Avi Kivity wrote:
Avi Kivity wrote:
Exactly. But it is better to be explicit about it and pass the page
directly like you did before. I hate to make you go back-and-fourth,
but I did not understand the issue completely before.
btw, the call to gfn_to_page() can happen
Andrew Morton wrote:
On Sun, 23 Dec 2007 10:59:22 +0200 Avi Kivity [EMAIL PROTECTED] wrote:
Avi Kivity wrote:
Avi Kivity wrote:
Exactly. But it is better to be explicit about it and pass the page
directly like you did before. I hate to make you go back-and-fourth
Dor Laor wrote:
This patchset updates kvm repository with Anthony's virtio
implementation along
with rx performance improvements and guest reset handling.
The original code was sent to qemu devel list 2 weeks ago.
It contains support for network block devices.
Using the performance
Dor Laor wrote:
From f244bcad756c4f761627557bb7f315b1d8f22fb2 Mon Sep 17 00:00:00 2001
From: Dor Laor [EMAIL PROTECTED]
Date: Thu, 20 Dec 2007 13:26:30 +0200
Subject: [PATCH] [VIRTIO-NET] Rx performance improvement
The current performance are not good enough, the problem lies
in qemu tap
Andrew Morton wrote:
On Sun, 23 Dec 2007 12:35:30 +0200 Avi Kivity [EMAIL PROTECTED] wrote:
Andrew Morton wrote:
On Sun, 23 Dec 2007 10:59:22 +0200 Avi Kivity [EMAIL PROTECTED] wrote:
Avi Kivity wrote:
Avi Kivity wrote:
Exactly
From: Anthony Liguori [EMAIL PROTECTED]
Add vmmcall/vmcall to x86_emulate. Future patch will implement functionality
for these instructions.
Signed-off-by: Anthony Liguori [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/x86_emulate.c | 23
Following is the first part of the 2.6.25 merge window submission. Since
there are 238 patches in the queue (and a few more expected), they'll be
sent in five batches of around 50 each.
-
This SF.net email is sponsored by:
.
[avi: fix #ud injection on vmx]
Signed-off-by: Anthony Liguori [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm.h |8 +-
drivers/kvm/kvm_main.c| 156 +---
drivers/kvm/svm.c | 19 +-
drivers
From: Laurent Vivier [EMAIL PROTECTED]
emulate_instruction() calls now x86_decode_insn() and x86_emulate_insn().
x86_emulate_insn() is x86_emulate_memop() without the decoding part.
Signed-off-by: Laurent Vivier [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm
an iso.
Signed-off-by: Ryan Harper [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm_main.c | 11 ---
1 files changed, 4 insertions(+), 7 deletions(-)
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index da057cf..b10fd7e 100644
When we allow guest page faults to reach the guests directly, we lose
the fault tracking which allows us to detect demand paging. So we provide
an alternate mechnism by clearing the accessed bit when we set a pte, and
checking it later to see if the guest actually used it.
Signed-off-by: Avi
-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm.h |8 -
drivers/kvm/kvm_main.c| 77 +---
drivers/kvm/svm.c |9 +++--
drivers/kvm/vmx.c |9 +++--
drivers/kvm/x86_emulate.c | 24 --
5 files
mode. This is because the syscall instruction is
not available in 32-bit mode on Intel processors, so the SCE bit is
effectively meaningless.
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/vmx.c | 61
1 files changed, 37
hardware, as AMD lacks a
way to discriminate among page faults based on error code. It is also
a little risky since it uses reserved bits which might become unreserved
in the future, so a module parameter is provided to disable it.
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm
From: Laurent Vivier [EMAIL PROTECTED]
Split the decoding process into a new function x86_decode_insn().
Signed-off-by: Laurent Vivier [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/x86_emulate.c | 77 +++--
1 files
From: Qing He [EMAIL PROTECTED]
Signed-off-by: Qing He [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/x86_emulate.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/kvm/x86_emulate.c b/drivers/kvm/x86_emulate.c
index 8b0186f..fe50317
[EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/x86_emulate.c | 24 +++-
1 files changed, 15 insertions(+), 9 deletions(-)
diff --git a/drivers/kvm/x86_emulate.c b/drivers/kvm/x86_emulate.c
index fe50317..e6b213b 100644
--- a/drivers/kvm
From: Izik Eidus [EMAIL PROTECTED](none)
Signed-off-by: Izik Eidus [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm.h |9 +++
drivers/kvm/kvm_main.c | 160 +++-
drivers/kvm/vmx.c | 43 ++---
3
From: Laurent Vivier [EMAIL PROTECTED]
The only valid case is on protected page access, other cases are errors.
Signed-off-by: Laurent Vivier [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm_main.c | 10 +++---
1 files changed, 7 insertions(+), 3 deletions
From: Laurent Vivier [EMAIL PROTECTED]
Remove _eflags and use directly ctxt-eflags. Caching eflags is not needed as
it is restored to vcpu by kvm_main.c:emulate_instruction() from ctxt-eflags
only if emulation doesn't fail.
Signed-off-by: Laurent Vivier [EMAIL PROTECTED]
Signed-off-by: Avi
Now that smp_call_function_single() knows how to call a function on the
current cpu, there's no need to check explicitly.
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/vmx.c |6 +-
1 files changed, 1 insertions(+), 5 deletions(-)
diff --git a/drivers/kvm/vmx.c b/drivers
;
};
Signed-off-by: Laurent Vivier [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/x86_emulate.c | 919 -
drivers/kvm/x86_emulate.h | 34 ++
2 files changed, 518 insertions(+), 435 deletions(-)
diff --git
From: Laurent Vivier [EMAIL PROTECTED]
Remove no_wb, use dst.type = OP_NONE instead, idea stollen from xen-3.1
Signed-off-by: Laurent Vivier [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/x86_emulate.c | 76 ++--
drivers
From: Laurent Vivier [EMAIL PROTECTED]
To improve readability, move push, writeback, and grp 1a/2/3/4/5/9 emulation
parts into functions.
Signed-off-by: Laurent Vivier [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/x86_emulate.c | 451
From: Rusty Russell [EMAIL PROTECTED]
Since vcpu-apic is of the correct type, there's not need to cast.
Signed-off-by: Rusty Russell [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/lapic.c | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
diff
This is more consistent with the accessed bit management, and makes the dirty
bit available earlier for other purposes.
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/mmu.c |5 +
drivers/kvm/paging_tmpl.h | 31 ---
2 files changed, 13
kvm_vcpu_init.
2) Don't set apic-regs_page to zero before freeing apic.
Signed-off-by: Rusty Russell [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/irq.h |2 +-
drivers/kvm/kvm_main.c |4 +---
drivers/kvm/lapic.c| 19 +--
3 files
-off-by: Rusty Russell [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm_main.c | 10 +-
drivers/kvm/svm.c |6 --
drivers/kvm/vmx.c |6 --
3 files changed, 9 insertions(+), 13 deletions(-)
diff --git a/drivers/kvm/kvm_main.c b
-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/lapic.c | 13 +++--
1 files changed, 3 insertions(+), 10 deletions(-)
diff --git a/drivers/kvm/lapic.c b/drivers/kvm/lapic.c
index e15b42e..8840f9d 100644
--- a/drivers/kvm/lapic.c
+++ b/drivers/kvm/lapic.c
@@ -395,10 +395,9 @@ static
Beside the obvious goodness of making code more common, this prevents
a livelock with the next patch which moves interrupt injection out of the
critical section.
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm_main.c |2 ++
drivers/kvm/svm.c |1 -
drivers/kvm/vmx.c
need to be
called on all VCPUs.
Signed-off-by: Anthony Liguori [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/mmu.c | 18 +-
drivers/kvm/paging_tmpl.h |4 ++--
2 files changed, 11 insertions(+), 11 deletions(-)
diff --git a/drivers/kvm
Since the mmu uses different shadow pages for dirty large pages and clean
large pages, this allows the mmu to drop ptes that are now invalid.
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/paging_tmpl.h |5 +
1 files changed, 5 insertions(+), 0 deletions(-)
diff --git
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/paging_tmpl.h | 23 +++
1 files changed, 15 insertions(+), 8 deletions(-)
diff --git a/drivers/kvm/paging_tmpl.h b/drivers/kvm/paging_tmpl.h
index 4538b15..a0f84a5 100644
--- a/drivers/kvm/paging_tmpl.h
+++ b
By forcing clean huge pages to be read-only, we have separate roles
for the shadow of a clean large page and the shadow of a dirty large
page. This is necessary because different ptes will be instantiated
for the two cases, even for read faults.
Signed-off-by: Avi Kivity [EMAIL PROTECTED
We must set the bit before the shift, otherwise the wrong bit gets set.
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/paging_tmpl.h |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/kvm/paging_tmpl.h b/drivers/kvm/paging_tmpl.h
index da36e48..e07cb2e
This is consistent with real-mode permissions.
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/mmu.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c
index e6616a6..f52604a 100644
--- a/drivers/kvm/mmu.c
+++ b/drivers/kvm
We now have a new namespace, KVM_REQ_*, for bits in vcpu-requests.
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm.h |2 +-
drivers/kvm/kvm_main.c |4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index
From: Nitin A Kamble [EMAIL PROTECTED]
Instructions:
inc r16/r32 (opcode 0x40-0x47)
dec r16/r32 (opcode 0x48-0x4f)
Signed-off-by: Nitin A Kamble [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/x86_emulate.c | 20 ++--
1 files
From: Laurent Vivier [EMAIL PROTECTED]
Add new-line at end of debug strings.
Signed-off-by: Laurent Vivier [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/ioapic.c | 25 ++---
1 files changed, 14 insertions(+), 11 deletions(-)
diff --git
From: Izik Eidus [EMAIL PROTECTED]
The user is now able to set how many mmu pages will be allocated to the guest.
Signed-off-by: Izik Eidus [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm.h |7 ++-
drivers/kvm/kvm_main.c | 47
the gfn of each gpte in
the shadow pages, since the memory slot is addressed by gfn, instead of
hfn like struct page.
Signed-off-by: Izik Eidus [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm.h |6 ++-
drivers/kvm/kvm_main.c| 11 +++-
drivers/kvm
PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm.h |9 ++
drivers/kvm/kvm_main.c | 200 ++--
drivers/kvm/x86.c | 219
3 files changed, 234 insertions(+), 194 deletions
From: Mike Day [EMAIL PROTECTED]
Signed-off-by: Mike D. Day [EMAIL PROTECTED]
Signed-off-by: Avi Kivity [EMAIL PROTECTED]
---
drivers/kvm/kvm.h | 32 +-
drivers/kvm/kvm_main.c| 58 ++
drivers/kvm/lapic.c |3 +-
drivers/kvm
Andrea Arcangeli wrote:
On Sat, Dec 22, 2007 at 09:13:44PM +0200, Avi Kivity wrote:
Unfortunately, this fails badly on Intel i386:
Hmm ok there's a definitive bug that I forgot a int1 kernel-kernel
switch on x86 has no special debug stack like on x86-64. This will
have a better
Avi Kivity wrote:
From: Izik Eidus [EMAIL PROTECTED]
Instead of having the kernel allocate memory to the guest, let userspace
allocate it and pass the address to the kernel.
This is required for s390 support, but also enables features like memory
sharing and using hugetlbfs backed memory
Marcelo Tosatti wrote:
On Sun, Dec 23, 2007 at 10:41:07AM +0200, Avi Kivity wrote:
Avi Kivity wrote:
Exactly. But it is better to be explicit about it and pass the page
directly like you did before. I hate to make you go back-and-fourth,
but I did not understand the issue
Marcelo Tosatti wrote:
It can acquire the pagetablelock in the callback handler. But then,
vm_normal_page() must also be exported.
Are you guys OK with this ?
Seems to me that requires fairly detailed mucking in mm details, just to
get at a page.
I believe that a new
avadh patel wrote:
Hi All,
I was looking into the source code of kvm and modified qemu and trying
to find the time spent between two VM-exit in kvm.
Is there a way to find this from user-space qemu, without modifying
the kernel module or libkvm?
What do you mean exactly? Average time
Andrea Arcangeli wrote:
[snip]
So in short with the below fix applied, after a write_tsc(0,0), the
UP guest never return any error anymore. Previously it would return
frequent errors because sched_in wasn't properly invoked by svm.c and
it would crash at boot every single time after a
Marcelo Tosatti wrote:
In preparation for a mmu spinlock, avoid schedules in mmu_set_spte()
by using follow_page() instead of get_user_pages().
The fault handling work by get_user_pages() now happens outside the lock.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index:
Marcelo Tosatti wrote:
follow_page() is required by KVM to find the struct page which maps
to a given address in spinlock protected code.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm.quilt/mm/memory.c
===
---
Marcelo Tosatti wrote:
The following patchset increases KVM shadow scalability by allowing concurrent
guest walking, allocation and instruction emulation.
Thanks; the split patchset it _much_ easier to review, and hopefully to
bisect in case the review misses something.
I believe we can
Carlo Marcelo Arenas Belon wrote:
The following patch series fixes compilation warnings from kvm 57 triggered
by 25a2e04e5689af63d104de2dea755f30bf643afd :
PATCH 1/9 : fix definition for vmchannel_init in qemu/vl.c
PATCH 2/9 : fix definition for migrate_incoming in qemu/vl.c
PATCH 3/9
Zhang, Xiantao wrote:
Hi, Avi
Due to last merge with Qemu upstream, some interfaces are changed, and
leads to build fail
, this patch fixed them.
Xiantao
From: Zhang Xiantao [EMAIL PROTECTED]
Date: Fri, 21 Dec 2007 22:47:52 +0800
Subject: [PATCH] kvm: qemu : Fix ia64 build on latest
Carlo Marcelo Arenas Belon wrote:
On Fri, Dec 21, 2007 at 07:22:57PM +0100, Andrea Arcangeli wrote:
On Fri, Dec 21, 2007 at 07:52:52PM +0200, Izik Eidus wrote:
oh, it was sent to the list, dont trust (in case you did) the source forge
site for the mails
But this time I
Carlo Marcelo Arenas Belon wrote:
This patch complements 684c6426c5d58f01f45cd45420df96c837d0feb4 by
deleting the tmp directory once the merge is completed so that it
doesn't get included by mistake in the release package
Applied, thanks.
--
Do not meddle in the internals of kernels,
Izik Eidus wrote:
This fixes my longstanding KVM instability and -smp 2 now works
flawlessy with svm too! -smp 2 -snapshot crashes in qemu userland but
that's not kernel related, must be some thread mutex lock recursion or
lock inversion in the qcow cow code. Removing -snapshot make -smp 2
Amit Shah wrote:
On Friday 21 December 2007 20:28:23 Zhang, Xiantao wrote:
Hi, Avi
Due to last merge with Qemu upstream, some interfaces are changed, and
leads to build fail
, this patch fixed them.
Xiantao
I've not checked the patch yet; but a curiosity question: Do you also
Marcelo Tosatti wrote:
Hi Avi,
Andrew, please see the comments on the need for a atomic get_user_pages()
below.
On Sat, Dec 22, 2007 at 09:42:27PM +0200, Avi Kivity wrote:
Marcelo Tosatti wrote:
In preparation for a mmu spinlock, avoid schedules in mmu_set_spte()
by using
Yang, Sheng wrote:
From 33d84813131809e962a4cd8cae9b66a9062fa8db Mon Sep 17 00:00:00 2001
From: Sheng Yang [EMAIL PROTECTED]
Date: Thu, 20 Dec 2007 14:22:18 +0800
Subject: [PATCH] kvm: external module: Remove 'include/asm' before
execute 'ln'
We use 'ln -sf' to refresh the symbolic now.
Dong, Eddie wrote:
I observed 1-2 seconds faster in Windows XPSP2 ACPI bootup with total of
23 seconds.
I can;t say it is exactly caused by this patch, but anyway it simplifies
the logic and
code w/o any additional complexity.
thx,eddie
Patch looks good, but will delay so that the
Anthony Liguori wrote:
Hi,
The attached patch is an option ROM that allows x86 guests to boot
from any block device supported by QEMU (and the guest kernel). It
works by hijacking the BIOS int13 function and then uses a very simply
protocol to communicate all block requests to QEMU.
This
Zhang, Xiantao wrote:
From: Zhang Xiantao [EMAIL PROTECTED]
Date: Thu, 20 Dec 2007 19:30:56 +0800
Subject: [PATCH] kvm: qemu: Enable cmos, since rtc has been added back.
Enable cmos, since code is mature and rtc added back
Applied this and the next patch; thanks.
--
error compiling
Jan Kiszka wrote:
The qemu update did take place for -57, but a bit too mechanically.
Please merge this patch to finally fix the gdbstub.
Thanks, applied.
--
error compiling committee.c: too many arguments to function
Christoph Hellwig wrote:
On Thu, Dec 20, 2007 at 01:57:01PM +0800, Zhang, Xiantao wrote:
Hi Tony,
In order to build kvm source successfully, we need to export
three symbols for kvm use. Please review. Maybe Avi can check-in to
kvm.git upstream first.
Thanks
Xiantao
From: Zhang
Avi Kivity wrote:
Ingo Molnar wrote:
While the change mentions that it fixes a time warp bug, it also says
it should be rare. So clearly kvm smp tsc handing is buggy.
Ingo/Thomas, (or anybody else), do you have any insight as to what kvm
can be doing wrong to trigger this behavior
Avi Kivity wrote:
Testing shows wrmsr and rdtsc function normally.
I'll try pinning the vcpus to cpus and see if that helps.
It does.
--
error compiling committee.c: too many arguments to function
-
SF.Net email
Ingo Molnar wrote:
* Avi Kivity [EMAIL PROTECTED] wrote:
Avi Kivity wrote:
Testing shows wrmsr and rdtsc function normally.
I'll try pinning the vcpus to cpus and see if that helps.
It does.
do we let the guest read the physical CPU's TSC? That would be trouble
David P. Reed wrote:
Avi Kivity wrote:
kvm will forward a virtual machine's writes to port 0x80 to the real
port. The reason is that the write is much faster than exiting and
emulating it; the difference is measurable when compiling kernels.
Now if the cause is simply writing to port 0x80
Carlo Marcelo Arenas Belon wrote:
so just to be clear on this?, would you rather have these changes in a patch
series, or as unrelated patches with the second one fixing a formatting issue
that the first one triggered?
Oh, I misunderstood. The changes are related and should be in one
Ingo Molnar wrote:
try this test perhaps in an SMP guest:
http://people.redhat.com/mingo/time-warp-test/time-warp-test.c
you can ignore TSC warps - but no GTOD or CLOCK warps should occur.
On a broken guest kernel, I see gtod and clock warps. On a good guest
kernel, I do not,
Glauber de Oliveira Costa wrote:
Changes in rate does not sound good. It's possibly what's screwing up
my paravirt clock implementation in smp.
You should renew the timebase on vcpu migration, and hook cpufreq so
that changes in frequency are reflected in the timebase.
Since the host
Carlo Marcelo Arenas Belon wrote:
Complements f2601fd2874b94d14dfcb21a3e1b84d7f3262b17 which made obsolete
the use of io_apic provided defines that were not being included for x86
when the host kernel didn't have IO_APIC support.
Reverts :
5d6ea6b370ff9aae1dd7671b71e89351749fbeee
Christian Ehrhardt wrote:
Subject: [PATCH][UPDATE] kvm-userspace: simplify mmio callback
From: Christian Ehrhardt [EMAIL PROTECTED]
Merging the read[bwlq]/write[bwlq] callback callback pointers to
mmio_read/write functions simplifies the callback interface.
On the qemu side it now uses the
Carlo Marcelo Arenas Belon wrote:
On Tue, Dec 18, 2007 at 06:15:07PM +0200, Avi Kivity wrote:
The major focus for this release is a qemu-cvs merge, which brings many
new features, some introduced by kvm developers.
but didn't include my changes to the IDE emulation to allow
Dor Laor wrote:
vmx (and svm) allow us to add an offset to the physical tsc. We
set it
on startup to -tsc (so that an rdtsc on boot would return 0), and
massage it on vcpu migration so that guest rdtsc is monotonic.
The net effect is that tsc on a vcpu can experience large
Carlo Marcelo Arenas Belon wrote:
The following patch series implements fixes to the IDE/ATAPI emulation
in qemu which affects access to the CD-ROM device preventing installation
of OpenSolaris guests because of an incorrect implementation of the MMC6
GET_CONFIGURATION command when it was
Marcelo Tosatti wrote:
Updated patch, now feature complete. Changes from last version:
- Use __gfn_to_page in cmpxchg_pte() to avoid potential deadlock
- Add kvm_read_guest_inatomic() and use it in fetch()
- Make prefetch_page() use copy_from_user_inatomic()
- Pass grabbed page down to
Dong, Eddie wrote:
Dong, Eddie wrote:
host side TLB flush can be merged together if multiple
spte need to be write-protected.
Signed-off-by: Yaozu (Eddie) Dong [EMAIL PROTECTED]
diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c
index 92ac0d1..749db46 100644
--- a/drivers/kvm/mmu.c
Marcelo Tosatti wrote:
Hi,
The following is an improvement on top of an earlier patch by Izik. It
increases pagefault scalability for SMP guests by allowing concurrent
guest walking, allocation and instruction emulation on the fault path.
The test being used is pft, which starts a number
The major focus for this release is a qemu-cvs merge, which brings many
new features, some introduced by kvm developers. Examples are
improved usb support and the new -drive qemu option.
While we've tested this carefully, this is a big merge so please handle
this release with care and
Zhao, Yunfeng wrote:
Save/restore still passes on IA32e host but fails on IA32 host.
This is now fixed.
--
error compiling committee.c: too many arguments to function
-
SF.Net email is sponsored by:
Check out the
Carlo Marcelo Arenas Belon wrote:
The following patch eliminates the uninitialized mem pointer, using
instead the corresponding entry from the slots array to fix :
libkvm.c:580: warning: 'mem' is used uninitialized in this function
Also changes the formatting type for phys_addr to long to
Ryan Harper wrote:
* Ryan Harper [EMAIL PROTECTED] [2007-12-14 09:36]:
* Avi Kivity [EMAIL PROTECTED] [2007-12-14 08:49]:
Ryan Harper wrote:
Add printk_ratelimit check in front of printk. This prevents spamming
of the message during 32-bit ubuntu 6.06server install
Carlo Marcelo Arenas Belon wrote:
The following patch (which is composed of 3 different patches from a series)
has been validated to solve compilation issues observed with kvm-56 in x86
with kernels older than 2.6.21 and are posted here for testing and debate.
The main points accomplished (in
Christian Ehrhardt wrote:
Subject: [PATCH][UPDATE] kvm-userspace: sync icache for more architectures
From: Christian Ehrhardt [EMAIL PROTECTED]
A ia64 patch introduced kvm_sync_icache within a ifdef __ia64__, but the
concept of split caches is not bound to ia64. This patch replaces the the
Booting RHEL 5 i386 in kvm with -no-kvm-irqchip -smp 4 will hang in
udev. I bisected this to a change in the _guest_ kernel:
commit 95492e4646e5de8b43d9a7908d6177fb737b61f0
Author: Ingo Molnar [EMAIL PROTECTED]
Date: Fri Feb 16 01:27:34 2007 -0800
[PATCH] x86: rewrite SMP TSC sync
Zhao, Yunfeng wrote:
Retested it on KVM-57.
It has been fixed already.
Actually, the fix is post kvm-57. What kvm-userspace.git commit did you
test?
--
error compiling committee.c: too many arguments to function
-
Ingo Molnar wrote:
While the change mentions that it fixes a time warp bug, it also says
it should be rare. So clearly kvm smp tsc handing is buggy.
Ingo/Thomas, (or anybody else), do you have any insight as to what kvm
can be doing wrong to trigger this behavior?
hm. Those time
Zhang, Xiantao wrote:
According to Avi's comments, Updated the patches.
[1/3] Create ioapic.h and lapic.h to hold their declarations.
[2/3] Moving ioapic{.h,.c} back to common directory.
[3/3] Moving out kvm_vcpu_kick from irq.c
Applied all, thanks.
--
error compiling committee.c:
Zhang, Xiantao wrote:
From: Zhang Xiantao [EMAIL PROTECTED]
Date: Mon, 17 Dec 2007 16:43:24 +0800
Subject: [PATCH] kvm: qemu: No reason to disallow (phys_ram_alloc_offset
+ size) = phys_ram_size
No reason to disallow (phys_ram_alloc_offset + size) = phys_ram_size.
Applied, thanks.
--
Zhang, Xiantao wrote:
From: [EMAIL PROTECTED] [EMAIL PROTECTED]
Date: Mon, 17 Dec 2007 20:27:27 +0800
Subject: [PATCH] kvm: make ioapic structure in irqchip seen by ia64
IA64 also needs to see ioapic structure in irqchip.
Patch doesn't apply due to whitespace corruption.
--
error
Zhang, Xiantao wrote:
Attached.
From: [EMAIL PROTECTED] [EMAIL PROTECTED]
Date: Mon, 17 Dec 2007 20:27:27 +0800
Subject: [PATCH] kvm: make ioapic structure in irqchip seen by ia64
IA64 also needs to see ioapic structure in irqchip.
Applied, thanks.
--
error compiling committee.c: too
Dong, Eddie wrote:
host side TLB flush can be merged together if multiple
spte need to be write-protected.
Signed-off-by: Yaozu (Eddie) Dong [EMAIL PROTECTED]
diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c
index 92ac0d1..749db46 100644
--- a/drivers/kvm/mmu.c
+++ b/drivers/kvm/mmu.c
Dong, Eddie wrote:
Avi/Anthony:
Following commit added #UD trap in KVM and do emulation in KVM.
It is good to avoid the mis hypercall instruction issue, but I am not
sure if it will have any serious side effect. I am blind to AMD
instruction, I assume, for example, AMD 3D now
Carlo Marcelo Arenas Belon wrote:
On Thu, Dec 13, 2007 at 03:32:46PM +0200, Avi Kivity wrote:
Notes:
If you use the modules bundled with kvm-56, you can use any version
of Linux from 2.6.9 upwards.
the external module (once patched) will only be able to build against 2.6.17
Carlo Marcelo Arenas Belon wrote:
The following patch fixes building the kvm module for kernels older than
2.6.21 and which were missing an include to asm/apicdef.h in asm/io_apic.h
resulting in the following error :
include/asm/io_apic.h:61: error: 'MAX_IO_APICS' undeclared here (not
Dong, Eddie wrote:
A little bit suspicion for the usage model too :(
The CPUID feature list can't be changed dynamically, so I doutb
if a real user will give up performance at beginning when he/she
starts the guest for maybe future migration cross host platform.
A VT or SVM enabled machine
Dong, Eddie wrote:
Since we don't emulate 3Dnow instructions, a #UD will be
injected instead.
OK, how many architecture specific instructiosn are emulated now?
Just vmcall/vmmcall (which is emulated once, then patched).
Or what is planed to be implemented? If no, then
Zhang, Xiantao wrote:
From: Zhang Xiantao [EMAIL PROTECTED]
Date: Sun, 16 Dec 2007 23:41:50 +0800
Subject: [PATCH] kvm: Split irq.h into two parts, and make ioapic
independent of
X86 header files.
This patch splits irq.h into two parts, and make ioapic independent of
X86 header files.
1301 - 1400 of 3824 matches
Mail list logo