family) (NOT WORKING)"
default n
endif
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-07 17:56:52
|
arch/vax/kernel/asm-offsets.c | 17 +++++++++++++++++
1 files changed, 17 insertions(+)
New commits:
commit bbca2b75fc4e0147e9e5bce87016ed83d1e285d4
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Fri Apr 7 19:56:11 2006 +0200
[VAX] Add another values asm-offsets, clean-up a bit.
This'll allow to zap another magic value from out boot code (head.S).
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-07 17:52:29
|
arch/vax/kernel/vmlinux.lds.S | 20 ++++++++++++--------
1 files changed, 12 insertions(+), 8 deletions(-)
New commits:
commit cf7b2faf15a0d6ee32b53703cef050fe66d2c494
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Fri Apr 7 19:52:08 2006 +0200
[VAX] Remove a magic value from vmlinux.lds.S
We used to have a magic constant for 0x80100000 (PAGE_OFFSET+PHYS_START),
which is the start address of the kernel image after relocation, as a
virtual address. Now it uses the applicable define.
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-05 05:18:32
|
arch/vax/kernel/ptrace.c | 23 +++++++++++++++++++++++
1 files changed, 23 insertions(+)
New commits:
commit 9c1a21663c41c85f0d72bbbef20c6626978e4b70
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Wed Apr 5 07:18:15 2006 +0200
[VAX] Implement the PTRACE_GETEVENTMSG case for sys_ptrace().
This is suggested by Dave S. Miller in
http://marc.theaimsgroup.com/?l=linux-arch&m=114420711312161&w=4 , but
it won't make our ptrace more useful, since it needs a real rework (and
AKAIK there's no working GDB right now.)
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@he...> - 2006-04-05 05:01:56
|
Documentation/vax/vxt2000+.txt | 36 ++++++++++++++++++++++++++++++------
1 files changed, 30 insertions(+), 6 deletions(-)
New commits:
commit d53fb96b0bea16b94b402f2ff11a81e9002ba4d5
Author: Jan-Benedict Glaw <jb...@d2...>
Date: Wed Apr 5 07:01:26 2006 +0200
[VAX] Better description of the VXT2000+
Signed-off-by: Jan-Benedict Glaw <jb...@lu...>
|
|
From: Jan-Benedict G. <jb...@lu...> - 2006-04-04 18:21:54
|
Hi! This is a test only, please ignore. MfG, JBG --=20 Jan-Benedict Glaw jb...@lu... . +49-172-7608481 = _ O _ "Eine Freie Meinung in einem Freien Kopf | Gegen Zensur | Gegen Krieg = _ _ O f=C3=BCr einen Freien Staat voll Freier B=C3=BCrger" | im Internet! | i= m Irak! O O O ret =3D do_actions((curr | FREE_SPEECH) & ~(NEW_COPYRIGHT_LAW | DRM | TCPA)= ); |
