want to establish matches that of the
PIT timer (in a not completely obvious way, though).
Having said that the proposed patch in my first mail is incomplete, as
the mod_64 does not work correctly for negative values. A fixed version
is below.
regards Christian
Signed-off-by: Christian
use the negative
difference unmodified.
regards Christian
Fix lapic time counter read for periodic mode.
Signed-off-by: Christian Ehrhardt l...@c--e.de
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 43e9fad..eff902d 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm
it mutually exclusive per vcpu and should ensure it
isn't running - by that it should be fine.
Reviewed-by: Christian Ehrhardt ehrha...@linux.vnet.ibm.com
Alex
Signed-off-by: Scott Woodscottw...@freescale.com
---
arch/powerpc/kvm/timing.c |2 --
1 files changed, 0 insertions(+), 2 deletions
Marcelo Tosatti wrote:
On Tue, Jun 02, 2009 at 04:26:11PM +0200, ehrha...@linux.vnet.ibm.com wrote:
From: Christian Ehrhardt ehrha...@linux.vnet.ibm.com
[...]
@@ -706,13 +713,19 @@ int kvm_arch_set_memory_region(struct kv
/* request update of sie control block for all available
Avi Kivity wrote:
Christian Ehrhardt wrote:
Really need that smp_mb__after_clear_bit ? AFAIK test_and_clear_bit
implies a barrier?
Well I agree that practically test_and_clear_bit has a barrier on
s390, but as far as I read Documentation/atomic_ops.txt at line
339-360 I think
Marcelo Tosatti wrote:
On Sun, May 31, 2009 at 11:22:58AM +0300, Avi Kivity wrote:
ehrha...@linux.vnet.ibm.com wrote:
From: Christian Ehrhardt ehrha...@linux.vnet.ibm.com
*updates in v6*
- ensure the wait_on_bit waiter is notified
- move the reset of requests to kvm_vcpu_release
Marcelo Tosatti wrote:
On Tue, May 26, 2009 at 10:02:59AM +0200, Christian Ehrhardt wrote:
Marcelo Tosatti wrote:
On Mon, May 25, 2009 at 01:40:49PM +0200, ehrha...@linux.vnet.ibm.com wrote:
From: Christian Ehrhardt ehrha...@linux.vnet.ibm.com
To ensure vcpu's come out
Avi Kivity wrote:
Christian Ehrhardt wrote:
So you _need_ a mechanism to kick all vcpus out of guest mode?
I have a mechanism to kick a vcpu, and I use it. Due to the fact that
smp_call_* don't work as kick for us the kick is an arch specific
function.
I hop ethat clarified this part
Marcelo Tosatti wrote:
On Mon, May 25, 2009 at 01:40:49PM +0200, ehrha...@linux.vnet.ibm.com wrote:
From: Christian Ehrhardt ehrha...@linux.vnet.ibm.com
To ensure vcpu's come out of guest context in certain cases this patch adds a
s390 specific way to kick them out of guest context
) and I try to come up with
something in the next few days - either a updated patch series or
additional discussion input :-).
--
Grüsse / regards, Christian Ehrhardt
IBM Linux Technology Center, Open Virtualization
--
To unsubscribe from this list: send the line unsubscribe kvm in
the body
Avi Kivity wrote:
ehrha...@linux.vnet.ibm.com wrote:
From: Christian Ehrhardt ehrha...@linux.vnet.ibm.com
This patch relocates the variables kvm-s390 uses to track guest mem
addr/size.
As discussed dropping the variables at struct kvm_arch level allows
to use the
common vcpu-request based
Christian Ehrhardt wrote:
Avi Kivity wrote:
ehrha...@linux.vnet.ibm.com wrote:
From: Christian Ehrhardt ehrha...@linux.vnet.ibm.com
[...]
-/* update sie control blocks, and unlock all vcpus */
+/* request update of sie control block for all available vcpus */
for (i = 0; i
Avi Kivity wrote:
Christian Ehrhardt wrote:
The bad thing on vcpu-request in that case is that I don't want
the async behaviour of vcpu-requests in that case, I want the
memory slot updated in all vcpu's when the ioctl is returning.
You mean, the hardware can access the vcpu control block
Avi Kivity wrote:
Christian Ehrhardt wrote:
The bad thing on vcpu-request in that case is that I don't want the
async behaviour of vcpu-requests in that case, I want the memory
slot updated in all vcpu's when the ioctl is returning.
You mean, the hardware can access the vcpu control block
Avi Kivity wrote:
Christian Ehrhardt wrote:
Avi Kivity wrote:
Christian Ehrhardt wrote:
The bad thing on vcpu-request in that case is that I don't want
the async behaviour of vcpu-requests in that case, I want the
memory slot updated in all vcpu's when the ioctl is returning.
You mean
)
+vcpu-kvm-arch.sca-cpu[vcpu-vcpu_id].sda = 0;
free_page((unsigned long)(vcpu-arch.sie_block));
If this is accessed by hardware on a different cpu, don't you need a
memory barrier here?
Right, will be in v2
--
Grüsse / regards,
Christian Ehrhardt
IBM Linux Technology Center, Open
I explicitly let
the vcpu exit from guest.
--
Grüsse / regards,
Christian Ehrhardt
IBM Linux Technology Center, Open Virtualization
--
To unsubscribe from this list: send the line unsubscribe kvm in
the body of a message to majord...@vger.kernel.org
More majordomo info at http
Avi Kivity wrote:
Christian Ehrhardt wrote:
On x86, we use slots_lock to protect memory slots. When we change
the global memory configuration, we set a bit in vcpu-requests, and
send an IPI to all cpus that are currently in guest mode for our
guest. This forces the cpu back to host mode
Avi Kivity wrote:
Christian Ehrhardt wrote:
I thought about implementing it with slots_lock, vcpu-request, etc
but it really looks like overkill for s390.
We could make (some of) it common code, so it won't look so bad.
There's value in having all kvm ports do things similarly; though
Andre Przywara wrote:
ehrha...@linux.vnet.ibm.com wrote:
From: Christian Ehrhardt ehrha...@linux.vnet.ibm.com
There is already a variable kvm_cflags which gets the path of the kernel
includes when using --kerneldir. But eventually with newer kernels we
all will
need arch/$arch/include too (my
Anthony Liguori wrote:
ehrha...@linux.vnet.ibm.com wrote:
From: Christian Ehrhardt ehrha...@linux.vnet.ibm.com
The patch applies to upstream qemu as well as kvm-userspace, but
since it is
the qemu configure script I think it should go to upstream qemu
(Anthony)
first and with the next merge
Avi Kivity wrote:
Christian Ehrhardt wrote:
Hi, this patch breaks all non x86 architectures as
libkvm/libkvm-x86.c has the only implementation of the alias
functionality.
Until now only qemu-kvm-x86 has called that functions, but since this
patch the generic qemu-kvm.c calls them which leads
# HG changeset patch
# User Christian Ehrhardt ehrha...@linux.vnet.ibm.com
# Date 1229345133 -3600
# Node ID b48b9d560f80037ab4e12eae128022622c7ccb34
# Parent 4b0ad05490115e4c6f31d2419c0e5b628040f90b
[PATCH] kvm-userspace: ppc: fix compatfd build decision v2
From: Christian Ehrhardt ehrha
if CONFIG_signalfd/CONFIG_eventfd are set. Therefore we can savely
remove the makefile guard completely and just always build compatfd.c.
This updated patch works for x86powerpc with/without --disable-aio in
my tests.
It should appear on the list shortly.
--
Grüsse / regards,
Christian Ehrhardt
IBM
Hollis Blanchard wrote:
On Thu, 2008-12-11 at 17:05 +0100, Jan Kiszka wrote:
Hollis Blanchard wrote:
On Thu, 2008-12-11 at 13:53 +0100, Christian Ehrhardt wrote:
This is v2 as version one had a type in it occured when splitting patches.
Mercurial somehow lost my changes
This patch series updates the gdbstub support for kvm.
Patch 12 introduce basic powerpc support while patch 3 fixes gdbstub generic
code that was broken in a qemu merge.
--
To unsubscribe from this list: send the line unsubscribe kvm in
the body of a message to [EMAIL PROTECTED]
More majordomo
# HG changeset patch
# User Christian Ehrhardt [EMAIL PROTECTED]
# Date 1228989956 -3600
# Node ID 6f228c807ad0b239b7342d2974debfc66418d784
# Parent 38846cef16e56c681da1ddc179e248972c8b2ff9
[PATCH] qemu: ppc: kvm-userspace: KVM PowerPC support for qemu gdbstub
From: Hollis Blanchard [EMAIL
# HG changeset patch
# User Christian Ehrhardt [EMAIL PROTECTED]
# Date 1228924564 -3600
# Node ID 38846cef16e56c681da1ddc179e248972c8b2ff9
# Parent 705d874ff7a24484eaa15ed75a748c4e1a70c2ef
[PATCH] kvm-userspace: ppc: Add kvm_translate wrapper
From: Hollis Blanchard [EMAIL PROTECTED]
Add
This is v2 as version one had a type in it occured when splitting patches.
Mercurial somehow lost my changes to the patch description explaining
that, but the patch is right this way.
Christian Ehrhardt wrote:
# HG changeset patch
# User Christian Ehrhardt [EMAIL PROTECTED]
# Date 1228999833
# HG changeset patch
# User Christian Ehrhardt [EMAIL PROTECTED]
# Date 1228999833 -3600
# Node ID dc1466c9077ab162f4637fffee1869f26be02299
# Parent 4c07fe2a56c7653a9113e05bb08c2de9aec210ce
[PATCH] qemu: ppc: kvm-userspace: KVM PowerPC support for qemu gdbstub
From: Hollis Blanchard [EMAIL
# HG changeset patch
# User Christian Ehrhardt [EMAIL PROTECTED]
# Date 1228924564 -3600
# Node ID 38846cef16e56c681da1ddc179e248972c8b2ff9
# Parent 705d874ff7a24484eaa15ed75a748c4e1a70c2ef
[PATCH] kvm-userspace: ppc: Add kvm_translate wrapper
From: Hollis Blanchard [EMAIL PROTECTED]
Add
# HG changeset patch
# User Christian Ehrhardt [EMAIL PROTECTED]
# Date 1228989958 -3600
# Node ID f80fb35de91fe69dae889c70948c9a53212ee444
# Parent 6f228c807ad0b239b7342d2974debfc66418d784
[PATCH] kvm-userspace: fix gdbstub kvm integration
From: Christian Ehrhardt [EMAIL PROTECTED]
Some recent
This is v2 as version one had a type in it occured when splitting patches.
Mercurial somehow lost my changes to the patch description explaining
that, but the patch is right this way.
Christian Ehrhardt wrote:
# HG changeset patch
# User Christian Ehrhardt [EMAIL PROTECTED]
# Date 1228999833
and there is an even better or trivial
approach :-)
Since this could be solved several very different ways I think its worth
a discussion.
As I prefer b) I attached a simple patch example how this could look
like :-).
--
Grüsse / regards,
Christian Ehrhardt
IBM Linux Technology Center, Open
From: Christian Ehrhardt [EMAIL PROTECTED]
*update to v5*
- add exittiming.c to diff content
- prefix all exit timing functions with kvmppc to prevent name collisions (was
already done for some of the functions, now its consequently done for all of
them)
- renamed header c-file and relocated
Hollis Blanchard wrote:
On Tue, 2008-11-11 at 16:43 +0100, Christian Ehrhardt wrote:
From: Christian Ehrhardt [EMAIL PROTECTED]
*update to v4*
- EMUL_CORE no longer had more than wrtee emulation, therefore it now accounts
for WRTEE in the output and set_exit_type calls are in the wrtee
From: Christian Ehrhardt [EMAIL PROTECTED]
Fix some whitespaces in kvm_ppc.h
Signed-off-by: Christian Ehrhardt [EMAIL PROTECTED]
---
[diffstat]
[diff]
diff --git a/arch/powerpc/include/asm/kvm_ppc.h
b/arch/powerpc/include/asm/kvm_ppc.h
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch
From: Christian Ehrhardt [EMAIL PROTECTED]
*update to v4*
- EMUL_CORE no longer had more than wrtee emulation, therefore it now accounts
for WRTEE in the output and set_exit_type calls are in the wrtee handlers to
let any residual core op be counted as EMULINST
*update to v3*
- ensure build
From: Christian Ehrhardt [EMAIL PROTECTED]
*update to v3*
- ensure build time optimization when calling exit accouting functions using
build time bug / constant check
- migrate most of the exit timing code from powerpc.c and
kvm_timing_stats.h to a separate exittiming.c file
- renamed a lot
And btw - the wiki page is updated with new versions of the exittimings
using all of our new memory management improvements.
The page now also has some extended descriptions about the simple
workloads used, holds the current version of the postprocessing script etc.
Christian Ehrhardt wrote
+2959,8 @@
env-cpu_model_str = cpu_model;
cpu_ppc_register_internal(env, def);
cpu_ppc_reset(env);
-#ifdef USE_KVM
if (kvm_enabled())
- kvm_init_new_ap(env-cpu_index, env);
-#endif
+ kvm_init_vcpu(env);
return env;
}
--
Grüsse / regards,
Christian Ehrhardt
Hollis Blanchard wrote:
On Thu, 2008-11-06 at 12:25 +0100, Christian Ehrhardt wrote:
Hi,
I just added the following page to our wiki based online documentation
to show some numbers on kvmppc as of today.
You can find that page at thew wiki entry:
PowerPC_Exittimings
http
/$3)*($9/$3))),
($9/('$sumsum'/100))); }' $1
This runs awk twice, because for % of overall time you need the sum the
durations and I personally don't like all those rewind hacks for awk.
Ehrhardt Christian wrote:
From: Christian Ehrhardt [EMAIL PROTECTED]
Other existing kvm stats are either
with the arch-platform-os-compiler after our discussion,
but I like the $PROCESSOR solution for our *powerpc* Makefiles too.
And a good catch with that AR usage in patch 7.
(full series)
Acked-by: Christian Ehrhardt [EMAIL PROTECTED]
--
Grüsse / regards,
Christian Ehrhardt
IBM Linux Technology Center
[EMAIL PROTECTED]
This patch adds a better error reporting for powerpc testcases.
It prints the bytes read in load_file so far until an error occured.
Signed-off-by: Hollis Blanchard [EMAIL PROTECTED]
Signed-off-by: Christian Ehrhardt [EMAIL PROTECTED]
---
[diffstat]
main-ppc.c |3 ++-
1 file
/ and a single patch I just
submitted while cleaning our userspace repo to get the missing things
upstream.
Avi, let me know if I confused you and I'll send them once again with
some time in between to order them easier.
[EMAIL PROTECTED] wrote:
From: Christian Ehrhardt [EMAIL PROTECTED]
This is a set
/ and a single patch I just
submitted while cleaning our userspace repo to get the missing things
upstream.
Avi, let me know if I confused you and I'll send them once again with
some time in between to order them easier.
[EMAIL PROTECTED] wrote:
From: Christian Ehrhardt [EMAIL PROTECTED]
This is a set
+unifdef_uflags = $(foreach arch, $(NONARCH_CONFIG), -UCONFIG_$(arch))
$(patsubst ...), or even $(NONARCH_CONFIG:%=-UCONFIG_%)
But I think NONARCH_CONFIG needs to be adjusted as well.
--
Grüsse / regards,
Christian Ehrhardt
IBM Linux Technology Center, Open Virtualization
--
To unsubscribe from
and EMULINST we
are at ~2.75%.
Therefore it should be ok to keep the generic find_next_bit there.
--
Grüsse / regards,
Christian Ehrhardt
IBM Linux Technology Center, Open Virtualization
--
To unsubscribe from this list: send the line unsubscribe kvm-ppc in
the body of a message to [EMAIL PROTECTED
into that section once again more in detail
first.
Christian Ehrhardt wrote:
Hollis Blanchard wrote:
On Wed, 2008-10-08 at 15:49 +0200, Christian Ehrhardt wrote:
Wondering about that 30.5% for postprocessing and
kvmppc_check_and_deliver_interrupts I quickly checked that in detail
- part d is now
malc wrote:
On Tue, 30 Sep 2008, [EMAIL PROTECTED] wrote:
From: Christian Ehrhardt [EMAIL PROTECTED]
*update*
further debugging according to some requests revealed that
ARCH_CFLAGS does
not contain all CFLAGS that might be needed, especially those
supplied via
extra-cflags. Therefore people
malc wrote:
On Tue, 30 Sep 2008, [EMAIL PROTECTED] wrote:
From: Christian Ehrhardt [EMAIL PROTECTED]
*update*
further debugging according to some requests revealed that
ARCH_CFLAGS does
not contain all CFLAGS that might be needed, especially those
supplied via
extra-cflags. Therefore people
guest code - sprg0-3
From: Christian Ehrhardt [EMAIL PROTECTED]
This patch adds the functionality of rewriting guest code
using the magic page
mechanism. If a paravirtual guest memory (magic page) is
registered the host
rewrites trapping emulations instead of emulating them. This
only works
for that.
I think we need to put a little more thought into all the corner cases
here.
yeah, lets do that in a brainstorm session next week if you have some time.
It's better to clean corners interactively ;-)
--
Grüsse / regards,
Christian Ehrhardt
IBM Linux Technology Center, Open
Kumar Gala wrote:
On Aug 19, 2008, at 5:36 AM, [EMAIL PROTECTED] wrote:
From: Christian Ehrhardt [EMAIL PROTECTED]
Dependent on the already existing CONFIG_KVM_GUEST config option this
patch
changes wrteei to wrtee allowing the hypervisor to rewrite those to
nontrapping
instructions. Maybe
)
+{
+ _tlbia();
+}
+
+#endif /* __KVM_POWERPC_TLB_H__ */
you defined it as __KVM_E500_TLB_H__ above, this might be the old 440
comment from Hollis
--
Grüsse / regards,
Christian Ehrhardt
IBM Linux Technology Center, Open Virtualization
--
To unsubscribe from this list: send the line unsubscribe
is that we don't need to flush the TLB on privilege
switches, but we do on guest context switches (which are far
more infrequent). Guest boot time performance improvement: about 30%.
--
Grüsse / regards,
Christian Ehrhardt
IBM Linux Technology Center, Open Virtualization
--
To unsubscribe from
for a few weeks starting Monday, but
Christian Ehrhardt should be able to take care of any technical issues.
Applied all; thanks. I prefer to only merge bug fixes at this time for
2.6.27. As far as I can tell, patch 2 is independent of the rest so
I'll queue that. Let me know if that works.
Yes
block until the for loop run in
if !defined(__s390__)
That way you would additionally stop modifying the tss_ext variable
which is not used in the s390 case anyway.
[...]
--
Grüsse / regards,
Christian Ehrhardt
IBM Linux Technology Center, Open Virtualization
--
To unsubscribe from
Hollis Blanchard wrote:
On Mon, 2008-07-07 at 15:56 +0200, [EMAIL PROTECTED] wrote:
From: Christian Ehrhardt [EMAIL PROTECTED]
The current implementation of kvmtrace uses always a 64 bit cycle variable,
but get_cycles() which is used to fill it is unsigned long which might be 32
bit
Hollis Blanchard wrote:
On Mon, 2008-07-07 at 15:56 +0200, [EMAIL PROTECTED] wrote:
From: Christian Ehrhardt [EMAIL PROTECTED]
The current implementation of kvmtrace uses always a 64 bit cycle variable,
but get_cycles() which is used to fill it is unsigned long which might be 32
bit
61 matches
Mail list logo