I am using qemu to run VMs on my computer with amd chip on board.
However, my computer doesn't support amd-v, so the performance of the VM is
low.
I do have experiences in optimization of program using sse technique.
Is it possible to speed up the qemu for those x86 chips that do not support
On Sun, Apr 22, 2012 at 04:07, 陳韋任 che...@iis.sinica.edu.tw wrote:
I am using qemu to run VMs on my computer with amd chip on board.
However, my computer doesn't support amd-v, so the performance of the VM is
low.
I do have experiences in optimization of program using sse technique.
Is it
In this rebased version I pushed the exception helper conversion first.
Helpers for cpu_{ld,st}_data are in common code.
Two new bonus patches last.
URL git://repo.or.cz/qemu/blueswirl.git
http://repo.or.cz/r/qemu/blueswirl.git
Blue Swirl (22):
ppc: fix coding style in
Move exception helpers from op_helper.c to excp_helper.c.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
Makefile.target |6 ++
target-ppc/excp_helper.c | 218 ++
target-ppc/op_helper.c | 190
Add an explicit CPUPPCState parameter instead of relying on AREG0.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
Makefile.target |3 ---
target-ppc/excp_helper.c | 60 --
target-ppc/helper.h | 28 +++---
Add an explicit CPUPPCState parameter instead of relying on AREG0.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
Makefile.target |3 ---
target-ppc/helper.h | 18 +-
target-ppc/misc_helper.c| 19 +--
target-ppc/translate.c
Move more misc helpers from helper.c to misc_helper.c.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
target-ppc/helper.c |9 -
target-ppc/misc_helper.c |9 +
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/target-ppc/helper.c b/target-ppc/helper.c
When the code is moved together by the next patch, compiler
detects a possible uninitialized variable use. Avoid the warning
by initializing the variables.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
target-ppc/mmu_helper.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
Move decrementer and timebase helpers to a dedicated file.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
Makefile.target |4 ++
target-ppc/op_helper.c | 135 ---
target-ppc/timebase_helper.c | 160
Add an explicit CPUPPCState parameter instead of relying on AREG0.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
Makefile.target |3 ---
target-ppc/helper.h | 42 -
target-ppc/timebase_helper.c | 43
On Thu, Apr 19, 2012 at 04:08:38PM +0200, Vasilis Liaskovitis wrote:
This is a prototype for ACPI memory hotplug on x86_64 target. Based on some
earlier work and comments from Gleb.
Memslot devices are modeled with a new qemu command line
-memslot id=name,start=start_addr,size=sz,node=pxm
Add an explicit CPUPPCState parameter instead of relying on AREG0.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
Makefile.target |4 +-
target-ppc/helper.h | 176 +++
target-ppc/int_helper.c | 120 +++-
On Thu, Apr 19, 2012 at 04:08:46PM +0200, Vasilis Liaskovitis wrote:
Hotplugged memory is not persistent in the e820 memory maps. After
hotplugging
a memslot and rebooting the VM, the hotplugged device is not present.
A possible solution is to add an e820 for the new memslot in the
On 04/22/2012 04:56 PM, Gleb Natapov wrote:
start. We will need it for migration anyway.
hotplug-able memory slots i.e. initial system memory is not modeled with
memslots. The concept could be generalized to include all memory though, or
it
could more closely follow kvm-memory slots.
Add an explicit CPUPPCState parameter instead of relying on AREG0.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
Makefile.target |3 --
target-ppc/helper.h | 74 +--
target-ppc/mmu_helper.c | 91
On Sun, Apr 22, 2012 at 05:06:43PM +0300, Avi Kivity wrote:
On 04/22/2012 04:56 PM, Gleb Natapov wrote:
start. We will need it for migration anyway.
hotplug-able memory slots i.e. initial system memory is not modeled with
memslots. The concept could be generalized to include all memory
Add an explicit CPUPPCState parameter instead of relying on AREG0
and rename op_helper.c (which only contains load and store helpers)
to mem_helper.c. Remove AREG0 swapping in
tlb_fill().
Switch to AREG0 free mode. Use cpu_ld{l,uw}_code in translation
and interrupt handling,
On 04/22/2012 05:09 PM, Gleb Natapov wrote:
On Sun, Apr 22, 2012 at 05:06:43PM +0300, Avi Kivity wrote:
On 04/22/2012 04:56 PM, Gleb Natapov wrote:
start. We will need it for migration anyway.
hotplug-able memory slots i.e. initial system memory is not modeled with
memslots. The
Remove useless wrappers. In some cases 'int' parameters are
changed to uint32_t.
Make internal functions static.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
target-ppc/cpu.h| 22 -
target-ppc/mmu_helper.c | 122 ++-
2 files
Add obviously missing 'break' statement.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
target-ppc/fpu_helper.c |1 +
1 file changed, 1 insertion(+)
diff --git a/target-ppc/fpu_helper.c b/target-ppc/fpu_helper.c
index d90636f..9d67926 100644
--- a/target-ppc/fpu_helper.c
+++
On Sun, Apr 22, 2012 at 05:13:27PM +0300, Avi Kivity wrote:
On 04/22/2012 05:09 PM, Gleb Natapov wrote:
On Sun, Apr 22, 2012 at 05:06:43PM +0300, Avi Kivity wrote:
On 04/22/2012 04:56 PM, Gleb Natapov wrote:
start. We will need it for migration anyway.
hotplug-able memory slots
Move MMU, TLB, SLB and BAT ops to mmu_helper.c.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
Makefile.target |5 +-
target-ppc/{op_helper.c = mmu_helper.c} | 543 +--
target-ppc/op_helper.c | 854
Lookup table 'hbrev' is never written to, so add a 'const' qualifier.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
target-ppc/int_helper.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
index 3173f11..f638b2a 100644
Move misc helpers from op_helper.c to misc_helpers.c.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
Makefile.target |5 +-
target-ppc/misc_helper.c | 116 ++
target-ppc/op_helper.c | 94 -
3
helper.c will be spilt by the next patches, fix
style issues before that.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
target-ppc/helper.c | 396 ++-
1 file changed, 231 insertions(+), 165 deletions(-)
diff --git a/target-ppc/helper.c
During normal operation, virtio host first writes a used index
and then checks whether it should interrupt the guest
by reading guest avail flag/used event index values.
Guest does the reverse: writes the index/flag,
then checks the used ring.
The ordering is important: if host avail flag read
I split the first patch and rebased.
Blue Swirl (4):
exec: prepare for splitting
cputlb: move TLB handling to a separate file
cputlb: prepare private memory API for public consumption
memory: move functions is_romd and section_addr to memory API
Makefile.target |2 +-
cpu-all.h
Make s_cputlb_empty_entry 'const'.
Rename tlb_flush_jmp_cache() to tb_flush_jmp_cache().
Refactor code to add cpu_tlb_reset_dirty_all(),
memory_region_section_get_iotlb() and
memory_region_is_unassigned().
Remove unused cpu_tlb_update_dirty().
Fix coding style in areas to be moved.
Move TLB handling and softmmu code load helpers to cputlb.c,
compile only for softmmu targets.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
Makefile.target |2 +-
cputlb.c| 362
cputlb.h| 63 +
exec-all.h
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
cputlb.h |6 --
exec.c |8
memory.h | 29 +
3 files changed, 29 insertions(+), 14 deletions(-)
diff --git a/cputlb.h b/cputlb.h
index 60479af..ce6e297 100644
--- a/cputlb.h
+++ b/cputlb.h
@@
On 04/22/2012 06:34 PM, Blue Swirl wrote:
I split the first patch and rebased.
The memory patch is fine from my POV. Please commit directly, no need
to Blue-Avi-Blue.
Inlining memory_region_section_addr is a bit premature, but doesn't
really matter. My feeling is that we should turn on LTO
Fold is_ram_rom and is_ram_rom_romd() into callers.
Change is_romd() and section_addr() to take MemoryRegion
instead of MemoryRegionSection for consistency and
use memory_region_ prefix.
Signed-off-by: Blue Swirl blauwir...@gmail.com
---
cputlb.c | 12 ++-
cputlb.h | 18
On 04/20/2012 11:00 PM, aaron.sto...@celestech.com wrote:
So my next question is, does qemu have a way that features can be
sponsored?
You can hire developers or consultants to work on a feature you like.
--
error compiling committee.c: too many arguments to function
i have an OS image that does not have network drivers. are there any
other ways to copy a benchmark onto the os image ?
Thanks
Xin
On Sat, Apr 21, 2012 at 18:29, Peter Maydell peter.mayd...@linaro.org wrote:
Hi; this is a pullreq for target-arm. It's got a trivial patch from Benoit
that I've been holding onto for ages, plus the first 13 of the QOM
subclass/reset patches [with the trivial fixes to the bits pointed out
by
*** This bug is a duplicate of bug 919242 ***
https://bugs.launchpad.net/bugs/919242
** This bug has been marked a duplicate of bug 919242
qemu-img convert to VDI corrupts image
** Changed in: qemu
Status: New = Fix Committed
--
You received this bug notification because you are
** Changed in: qemu
Status: Confirmed = Fix Committed
** Changed in: qemu
Assignee: (unassigned) = Stefan Weil (ubuntu-weilnetz)
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/984476
On 18/04/12 21:31, Jakub Jermar wrote:
Does HelenOS break without the patch? It worked fine for me.
Hi Alex,
I've just tested QEMU git (which includes the TLB invalidation fix) and
it seems to work with HelenOS mainline quite nice. Not sure if we can
conclude the other fix is not needed
Am 20.04.2012 17:44, schrieb Peter Portante:
Signed-off-by: Peter Portante peter.porta...@redhat.com
Fix itself looks okay, but author, maintainer and qemu-ppc were missing
in CC, and a better commit message would be:
---8---
pseries: Fix use of global CPU state
Commit
On 22 April 2012 20:56, Andreas Färber afaer...@suse.de wrote:
Commit ed120055c7f9b26b5707d3ceabbe5a3f06aaf937 (Implement PAPR VPA
functions for pSeries shared processor partitions) introduced the
register_dtl() function and typo emv as name of its argument.
This went unnoticed because the
Am 22.04.2012 22:06, schrieb Peter Maydell:
On 22 April 2012 20:56, Andreas Färber afaer...@suse.de wrote:
Commit ed120055c7f9b26b5707d3ceabbe5a3f06aaf937 (Implement PAPR VPA
functions for pSeries shared processor partitions) introduced the
register_dtl() function and typo emv as name of its
On Fri, Apr 20, 2012 at 03:56:01PM +0200, Kevin Wolf wrote:
Refcount block allocation and refcount table growth rely on
s-free_cluster_index pointing to somewhere after the current
allocation. Change qcow2_allocate_cluster_at() to fulfill this
assumption.
Without this change it could happen
Hi all,
Most of the files are unchanged since last time. Main differences
are:
-- Change the way the timers are hooked up to the board to address
Peter M's review comments.
-- Move timer tick rate limiting out of the imx-timers implementation into a
separate patch (not in
For Linux to be able to work out how fast its clocks are going, so
that timer ticks come approximately at the right time, it needs to
be able to query the clock control module (CCM).
This is the start of a CCM implementation. It currently knows only about
the MCU, HSP and IPG clocks --- i.e.,
Board support for Kyoto Micro's KZM-ARM11-01, an evaluation board built
around the FreeScale i.MX31.
Signed-off-by: Philip O'Sullivan phil...@ok-labs.com
Signed-off-by: Peter Chubb peter.ch...@nicta.com.au
---
Makefile.target |2
hw/kzm.c| 161
Implement the FreeScale i.MX UART. This uart is used in a variety of
SoCs, including some by Motorola, as well as in the FreeScale i.MX
series.
This patch gives only a `bare-bones' implementation, enough to run Linux
or OKL4, but that's about it.
Signed-off-by: Philip O'Sullivan
Implement the timers on the FreeScale i.MX31 SoC.
This is not a complete implementation, but gives enough for
Linux to boot and run. In particular external triggers, which are
not useful under QEMU, are not implemented.
Signed-off-by: Philip O'Sullivan phil...@ok-labs.com
Signed-off-by: Peter
Implement the FreeSCALE i.MX31 advanced vectored interrupt controller, at least
to the extent it is used by Linux 3.0.x
Vectors are not implemented.
Signed-off-by: Philip O'Sullivan phil...@ok-labs.com
Signed-off-by: Peter Chubb peter.ch...@nicta.com.au
Reviewed-by: Peter Maydell
On Sun, Apr 22, 2012 at 08:18:49PM -0300, Marcelo Tosatti wrote:
On Fri, Apr 20, 2012 at 03:56:01PM +0200, Kevin Wolf wrote:
Refcount block allocation and refcount table growth rely on
s-free_cluster_index pointing to somewhere after the current
allocation. Change
Public bug reported:
with latest qemu-kvm from git i have following problem:
/usr/bin/kvm -daemonize -smp 4 -cpu host -vga qxl -enable-kvm -m 4096
-localtime -drive
file=/home/sss/vm/win_xp/tmp_sys.img,if=virtio,media=disk,cache=none -drive
Only proxy helper does.
Signed-off-by: Kusanagi Kouichi sl...@ac.auone-net.jp
---
configure |6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/configure b/configure
index 2d62d12..c8e6fe4 100755
--- a/configure
+++ b/configure
@@ -2860,9 +2860,11 @@ tools=
if test
QEMU uses this very hard to understand setjmp and longjmp to handle
asynchronous interrupt. what i do not understand is that say one of
the process is doing a context switching in the guest os. tlb_flush
gets called, what if in the middle of tlb_flush, an interrupt comes in
( longjmp executed ).
On Sun, Apr 22, 2012 at 3:56 PM, Andreas Färber afaer...@suse.de wrote:
Am 20.04.2012 17:44, schrieb Peter Portante:
Signed-off-by: Peter Portante peter.porta...@redhat.com
Fix itself looks okay, but author, maintainer and qemu-ppc were missing
in CC, and a better commit message would be:
On Sun, Apr 22, 2012 at 9:40 PM, Peter Portante
peter.a.porta...@gmail.comwrote:
On Sun, Apr 22, 2012 at 3:56 PM, Andreas Färber afaer...@suse.de wrote:
Am 20.04.2012 17:44, schrieb Peter Portante:
Signed-off-by: Peter Portante peter.porta...@redhat.com
Fix itself looks okay, but author,
On Sun, Apr 22, 2012 at 09:56:09PM +0200, Andreas Färber wrote:
Am 20.04.2012 17:44, schrieb Peter Portante:
Signed-off-by: Peter Portante peter.porta...@redhat.com
Fix itself looks okay, but author, maintainer and qemu-ppc were missing
in CC, and a better commit message would be:
On 04/22/2012 09:22 PM, David Gibson wrote:
On Sun, Apr 22, 2012 at 09:56:09PM +0200, Andreas Färber wrote:
Am 20.04.2012 17:44, schrieb Peter Portante:
Signed-off-by: Peter Portantepeter.porta...@redhat.com
Fix itself looks okay, but author, maintainer and qemu-ppc were missing
in CC, and a
On 20.04.2012 15:51, Evgeny Voevodin wrote:
On 20.04.2012 15:13, Peter Maydell wrote:
On 17 April 2012 06:41, Evgeny Voevodine.voevo...@samsung.com wrote:
+static Property exynos4210_irq_gate_properties[] = {
+ DEFINE_PROP_UINT32(n_in, Exynos4210IRQGateState, n_in, 1),
+
57 matches
Mail list logo