It is an optimization and a cleanup, and adds the following new
generic percpu methods:
percpu_read()
percpu_write()
percpu_add()
percpu_sub()
percpu_and()
percpu_or()
percpu_xor()
and implements support for them on x86. (other architectures will fall
back to a default implementation)
The advantage is that for example to read a local percpu variable,
instead of this sequence:
return __get_cpu_var(var);
ffffffff8102ca2b: 48 8b 14 fd 80 09 74 mov -0x7e8bf680(,%rdi,8),%rdx
ffffffff8102ca32: 81
ffffffff8102ca33: 48 c7 c0 d8 59 00 00 mov $0x59d8,%rax
ffffffff8102ca3a: 48 8b 04 10 mov (%rax,%rdx,1),%rax
We can get a single instruction by using the optimized variants:
return percpu_read(var);
ffffffff8102ca3f: 65 48 8b 05 91 8f fd mov %gs:0x7efd8f91(%rip),%rax
I also cleaned up the x86-specific APIs and made the x86 code use
these new generic percpu primitives.
tj: * fixed generic percpu_sub() definition as Roel Kluin pointed out
* added percpu_and() for completeness's sake
* made generic percpu ops atomic against preemption
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Tejun Heo <tj@kernel.org>
* 'cpus4096-for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (66 commits)
x86: export vector_used_by_percpu_irq
x86: use logical apicid in x2apic_cluster's x2apic_cpu_mask_to_apicid_and()
sched: nominate preferred wakeup cpu, fix
x86: fix lguest used_vectors breakage, -v2
x86: fix warning in arch/x86/kernel/io_apic.c
sched: fix warning in kernel/sched.c
sched: move test_sd_parent() to an SMP section of sched.h
sched: add SD_BALANCE_NEWIDLE at MC and CPU level for sched_mc>0
sched: activate active load balancing in new idle cpus
sched: bias task wakeups to preferred semi-idle packages
sched: nominate preferred wakeup cpu
sched: favour lower logical cpu number for sched_mc balance
sched: framework for sched_mc/smt_power_savings=N
sched: convert BALANCE_FOR_xx_POWER to inline functions
x86: use possible_cpus=NUM to extend the possible cpus allowed
x86: fix cpu_mask_to_apicid_and to include cpu_online_mask
x86: update io_apic.c to the new cpumask code
x86: Introduce topology_core_cpumask()/topology_thread_cpumask()
x86: xen: use smp_call_function_many()
x86: use work_on_cpu in x86/kernel/cpu/mcheck/mce_amd_64.c
...
Fixed up trivial conflict in kernel/time/tick-sched.c manually
Impact: use new API, remove cpumask from stack.
Change smp_call_function_mask() callers to smp_call_function_many().
This removes a cpumask from the stack, and falls back should allocating
the cpumask var fail (only possible with CONFIG_CPUMASKS_OFFSTACK).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Cc: jeremy@xensource.com
* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86, xen: fix use of pgd_page now that it really does return a page
Xen requires that all mappings of pagetable pages are read-only, so
that they can't be updated illegally. As a result, if a page is being
turned into a pagetable page, we need to make sure all its mappings
are RO.
If the page had been used for ioremap or vmalloc, it may still have
left over mappings as a result of not having been lazily unmapped.
This change makes sure we explicitly mop them all up before pinning
the page.
Unlike aliases created by kmap, the there can be vmalloc aliases even
for non-high pages, so we must do the flush unconditionally.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Linux Memory Management List <linux-mm@kvack.org>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix 32-bit Xen guest boot crash
On 32-bit PAE, pud_page, for no good reason, didn't really return a
struct page *. Since Jan Beulich's fix "i386/PAE: fix pud_page()",
pud_page does return a struct page *.
Because PAE has 3 pagetable levels, the pud level is folded into the
pgd level, so pgd_page() is the same as pud_page(), and now returns
a struct page *. Update the xen/mmu.c code which uses pgd_page()
accordingly.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix guest kernel boot crash on certain configs
Recent i686 2.6.27 kernels with a certain amount of memory (between
736 and 855MB) have a problem booting under a hypervisor that supports
batched mprotect (this includes the RHEL-5 Xen hypervisor as well as
any 3.3 or later Xen hypervisor).
The problem ends up being that xen_ptep_modify_prot_commit() is using
virt_to_machine to calculate which pfn to update. However, this only
works for pages that are in the p2m list, and the pages coming from
change_pte_range() in mm/mprotect.c are kmap_atomic pages. Because of
this, we can run into the situation where the lookup in the p2m table
returns an INVALID_MFN, which we then try to pass to the hypervisor,
which then (correctly) denies the request to a totally bogus pfn.
The right thing to do is to use arbitrary_virt_to_machine, so that we
can be sure we are modifying the right pfn. This unfortunately
introduces a performance penalty because of a full page-table-walk,
but we can avoid that penalty for pages in the p2m list by checking if
virt_addr_valid is true, and if so, just doing the lookup in the p2m
table.
The attached patch implements this, and allows my 2.6.27 i686 based
guest with 768MB of memory to boot on a RHEL-5 hypervisor again.
Thanks to Jeremy for the suggestions about how to fix this particular
issue.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Chris Lalancette <clalance@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Rewrite the vmap allocator to use rbtrees and lazy tlb flushing, and
provide a fast, scalable percpu frontend for small vmaps (requires a
slightly different API, though).
The biggest problem with vmap is actually vunmap. Presently this requires
a global kernel TLB flush, which on most architectures is a broadcast IPI
to all CPUs to flush the cache. This is all done under a global lock. As
the number of CPUs increases, so will the number of vunmaps a scaled
workload will want to perform, and so will the cost of a global TLB flush.
This gives terrible quadratic scalability characteristics.
Another problem is that the entire vmap subsystem works under a single
lock. It is a rwlock, but it is actually taken for write in all the fast
paths, and the read locking would likely never be run concurrently anyway,
so it's just pointless.
This is a rewrite of vmap subsystem to solve those problems. The existing
vmalloc API is implemented on top of the rewritten subsystem.
The TLB flushing problem is solved by using lazy TLB unmapping. vmap
addresses do not have to be flushed immediately when they are vunmapped,
because the kernel will not reuse them again (would be a use-after-free)
until they are reallocated. So the addresses aren't allocated again until
a subsequent TLB flush. A single TLB flush then can flush multiple
vunmaps from each CPU.
XEN and PAT and such do not like deferred TLB flushing because they can't
always handle multiple aliasing virtual addresses to a physical address.
They now call vm_unmap_aliases() in order to flush any deferred mappings.
That call is very expensive (well, actually not a lot more expensive than
a single vunmap under the old scheme), however it should be OK if not
called too often.
The virtual memory extent information is stored in an rbtree rather than a
linked list to improve the algorithmic scalability.
There is a per-CPU allocator for small vmaps, which amortizes or avoids
global locking.
To use the per-CPU interface, the vm_map_ram / vm_unmap_ram interfaces
must be used in place of vmap and vunmap. Vmalloc does not use these
interfaces at the moment, so it will not be quite so scalable (although it
will use lazy TLB flushing).
As a quick test of performance, I ran a test that loops in the kernel,
linearly mapping then touching then unmapping 4 pages. Different numbers
of tests were run in parallel on an 4 core, 2 socket opteron. Results are
in nanoseconds per map+touch+unmap.
threads vanilla vmap rewrite
1 14700 2900
2 33600 3000
4 49500 2800
8 70631 2900
So with a 8 cores, the rewritten version is already 25x faster.
In a slightly more realistic test (although with an older and less
scalable version of the patch), I ripped the not-very-good vunmap batching
code out of XFS, and implemented the large buffer mapping with vm_map_ram
and vm_unmap_ram... along with a couple of other tricks, I was able to
speed up a large directory workload by 20x on a 64 CPU system. I believe
vmap/vunmap is actually sped up a lot more than 20x on such a system, but
I'm running into other locks now. vmap is pretty well blown off the
profiles.
Before:
1352059 total 0.1401
798784 _write_lock 8320.6667 <- vmlist_lock
529313 default_idle 1181.5022
15242 smp_call_function 15.8771 <- vmap tlb flushing
2472 __get_vm_area_node 1.9312 <- vmap
1762 remove_vm_area 4.5885 <- vunmap
316 map_vm_area 0.2297 <- vmap
312 kfree 0.1950
300 _spin_lock 3.1250
252 sn_send_IPI_phys 0.4375 <- tlb flushing
238 vmap 0.8264 <- vmap
216 find_lock_page 0.5192
196 find_next_bit 0.3603
136 sn2_send_IPI 0.2024
130 pio_phys_write_mmr 2.0312
118 unmap_kernel_range 0.1229
After:
78406 total 0.0081
40053 default_idle 89.4040
33576 ia64_spinlock_contention 349.7500
1650 _spin_lock 17.1875
319 __reg_op 0.5538
281 _atomic_dec_and_lock 1.0977
153 mutex_unlock 1.5938
123 iget_locked 0.1671
117 xfs_dir_lookup 0.1662
117 dput 0.1406
114 xfs_iget_core 0.0268
92 xfs_da_hashname 0.1917
75 d_alloc 0.0670
68 vmap_page_range 0.0462 <- vmap
58 kmem_cache_alloc 0.0604
57 memset 0.0540
52 rb_next 0.1625
50 __copy_user 0.0208
49 bitmap_find_free_region 0.2188 <- vmap
46 ia64_sn_udelay 0.1106
45 find_inode_fast 0.1406
42 memcmp 0.2188
42 finish_task_switch 0.1094
42 __d_lookup 0.0410
40 radix_tree_lookup_slot 0.1250
37 _spin_unlock_irqrestore 0.3854
36 xfs_bmapi 0.0050
36 kmem_cache_free 0.0256
35 xfs_vn_getattr 0.0322
34 radix_tree_lookup 0.1062
33 __link_path_walk 0.0035
31 xfs_da_do_buf 0.0091
30 _xfs_buf_find 0.0204
28 find_get_page 0.0875
27 xfs_iread 0.0241
27 __strncpy_from_user 0.2812
26 _xfs_buf_initialize 0.0406
24 _xfs_buf_lookup_pages 0.0179
24 vunmap_page_range 0.0250 <- vunmap
23 find_lock_page 0.0799
22 vm_map_ram 0.0087 <- vmap
20 kfree 0.0125
19 put_page 0.0330
18 __kmalloc 0.0176
17 xfs_da_node_lookup_int 0.0086
17 _read_lock 0.0885
17 page_waitqueue 0.0664
vmap has gone from being the top 5 on the profiles and flushing the crap
out of all TLBs, to using less than 1% of kernel time.
[akpm@linux-foundation.org: cleanups, section fix]
[akpm@linux-foundation.org: fix build on alpha]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When pinning/unpinning a pagetable with split pte locks, we can end up
holding multiple pte locks at once (we need to hold the locks while
there's a pending batched hypercall affecting the pte page). Because
all the pte locks are in the same lock class, lockdep thinks that
we're potentially taking a lock recursively.
This warning is spurious because we always take the pte locks while
holding mm->page_table_lock. lockdep now has spin_lock_nest_lock to
express this kind of dominant lock use, so use it here so that lockdep
knows what's going on.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Define USE_SPLIT_PTLOCKS as a constant expression rather than repeating
"NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS" all over the place.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Add support for exporting statistics on mmu updates, multicall
batching and pv spinlocks into debugfs. The base path is xen/ and
each subsystem adds its own directory: mmu, multicalls, spinlocks.
In each directory, writing 1 to "zero_stats" will cause the
corresponding stats to be zeroed the next time they're updated.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
It's easier to pattern match on Xen function if they all start with xen_.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Add some comments explaining the locking and pinning algorithm when
using split pte locks. Also implement a minor optimisation of not
pinning the PTE when not using split pte locks.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Xen-devel <xen-devel@lists.xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
PTE_PFN_MASK was getting lonely, so I made it a friend.
Signed-off-by: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Rusty, in his peevish way, complained that macros defining constants
should have a name which somewhat accurately reflects the actual
purpose of the constant.
Aside from the fact that PTE_MASK gives no clue as to what's actually
being masked, and is misleadingly similar to the functionally entirely
different PMD_MASK, PUD_MASK and PGD_MASK, I don't really see what the
problem is.
But if this patch silences the incessent noise, then it will have
achieved its goal (TODO: write test-case).
Signed-off-by: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Because the x86_64 architecture does not enforce segment limits, Xen
cannot protect itself with them as it does in 32-bit mode. Therefore,
to protect itself, it runs the guest kernel in ring 3. Since it also
runs the guest userspace in ring3, the guest kernel must maintain a
second pagetable for its userspace, which does not map kernel space.
Naturally, the guest kernel pagetables map both kernel and userspace.
The userspace pagetable is attached to the corresponding kernel
pagetable via the pgd's page->private field. It is allocated and
freed at the same time as the kernel pgd via the
paravirt_pgd_alloc/free hooks.
Fortunately, the user pagetable is almost entirely shared with the
kernel pagetable; the only difference is the pgd page itself. set_pgd
will populate all entries in the kernel pagetable, and also set the
corresponding user pgd entry if the address is less than
STACK_TOP_MAX.
The user pagetable must be pinned and unpinned with the kernel one,
but because the pagetables are aliased, pgd_walk() only needs to be
called on the kernel pagetable. The user pgd page is then
pinned/unpinned along with the kernel pgd page.
xen_write_cr3 must write both the kernel and user cr3s.
The init_mm.pgd pagetable never has a user pagetable allocated for it,
because it can never be used while running usermode.
One awkward area is that early in boot the page structures are not
available. No user pagetable can exist at that point, but it
complicates the logic to avoid looking at the page structure.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Rewrite pgd_walk to deal with 64-bit address spaces. There are two
notible features of 64-bit workspaces:
1. The physical address is only 48 bits wide, with the upper 16 bits
being sign extension; kernel addresses are negative, and userspace is
positive.
2. The Xen hypervisor mapping is at the negative-most address, just above
the sign-extension hole.
1. means that we can't easily use addresses when traversing the space,
since we must deal with sign extension. This rewrite expresses
everything in terms of pgd/pud/pmd indices, which means we don't need
to worry about the exact configuration of the virtual memory space.
This approach works equally well in 32-bit.
To deal with 2, assume the hole is between the uppermost userspace
address and PAGE_OFFSET. For 64-bit this skips the Xen mapping hole.
For 32-bit, the hole is zero-sized.
In all cases, the uppermost kernel address is FIXADDR_TOP.
A side-effect of this patch is that the upper boundary is actually
handled properly, exposing a long-standing bug in 32-bit, which failed
to pin kernel pmd page. The kernel pmd is not shared, and so must be
explicitly pinned, even though the kernel ptes are shared and don't
need pinning.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Make Xen's set_pte_mfn() use set_pte_vaddr rather than copying it.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When building initial pagetables in 64-bit kernel the pud/pmd pointer may
be in ioremap/fixmap space, so we need to walk the pagetable to look up the
physical address.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
arbitrary_virt_to_machine can truncate a machine address if its above
4G. Cast the problem away.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
x86_64 stores the active_mm in the pda, so fetch it from there.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We need extra pv_mmu_ops for 64-bit, to deal with the extra level of
pagetable.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Update arch/x86's use of page-aligned variables. The change to
arch/x86/xen/mmu.c fixes an actual bug, but the rest are cleanups
and to set a precedent.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When converting the page number in a pte/pmd/pud/pgd between
machine and pseudo-physical addresses, the converted result was
being truncated at 32-bits. This caused failures on machines
with more than 4G of physical memory.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: "Christopher S. Aker" <caker@theshore.net>
Cc: Ian Campbell <Ian.Campbell@eu.citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This converts x86, x86-64, and xen to use the new helpers for
smp_call_function() and friends, and adds support for
smp_call_function_single().
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Some Xen hypercalls accept an array of operations to work on. In
general this is because its more efficient for the hypercall to the
work all at once rather than as separate hypercalls (even batched as a
multicall).
This patch adds a mechanism (xen_mc_extend_args()) to allocate more
argument space to the last-issued multicall, in order to extend its
argument list.
The user of this mechanism is xen/mmu.c, which uses it to extend the
args array of mmu_update. This is particularly valuable when doing
the update for a large mprotect, which goes via
ptep_modify_prot_commit(), but it also manages to batch updates to
pgd/pmds as well.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Xen has a pte update function which will update a pte while preserving
its accessed and dirty bits. This means that ptep_modify_prot_start() can be
implemented as a simple read of the pte value. The hardware may
update the pte in the meantime, but ptep_modify_prot_commit() updates it while
preserving any changes that may have happened in the meantime.
The updates in ptep_modify_prot_commit() are batched if we're currently in lazy
mmu mode.
The mmu_update hypercall can take a batch of updates to perform, but
this code doesn't make particular use of that feature, in favour of
using generic multicall batching to get them all into the hypervisor.
The net effect of this is that each mprotect pte update turns from two
expensive trap-and-emulate faults into they hypervisor into a single
hypercall whose cost is amortized in a batched multicall.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Non-PAE operation has been deprecated in Xen for a while, and is
rarely tested or used. xen-unstable has now officially dropped
non-PAE support. Since Xen/pvops' non-PAE support has also been
broken for a while, we may as well completely drop it altogether.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Because NX is now enforced properly, we must put the hypercall page
into the .text segment so that it is executable.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stable Kernel <stable@kernel.org>
Cc: the arch/x86 maintainers <x86@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
[ Stable: this isn't a bugfix in itself, but it's a pre-requiste
for "xen: don't drop NX bit" ]
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stable Kernel <stable@kernel.org>
Cc: the arch/x86 maintainers <x86@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Because NX is now enforced properly, we must put the hypercall page
into the .text segment so that it is executable.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stable Kernel <stable@kernel.org>
Cc: the arch/x86 maintainers <x86@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
[ Stable: this isn't a bugfix in itself, but it's a pre-requiste
for "xen: don't drop NX bit" ]
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stable Kernel <stable@kernel.org>
Cc: the arch/x86 maintainers <x86@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When operating on an unpinned pagetable (ie, one under construction or
destruction), it isn't necessary to use a hypercall to update a
pud/pmd entry. Jan Beulich observed that a similar optimisation
avoided many thousands of hypercalls while doing a kernel build.
One tricky part is that early in the kernel boot there's no page
structure, so we can't check to see if the page is pinned. In that
case, we just always use the hypercall.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
-tip tree auto-testing found the following early bootup hang:
-------------->
get_memcfg_from_srat: assigning address to rsdp
RSD PTR v0 [Nvidia]
BUG: Int 14: CR2 ffd00040
EDI 8092fbfe ESI ffd00040 EBP 80b0aee8 ESP 80b0aed0
EBX 000f76f0 EDX 0000000e ECX 00000003 EAX ffd00040
err 00000000 EIP 802c055a CS 00000060 flg 00010006
Stack: ffd00040 80bc78d0 80b0af6c 80b1dbfe 8093d8ba 00000008 80b42810 80b4ddb4
80b42842 00000000 80b0af1c 801079c8 808e724e 00000000 80b42871 802c0531
00000100 00000000 0003fff0 80b0af40 80129999 00040100 00040100 00000000
Pid: 0, comm: swapper Not tainted 2.6.26-rc4-sched-devel.git #570
[<802c055a>] ? strncmp+0x11/0x25
[<80b1dbfe>] ? get_memcfg_from_srat+0xb4/0x568
[<801079c8>] ? mcount_call+0x5/0x9
[<802c0531>] ? strcmp+0xa/0x22
[<80129999>] ? printk+0x38/0x3a
[<80129999>] ? printk+0x38/0x3a
[<8011b122>] ? memory_present+0x66/0x6f
[<80b216b4>] ? setup_memory+0x13/0x40c
[<80b16b47>] ? propagate_e820_map+0x80/0x97
[<80b1622a>] ? setup_arch+0x248/0x477
[<80129999>] ? printk+0x38/0x3a
[<80b11759>] ? start_kernel+0x6e/0x2eb
[<80b110fc>] ? i386_start_kernel+0xeb/0xf2
=======================
<------
with this config:
http://redhat.com/~mingo/misc/config-Wed_May_28_01_33_33_CEST_2008.bad
The thing is, the crash makes little sense at first sight. We crash on a
benign-looking printk. The code around it got changed in -tip but
checking those topic branches individually did not reproduce the bug.
Bisection led to this commit:
| d5edbc1f75 is first bad commit
| commit d5edbc1f75
| Author: Jeremy Fitzhardinge <jeremy@goop.org>
| Date: Mon May 26 23:31:22 2008 +0100
|
| xen: add p2m mfn_list_list
Which is somewhat surprising, as on native hardware Xen client side
should have little to no side-effects.
After some head scratching, it turns out the following happened:
randconfig enabled the following Xen options:
CONFIG_XEN=y
CONFIG_XEN_MAX_DOMAIN_MEMORY=8
# CONFIG_XEN_BLKDEV_FRONTEND is not set
# CONFIG_XEN_NETDEV_FRONTEND is not set
CONFIG_HVC_XEN=y
# CONFIG_XEN_BALLOON is not set
which activated this piece of code in arch/x86/xen/mmu.c:
> @@ -69,6 +69,13 @@
> __attribute__((section(".data.page_aligned"))) =
> { [ 0 ... TOP_ENTRIES - 1] = &p2m_missing[0] };
>
> +/* Arrays of p2m arrays expressed in mfns used for save/restore */
> +static unsigned long p2m_top_mfn[TOP_ENTRIES]
> + __attribute__((section(".bss.page_aligned")));
> +
> +static unsigned long p2m_top_mfn_list[TOP_ENTRIES / P2M_ENTRIES_PER_PAGE]
> + __attribute__((section(".bss.page_aligned")));
The problem is, you must only put variables into .bss.page_aligned that
have a _size_ that is _exactly_ page aligned. In this case the size of
p2m_top_mfn_list is not page aligned:
80b8d000 b p2m_top_mfn
80b8f000 b p2m_top_mfn_list
80b8f008 b softirq_stack
80b97008 b hardirq_stack
80b9f008 b bm_pte
So all subsequent variables get unaligned which, depending on luck,
breaks the kernel in various funny ways. In this case what killed the
kernel first was the misaligned bootmap pte page, resulting in that
creative crash above.
Anyway, this was a fun bug to track down :-)
I think the moral is that .bss.page_aligned is a dangerous construct in
its current form, and the symptoms of breakage are very non-trivial, so
i think we need build-time checks to make sure all symbols in
.bss.page_aligned are truly page aligned.
The Xen fix below gets the kernel booting again.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch implements Xen save/restore and migration.
Saving is triggered via xenbus, which is polled in
drivers/xen/manage.c. When a suspend request comes in, the kernel
prepares itself for saving by:
1 - Freeze all processes. This is primarily to prevent any
partially-completed pagetable updates from confusing the suspend
process. If CONFIG_PREEMPT isn't defined, then this isn't necessary.
2 - Suspend xenbus and other devices
3 - Stop_machine, to make sure all the other vcpus are quiescent. The
Xen tools require the domain to run its save off vcpu0.
4 - Within the stop_machine state, it pins any unpinned pgds (under
construction or destruction), performs canonicalizes various other
pieces of state (mostly converting mfns to pfns), and finally
5 - Suspend the domain
Restore reverses the steps used to save the domain, ending when all
the frozen processes are thawed.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
When saving a domain, the Xen tools need to remap all our mfns to
portable pfns. In order to remap our p2m table, it needs to know
where all its pages are, so maintain the references to the p2m table
for it to use.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
When using sparsemem and memory hotplug, the kernel's pseudo-physical
address space can be discontigious. Previously this was dealt with by
having the upper parts of the radix tree stubbed off. Unfortunately,
this is incompatible with save/restore, which requires a complete p2m
table.
The solution is to have a special distinguished all-invalid p2m leaf
page, which we can point all the hole areas at. This allows the tools
to see a complete p2m table, but it only costs a page for all memory
holes.
It also simplifies the code since it removes a few special cases.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Add a config option to set the max size of a Xen domain. This is used
to scale the size of the physical-to-machine array; it ends up using
around 1 page/GByte, so there's no reason to be very restrictive.
For a 32-bit guest, the default value of 8GB is probably sufficient;
there's not much point in giving a 32-bit machine much more memory
than that.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
We now support the use of memory hotplug, so the physical to machine
page mapping structure must be dynamic. This is implemented as a
two-level radix tree structure, which allows us to efficiently
incrementally allocate memory for the p2m table as new pages are
added.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
While I realize that the function isn't currently being used, I still
think an obvious mistake like this should be corrected.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Acked-by: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Non-PAE operation has been deprecated in Xen for a while, and is
rarely tested or used. xen-unstable has now officially dropped
non-PAE support. Since Xen/pvops' non-PAE support has also been
broken for a while, we may as well completely drop it altogether.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The usual pagetable locking protocol doesn't seem to apply to updates
to init_mm, so don't rely on preemption being disabled in xen_set_pte_at
on init_mm.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
We can fold the essentially common pte functions together now.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
pte_t always contains a "pte" field for the whole pte value, so make
use of it.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Adjust the definition of lookup_address to take an unsigned long
level argument. Adjust callers in xen/mmu.c that pass in a
dummy variable.
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
based on this patch from Andi Kleen:
| Subject: CPA: Return the page table level in lookup_address()
| From: Andi Kleen <ak@suse.de>
|
| Needed for the next change.
|
| And change all the callers.
and ported it to x86.git.
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Looks like a mismerge/misapply dropped one of the cases of pte flag
masking for Xen. Also, only mask the flags for present ptes.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The hypervisor doesn't allow PCD or PWT to be set on guest ptes, so
make sure they're masked out. Also, fix up some previous mispatching.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Make sure pte_t, whatever its definition, has a pte element with type
pteval_t. This allows common code to access it without needing to be
specifically parameterised on what pagetable mode we're compiling for.
For 32-bit, this means that pte_t becomes a union with "pte" and "{
pte_low, pte_high }" (PAE) or just "pte_low" (non-PAE).
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
_PAGE_PCD maps a page with caching disabled, which is typically used for
mapping harware registers. Xen never allows it to be set on a mapping, and
unprivileged guests never need it since they can't see the real underlying
hardware. However, some uncached mappings are made early when probing the
(non-existent) APIC, and its OK to mask off the PCD flag in these cases.
This became necessary because Xen started checking for this bit, rather
than silently masking it off.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When a pagetable is created, it is made globally visible in the rmap
prio tree before it is pinned via arch_dup_mmap(), and remains in the
rmap tree while it is unpinned with arch_exit_mmap().
This means that other CPUs may race with the pinning/unpinning
process, and see a pte between when it gets marked RO and actually
pinned, causing any pte updates to fail with write-protect faults.
As a result, all pte pages must be properly locked, and only unlocked
once the pinning/unpinning process has finished.
In order to avoid taking spinlocks for the whole pagetable - which may
overflow the PREEMPT_BITS portion of preempt counter - it locks and pins
each pte page individually, and then finally pins the whole pagetable.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Hugh Dickens <hugh@veritas.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andi Kleen <ak@suse.de>
Cc: Keir Fraser <keir@xensource.com>
Cc: Jan Beulich <jbeulich@novell.com>
When a pagetable is no longer in use, it must be unpinned so that its
pages can be freed. However, this is only possible if there are no
stray uses of the pagetable. The code currently deals with all the
usual cases, but there's a rare case where a vcpu is changing cr3, but
is doing so lazily, and the change hasn't actually happened by the time
the pagetable is unpinned, even though it appears to have been completed.
This change adds a second per-cpu cr3 variable - xen_current_cr3 -
which tracks the actual state of the vcpu cr3. It is only updated once
the actual hypercall to set cr3 has been completed. Other processors
wishing to unpin a pagetable can check other vcpu's xen_current_cr3
values to see if any cross-cpu IPIs are needed to clean things up.
[ Stable folks: 2.6.23 bugfix ]
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Stable Kernel <stable@kernel.org>
This patch cleans up duplicate includes in
arch/i386/xen/
Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Currently, the set_lazy_mode pv_op is overloaded with 5 functions:
1. enter lazy cpu mode
2. leave lazy cpu mode
3. enter lazy mmu mode
4. leave lazy mmu mode
5. flush pending batched operations
This complicates each paravirt backend, since it needs to deal with
all the possible state transitions, handling flushing, etc. In
particular, flushing is quite distinct from the other 4 functions, and
seems to just cause complication.
This patch removes the set_lazy_mode operation, and adds "enter" and
"leave" lazy mode operations on mmu_ops and cpu_ops. All the logic
associated with enter and leaving lazy states is now in common code
(basically BUG_ONs to make sure that no mode is current when entering
a lazy mode, and make sure that the mode is current when leaving).
Also, flush is handled in a common way, by simply leaving and
re-entering the lazy mode.
The result is that the Xen, lguest and VMI lazy mode implementations
are much simpler.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Zach Amsden <zach@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Avi Kivity <avi@qumranet.com>
Cc: Anthony Liguory <aliguori@us.ibm.com>
Cc: "Glauber de Oliveira Costa" <glommer@gmail.com>
Cc: Jun Nakajima <jun.nakajima@intel.com>
Brian Gerst