On 2012-10-21 14:26, Aurelien Jarno wrote:
-if (TCG_TARGET_REG_BITS == 64) {
-/* ??? We assume all operations have left us with register
- contents that are zero extended. So far this appears to
- be true. If we want to enforce this, we can
On 2012-10-21 03:28, Aurelien Jarno wrote:
Now that CONFIG_TCG_PASS_AREG0 has been removed, it's easier to get
an optimal code for the load/store functions.
First swap the two registers used in tcg_out_tlb_load() so that the
address end-up in the second register instead of the first one.
On 2012-10-21 03:28, Aurelien Jarno wrote:
On x86_64, remove the constraint on the third argument register which
is not needed:
- For loads the helper arguments are env, addr, mem_idx. The addr
value should not be in the two first argument registers as they are
used in
On 20.10.2012 00:43, Jason Baron wrote:
From: Jason Baron jba...@redhat.com
If -L dir is specified, and qemu does not find the bios file in dir, then
the search fails. Add infrastructure such that the search will continue in
the default paths, if not found in the -L path.
Reviewed-by:
On 21 October 2012 08:26, Michael Tokarev m...@tls.msk.ru wrote:
On 20.10.2012 00:43, Jason Baron wrote:
+char *qemu_find_file(int type, const char *name)
+{
+char *filename;
+
+/* Try the name as a straight path first */
+if (access(name, R_OK) == 0) {
+return
On Fri, Oct 19, 2012 at 04:43:38PM -0400, Jason Baron wrote:
From: Jan Kiszka jan.kis...@siemens.com
Both bits are added to the write-1-to-clear mask by default. As the
smbus device does not allow writes at all, we have to remove it from
that mask, also to avoid triggering a runtime
On Fri, Oct 19, 2012 at 04:43:25PM -0400, Jason Baron wrote:
Hi,
Qemu bits for q35 support, I'm posting the seabios changes separately. The
patches require '-M pc_q35' and -L 'seabios dir with q35 changes' on the
qemu command line. Hopefully, we can make it the default for x86 at some
On Fri, Oct 19, 2012 at 04:43:25PM -0400, Jason Baron wrote:
Hi,
Qemu bits for q35 support, I'm posting the seabios changes separately. The
patches require '-M pc_q35' and -L 'seabios dir with q35 changes' on the
qemu command line. Hopefully, we can make it the default for x86 at some
On 10/18/2012 09:29 AM, Juan Quintela wrote:
From: Paolo Bonzini pbonz...@redhat.com
Outside the execution threads the normal, non-MRU-ized order of
the RAM blocks should always be enough. So manage two separate
lists, which will have separate locking rules.
Signed-off-by: Paolo Bonzini
On 10/18/2012 09:29 AM, Juan Quintela wrote:
From: Umesh Deshpande udesh...@redhat.com
Add the new mutex that protects shared state between ram_save_live
and the iothread. If the iothread mutex has to be taken together
with the ramlist mutex, the iothread shall always be _outside_.
On 10/18/2012 09:30 AM, Juan Quintela wrote:
We still protect everything except the wait with the iothread lock.
But we moved from a timer to a thread. Steps one by one.
We also need to detect when we have finished with a variable complete.
Signed-off-by: Juan Quintela quint...@redhat.com
On 10/18/2012 09:30 AM, Juan Quintela wrote:
We want the file assignment to happen before the thread is created to
avoid locking, so we just do it before creating the thread.
Signed-off-by: Juan Quintela quint...@redhat.com
---
buffered_file.c | 13 ++---
buffered_file.h | 2 +-
On 10/18/2012 09:29 AM, Juan Quintela wrote:
From: Umesh Deshpande udesh...@redhat.com
This will be used to detect if last_block might have become invalid
across different calls to ram_save_live.
Signed-off-by: Paolo Bonzini pbonz...@redhat.com
Signed-off-by: Umesh Deshpande
On 10/18/2012 09:30 AM, Juan Quintela wrote:
Instead of testing each page individually, we search what is the next
dirty page with a bitmap operation. We have to reorganize the code to
move from a for loop, to a while(dirty) loop.
Signed-off-by: Juan Quintela quint...@redhat.com
---
Signed-off-by: Lei Li li...@linux.vnet.ibm.com
---
qemu-char.c | 72 +++
1 files changed, 72 insertions(+), 0 deletions(-)
diff --git a/qemu-char.c b/qemu-char.c
index b082bae..b174da1 100644
--- a/qemu-char.c
+++ b/qemu-char.c
@@ -2588,6
Signed-off-by: Lei Li li...@linux.vnet.ibm.com
---
qemu-char.c | 36
qemu-config.c |3 +++
qemu-options.hx | 10 ++
3 files changed, 49 insertions(+), 0 deletions(-)
diff --git a/qemu-char.c b/qemu-char.c
index b174da1..381bf60 100644
This RFC series attempts to convert the MemCharDriver to use a circular
buffer for input and output, expose it to users by introducing QMP commands
memchar_write and memchar_read and via the command line like the other
CharDriverStates.
Serial ports in qemu always use CharDriverStates as there
Signed-off-by: Lei Li li...@linux.vnet.ibm.com
---
hmp-commands.hx | 25 +
hmp.c| 17 +
hmp.h|1 +
qapi-schema.json | 27 +++
qemu-char.c | 52
Signed-off-by: Lei Li li...@linux.vnet.ibm.com
---
hmp-commands.hx | 22 +
hmp.c| 19 +++
hmp.h|1 +
qapi-schema.json | 69 ++
qemu-char.c | 46
Signed-off-by: Lei Li li...@linux.vnet.ibm.com
---
hmp-commands.hx | 23 +++
hmp.c | 53 +
hmp.h |1 +
monitor.c | 15 +++
monitor.h |3 +++
5 files changed, 95
On 18 October 2012 08:29, Juan Quintela quint...@redhat.com wrote:
-RAMList ram_list = { .blocks = QLIST_HEAD_INITIALIZER(ram_list.blocks) };
+RAMList ram_list = {
+.blocks = QLIST_HEAD_INITIALIZER(ram_list.blocks),
+.blocks_mru = QLIST_HEAD_INITIALIZER(ram_list.blocks_mru)
+};
Maybe
On 19.10.2012 23:33, Aurelien Jarno wrote:
On Wed, Oct 17, 2012 at 07:09:25PM +0200, Jan Kiszka wrote:
We need to evaluate $libexecdir in configure, otherwise we literally end
up with ${prefix}/libexec instead of the absolute path as
CONFIG_QEMU_HELPERDIR.
Signed-off-by: Jan Kiszka
On 2012-10-20 07:38, Aurelien Jarno wrote:
This patch series rework the liveness analysis and register allocator
in order to generate more optimized code, by avoiding a lot of move
instructions. I have measured a 9% performance improvement in user mode
and 4% in system mode.
The idea behind
On 2012-10-21 14:26, Aurelien Jarno wrote:
+static inline void setup_guest_base_seg(void)
+{
+if (syscall(__NR_arch_prctl, ARCH_SET_GS, GUEST_BASE) == 0) {
+guest_base_flags = P_GS;
+}
Why calling the syscall directly instead of using arch_prctl(2)?
Err... there is no
On 2012-10-20 07:25, Aurelien Jarno wrote:
MIPS32 and later instruction sets have a multiplication instruction
directly operating on GPRs. It only produces a 32-bit result but
it is exactly what is needed by QEMU.
Signed-off-by: Aurelien Jarno aurel...@aurel32.net
---
Reviewed-by: Richard
Ping? It's been Reviewed-by...
r~
On 2012-10-13 02:26, Richard Henderson wrote:
Older glibc (RHEL 5.x, Debian 5.x) does not have the _sigev_un._tid
member in its structure definition, while the accompanying kernel
headers do define SIGEV_THREAD_ID. We need configure to check for
both
On 2012-10-21 09:17, Aurelien Jarno wrote:
On Sun, Oct 21, 2012 at 08:48:52AM +1000, Richard Henderson wrote:
On 2012-10-21 00:48, Aurelien Jarno wrote:
I am not sure it is the real problem, but at least the optimization of
using the destination register as a temporary is wrong when the
On Mon, Oct 22, 2012 at 07:29:19AM +1000, Richard Henderson wrote:
On 2012-10-21 09:17, Aurelien Jarno wrote:
On Sun, Oct 21, 2012 at 08:48:52AM +1000, Richard Henderson wrote:
On 2012-10-21 00:48, Aurelien Jarno wrote:
I am not sure it is the real problem, but at least the optimization of
On Fri, Oct 12, 2012 at 09:26:55AM -0700, Richard Henderson wrote:
Older glibc (RHEL 5.x, Debian 5.x) does not have the _sigev_un._tid
member in its structure definition, while the accompanying kernel
headers do define SIGEV_THREAD_ID. We need configure to check for
both before using it.
OpenBIOS on sparc64 only support Standard VGA and not Cirrus VGA. Don't
build Cirrus VGA support so that it can't be selected.
This fixes the breakage introduced by commit f2898771.
Reported-by: Richard Henderson r...@twiddle.net
Cc: Blue Swirl blauwir...@gmail.com
Signed-off-by: Aurelien Jarno
When we allocate a reserved_va for the guest, the kernel will likely
choose an address well above 4G. At which point we must use a pair
of movabsq+addq to form the host address. If we have OS support,
set up a segment register to point to guest_base instead.
Signed-off-by: Richard Henderson
On Thu, Oct 18, 2012 at 04:50:25PM +1100, David Gibson wrote:
At present, using 'system_powerdown' from the monitor or otherwise
instructing qemu to (cleanly) shut down a pseries guest will not work,
because we did not have a method of signalling the shutdown request to the
guest.
PAPR does
On 10/19/2012 09:55 PM, Blue Swirl wrote:
On Fri, Oct 19, 2012 at 12:42 PM, Evgeny e.voevo...@samsung.com wrote:
This set of patches moves global variables to tcg_ctx:
gen_opc_ptr
gen_opparam_ptr
gen_opc_buf
gen_opparam_buf
Where it was possible I used s-...
Where we don't have an interface to
On 10/19/2012 09:53 PM, Blue Swirl wrote:
On Fri, Oct 19, 2012 at 12:42 PM, Evgeny e.voevo...@samsung.com wrote:
Signed-off-by: Evgeny e.voevo...@samsung.com
---
tcg/tcg.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/tcg/tcg.c b/tcg/tcg.c
index
On 10/22/2012 07:40 AM, Evgeny Voevodin wrote:
On 10/19/2012 09:55 PM, Blue Swirl wrote:
On Fri, Oct 19, 2012 at 12:42 PM, Evgeny e.voevo...@samsung.com wrote:
This set of patches moves global variables to tcg_ctx:
gen_opc_ptr
gen_opparam_ptr
gen_opc_buf
gen_opparam_buf
Where it was possible
On 10/19/12 19:02, Stefano Stabellini wrote:
On Wed, 17 Oct 2012, Gerd Hoffmann wrote:
Currently it is inconstent, PixelFormat-amask is left unset whereas
abits and amax and ashift are filled. As an alpha channel doesn't make
sense for the vga framebuffer leave all alpha fields clear.
On 10/19/12 20:04, Stefano Stabellini wrote:
On Wed, 17 Oct 2012, Gerd Hoffmann wrote:
The vnc code uses *three* DisplaySurfaces:
First is the surface of the actual QemuConsole, usually the guest
screen, but could also be a text console (monitor/serial reachable via
Ctrl-Alt-nr keys). This
Hi,
+pci_create_simple_multifunction(
+host_bus, PCI_DEVFN(ICH9_USB_DEV, ICH9_USB_EHCI1_FUNC),
+true, ich9-usb-ehci1);
/* XXX: EHCI */
You can remove that comment now ;)
cheers,
Gerd
Hi,
Would it make sense to temporarily rename the machine type e.g.
pc-q35-experimental to stress it's not fully supported?
I don't think this is needed as piix will continue to be the default.
It will also cause trouble with libvirt when pc-q35-experimental goes
away some day.
cheers,
On Mon, Oct 22, 2012 at 12:11:07PM +1000, Richard Henderson wrote:
When we allocate a reserved_va for the guest, the kernel will likely
choose an address well above 4G. At which point we must use a pair
of movabsq+addq to form the host address. If we have OS support,
set up a segment
40 matches
Mail list logo
