Tetsuo Handa reported this link bug:
| arch/x86/mm/built-in.o(.init.text+0x1831): In function `early_ioremap_init':
| : undefined reference to `__this_fixmap_does_not_exist'
| make: *** [.tmp_vmlinux1] Error 1
Commit:8827247ffcc9e880cbe4705655065cf011265157 used a variable (which
would be optimized to constant) as fix_to_virt()'s parameter.
It's depended on gcc's optimization and fails on old gcc. (Tetsuo used gcc 3.3)
We can use __fix_to_vir() instead, because we know it's safe and
don't need link time error reporting.
Reported-by: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Cc: sfr@canb.auug.org.au
LKML-Reference: <49C9FFEA.7060908@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix bug with irq-descriptor moving when logical flat
Rusty observed:
> The effect of setting desc->affinity (ie. from userspace via sysfs) has varied
> over time. In 2.6.27, the 32-bit code anded the value with cpu_online_map,
> and both 32 and 64-bit did that anding whenever a cpu was unplugged.
>
> 2.6.29 consolidated this into one routine (and fixed hotplug) but introduced
> another variation: anding the affinity with cfg->domain. Is this right, or
> should we just set it to what the user said? Or as now, indicate that we're
> restricting it.
Eric pointed out that desc->affinity should be what the user requested,
if it is at all possible to honor the user space request.
This bug got introduced by commit 22f65d31b "x86: Update io_apic.c to use
new cpumask API".
Fix it by moving the masking to before the descriptor moving ...
Reported-by: Rusty Russell <rusty@rustcorp.com.au>
Reported-by: Eric W. Biederman <ebiederm@xmission.com>
LKML-Reference: <49C94134.4000408@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch move the timestamp from happening in the arch specific
code into the general code. This allows for better control by the tracer
to time manipulation.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
This iommu_op can tell if domain have a specific capability, like snooping
control for Intel IOMMU, which can be used by other components of kernel to
adjust the behaviour.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
While looking at the issue in the thread:
http://marc.info/?l=dri-devel&m=123606627824556&w=2
noticed a bug in pci PAT code and memory type setting.
PCI mmap code did not set the proper protection in vma, when it
inherited protection in reserve_memtype. This bug only affects
the case where there exists a WC mapping before X does an mmap
with /proc or /sys pci interface. This will cause X userlevel
mmap from /proc or /sysfs to fail on fork.
Reported-by: Kevin Winchester <kjwinchester@gmail.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: Dave Airlie <airlied@redhat.com>
Cc: <stable@kernel.org>
LKML-Reference: <20090323190720.GA16831@linux-os.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
vmx_set_msr() does not allow i386 guests to touch EFER, but they can still
do so through the default: label in the switch. If they set EFER_LME, they
can oops the host.
Fix by having EFER access through the normal channel (which will check for
EFER_LME) even on i386.
Reported-and-tested-by: Benjamin Gilbert <bgilbert@cs.cmu.edu>
Cc: stable@kernel.org
Signed-off-by: Avi Kivity <avi@redhat.com>
When kvm emulates an invlpg instruction, it can drop a shadow pte, but
leaves the guest tlbs intact. This can cause memory corruption when
swapping out.
Without this the other cpu can still write to a freed host physical page.
tlb smp flush must happen if rmap_remove is called always before mmu_lock
is released because the VM will take the mmu_lock before it can finally add
the page to the freelist after swapout. mmu notifier makes it safe to flush
the tlb after freeing the page (otherwise it would never be safe) so we can do
a single flush for multiple sptes invalidated.
Cc: stable@kernel.org
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Acked-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Impact: Make symbols static.
Fix this sparse warnings:
arch/x86/kvm/mmu.c:992:5: warning: symbol 'mmu_pages_add' was not declared. Should it be static?
arch/x86/kvm/mmu.c:1124:5: warning: symbol 'mmu_pages_next' was not declared. Should it be static?
arch/x86/kvm/mmu.c:1144:6: warning: symbol 'mmu_pages_clear_parents' was not declared. Should it be static?
arch/x86/kvm/x86.c:2037:5: warning: symbol 'kvm_read_guest_virt' was not declared. Should it be static?
arch/x86/kvm/x86.c:2067:5: warning: symbol 'kvm_write_guest_virt' was not declared. Should it be static?
virt/kvm/irq_comm.c:220:5: warning: symbol 'setup_routing_entry' was not declared. Should it be static?
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: Avi Kivity <avi@redhat.com>
Impact: Attribute function with __acquires(...) resp. __releases(...).
Fix this sparse warnings:
arch/x86/kvm/i8259.c:34:13: warning: context imbalance in 'pic_lock' - wrong count at exit
arch/x86/kvm/i8259.c:39:13: warning: context imbalance in 'pic_unlock' - unexpected unlock
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: Avi Kivity <avi@redhat.com>
is_long_mode currently checks the LongModeEnable bit in
EFER instead of the LongModeActive bit. This is wrong, but
we survived this till now since it wasn't triggered. This
breaks guests that go from long mode to compatibility mode.
This is noticed on a solaris guest and fixes bug #1842160
Signed-off-by: Amit Shah <amit.shah@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
setup_msrs() should be called when entering long mode to save the
shadow state for the 64-bit guest state.
Using vmx_set_efer() in enter_lmode() removes some duplicated code
and also ensures we call setup_msrs(). We can safely pass the value
of shadow_efer to vmx_set_efer() as no other bits in the efer change
while enabling long mode (guest first sets EFER.LME, then sets CR0.PG
which causes a vmexit where we activate long mode).
With this fix, is_long_mode() can check for EFER.LMA set instead of
EFER.LME and 5e23049e86dd298b72e206b420513dbc3a240cd9 can be reverted.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
In the paging_fetch function rmap_remove is called after setting a large
pte to non-present. This causes rmap_remove to not drop the reference to
the large page. The result is a memory leak of that page.
Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
In the segment descriptor _cache_ the accessed bit is always set
(although it can be cleared in the descriptor itself). Since Intel
checks for this condition on a VMENTRY, set this bit in the AMD path
to enable cross vendor migration.
Cc: stable@kernel.org
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Acked-By: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
IRQ injection status is either -1 (if there was no CPU found
that should except the interrupt because IRQ was masked or
ioapic was misconfigured or ...) or >= 0 in that case the
number indicates to how many CPUs interrupt was injected.
If the value is 0 it means that the interrupt was coalesced
and probably should be reinjected.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The assertion no longer makes sense since we don't clear page tables on
allocation; instead we clear them during prefetch.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The following code flow is unnecessary:
if (largepage)
was_rmapped = is_large_pte(*shadow_pte);
else
was_rmapped = 1;
The is_large_pte() function will always evaluate to one here because the
(largepage && !is_large_pte) case is already handled in the first
if-clause. So we can remove this check and set was_rmapped to one always
here.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
kvmclock currently falls apart on machines without constant tsc.
This patch fixes it. Changes:
* keep tsc frequency in a per-cpu variable.
* handle kvmclock update using a new request flag, thus checking
whenever we need an update each time we enter guest context.
* use a cpufreq notifier to track frequency changes and force
kvmclock updates.
* send ipis to kick cpu out of guest context if needed to make
sure the guest doesn't see stale values.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Looks like neither the direction nor the rep prefix are used anymore.
Drop related evaluations from SVM's and VMX's I/O exit handlers.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
AMD K10 CPUs implement the FFXSR feature that gets enabled using
EFER. Let's check if the virtual CPU description includes that
CPUID feature bit and allow enabling it then.
This is required for Windows Server 2008 in Hyper-V mode.
v2 adds CPUID capability exposure
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
AMD k10 includes support for the FFXSR feature, which leaves out
XMM registers on FXSAVE/FXSAVE when the EFER_FFXSR bit is set in
EFER.
The CPUID feature bit exists already, but the EFER bit is missing
currently, so this patch adds it to the list of known EFER bits.
Signed-off-by: Alexander Graf <agraf@suse.de>
CC: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
IRQ ack notifications assume an identity mapping between pin->gsi,
which might not be the case with, for example, HPET.
Translate before acking.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Acked-by: Gleb Natapov <gleb@redhat.com>
Kconfig symbols are not available in userspace, and are not stripped by
headers-install. Avoid their use by adding #defines in <asm/kvm.h> to
suit each architecture.
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently KVM has a static routing from GSI numbers to interrupts (namely,
0-15 are mapped 1:1 to both PIC and IOAPIC, and 16:23 are mapped 1:1 to
the IOAPIC). This is insufficient for several reasons:
- HPET requires non 1:1 mapping for the timer interrupt
- MSIs need a new method to assign interrupt numbers and dispatch them
- ACPI APIC mode needs to be able to reassign the PCI LINK interrupts to the
ioapics
This patch implements an interrupt routing table (as a linked list, but this
can be easily changed) and a userspace interface to replace the table. The
routing table is initialized according to the current hardwired mapping.
Signed-off-by: Avi Kivity <avi@redhat.com>
Some typos, comments, whitespace errors corrected in the cpuid code
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This actually describes what is going on, rather than alerting the reader
that something strange is going on.
Signed-off-by: Avi Kivity <avi@redhat.com>
Zeroing on mmu_memory_cache_alloc is unnecessary since:
- Smaller areas are pre-allocated with kmem_cache_zalloc.
- Page pointed by ->spt is overwritten with prefetch_page
and entries in page pointed by ->gfns are initialized
before reading.
[avi: zeroing pages is unnecessary]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
While the PIT is masked the guest cannot ack the irq, so the reinject logic
will never allow the interrupt to be injected.
Fix by resetting the reinjection counters on unmask.
Unbreaks Xen.
Signed-off-by: Avi Kivity <avi@redhat.com>
Two KVM archs support irqchips and two don't. Add a Kconfig item to
make selecting between the two models easier.
Signed-off-by: Avi Kivity <avi@redhat.com>
Using kvm_mmu_lookup_page() will result in multiple scans of the hash chains;
use hlist_for_each_entry_safe() to achieve a single scan instead.
Signed-off-by: Avi Kivity <avi@redhat.com>
VMware ESX checks if the microcode level is correct when using a barcelona
CPU, in order to see if it actually can use SVM. Let's tell it we're on the
safe side...
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Allow emulate_pop() to read into arbitrary memory rather than just the
source operand. Needed for complicated instructions like far returns.
Signed-off-by: Avi Kivity <avi@redhat.com>
If we've just emulated an instruction, we won't have any valid exit
reason and associated information.
Fix by moving the clearing of the emulation_required flag to the exit handler.
This way the exit handler can notice that we've been emulating and abort
early.
Signed-off-by: Avi Kivity <avi@redhat.com>
Certain clocks (such as TSC) in older 2.6 guests overaccount for lost
ticks, causing severe time drift. Interrupt reinjection magnifies the
problem.
Provide an option to disable it.
[avi: allow room for expansion in case we want to disable reinjection
of other timers]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Since we advertise MSR_VM_HSAVE_PA, userspace will attempt to read it
even on Intel. Implement fake support for this MSR to avoid the
warnings.
Signed-off-by: Avi Kivity <avi@redhat.com>
vmap() on guest pages hides those pages from the Linux mm for an extended
(userspace determined) amount of time. Get rid of it.
Signed-off-by: Izik Eidus <ieidus@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This commit change the name of emulator_read_std into kvm_read_guest_virt,
and add new function name kvm_write_guest_virt that allow writing into a
guest virtual address.
Signed-off-by: Izik Eidus <ieidus@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
VMX initializes the TSC offset for each vcpu at different times, and
also reinitializes it for vcpus other than 0 on APIC SIPI message.
This bug causes the TSC's to appear unsynchronized in the guest, even if
the host is good.
Older Linux kernels don't handle the situation very well, so
gettimeofday is likely to go backwards in time:
http://www.mail-archive.com/kvm@vger.kernel.org/msg02955.htmlhttp://sourceforge.net/tracker/index.php?func=detail&aid=2025534&group_id=180599&atid=893831
Fix it by initializating the offset of each vcpu relative to vm creation
time, and moving it from vmx_vcpu_reset to vmx_vcpu_setup, out of the
APIC MP init path.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Using a for_each loop style removes the need to write callback and nasty
casts.
Implement the walk_shadow() using the for_each_shadow_entry().
Signed-off-by: Avi Kivity <avi@redhat.com>
The AMD SVM instruction family all overload the 0f 01 /3 opcode, further
multiplexing on the three r/m bits. But the code decided that anything that
isn't a vmmcall must be an lidt (which shares the 0f 01 /3 opcode, for the
case that mod = 3).
Fix by aborting emulation if this isn't a vmmcall.
Signed-off-by: Avi Kivity <avi@redhat.com>
If cr4.pge is cleared, we ought to treat any ptes in the page as non-global.
This allows us to remove the check from set_spte().
Signed-off-by: Avi Kivity <avi@redhat.com>
Don't allow a vcpu with cr4.pge cleared to use a shadow page created with
cr4.pge set; this might cause a cr3 switch not to sync ptes that have the
global bit set (the global bit has no effect if !cr4.pge).
This can only occur on smp with different cr4.pge settings for different
vcpus (since a cr4 change will resync the shadow ptes), but there's no
cost to being correct here.
Signed-off-by: Avi Kivity <avi@redhat.com>
Instead of "calculating" it on every shadow page allocation, set it once
when switching modes, and copy it when allocating pages.
This doesn't buy us much, but sets up the stage for inheriting more
information related to the mmu setup.
Signed-off-by: Avi Kivity <avi@redhat.com>
Add the remaining bits to make use of debug registers also for guest
debugging, thus enabling the use of hardware breakpoints and
watchpoints.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
So far KVM only had basic x86 debug register support, once introduced to
realize guest debugging that way. The guest itself was not able to use
those registers.
This patch now adds (almost) full support for guest self-debugging via
hardware registers. It refactors the code, moving generic parts out of
SVM (VMX was already cleaned up by the KVM_SET_GUEST_DEBUG patches), and
it ensures that the registers are properly switched between host and
guest.
This patch also prepares debug register usage by the host. The latter
will (once wired-up by the following patch) allow for hardware
breakpoints/watchpoints in guest code. If this is enabled, the guest
will only see faked debug registers without functionality, but with
content reflecting the guest's modifications.
Tested on Intel only, but SVM /should/ work as well, but who knows...
Known limitations: Trapping on tss switch won't work - most probably on
Intel.
Credits also go to Joerg Roedel - I used his once posted debugging
series as platform for this patch.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
When single-stepping over STI and MOV SS, we must clear the
corresponding interruptibility bits in the guest state. Otherwise
vmentry fails as it then expects bit 14 (BS) in pending debug exceptions
being set, but that's not correct for the guest debugging case.
Note that clearing those bits is safe as we check for interruptibility
based on the original state and do not inject interrupts or NMIs if
guest interruptibility was blocked.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This rips out the support for KVM_DEBUG_GUEST and introduces a new IOCTL
instead: KVM_SET_GUEST_DEBUG. The IOCTL payload consists of a generic
part, controlling the "main switch" and the single-step feature. The
arch specific part adds an x86 interface for intercepting both types of
debug exceptions separately and re-injecting them when the host was not
interested. Moveover, the foundation for guest debugging via debug
registers is layed.
To signal breakpoint events properly back to userland, an arch-specific
data block is now returned along KVM_EXIT_DEBUG. For x86, the arch block
contains the PC, the debug exception, and relevant debug registers to
tell debug events properly apart.
The availability of this new interface is signaled by
KVM_CAP_SET_GUEST_DEBUG. Empty stubs for not yet supported archs are
provided.
Note that both SVM and VTX are supported, but only the latter was tested
yet. Based on the experience with all those VTX corner case, I would be
fairly surprised if SVM will work out of the box.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
VMX differentiates between processor and software generated exceptions
when injecting them into the guest. Extend vmx_queue_exception
accordingly (and refactor related constants) so that we can use this
service reliably for the new guest debugging framework.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Userspace has to tell the kernel module somehow that nested SVM should be used.
The easiest way that doesn't break anything I could think of is to implement
if (cpuid & svm)
allow write to efer
else
deny write to efer
Old userspaces mask the SVM capability bit, so they don't break.
In order to find out that the SVM capability is set, I had to split the
kvm_emulate_cpuid into a finding and an emulating part.
(introduced in v6)
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Normally setting the SVME bit in EFER is not allowed, as we did
not support SVM. Not since we do, we should also allow enabling
SVM mode.
v2 comes as last patch, so we don't enable half-ready code
v4 introduces a module option to enable SVM
v6 warns that nesting is enabled
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
KVM tries to read the VM_CR MSR to find out if SVM was disabled by
the BIOS. So implement read support for this MSR to make nested
SVM running.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This adds the #VMEXIT intercept, so we return to the level 1 guest
when something happens in the level 2 guest that should return to
the level 1 guest.
v2 implements HIF handling and cleans up exception interception
v3 adds support for V_INTR_MASKING_MASK
v4 uses the host page hsave
v5 removes IOPM merging code
v6 moves mmu code out of the atomic section
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch implements VMRUN. VMRUN enters a virtual CPU and runs that
in the same context as the normal guest CPU would run.
So basically it is implemented the same way, a normal CPU would do it.
We also prepare all intercepts that get OR'ed with the original
intercepts, as we do not allow a level 2 guest to be intercepted less
than the first level guest.
v2 implements the following improvements:
- fixes the CPL check
- does not allocate iopm when not used
- remembers the host's IF in the HIF bit in the hflags
v3:
- make use of the new permission checking
- add support for V_INTR_MASKING_MASK
v4:
- use host page backed hsave
v5:
- remove IOPM merging code
v6:
- save cr4 so PAE l1 guests work
v7:
- return 0 on vmrun so we check the MSRs too
- fix MSR check to use the correct variable
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This implements the VMLOAD and VMSAVE instructions, that usually surround
the VMRUN instructions. Both instructions load / restore the same elements,
so we only need to implement them once.
v2 fixes CPL checking and replaces memcpy by assignments
v3 makes use of the new permission checking
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Implement the hsave MSR, that gives the VCPU a GPA to save the
old guest state in.
v2 allows userspace to save/restore hsave
v4 dummys out the hsave MSR, so we use a host page
v6 remembers the guest's hsave and exports the MSR
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch implements the GIF flag and the clgi and stgi instructions that
set this flag. Only if the flag is set (default), interrupts can be received by
the CPU.
To keep the information about that somewhere, this patch adds a new hidden
flags vector. that is used to store information that does not go into the
vmcb, but is SVM specific.
I tried to write some code to make -no-kvm-irqchip work too, but the first
level guest won't even boot with that atm, so I ditched it.
v2 moves the hflags to x86 generic code
v3 makes use of the new permission helper
v6 only enables interrupt_window if GIF=1
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
These are helpers for the nested SVM implementation.
- nsvm_printk implements a debug printk variant
- nested_svm_do calls a handler that can accesses gpa-based memory
v3 makes use of the new permission checker
v6 changes:
- streamline nsvm_debug()
- remove printk(KERN_ERR)
- SVME check before CPL check
- give GP error code
- use new EFER constant
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
MSR_EFER_SVME_MASK, MSR_VM_CR and MSR_VM_HSAVE_PA are set in KVM
specific headers. Linux does have nice header files to collect
EFER bits and MSR IDs, so IMHO we should put them there.
While at it, I also changed the naming scheme to match that
of the other defines.
(introduced in v6)
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
The current VINTR intercept setters don't look clean to me. To make
the code easier to read and enable the possibilty to trap on a VINTR
set, this uses a helper function to set the VINTR intercept.
v2 uses two distinct functions for setting and clearing the bit
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Impact: section mismatch fix
Ingo reports these warnings:
> WARNING: vmlinux.o(.text+0x6a288e): Section mismatch in reference from
> the function dmi_alloc() to the function .init.text:extend_brk()
> The function dmi_alloc() references
> the function __init extend_brk().
> This is often because dmi_alloc lacks a __init annotation or the
> annotation of extend_brk is wrong.
dmi_alloc() is a static inline, and so should be immune to this
kind of error. But force it to be inlined and make it __init
anyway, just to be extra sure.
All of dmi_alloc()'s callers are already __init.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <49C6B23C.2040308@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
This fixed various signedness issues in setup.c and e820.c:
arch/x86/kernel/setup.c:455:53: warning: incorrect type in argument 3 (different signedness)
arch/x86/kernel/setup.c:455:53: expected int *pnr_map
arch/x86/kernel/setup.c:455:53: got unsigned int extern [toplevel] *<noident>
arch/x86/kernel/setup.c:639:53: warning: incorrect type in argument 3 (different signedness)
arch/x86/kernel/setup.c:639:53: expected int *pnr_map
arch/x86/kernel/setup.c:639:53: got unsigned int extern [toplevel] *<noident>
arch/x86/kernel/setup.c:820:54: warning: incorrect type in argument 3 (different signedness)
arch/x86/kernel/setup.c:820:54: expected int *pnr_map
arch/x86/kernel/setup.c:820:54: got unsigned int extern [toplevel] *<noident>
arch/x86/kernel/e820.c:670:53: warning: incorrect type in argument 3 (different signedness)
arch/x86/kernel/e820.c:670:53: expected int *pnr_map
arch/x86/kernel/e820.c:670:53: got unsigned int [toplevel] *<noident>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
move out msi_ir_chip and ir_ioapic_chip from CONFIG_INTR_REMAP shadow
Fix:
arch/x86/kernel/apic/io_apic.c:1431: warning: ‘msi_ir_chip’ defined but not used
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Impact: cleanup
This patch fixes the following sparse warnings:
arch/x86/kernel/apic/io_apic.c:3602:17: warning: symbol 'hpet_msi_type'
was not declared. Should it be static?
arch/x86/kernel/apic/io_apic.c:3467:30: warning: Using plain integer as
NULL pointer
Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@movial.com>
LKML-Reference: <1237741871-5827-2-git-send-email-dmitri.vorobiev@movial.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This reverts commit 698609bdcd.
69860 breaks Xen booting, as it relies on head*.S to set up the fixmap
pagetables (as a side-effect of initializing the USB debug port).
Xen, however, does not boot via head*.S, and so the fixmap area is
not initialized.
The specific symptom of the crash is a fault in dmi_scan(), because
the pointer that early_ioremap returns is not actually present.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Jan Beulich <jbeulich@novell.com>
LKML-Reference: <49C43A8E.5090203@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
To reduce the size of the oversized function __get_smp_config()
There should be no impact to functionality.
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Impact: fix incorrect error message
- IO APIC resource allocation error message contains one too many "be".
- Print the error message iff there are IO APICs in the system.
I've seen this error message for some time on my x86-32 laptop...
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Cc: Alan Bartlett <ajb.stxsl@googlemail.com>
LKML-Reference: <200903202100.30789.bzolnier@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix mmconfig detection to not assume a single mmconfig space in the
northbridge, paving the way for AMD fam10h + mcp55 CPUs. On those, the
MSR has some range, but the mcp55 pci config will have another one.
Also helps the mcp55 + io55 case, where every one will have one range.
If it is mcp55, exclude the range that is used by CPU MSR, in other
words , if the CPU claims busses 0-255, the range in mcp55 is dropped,
because CPU HW will not route those ranges to mcp55 mmconfig to handle
it.
Signed-off-by: Yinghai Lu <yinghai.lu@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Detect and enable memory-mapped PCI configuration space on the nVidia
MCP55 southbridge. Tested against 2.6.27.4 on an Arista Networks
development board with one MCP55, Coreboot firmware, no ACPI.
Signed-off-by: Ed Swierk <eswierk@aristanetworks.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Impact: cleanup
Check alternate signal stack overflow with proper stack pointer.
The stack pointer of the next signal frame is different if that
task has i387 state.
On x86_64, redzone would be included.
No need to check SA_ONSTACK if we're already using alternate signal stack.
Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Cc: Roland McGrath <roland@redhat.com>
LKML-Reference: <49C2874D.3080002@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Many host bridges support a 4k config space, so check them directy
instead of using quirks to add them.
We only need to do this extra check for host bridges at this point,
because only host bridges are known to have extended address space
without also having a PCI-X/PCI-E caps. Other devices with this
property could be done with quirks (if there are any).
As a bonus, we can remove the quirks for AMD host bridges with family
10h and 11h since they're not needed any more.
With this patch, we can get correct pci cfg size of new Intel CPUs/IOHs
with host bridges.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Reviewed-by: Matthew Wilcox <willy@linux.intel.com>
Cc: <stable@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Add the new API pci_enable_msi_block() to allow drivers to
request multiple MSI and reimplement pci_enable_msi in terms of
pci_enable_msi_block. Ensure that the architecture back ends don't
have to know about multiple MSI.
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This patch changes a VIA PCI quirk to use dev_info() rather than printk().
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgek.org>
Add new interfaces:
set_pages_array_uc()
set_pages_array_wb()
that can be used change the page attribute for a bunch of pages with
flush etc done once at the end of all the changes. These interfaces
are similar to existing set_memory_array_uc() and set_memory_array_wc().
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: arjan@infradead.org
Cc: eric@anholt.net
Cc: airlied@redhat.com
LKML-Reference: <20090319215358.901545000@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Add struct page array pointer to cpa struct and CPA_PAGES_ARRAY.
With that we can add change_page_attr_set_clr() a parameter to pass
struct page array pointer and that can be handled by the underlying
cpa code.
cpa_flush_array() is also changed to support both addr array or
struct page pointer array, depending on the flag.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: arjan@infradead.org
Cc: eric@anholt.net
Cc: airlied@redhat.com
LKML-Reference: <20090319215358.758513000@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Change change_page_attr_set_clr() array parameter to a flag. This helps
following patches which adds an interface to change attr to uc/wb over a
set of pages referred by struct page.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: arjan@infradead.org
Cc: eric@anholt.net
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: airlied@redhat.com
LKML-Reference: <20090319215358.611346000@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Weak functions aren't all they're cracked up to be. They lead to
incorrect binaries with some toolchains, they require us to have empty
functions we otherwise wouldn't, and the unused code is not elided
(as of gcc 4.3.2 anyway).
So replace the weak MSI arch hooks with the #define foo foo idiom. We no
longer need empty versions of arch_setup/teardown_msi_irq().
This is less source (by 1 line!), and results in smaller binaries too:
text data bss dec hex filename
9354300 1693916 678424 11726640 b2ef30 build/powerpc/vmlinux-before
9354052 1693852 678424 11726328 b2edf8 build/powerpc/vmlinux-after
Also smaller on x86_64 and arm (iop13xx).
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Use %02x:%02x.%d rather than %02x:%02x:%02x so PCI addresses
look the same as in other parts of the kernel.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
The early "dump PCI config space" code skips many multi-function
devices. This patch fixes that, so it dumps all devices in PCI
domain 0.
We should not skip the rest of the functions if CLASS_REVISION is
0xffffffff. Often multi-function devices have gaps in the function ID
space, e.g., 1c.0 and 1c.2 exist but 1c.1 doesn't. The CLASS_REVISION
of the non-existent 1c.1 function will appear to be 0xffffffff.
We should only look at the HEADER_TYPE of function zero. Often the
"multi-function" is set in function zero, but not in other functions.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Impact: cleanup, remove last user of set_pte_present
set_pte_vaddr() is only used to install ptes in fixmaps, and
should never be used to overwrite a present mapping.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Xen-devel <xen-devel@lists.xensource.com>
LKML-Reference: <1237406613-2929-1-git-send-email-jeremy@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When I review the sensitive code ftrace_nmi_enter(), I found
the atomic variable nmi_running does protect NMI VS do_ftrace_mod_code(),
but it can not protects NMI(entered nmi) VS NMI(ftrace_nmi_enter()).
cpu#1 | cpu#2 | cpu#3
ftrace_nmi_enter() | do_ftrace_mod_code() |
not modify | |
------------------------|-----------------------|--
executing | set mod_code_write = 1|
executing --|-----------------------|--------------------
executing | | ftrace_nmi_enter()
executing | | do modify
------------------------|-----------------------|-----------------
ftrace_nmi_exit() | |
cpu#3 may be being modified the code which is still being executed on cpu#1,
it will have undefined results and possibly take a GPF, this patch
prevents it occurred.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
LKML-Reference: <49C0B411.30003@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Impact: fix rarely-used feature
The VGA Miscellaneous Output Register is read from address 0x3CC but
written to address 0x3C2. This was missed when this code was
converted from assembly to C. While we're at it, clean up the code by
making the overflow bits and the math used to set the bits explicit.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Impact: cleanup
Refactor the MP-table parsing code via the introduction of the
following helper functions:
skip_entry()
smp_reserve_bootmem()
check_irq_src()
check_slot()
To simplify the code flow and to reduce the size of the
following oversized functions: smp_read_mpc(), smp_scan_config().
There should be no impact to functionality.
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup, reduce memory usage for CONFIG_CPUMASK_OFFSTACK=y
Part of the "getting rid of obsolete cpumask_t" patch:
1) Use cpumask_var_t: this is a pointer if CONFIG_CPUMASK_OFFSTACK=y
2) Call alloc_cpumask_var() on first entry into enter_uniprocessor()
3) Use modern cpumask_* functions.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Pekka Paalanen <pq@iki.fi>
LKML-Reference: <200903111633.55952.rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup, avoid cpumask games
The APM code wants to run on CPU 0: we create an "on_cpu0" wrapper
which uses work_on_cpu() if we're not already on cpu 0.
This introduces a new failure mode: -ENOMEM, so we add an explicit
err arg and handle Linux-style errnos in apm_err().
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
LKML-Reference: <200903111631.29787.rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: don't play with current's cpumask
Straightforward indirection through work_on_cpu(). One change is
that the error code from microcode_update_cpu() is now actually
plumbed back to microcode_init_cpu(), so now we printk if it fails
on cpu hotplug.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Dmitry Adamushko <dmitry.adamushko@gmail.com>
Cc: Peter Oruba <peter.oruba@amd.com>
LKML-Reference: <200903111632.37279.rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
arch/x86/kernel/cpu/mtrr/cleanup.c:197: warning: format ‘%d’ expects type ‘int’, but argument 2 has type ‘long unsigned int’
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <1237378015.13488.1.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix boot crash on UV systems
Commit 76ba0ecda0 "cpumask: use
cpumask_var_t in uv_flush_tlb_others" used cur_cpu as an iterator;
it was supposed to be zero for the code below it.
Reported-by: Cliff Wickman <cpw@sgi.com>
Original-From: Cliff Wickman <cpw@sgi.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Mike Travis <travis@sgi.com>
Cc: steiner@sgi.com
Cc: <stable@kernel.org>
LKML-Reference: <200903180822.31196.rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: Fix cpu offline when CONFIG_MAXSMP=y
Changeset bc9b83dd1f "cpumask: convert
c1e_mask in arch/x86/kernel/process.c to cpumask_var_t" contained a
bug: c1e_mask is manipulated even if C1E isn't detected (and hence
not allocated).
This is simply fixed by checking for NULL (which gcc optimizes out
anyway of CONFIG_CPUMASK_OFFSTACK=n, since it knows ce1_mask can never
be NULL).
In addition, fix a leak where select_idle_routine re-allocates
(and re-clears) c1e_mask on every cpu init.
Reported-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Mike Travis <travis@sgi.com>
LKML-Reference: <200903171450.34549.rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: optimize APIC IPI related barriers
Uncached MMIO accesses for xapic are inherently serializing and hence
we don't need explicit barriers for xapic IPI paths.
x2apic MSR writes/reads don't have serializing semantics and hence need
a serializing instruction or mfence, to make all the previous memory
stores globally visisble before the x2apic msr write for IPI.
Add x2apic_wrmsr_fence() in flush tlb path to x2apic specific paths.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: "steiner@sgi.com" <steiner@sgi.com>
Cc: Nick Piggin <npiggin@suse.de>
LKML-Reference: <1237313814.27006.203.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Attempting to rid us of the problematic work_on_cpu(). Just use
smp_call_function_single() here.
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
LKML-Reference: <20090318042217.EF3F1DDF39@ozlabs.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix spurious IRQs
During irq migration, we send a low priority interrupt to the previous
irq destination. This happens in non interrupt-remapping case after interrupt
starts arriving at new destination and in interrupt-remapping case after
modifying and flushing the interrupt-remapping table entry caches.
This low priority irq cleanup handler can cleanup multiple vectors, as
multiple irq's can be migrated at almost the same time. While
there will be multiple invocations of irq cleanup handler (one cleanup
IPI for each irq migration), first invocation of the cleanup handler
can potentially cleanup more than one vector (as the first invocation can
see the requests for more than vector cleanup). When we cleanup multiple
vectors during the first invocation of the smp_irq_move_cleanup_interrupt(),
other vectors that are to be cleanedup can still be pending in the local
cpu's IRR (as smp_irq_move_cleanup_interrupt() runs with interrupts disabled).
When we are ready to unhook a vector corresponding to an irq, check if that
vector is registered in the local cpu's IRR. If so skip that cleanup and
do a self IPI with the cleanup vector, so that we give a chance to
service the pending vector interrupt and then cleanup that vector
allocation once we execute the lowest priority handler.
This fixes spurious interrupts seen when migrating multiple vectors
at the same time.
[ This is apparently possible even on conventional xapic, although to
the best of our knowledge it has never been seen. The stable
maintainers may wish to consider this one for -stable. ]
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: stable@kernel.org
Impact: fix possible race
save_mask_IO_APIC_setup() was using non atomic memory allocation while getting
called with interrupts disabled. Fix this by splitting this into two different
function. Allocation part save_IO_APIC_setup() now happens before
disabling interrupts.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Impact: cleanup
Clean up #ifdefs and replace them with helper functions.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Impact: simplification
In the current code, for level triggered migration, we need to modify the
io-apic RTE with the update vector information, along with modifying interrupt
remapping table entry(IRTE) with vector and destination. This is to ensure that
remote IRR bit inthe IOAPIC RTE gets cleared when the cpu does EOI.
With this patch, for level triggered, we eliminate the io-apic RTE modification
(with the updated vector information), by using a virtual vector (io-apic pin
number). Real vector that is used for interrupting cpu will be coming from
the interrupt-remapping table entry. Trigger mode in the IRTE will always be
edge, and the actual level or edge trigger will be setup in the IO-APIC RTE.
So a level triggered interrupt will appear as an edge to the local apic
cpu but still as level to the IO-APIC.
With this change, level irq migration can be done by simply modifying
the interrupt-remapping table entry with out changing the io-apic RTE.
And as the interrupt appears as edge at the cpu, in addition to do the
local apic EOI, we need to do IO-APIC directed EOI to clear the remote
IRR bit in the IO-APIC RTE.
This simplies the irq migration in the presence of interrupt-remapping.
Idea-by: Rajesh Sankaran <rajesh.sankaran@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Impact: cleanup, paranoia
We were not clearing the local APIC in clear_local_APIC() in the
presence of x2apic. Fix it.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Impact: make kexec work with x2apic
disable_IO_APIC() gets called during crashdump aswell, which configures the
IO-APIC/LAPIC so that legacy interrupts can be delivered for the kexec'd kernel.
In the presence of interrupt-remapping, we need to change the
interrupt-remapping configuration aswell as modifying IO-APIC for virtual wire
B mode.
To keep things simple during the crash, use virtual wire A mode
(for which we don't need to touch io-apic and interrupt-remapping tables).
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Impact: interface augmentation (not yet used)
Enable fault handling flow for intr-remapping aswell. Fault handling
code now shared by both dma-remapping and intr-remapping.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Impact: cleanup
The setup code is mostly 16-bit code, but there is a small stub of
32-bit code at the end. Move the 32-bit code to a separate segment,
.text32, to avoid scrambling the disassembly.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Impact: build fix with CONFIG_RELOCATABLE
Move _end into a dummy section, so that relocs.c will know it is a
relocatable symbol.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Impact: disambiguate real .bss variables from .brk storage
Add a .brk section after the .bss section. This has no effect
on the final vmlinux, but it more clearly distinguishes the space
taken by actual .bss symbols, and the variable space reserved
by .brk users.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Impact: bulletproofing, clarification
The brk reservation symbols are just there to document the amount
of space reserved by brk users in the final vmlinux file. Their
addresses are irrelevent, and using their addresses will cause
certain havok. Name them ".brk.NAME", which is a valid asm symbol
but C can't reference it; it also highlights their special
role in the symbol table.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Impact: Tighten bound to avoid masking errors
The definition of MAPPING_BEYOND_END was excessive; this has a nasty
tendency to mask bugs. We have learned over time that this kind of
bug hiding can cause some very strange errors. Therefore, tighten the
bound to only need to map the actual kernel area.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Impact: cleanup
ALLOCATOR_SLOP is a vestigial remain from when we used the
bootmem allocator to allocate the kernel's linear memory mapping.
Now we directly reserve pages from the e820 mapping, and no
longer require secondary structures to keep track of allocated
pages.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Impact: crash fix
head_32.S needs to map the kernel itself, and enough space so
that mm/init.c can allocate space from the e820 allocator
for the linear map of low memory.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Don't boost at the addresses which are listed on exception tables,
because major page fault will occur on those addresses. In that case,
kprobes can not ensure that when instruction buffer can be freed since
some processes will sleep on the buffer.
kprobes-ia64 already has same check.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In order for ntpd to correctly synchronize the clocks, the frequency of
the system clock must not be off by more than 500 ppm (or, put another
way, 1:2000), or ntpd will end up giving up on trying to synchronize
properly, and ends up reseting the clock in jumps instead.
The fast TSC PIT calibration sometimes failed this test - it was
assuming that the PIT reads always took about one microsecond each (2us
for the two reads to get a 16-bit timer), and that calibrating TSC to
the PIT over 15ms should thus be sufficient to get much closer than
500ppm (max 2us error on both sides giving 4us over 15ms: a 270 ppm
error value).
However, that assumption does not always hold: apparently some hardware
is either very much slower at reading the PIT registers, or there was
other noise causing at least one machine to get 700+ ppm errors.
So instead of using a fixed 15ms timing loop, this changes the fast PIT
calibration to read the TSC delta over the individual PIT timer reads,
and use the result to calculate the error bars on the PIT read timing
properly. We then successfully calibrate the TSC only if the maximum
error bars fall below 500ppm.
In the process, we also relax the timing to allow up to 25ms for the
calibration, although it can happen much faster depending on hardware.
Reported-and-tested-by: Jesper Krogh <jesper@krogh.cc>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
During bootup, when we reprogram the PIT (programmable interval timer)
to start counting down from 0xffff in order to use it for the fast TSC
calibration, we should also make sure to delay a bit afterwards to allow
the PIT hardware to actually start counting with the new value.
That will happens at the next CLK pulse (1.193182 MHz), so the easiest
way to do that is to just wait at least one microsecond after
programming the new PIT counter value. We do that by just reading the
counter value back once - which will take about 2us on PC hardware.
Reported-and-tested-by: john stultz <johnstul@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Impact: don't trim e820 according to wrong mtrr
Ozan reports that his server emits strange warning.
it turns out the BIOS sets the MTRRs incorrectly.
Ignore those strange ranges, and don't trim e820,
just emit one warning about BIOS
Reported-by: Ozan Çağlayan <ozan@pardus.org.tr>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <49BEE1E7.7020706@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix crash on VMI (VMware)
When we generate a call sequence for calling a paravirtualized
function, we presume that the generated code is "call *0xXXXXX",
which is a 6 byte opcode; this is larger than a normal
direct call, and so we can patch a direct call over it.
At the moment, however we give gcc enough rope to hang us by
putting the address in a register and generating a two byte
indirect-via-register call. Prevent this by explicitly
dereferencing the function pointer and passing it into the
asm as a constant.
This prevents crashes in VMI, as it cannot handle unpatchable
callsites.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Alok Kataria <akataria@vmware.com>
LKML-Reference: <49BEEDC2.2070809@goop.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: Bug fix on UP
Referring commit cc3ca22063,
Peter removed __cpuinit annotations for mce_cpu_features()
and its successor functions, which caused troubles on UP
configurations.
However the intel_init_cmci() was introduced after that and
it also has __cpuinit annotation even though it is called from
mce_cpu_features(). Remove the annotation from that function
too.
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
asm/highmem.h inclusion is added to use kmap_atomic_prot_pfn()
by commit bb6d59ca92
Now kmap_atomic_prot_pfn is moved to iomap_32.c
by commit dd63fdcc63
So the asm/highmem.h inclusion in iomap_32.c is unnecessary now.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
LKML-Reference: <20090315151517.GA29074@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: help debug e820 bugs
Try to print out more info, to catch wrong call parameters.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <49BCB557.3030000@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix boot crash
Need to exit early if the addr is far above 64k.
The crash got exposed by:
78a8b35: x86: make e820_update_range() handle small range update
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: <stable@kernel.org>
LKML-Reference: <49BC2279.2030101@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: makes vmlinux section information more useful
Don't use ram after _end blindly for pagetables. aka init pages is before _end
put those pg table into .bss
[Adapted to use brk segment - Jeremy]
v2: keep initial page table up to 512M only.
v4: put initial page tables just before _end
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: new interface; remove hard-coded limit
Add RESERVE_BRK(name, size) macro to reserve space in the brk
area. This should be a conservative (ie, larger) estimate of
how much space might possibly be required from the brk area.
Any unused space will be freed, so there's no real downside
on making the reservation too large (within limits).
The name should be unique within a given file, and somewhat
descriptive.
The C definition of RESERVE_BRK() ends up being more complex than
one would expect to work around a cluster of gcc infelicities:
The first attempt was to simply try putting __section(.brk_reservation)
on a variable. This doesn't work because it ends up making it a
@progbits section, which gets actual space allocated in the vmlinux
executable.
The second attempt was to emit the space into a section using asm,
but gcc doesn't allow arguments to be passed to file-level asm()
statements, making it hard to pass in the size.
The final attempt is to wrap the asm() in a function to allow
it to have arguments, and put the function itself into the
.discard section, which vmlinux*.lds drops entirely from the
emitted vmlinux.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: simplification
We only need to map the kernel in head_32.S, not the whole of
lowmem. We use 512MB as a reasonable (but arbitrary) limit on
the maximum size of the kernel image.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: use new interface instead of previous ad hoc implementation
Use extend_brk() to allocate memory for DMI rather than having an
ad-hoc allocator.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: use new interface instead of previous ad hoc implementation
Rather than having special purpose init_pg_table_start/end variables
to delimit the kernel pagetable built by head_32.S, just use the brk
mechanism to extend the bss for the new pagetable.
This patch removes init_pg_table_start/end and pg0, defines __brk_base
(which is page-aligned and immediately follows _end), initializes
the brk region to start there, and uses it for the 32-bit pagetable.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: build fix
The brk initialization functions were incorrectly located inside
an #ifdef CONFIG_VLK_DEV_INITRD block, causing the obvious build failure in
minimal configurations.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Impact: new interface
Add a brk()-like allocator which effectively extends the bss in order
to allow very early code to do dynamic allocations. This is better than
using statically allocated arrays for data in subsystems which may never
get used.
The space for brk allocations is in the bss ELF segment, so that the
space is mapped properly by the code which maps the kernel, and so
that bootloaders keep the space free rather than putting a ramdisk or
something into it.
The bss itself, delimited by __bss_stop, ends before the brk area
(__brk_base to __brk_limit). The kernel text, data and bss is reserved
up to __bss_stop.
Any brk-allocated data is reserved separately just before the kernel
pagetable is built, as that code allocates from unreserved spaces
in the e820 map, potentially allocating from any unused brk memory.
Ultimately any unused memory in the brk area is used in the general
kernel memory pool.
Initially the brk space is set to 1MB, which is probably much larger
than any user needs (the largest current user is i386 head_32.S's code
to build the pagetables to map the kernel, which can get fairly large
with a big kernel image and no PSE support). So long as the system
has sufficient memory for the bootloader to reserve the kernel+1MB brk,
there are no bad effects resulting from an over-large brk.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: cleanup
Move the symbols delimiting a section part of the section
(section relative) rather than absolute. This avoids any
unexpected gaps between the section-start symbol and the first
data in the section, which could be caused by implicit
alignment of the section data. It also makes the general
form of vmlinux_64.lds.S consistent with vmlinux_32.lds.S.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
there should be no difference, except:
* the 64bit variant now also initializes the padlock unit.
* ->c_early_init() is executed again from ->c_init()
* the 64bit fixups made into 32bit path.
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Cc: herbert@gondor.apana.org.au
LKML-Reference: <1237029843-28076-2-git-send-email-sebastian@breakpoint.cc>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: print out fewer lines
1. print continuous range with same type together
2. change _INFO to _DEBUG
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <49BACB61.8000302@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix debug warning
Jaswinder noticed that there is a warning about smp_processor_id()
in get_mtrr().
Fix it by wrapping the printout into a get/put_cpu() pair.
Reported-by: Jaswinder Singh Rajput <jaswinder@kernel.org>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
LKML-Reference: <49BAB7FF.4030107@kernel.org>
[ changed to get/put_cpu(), cleaned up surrounding code a it. ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: enhance e820 code to handle more cases
Try to handle new range which could be covered by one entry.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: jbeulich@novell.com
LKML-Reference: <49B9F0C1.10402@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Complete/fix the cleanups of cpu/common.c:
- fix ugly warning due to asm/topology.h -> linux/topology.h change
- standardize the style across the file
- simplify/refactor the code flow where possible
Cc: Jaswinder Singh Rajput <jaswinder@kernel.org>
LKML-Reference: <1237009789.4387.2.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix:
arch/x86/kernel/entry_32.S:446: Warning: 00000000080001d1 shortened to 00000000000001d1
arch/x86/kernel/entry_32.S:457: Warning: 000000000800feff shortened to 000000000000feff
arch/x86/kernel/entry_32.S:527: Warning: 00000000080001d1 shortened to 00000000000001d1
arch/x86/kernel/entry_32.S:541: Warning: 000000000800feff shortened to 000000000000feff
arch/x86/kernel/entry_32.S:676: Warning: 0000000008000091 shortened to 0000000000000091
TIF_SYSCALL_FTRACE is 0x08000000 and until now we checked the
first 16 bits of the work mask - bit 27 falls outside of that.
Update the entry_32.S code to check the full 32-bit mask.
[ %cx => %ecx fix from Cyrill Gorcunov <gorcunov@gmail.com> ]
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: "H. Peter Anvin" <hpa@kernel.org>
LKML-Reference: <1237012693.18733.3.camel@ht.satnam>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: build fix
kernel/built-in.o: In function `ftrace_syscall_exit':
(.text+0x76667): undefined reference to `syscall_nr_to_meta'
ftrace.o is built:
obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o
obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += ftrace.o
But now a CONFIG_FTRACE_SYSCALLS dependency is needed too.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
LKML-Reference: <1236401580-5758-3-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: disable unused code
x86 is fully converted to flow handlers. No need to keep the
deprecated __do_IRQ() support active.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Impact: fix (CONFIG_MAXSMP=y only) boot crash
c032ef60d1 "cpumask: convert
node_to_cpumask_map[] to cpumask_var_t" didn't get this one
conversion. There was a compile warning, but I missed it.
Reported-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Mike Travis <travis@sgi.com>
LKML-Reference: <200903132342.42813.rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Commit c2724775ce put a statement
after return, which makes that statement unreachable.
Move that statement before return.
Signed-off-by: WANG Cong <xiyou.wangcong@gmail.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Markus Metzger <markus.t.metzger@intel.com>
LKML-Reference: <20090313075622.GB8933@hack>
Cc: <stable@kernel.org> # .29 only
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: bug fix + BIOS workaround
BIOS is expected to clear the SYSCFG[MtrrFixDramModEn] on AMD CPUs
after fixed MTRRs are configured.
Some BIOSes do not clear SYSCFG[MtrrFixDramModEn] on BP (and on APs).
This can lead to obfuscation in Linux when this bit is not cleared on
BP but cleared on APs. A consequence of this is that the saved
fixed-MTRR state (from BP) differs from the fixed-MTRRs of APs --
because RdDram/WrDram bits are read as zero when
SYSCFG[MtrrFixDramModEn] is cleared -- and Linux tries to sync
fixed-MTRR state from BP to AP. This implies that Linux sets
SYSCFG[MtrrFixDramEn] and activates those bits.
More important is that (some) systems change these bits in SMM when
ACPI is enabled. Hence it is racy if Linux modifies RdMem/WrMem bits,
too.
(1) The patch modifies an old fix from Bernhard Kaindl to get
suspend/resume working on some Acer Laptops. Bernhard's patch
tried to sync RdMem/WrMem bits of fixed MTRR registers and that
helped on those old Laptops. (Don't ask me why -- can't test it
myself). But this old problem was not the motivation for the
patch. (See http://lkml.org/lkml/2007/4/3/110)
(2) The more important effect is to fix issues on some more current systems.
On those systems Linux panics or just freezes, see
http://bugzilla.kernel.org/show_bug.cgi?id=11541
(and also duplicates of this bug:
http://bugzilla.kernel.org/show_bug.cgi?id=11737http://bugzilla.kernel.org/show_bug.cgi?id=11714)
The affected systems boot only using acpi=ht, acpi=off or
when the kernel is built with CONFIG_MTRR=n.
The acpi options prevent full enablement of ACPI. Obviously when
ACPI is enabled the BIOS/SMM modfies RdMem/WrMem bits. When
CONFIG_MTRR=y Linux also accesses and modifies those bits when it
needs to sync fixed-MTRRs across cores (Bernhard's fix, see (1)).
How do you synchronize that? You can't. As a consequence Linux
shouldn't touch those bits at all (Rationale are AMD's BKDGs which
recommend to clear the bit that makes RdMem/WrMem accessible).
This is the purpose of this patch. And (so far) this suffices to
fix (1) and (2).
I suggest not to touch RdDram/WrDram bits of fixed-MTRRs and
SYSCFG[MtrrFixDramEn] and to clear SYSCFG[MtrrFixDramModEn] as
suggested by AMD K8, and AMD family 10h/11h BKDGs.
BIOS is expected to do this anyway. This should avoid that
Linux and SMM tread on each other's toes ...
Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: trenn@suse.de
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090312163937.GH20716@alberich.amd.com>
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Provide the x86 trace callbacks to trace syscalls.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
LKML-Reference: <1236401580-5758-3-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix left range size on head
| commit 5c0e6f035d
| x86: fix code paths used by update_mptable
| Impact: fix crashes under Xen due to unrobust e820 code
fixes one e820 bug, but introduces another bug.
Need to update size for left range at first in case it is header.
also add __e820_add_region take more parameter.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: jbeulich@novell.com
LKML-Reference: <49B9E286.502@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
We are removing cpumask_t in favour of struct cpumask: mainly as a
marker of what code is now CONFIG_CPUMASK_OFFSTACK-safe.
The only non-trivial change here is vector_allocation_domain():
explicitly clear the mask and set the first word, rather than using
assignment.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: cleanup
It's not legal to do assignments into cpumask_var_t; they will soon be of
variable length.
So explicitly clear the mask and set the first word, rather than using
assignment.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: cleanup, remove cpumask from stack
summit_send_IPI_allbutself might as well call
default_send_IPI_mask_allbutself_logical(). Also change cpumask_t to
struct cpumask and &cpu_online_map to cpu_online_mask while here.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: reduce kernel memory usage when CONFIG_CPUMASK_OFFSTACK=y
Straightforward conversion: done for 32 and 64 bit kernels.
node_to_cpumask_map is now a cpumask_var_t array.
64-bit used to be a dynamic cpumask_t array, and 32-bit used to be a
static cpumask_t array.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: cleanup
We take the 64-bit code and use it on 32-bit as well. The new file
is called mm/numa.c.
In a minor cleanup, we use cpu_none_mask instead of declaring a local
cpu_mask_none.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: implement new API
We define arch_send_call_function_ipi_mask and generic kernel/smp.c
code creates arch_send_call_function_ipi() as a wrapper.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: reduce kernel memory usage when CONFIG_CPUMASK_OFFSTACK=y
Simple conversion of mce_device_initialized to cpumask_var_t. We don't
check the alloc_cpumask_var() return since it's boot-time only, and
the misc_register() in that same function isn't checked.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: reduce per-cpu size for CONFIG_CPUMASK_OFFSTACK=y
In most places it's cleaner to use the accessors cpu_sibling_mask()
and cpu_core_mask() wrappers which already exist.
I couldn't avoid cleaning up the access in oprofile, either.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: cleanup, reduce memory usage for CONFIG_CPUMASK_OFFSTACK=y
I *think* every path calls check_nmi_watchdog before using the
watchdog, so that's the right place for the initialization.
If that's wrong, we'll get a nice NULL-deref with
CONFIG_CPUMASK_OFFSTACK=y, and have uncovered another bug.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
