Commit graph

64,840 commits

Author SHA1 Message Date
Yinghai Lu
a2f9f43858 x86_64: separate irq_cfgx from irq_cfgx_free
so later don't need to compare with -1U

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:51 +02:00
Yinghai Lu
cb5bc83225 x86_64: rename irq_desc/irq_desc_alloc
change names:

          irq_desc() ==> irq_desc_alloc
	__irq_desc() ==> irq_desc

Also split a few of the uses in lowlevel x86 code.

v2: need to check if desc is null in smp_irq_move_cleanup

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:51 +02:00
Yinghai Lu
1d5f6b36c4 x86: check with without_new in show_interrupts
so we don't get new one that we don't need it.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:51 +02:00
Yinghai Lu
46926b67fc generic: add irq_desc in function in parameter
So we could remove some duplicated calling to irq_desc

v2: make sure irq_desc in  init/main.c is not used without generic_hardirqs

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:50 +02:00
Yinghai Lu
7d94f7ca40 irq: remove >= nr_irqs checking with config_have_sparse_irq
remove irq limit checks - nr_irqs is dynamic and we expand anytime.

v2: fix checking about result irq_cfg_without_new, so could use msi again
v3: use irq_desc_without_new to check irq is valid

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:50 +02:00
Yinghai Lu
46b8214d12 x86, ioapic: replace loop with nr_irqs with for_each_irq_icfg
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:33 +02:00
Yinghai Lu
2c6927a38f irq: replace loop with nr_irqs with for_each_irq_desc
There are a handful of loops that go from 0 to nr_irqs and use
get_irq_desc() on them. These would allocate all the irq_desc
entries, regardless of the need for them.

Use the smarter for_each_irq_desc() iterator that will only iterate
over the present ones.

v2: make sure arch without GENERIC_HARDIRQS work too

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:33 +02:00
Yinghai Lu
7f95ec9e4c x86: move kstat_irqs from kstat to irq_desc
based on Eric's patch ...

together mold it with dyn_array for irq_desc, will allcate kstat_irqs for
nr_irq_desc alltogether if needed. -- at that point nr_cpus is known already.

v2: make sure system without generic_hardirqs works they don't have irq_desc
v3: fix merging
v4: [mingo@elte.hu] fix typo

[ mingo@elte.hu ] irq: build fix

fix:

 arch/x86/xen/spinlock.c: In function 'xen_spin_lock_slow':
 arch/x86/xen/spinlock.c:90: error: 'struct kernel_stat' has no member named 'irqs'

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:32 +02:00
Yinghai Lu
e5a53714ac x86: put irq_2_pin pointer into irq_cfg
preallocate 32 irq_2_pin entries, and use get_one_free_irq_2_pin() to get
one more and link to irq_cfg if needed.

so don't waste one where no irq is enabled.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:31 +02:00
Yinghai Lu
3ac2de48ed x86: add irq_cfg in io_apic_64.c
preallocate size is 32, and if it is not enough, irq_cfg will more
via alloc_bootmem() or kzalloc(). (depending on how early we are in
system setup)

v2: fix typo about size of init_one_irq_cfg ... should use sizeof(struct irq_cfg)
v3: according to Eric, change get_irq_cfg() to irq_cfg()
v4: squash add irq_cfg_alloc in

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:30 +02:00
Yinghai Lu
08678b0841 generic: sparse irqs: use irq_desc() together with dyn_array, instead of irq_desc[]
add CONFIG_HAVE_SPARSE_IRQ to for use condensed array.
Get rid of irq_desc[] array assumptions.

Preallocate 32 irq_desc, and irq_desc() will try to get more.

( No change in functionality is expected anywhere, except the odd build
  failure where we missed a code site or where a crossing commit itroduces
  new irq_desc[] usage. )

v2: according to Eric, change get_irq_desc() to irq_desc()

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:29 +02:00
Yinghai Lu
71f521bbaf x86, irq: get nr_irqs from madt
Until now, NR_IRQS was derived from black magic defines that had to
be "large enough" to both accomodate NR_CPUS and MAX_NR_IO_APICs.

This resulted in a way too large irq_desc[] array on most x86 systems.
Especially with larger CPU masks, the size of irq_desc can spiral out
of control quickly.

So be smarter about it and use precise allocation instead: determine the
default maximum possible IRQ number from the ACPI MADT. Use a minimum limit
of at least 32 IRQs for broken BIOSes.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:08 +02:00
Ingo Molnar
a84488c213 irq: sparse irqs, fix #3
fix non-APIC UP build:

 arch/x86/kernel/built-in.o: In function `setup_arch':
 : undefined reference to `pin_map_size'
 arch/x86/kernel/built-in.o: In function `setup_arch':
 : undefined reference to `first_free_entry'

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:08 +02:00
Yinghai Lu
301e619020 x86: use dyn_array in io_apic_xx.c
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:08 +02:00
Yinghai Lu
0799e432ac x86: use nr_irqs
also add first_free_entry and pin_map_size, which were NR_IRQS derived
constants.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:05 +02:00
Yinghai Lu
85c0f90978 irq: introduce nr_irqs
at this point nr_irqs is equal NR_IRQS

convert a few easy users from NR_IRQS to dynamic nr_irqs.

v2: according to Eric, we need to take care of arch without generic_hardirqs

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:05 +02:00
Yinghai Lu
6da55c3e8d x86: enable dyn_array support
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:04 +02:00
Yinghai Lu
1f8ff037a8 x86: alloc dyn_array all together
so could spare some memory with small alignment in bootmem

also tighten the alignment checking, and make print out less debug info.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:03 +02:00
Yinghai Lu
1f3fcd4b1a add per_cpu_dyn_array support
allow dyn-array in per_cpu area, allocated dynamically.

usage:

|  /* in .h */
| struct kernel_stat {
|        struct cpu_usage_stat   cpustat;
|        unsigned int *irqs;
| };
|
|  /* in .c */
| DEFINE_PER_CPU(struct kernel_stat, kstat);
|
| DEFINE_PER_CPU_DYN_ARRAY_ADDR(per_cpu__kstat_irqs, per_cpu__kstat.irqs, sizeof(unsigned int), nr_irqs, sizeof(unsigned long), NULL);

after setup_percpu()/per_cpu_alloc_dyn_array(), the dyn_array in
per_cpu area is ready to use.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:03 +02:00
Yinghai Lu
fe648be019 x86: add after_bootmem flag for 32bit
to prepare to use dyn_array support etc.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 16:52:02 +02:00
Graf Yang
b4f5c58fd1 Blackfin arch: Zero out bss region in L1/L2 memory.
Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
2008-10-16 22:50:46 +08:00
Bryan Wu
121e598f5d Blackfin arch: add read/write IO accessor functions to Blackfin
This is to kill some compiling warning on DM9000 netdev driver.

Signed-off-by: Bryan Wu <cooloney@kernel.org>
2008-10-16 22:31:56 +08:00
Martin Schwidefsky
0b59268285 [PATCH] remove unused ibcs2/PER_SVR4 in SET_PERSONALITY
The SET_PERSONALITY macro is always called with a second argument of 0.
Remove the ibcs argument and the various tests to set the PER_SVR4
personality.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2008-10-16 15:40:05 +02:00
Stephen Rothwell
463baa8a09 powerpc: fix linux-next build failure
Today's linux-next build (powerpc allyesconfig) failed like this:

In file included from arch/powerpc/include/asm/mmu-hash64.h:17,
                 from arch/powerpc/include/asm/mmu.h:8,
                 from arch/powerpc/include/asm/pgtable.h:8,
                 from arch/powerpc/mm/slb.c:20:
arch/powerpc/include/asm/page.h:76: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'memstart_addr'
arch/powerpc/include/asm/page.h:77: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'kernstart_addr'

Caused by commit 600715dcdf ("generic: add
phys_addr_t for holding physical addresses") from the tip-core tree.
This only fails if CONFIG_RELOCATABLE is set.

So include that instead of asm/types.h in asm/page.h for
the CONFIG_RELOCATABLE case.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: ppc-dev <linuxppc-dev@ozlabs.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-10-16 12:38:19 +02:00
Dominik Brodowski
459fc208ab cpufreq: remove policy->governor setting in drivers initialization
As policy->governor is already set to CPUFREQ_DEFAULT_GOVERNOR in the
(always built-in) cpufreq core, we do not need to set it in the drivers.
This fixes the sparc64 allmodconfig build failure.

Also, remove a totally useles setting of ->policy in cpufreq-pxa3xx.c.

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-15 16:42:47 -07:00
Russell King
2502991560 Merge branch 'fixes' into for-linus
Conflicts:

	arch/arm/mach-versatile/core.c
2008-10-15 23:16:07 +01:00
Pavel Vasilyev
1339c367a8 fix CONFIG_MMCONFIG=n build warning
arch/x86/kernel/acpi/boot.c💯 warning: 'acpi_mcfg_64bit_base_addr' defined
but not used

http://bugzilla.kernel.org/show_bug.cgi?id=11743

Signed-off-by: Pavel Vasilyev <linuxoid@tochka.ru>
Signed-off-by: Len Brown <len.brown@intel.com>
2008-10-15 17:33:48 -04:00
Robert Richter
5a289395bf Merge branch 'oprofile/x86-oprofile-for-tip' into oprofile/oprofile-for-tip
Conflicts:
	arch/x86/oprofile/op_model_ppro.c
2008-10-15 22:19:41 +02:00
Suravee Suthikulpanit
5f87dfb79f x86/oprofile: add the logic for enabling additional IBS bits
This patch adds the logic for enabling additional IBS control bits :
* IBS-Fetch IbsRandEn bit (bit 57)
* IBS-Op IbsOpCntCtl bit (bit 19)

Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
2008-10-15 20:56:56 +02:00
Robert Richter
69046d4304 x86/oprofile: reordering functions in nmi_int.c
No functional changes. The intension is to remove static function
declarations.

Signed-off-by: Robert Richter <robert.richter@amd.com>
2008-10-15 20:56:53 +02:00
Robert Richter
25ad2913ca oprofile: more whitespace fixes
Signed-off-by: Robert Richter <robert.richter@amd.com>
2008-10-15 20:55:51 +02:00
Robert Richter
c92960fccb oprofile: whitespace fixes
Signed-off-by: Robert Richter <robert.richter@amd.com>
2008-10-15 20:47:41 +02:00
Robert Richter
ccd755c2d9 OProfile: Rename IBS sysfs dir into "ibs_op"
The new name is now more close to those used in the spec.

Signed-off-by: Robert Richter <robert.richter@amd.com>
2008-10-15 20:47:38 +02:00
Robert Richter
2d55a47882 OProfile: Rework string handling in setup_ibs_files()
Signed-off-by: Robert Richter <robert.richter@amd.com>
2008-10-15 20:47:34 +02:00
Robert Richter
e2fee2761a OProfile: Rework oprofile_add_ibs_sample() function
Code looks much more cleaner now.

Signed-off-by: Robert Richter <robert.richter@amd.com>
2008-10-15 20:47:31 +02:00
Linus Torvalds
04ab591808 Merge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus
* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus:
  MIPS: Kill unused <asm/debug.h> inclusions
  MIPS: IP32: Add platform device for CMOS RTC; remove dead code
  RTC: M48T35: new RTC driver
  MIPS: IP27: Switch over to RTC class driver
  MIPS: DS1286: New RTC driver
  MIPS: IP22/28: Switch over to RTC class driver
  MIPS: PCI: Scan busses when they are registered
  MIPS: WGT634U: Add reset button support
  MIPS: BCM47xx: Use the new SSB GPIO API
  MIPS: BCM47xx: Remove references to BCM947XX
  MIPS: WGT634U: Add machine detection message
  MIPS: Align .data.cacheline_aligned based on CONFIG_MIPS_L1_CACHE_SHIFT
  MIPS: show_cpuinfo prints the type of the calling CPU
  MIPS: Fix wrong branch target in new spin_lock code.
  MIPS: Have a heart for a lonely, lost header file ...
2008-10-15 10:22:21 -07:00
Wolfram Sang
9629095761 powerpc/mpc5200: Don't touch pipelining for MPC5200B
MPC5200 needs to have pipelining disabled for ATA to work. MPC5200B does not.
So, for the latter, don't touch the original setting from the bootloader.

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
2008-10-15 11:09:59 -06:00
Grant Likely
40847993d8 powerpc/52xx: Make cuImage more robust in locating immr node.
Current device trees do not have the device_type = soc property set
anymore.  Fix up the cuImage bootwrapper fragment to still find the IMMR
nodes.

Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
2008-10-15 11:09:59 -06:00
Linus Torvalds
5f2434a66d Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: (158 commits)
  powerpc: Fix CHRP PCI config access for indirect_pci
  powerpc/chrp: Fix detection of Python PCI host bridge on IBM CHRPs
  powerpc: Fix 32-bit SMP boot on CHRP
  powerpc: Fix link errors on 32-bit machines using legacy DMA
  powerpc/pci: Improve detection of unassigned bridge resources
  hvc_console: Fix free_irq in spinlocked section
  powerpc: Get USE_STRICT_MM_TYPECHECKS working again
  powerpc: Reflect the used arguments in machine_init() prototype
  powerpc: Fix DMA offset for non-coherent DMA
  powerpc: fix fsl_upm nand driver modular build
  powerpc/83xx: add NAND support for the MPC8360E-RDK boards
  powerpc: FPGA support for GE Fanuc SBC610
  i2c: MPC8349E-mITX Power Management and GPIO expander driver
  powerpc: reserve two DMA channels for audio in MPC8610 HPCD device tree
  powerpc: document the "fsl,ssi-dma-channel" compatible property
  powerpc: disable CHRP and PMAC support in various defconfigs
  OF: add fsl,mcu-mpc8349emitx to the exception list
  powerpc/83xx: add DS1374 RTC support for the MPC837xE-MDS boards
  powerpc: remove support for bootmem-allocated memory for the DIU driver
  powerpc: remove non-dependent load fsl_booke PTE_64BIT
  ...
2008-10-15 08:07:35 -07:00
Xiantao Zhang
2381ad241d KVM: ia64: Add intel iommu support for guests.
With intel iommu hardware, we can assign devices to kvm/ia64 guests.

Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-10-15 14:25:39 +02:00
Xiantao Zhang
b010eb5103 KVM: ia64: add directed mmio range support for kvm guests
Using vt-d, kvm guests can be assigned physcial devices, so
this patch introduce a new mmio type (directed mmio)
to handle its mmio access.

Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-10-15 14:25:38 +02:00
Xiantao Zhang
1cbea809c4 KVM: ia64: Make pmt table be able to hold physical mmio entries.
Don't try to do put_page once the entries are mmio.
Set the tag to indicate the mmio space for vmm setting
TLB's memory attribute.

Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-10-15 14:25:37 +02:00
Xiantao Zhang
2f74977191 KVM: Move irqchip_in_kernel() from ioapic.h to irq.h
Moving irqchip_in_kernel() from ioapic.h to irq.h.

Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-10-15 14:25:36 +02:00
Xiantao Zhang
3de42dc094 KVM: Separate irq ack notification out of arch/x86/kvm/irq.c
Moving irq ack notification logic as common, and make
it shared with ia64 side.

Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-10-15 14:25:35 +02:00
Xiantao Zhang
8a98f6648a KVM: Move device assignment logic to common code
To share with other archs, this patch moves device assignment
logic to common parts.

Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-10-15 14:25:33 +02:00
Zhang xiantao
371c01b28e KVM: Device Assignment: Move vtd.c from arch/x86/kvm/ to virt/kvm/
Preparation for kvm/ia64 VT-d support.

Signed-off-by: Zhang xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-10-15 14:25:32 +02:00
Marcelo Tosatti
83dbc83a0d KVM: VMX: enable invlpg exiting if EPT is disabled
Manually disabling EPT via module option fails to re-enable INVLPG
exiting.

Reported-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-10-15 14:25:31 +02:00
Jan Kiszka
1b10bf31a5 KVM: x86: Silence various LAPIC-related host kernel messages
KVM-x86 dumps a lot of debug messages that have no meaning for normal
operation:
 - INIT de-assertion is ignored
 - SIPIs are sent and received
 - APIC writes are unaligned or < 4 byte long
   (Windows Server 2003 triggers this on SMP)

Degrade them to true debug messages, keeping the host kernel log clean
for real problems.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2008-10-15 14:25:30 +02:00
Weidong Han
e5fcfc821a KVM: Device Assignment: Map mmio pages into VT-d page table
Assigned device could DMA to mmio pages, so also need to map mmio pages
into VT-d page table.

Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2008-10-15 14:25:29 +02:00
Marcelo Tosatti
e48258009d KVM: PIC: enhance IPI avoidance
The PIC code makes little effort to avoid kvm_vcpu_kick(), resulting in
unnecessary guest exits in some conditions.

For example, if the timer interrupt is routed through the IOAPIC, IRR
for IRQ 0 will get set but not cleared, since the APIC is handling the
acks.

This means that everytime an interrupt < 16 is triggered, the priority
logic will find IRQ0 pending and send an IPI to vcpu0 (in case IRQ0 is
not masked, which is Linux's case).

Introduce a new variable isr_ack to represent the IRQ's for which the
guest has been signalled / cleared the ISR. Use it to avoid more than
one IPI per trigger-ack cycle, in addition to the avoidance when ISR is
set in get_priority().

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2008-10-15 14:25:28 +02:00