This version requires that host and guest have the same PAE status.
NX cap is not offered to the guest, yet.
Signed-off-by: Matias Zabaljauregui <zabaljauregui@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Add support for kvm_hypercall4(); PAE wants it.
Signed-off-by: Matias Zabaljauregui <zabaljauregui at gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
replace LHCALL_SET_PMD with LHCALL_SET_PGD hypercall name
(That's really what it is, and the confusion gets worse with PAE support)
Signed-off-by: Matias Zabaljauregui <zabaljauregui@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Reported-by: Jeremy Fitzhardinge <jeremy@goop.org>
Some cleanups and replace direct assignment with native_set_* macros which properly handle 64-bit entries when PAE is activated
Signed-off-by: Matias Zabaljauregui <zabaljauregui@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The downside of the last patch which made restore_flags and irq_enable
check interrupts is that they are now too big to be patched directly
into the callsites, so the C versions are always used.
But the C versions go via PV_CALLEE_SAVE_REGS_THUNK which saves all
the registers. In fact, we don't need any registers in the fast path,
so we can do better than this if we actually code them in assembler.
The results are in the noise, but since it's about the same amount of
code, it's worth applying.
1GB Guest->Host: input(suppressed),output(suppressed)
Before:
Seconds: 0:16.53
Packets: 377268,753673
Interrupts: 22461,24297
Notifications: 1(5245),21303(732370)
Net IRQs triggered: 377023(245),42578(711095)
After:
Seconds: 0:16.48
Packets: 377289,753673
Interrupts: 22281,24465
Notifications: 1(5245),21296(732377)
Net IRQs triggered: 377060(229),42564(711109)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
lguest never checked for pending interrupts when enabling interrupts, and
things still worked. However, it makes a significant difference to TCP
performance, so it's time we fixed it by introducing a pending_irq flag
and checking it on irq_restore and irq_enable.
These two routines are now too big to patch into the 8/10 bytes
patch space, so we drop that code.
Note: The high latency on interrupt delivery had a very curious
effect: once everything else was optimized, networking without GSO was
faster than networking with GSO, since more interrupts were sent and
hence a greater chance of one getting through to the Guest!
Note2: (Almost) Closing the same loophole for iret doesn't have any
measurable effect, so I'm leaving that patch for the moment.
Before:
1GB tcpblast Guest->Host: 30.7 seconds
1GB tcpblast Guest->Host (no GSO): 76.0 seconds
After:
1GB tcpblast Guest->Host: 6.8 seconds
1GB tcpblast Guest->Host (no GSO): 27.8 seconds
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Copy from arch/x86/kernel/irqinit_32.c: we don't use the vectors beyond
LGUEST_IRQS (if any), but we might as well set them all.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We don't set up the canary; let's disable stack protector on boot.c so
we can get into lguest_init, then set it up. As a side effect,
switch_to_new_gdt() sets up %fs for us properly too.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This simplifies the node awareness of the code. All our allocators
only deal with a NUMA node ID locality not with CPU ids anyway - so
there's no need to maintain (and transform) a CPU id all across the
IRq layer.
v2: keep move_irq_desc related
[ Impact: cleanup, prepare IRQ code to be NUMA-aware ]
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
LKML-Reference: <49F65536.2020300@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Pass clocksource pointer to the read() callback for clocksources. This
allows us to share the callback between multiple instances.
[hugh@veritas.com: fix powerpc build of clocksource pass clocksource mods]
[akpm@linux-foundation.org: cleanup]
Signed-off-by: Magnus Damm <damm@igel.co.jp>
Acked-by: John Stultz <johnstul@us.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fixes guest crash 'lguest: bad read address 0x4800000 len 256'
The new per-cpu allocator ends up handing a non-linear address to
write_gdt_entry. We do __pa() on it, and hand it to the host, which
kills us.
I've long wanted to make the hypercall "LOAD_GDT_ENTRY" to match the IDT
code, but had no pressing reason until now.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: lguest@ozlabs.org
Duplicate hcall -> kvm_hypercall0 convertion from "lguest: use KVM
hypercalls".
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Matias Zabaljauregui <zabaljauregui at gmail.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Impact: cleanup
This patch allow us to use KVM hypercalls
Signed-off-by: Matias Zabaljauregui <zabaljauregui at gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: intermittent guest segv/crash fix
I've been seeing random guest bad address crashes and segmentation faults:
bisect led to 4f98a2fee8 (vmscan: split LRU lists into anon & file sets),
but that's a red herring.
It turns out that lguest never hooked up the pte_update/pte_update_defer
calls, so our ptes were not always in sync. After the vmscan commit, the
bug became reproducible; now a fsck in a 64MB guest causes reproducible
pagetable corruption.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: jeremy@xensource.com
Cc: virtualization@lists.osdl.org
Cc: stable@kernel.org
Impact: fix lazy context switch API
Pass the previous and next tasks into the context switch start
end calls, so that the called functions can properly access the
task state (esp in end_context_switch, in which the next task
is not yet completely current).
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Impact: allow preemption during lazy mmu updates
If we're in lazy mmu mode when context switching, leave
lazy mmu mode, but remember the task's state in
TIF_LAZY_MMU_UPDATES. When we resume the task, check this
flag and re-enter lazy mmu mode if its set.
This sets things up for allowing lazy mmu mode while preemptible,
though that won't actually be active until the next change.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
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: remove lots of lguest boot WARN_ON() when CONFIG_SPARSE_IRQ=y
We now need to call irq_to_desc_alloc_cpu() before
set_irq_chip_and_handler_name(), but we can't do that from init_IRQ (no
kmalloc available).
So do it as we use interrupts instead. Also means we only alloc for
irqs we use, which was the intent of CONFIG_SPARSE_IRQ anyway.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Ingo Molnar <mingo@redhat.com>
Impact: fix lguest boot crash on modern Intel machines
The code in early_init_intel does:
if (c->x86 > 6 || (c->x86 == 6 && c->x86_model >= 0xd)) {
u64 misc_enable;
rdmsrl(MSR_IA32_MISC_ENABLE, misc_enable);
And that rdmsr faults (not allowed from non-0 PL). We can get around
this by mugging the family ID part of the cpuid. 5 seems like a good
number.
Of course, this is a hack (how very lguest!). We could just indicate
that we don't support MSRs, or implement lguest_rdmst.
Reported-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Tested-by: Patrick McHardy <kaber@trash.net>
Impact: remove unused/broken code
The Voyager subarch last built successfully on the v2.6.26 kernel
and has been stale since then and does not build on the v2.6.27,
v2.6.28 and v2.6.29-rc5 kernels.
No actual users beyond the maintainer reported this breakage.
Patches were sent and most of the fixes were accepted but the
discussion around how to do a few remaining issues cleanly
fizzled out with no resolution and the code remained broken.
In the v2.6.30 x86 tree development cycle 32-bit subarch support
has been reworked and removed - and the Voyager code, beyond the
build problems already known, needs serious and significant
changes and probably a rewrite to support it.
CONFIG_X86_VOYAGER has been marked BROKEN then. The maintainer has
been notified but no patches have been sent so far to fix it.
While all other subarchs have been converted to the new scheme,
voyager is still broken. We'd prefer to receive patches which
clean up the current situation in a constructive way, but even in
case of removal there is no obstacle to add that support back
after the issues have been sorted out in a mutually acceptable
fashion.
So remove this inactive code for now.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
make it simpler, don't need have one extra struct.
v2: fix the sgi_uv build
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: pt_regs changed, lazy gs handling made optional, add slight
overhead to SAVE_ALL, simplifies error_code path a bit
On x86_32, %gs hasn't been used by kernel and handled lazily. pt_regs
doesn't have place for it and gs is saved/loaded only when necessary.
In preparation for stack protector support, this patch makes lazy %gs
handling optional by doing the followings.
* Add CONFIG_X86_32_LAZY_GS and place for gs in pt_regs.
* Save and restore %gs along with other registers in entry_32.S unless
LAZY_GS. Note that this unfortunately adds "pushl $0" on SAVE_ALL
even when LAZY_GS. However, it adds no overhead to common exit path
and simplifies entry path with error code.
* Define different user_gs accessors depending on LAZY_GS and add
lazy_save_gs() and lazy_load_gs() which are noop if !LAZY_GS. The
lazy_*_gs() ops are used to save, load and clear %gs lazily.
* Define ELF_CORE_COPY_KERNEL_REGS() which always read %gs directly.
xen and lguest changes need to be verified.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: Optimization
One of the problems with inserting a pile of C calls where previously
there were none is that the register pressure is greatly increased.
The C calling convention says that the caller must expect a certain
set of registers may be trashed by the callee, and that the callee can
use those registers without restriction. This includes the function
argument registers, and several others.
This patch seeks to alleviate this pressure by introducing wrapper
thunks that will do the register saving/restoring, so that the
callsite doesn't need to worry about it, but the callee function can
be conventional compiler-generated code. In many cases (particularly
performance-sensitive cases) the callee will be in assembler anyway,
and need not use the compiler's calling convention.
Standard calling convention is:
arguments return scratch
x86-32 eax edx ecx eax ?
x86-64 rdi rsi rdx rcx rax r8 r9 r10 r11
The thunk preserves all argument and scratch registers. The return
register is not preserved, and is available as a scratch register for
unwrapped callee code (and of course the return value).
Wrapped function pointers are themselves wrapped in a struct
paravirt_callee_save structure, in order to get some warning from the
compiler when functions with mismatched calling conventions are used.
The most common paravirt ops, both statically and dynamically, are
interrupt enable/disable/save/restore, so handle them first. This is
particularly easy since their calls are handled specially anyway.
XXX Deal with VMI. What's their calling convention?
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
This patch moves the initial guest page table creation code to the host,
so the launcher keeps working with PAE enabled configs.
Signed-off-by: Matias Zabaljauregui <zabaljauregui@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: change calling convention of existing clock_event APIs
struct clock_event_timer's cpumask field gets changed to take pointer,
as does the ->broadcast function.
Another single-patch change. For safety, we BUG_ON() in
clockevents_register_device() if it's not set.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Ingo Molnar <mingo@elte.hu>
Don't generate interrupt stubs for interrupt vectors below
FIRST_EXTERNAL_VECTOR, and make the table of interrupt vectors
(interrupt[]) __initconst. Both of these changes both conserve memory
and improve consistency with 64 bits.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
do_IRQ: cannot handle IRQ -1 vector 0x20 cpu 0
------------[ cut here ]------------
kernel BUG at arch/x86/kernel/irq_32.c:219!
We're not ISA: we have a 1:1 mapping from vectors to irqs.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
dmi_scan_machine breaks under lguest:
lguest: unhandled trap 14 at 0xc04edeae (0xffa00000)
This is because we use current_cr3 for the read_cr3() paravirt
function, and it isn't set until the first cr3 change. We got away
with it until this happened.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
so we can merge io_apic_32.c and io_apic_64.c
v2: Use cpu_online_map as target cpus for bigsmp, just like 64-bit is doing.
Also remove some unused TARGET_CPUS macro.
v3: need to check if desc is null in smp_irq_move_cleanup
also migration needs to reset vector too, so copy __target_IO_APIC_irq
from 64bit.
(the duplication will go away once the two files are unified.)
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
6af61a7614 'x86: clean up max_pfn_mapped
usage - 32-bit' makes the following comment:
XEN PV and lguest may need to assign max_pfn_mapped too.
But no CC. Yinghai, wasting fellow developers' time is a VERY bad
habit. If you do it again, I will hunt you down and try to extract
the three hours of my life I just lost :)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
fix:
arch/x86/lguest/boot.c:816: error: variable ‘lguest_basic_apic_ops’ has initializer but incomplete type
arch/x86/lguest/boot.c:817: error: unknown field ‘read’ specified in initializer
[...]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Use alternatives to select the workaround for the 11AP Pentium erratum
for the affected steppings on the fly rather than build time. Remove the
X86_GOOD_APIC configuration option and replace all the calls to
apic_write_around() with plain apic_write(), protecting accesses to the
ESR as appropriate due to the 3AP Pentium erratum. Remove
apic_read_around() and all its invocations altogether as not needed.
Remove apic_write_atomic() and all its implementing backends. The use of
ASM_OUTPUT2() is not strictly needed for input constraints, but I have
used it for readability's sake.
I had the feeling no one else was brave enough to do it, so I went ahead
and here it is. Verified by checking the generated assembly and tested
with both a 32-bit and a 64-bit configuration, also with the 11AP
"feature" forced on and verified with gdb on /proc/kcore to work as
expected (as an 11AP machines are quite hard to get hands on these days).
Some script complained about the use of "volatile", but apic_write() needs
it for the same reason and is effectively a replacement for writel(), so I
have disregarded it.
I am not sure what the policy wrt defconfig files is, they are generated
and there is risk of a conflict resulting from an unrelated change, so I
have left changes to them out. The option will get removed from them at
the next run.
Some testing with machines other than mine will be needed to avoid some
stupid mistake, but despite its volume, the change is not really that
intrusive, so I am fairly confident that because it works for me, it will
everywhere.
Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Define the Xen specific basic apic ops, in additon to paravirt apic ops,
with some misc warning fixes.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: akpm@linux-foundation.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Rename the paravirtualized calculate_cpu_khz to calibrate_tsc.
In all cases, we actually calibrate_tsc and use that as the cpu_khz value.
Signed-off-by: Alok N Kataria <akataria@vmware.com>
Signed-off-by: Dan Hecht <dhecht@vmware.com>
Cc: Dan Hecht <dhecht@vmware.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
rename update_memory_range to e820_update_range
rename add_memory_region to e820_add_region
to make it more clear that they are about e820 map operations.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
introduce init_pg_table_start, so xen PV could specify the value.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Took some cycles to re-read the Lguest Journey end-to-end, fix some
rot and tighten some phrases.
Only comments change. No new jokes, but a couple of recycled old jokes.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Ahmed managed to crash the Host in release_pgd(), which cannot be a Guest
bug, and indeed it wasn't.
The bug was that handing a 0 as the address of the toplevel page table
being manipulated can cause the lookup code in find_pgdir() to return
an uninitialized cache entry (we shadow up to 4 top level page tables
for each Guest).
Commit 37cc8d7f96 introduced this
behaviour in the Guest, uncovering the bug.
The patch which he submitted (which removed the /4 from the index
calculation) simply ensured that these high-indexed entries hit the
early exit path of guest_set_pmd(). But you get lots of segfaults in
guest userspace as the PMDs aren't being updated.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Now the TSC code handles a zero return from calculate_cpu_khz(),
lguest can simply pass through the value it gets from the Host: if
non-zero, all the normal TSC code applies.
Otherwise (or if the Host really doesn't support TSC), the clocksource
code will fall back to the slower but reasonable lguest clock.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Hi all,
Beginning from commits close to v2.6.25-rc2, running lguest always oopses
the host kernel. Oops is at [1].
Bisection led to the following commit:
commit 37cc8d7f96
x86/early_ioremap: don't assume we're using swapper_pg_dir
At the early stages of boot, before the kernel pagetable has been
fully initialized, a Xen kernel will still be running off the
Xen-provided pagetables rather than swapper_pg_dir[]. Therefore,
readback cr3 to determine the base of the pagetable rather than
assuming swapper_pg_dir[].
static inline pmd_t * __init early_ioremap_pmd(unsigned long addr)
{
- pgd_t *pgd = &swapper_pg_dir[pgd_index(addr)];
+ /* Don't assume we're using swapper_pg_dir at this point */
+ pgd_t *base = __va(read_cr3());
+ pgd_t *pgd = &base[pgd_index(addr)];
pud_t *pud = pud_offset(pgd, addr);
pmd_t *pmd = pmd_offset(pud, addr);
Trying to analyze the problem, it seems on the guest side of lguest,
%cr3 has a different value from &swapper_pg-dir (which
is AFAIK fine on a pravirt guest):
Putting some debugging messages in early_ioremap_pmd:
/* Appears 3 times */
[ 0.000000] ***************************
[ 0.000000] __va(%cr3) = c0000000, &swapper_pg_dir = c02cc000
[ 0.000000] ***************************
After 8 hours of debugging and staring on lguest code, I noticed something
strange in paravirt_ops->set_pmd hypercall invocation:
static void lguest_set_pmd(pmd_t *pmdp, pmd_t pmdval)
{
*pmdp = pmdval;
lazy_hcall(LHCALL_SET_PMD, __pa(pmdp)&PAGE_MASK,
(__pa(pmdp)&(PAGE_SIZE-1))/4, 0);
}
The first hcall parameter is global pgdir which looks fine. The second
parameter is the pmd index in the pgdir which is suspectful.
AFAIK, calculating the index of pmd does not need a divisoin over four.
Removing the division made lguest work fine again . Patch is at [2].
I am not sure why the division over four existed in the first place. It
seems bogus, maybe the Xen patch just made the problem appear ?
[2]: The patch:
[PATCH] lguest: fix pgdir pmd index cacluation
Remove an error in index calculation which leads to removing
a not existing shadow page table (leading to a Null dereference).
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Added a declaration to asm-x86/lguest.h and moved the extern arrays there
as well. As an alternative to including asm/lguest.h directly, an
include could be put in linux/lguest.h
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: "rusty@rustcorp.com.au" <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
With this, the paravirt_ops code can be enabled on x86_64 also.
Each guest implementation (Xen, VMI, lguest) now depends on X86_32. The
dependencies can be dropped for each one when they start to support
x86_64.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This patch changes the write_gdt_entry function signature.
Instead of the old "a" and "b" parameters, it now receives
a pointer to a desc_struct, and the size of the entry being
handled. This is because x86_64 can have some 16-byte entries
as well as 8-byte ones.
Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
CC: Zachary Amsden <zach@vmware.com>
CC: Jeremy Fitzhardinge <Jeremy.Fitzhardinge.citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
this patch changes write_idt_entry signature. It now takes a gate_desc
instead of the a and b parameters. It will allow it to be later unified
between i386 and x86_64.
Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
CC: Zachary Amsden <zach@vmware.com>
CC: Jeremy Fitzhardinge <Jeremy.Fitzhardinge.citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Matias Zabaljauregui