For background of CELF project proposals, see:
http://elinux.org/CELF_Open_Project_Proposal_2010
Summary:
Integrate a flattened device tree parser into the emulator QEMU, so QEMU can
create board emulations on the fly (at runtime) from the same data files the
Linux kernel uses to attach driv
Everytime I try to send a patch using 'git send-email', I receive
this error:
: host mx10.gnu.org[199.232.76.166] said:
550-Verification failed for
550-Unrouteable address 550 Sender verify failed (in reply to
RCPT TO
command)
How do I make it so that I can send patches using g
This patch allows for the file vl.c to compile without any warnings.
Signed-off-by: John Arbuckle
---
vl.c |5 -
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/vl.c b/vl.c
index c0d98f5..b466fe0 100644
--- a/vl.c
+++ b/vl.c
@@ -107,16 +107,19 @@ extern int madvise(cadd
An initial cut at conditional moves for the sparc backend.
Untested, as I don't have sparc hardware and the build system
resists attempts at cross-compilation.
Note fixes to tcg_out_movi_imm32 (wrong check_fit_tl width),
use of TCG_TARGET_REG_BITS == 64 tests instead of explicitly
checking for __
Uses setcond in the many branch condition generators and movcond
in the conditional move expanders.
Signed-off-by: Richard Henderson
---
target-mips/translate.c | 124 +--
1 files changed, 66 insertions(+), 58 deletions(-)
diff --git a/target-mips/tr
Splits out the condition code handling into a new function that's
directly callable from setcc and cmov expanders. From there we can
directly emit the operation we care about.
Signed-off-by: Richard Henderson
---
target-i386/translate.c | 192 +-
tcg
An initial cut at conditional moves for the i386 backend.
Signed-off-by: Richard Henderson
---
elf.h |2 +
tcg/i386/tcg-target.c | 280 -
2 files changed, 233 insertions(+), 49 deletions(-)
diff --git a/elf.h b/elf.h
index 116
Limited usage of setcond/movcond to enable testing in the code generator.
Signed-off-by: Richard Henderson
---
target-alpha/translate.c | 66 +
1 files changed, 31 insertions(+), 35 deletions(-)
diff --git a/target-alpha/translate.c b/target-alpha/t
This patch series adds support for setcond (aka setcc) and
movcond (aka cmov) opcodes to TCG.
These new opcodes are considered "required" by the backend,
because expanding them at the tcg level breaks the basic block.
There might be some way to emulate within tcg internals, but
that doesn't seem w
Implement conditional moves in the x86_64 backend.
Signed-off-by: Richard Henderson
---
tcg/x86_64/tcg-target.c | 65 --
1 files changed, 62 insertions(+), 3 deletions(-)
diff --git a/tcg/x86_64/tcg-target.c b/tcg/x86_64/tcg-target.c
index 2339091..
Defines setcond and movcond for implementing conditional moves at
the tcg opcode level. 64-bit-on-32-bit is expanded via a setcond2
primitive plus other operations.
Signed-off-by: Richard Henderson
---
tcg/README| 16 ++-
tcg/tcg-op.h | 87 ++
On Wed, Dec 16, 2009 at 05:22:41PM +0100, Gerd Hoffmann wrote:
>> Right now, qemu cfg uses two ports - a file select port (0x510) and a
>> data port (0x511). Perhaps two new ports could be added - a file name
>> port (0x0512) and a file length port (0x513).
>>
>> Basically, if there is some way fo
Am 16.12.2009 um 18:46 schrieb Gerd Hoffmann :
Hi,
quick draft of a file xfer api for fw_cfg, qemu side only, not tested
yet, to be used for option roms.
comments?
2/3 is missing here.
How does elf rom loading come into play here? We'll need both - a
firmware rom blob (openbios) and a f
Hi,
Spice Windows binaries are now available for download from
http://spice-space.org/download.html. The binaries include among
others the QXL driver, vdi_port driver, and the Spice agent
vd_agent (windows service). In order to evaluate Spice you need
to install all three on WinXP. A user manual
The function do_unassigned_access in op_helper.c looks very similar
to the function cpu_sparc_handle_mmu_fault in helper.c and it is
called after the later one.
Why do we need to set env->mmuregs[3] and env->mmuregs[4] twice?
Setting the SFSR and SFAR in do_unassigned_access looks redundant to
m
Signed-off-by: Gerd Hoffmann
---
hw/fw_cfg.c | 26 --
hw/fw_cfg.h | 16 +---
2 files changed, 21 insertions(+), 21 deletions(-)
diff --git a/hw/fw_cfg.c b/hw/fw_cfg.c
index b25afff..2e3662d 100644
--- a/hw/fw_cfg.c
+++ b/hw/fw_cfg.c
@@ -45,11 +45,11 @@ ty
Signed-off-by: Gerd Hoffmann
---
hw/fw_cfg.c | 22 ++
hw/fw_cfg.h | 22 +-
2 files changed, 43 insertions(+), 1 deletions(-)
diff --git a/hw/fw_cfg.c b/hw/fw_cfg.c
index 2e3662d..1dd7d6a 100644
--- a/hw/fw_cfg.c
+++ b/hw/fw_cfg.c
@@ -47,6 +47,7 @@ typ
Signed-off-by: Gerd Hoffmann
---
hw/loader.c | 23 ---
hw/loader.h |5 +++--
hw/pc.c |2 ++
3 files changed, 25 insertions(+), 5 deletions(-)
diff --git a/hw/loader.c b/hw/loader.c
index 2d7a2c4..d1d05ed 100644
--- a/hw/loader.c
+++ b/hw/loader.c
@@ -48,6 +48,7
Hi,
quick draft of a file xfer api for fw_cfg, qemu side only, not tested
yet, to be used for option roms.
comments?
cheers,
Gerd
2009/11/4 Blue Swirl :
> On Wed, Nov 4, 2009 at 1:58 AM, Artyom Tarasenko
> wrote:
>> The page 108 of the SPARC Version 8 Architecture Manual describes
>> that addcc and addxcc shall compute carry flag the same way.
>> The page 110 claims the same about subcc and subxcc instructions.
>> This patch
Right now, qemu cfg uses two ports - a file select port (0x510) and a
data port (0x511). Perhaps two new ports could be added - a file name
port (0x0512) and a file length port (0x513).
Basically, if there is some way for SeaBIOS to walk a list of "files"
in the "qemu cfg" space, then it should
On Wed, Dec 16, 2009 at 04:41:33PM +0200, Michael S. Tsirkin wrote:
> On Wed, Dec 16, 2009 at 03:41:22PM +0100, Gerd Hoffmann wrote:
> > On 12/15/09 22:41, Anthony Liguori wrote:
> >> BTW, I'm pretty sure this style of option rom loading (from a PCI
> >> device) is going to be required for device p
On 12/15/09 23:51, Sebastian Herbszt wrote:
Qemu will map rom1.bin to PC_ROM_MIN_OPTION (0xc8000) and map rom2.bin
to 0xd.
No.
rom1.bin will be loaded to max(0xc8000, 0xc + cirrus-bios-size)
aligned to 2k, which (with cirrus) is 0xc9.
rom2.bin will be loaded after rom1.bin (also 2
Gerd Hoffmann writes:
> On 12/16/09 11:03, Markus Armbruster wrote:
>> This fixes "-global isa-fdc,driveB=DRIVE-ID" for me, but problems
>> remain:
>>
>> * -drive if=none,id=fd,file=fd.img -global isa-fdc.driveA=fd
>>
>>dev: isa-fdc, id ""
>> dev-prop: driveA = floppy0
>>
On Wed, Dec 16, 2009 at 03:41:22PM +0100, Gerd Hoffmann wrote:
> On 12/15/09 22:41, Anthony Liguori wrote:
>> BTW, I'm pretty sure this style of option rom loading (from a PCI
>> device) is going to be required for device passthrough if we want to
>> support running those roms in the guests.
>
> We
On 12/15/09 22:41, Anthony Liguori wrote:
BTW, I'm pretty sure this style of option rom loading (from a PCI
device) is going to be required for device passthrough if we want to
support running those roms in the guests.
Well, qemu-kvm has quite some code to poke the rom out of /proc/bus/pci
and
Gerd Hoffmann wrote:
> On 12/15/09 18:30, Wolfgang Mauerer wrote:
>> Hi Gerd,
>>
>> in commit 5b684b5a56e81f6f, you introduced an explicit check
>> to prevent adding SCSI controllers without attached disks
>> to the system.
>
> There was a patch from Daniel removing that check, isn't that one merg
On Wed, Dec 16, 2009 at 04:28:49PM +0200, Gleb Natapov wrote:
> On Wed, Dec 16, 2009 at 04:24:59PM +0200, Michael S. Tsirkin wrote:
> > On Wed, Dec 16, 2009 at 04:18:49PM +0200, Gleb Natapov wrote:
> > > On Wed, Dec 16, 2009 at 04:15:16PM +0200, Michael S. Tsirkin wrote:
> > > > On Wed, Dec 16, 200
On Wed, Dec 16, 2009 at 04:24:59PM +0200, Michael S. Tsirkin wrote:
> On Wed, Dec 16, 2009 at 04:18:49PM +0200, Gleb Natapov wrote:
> > On Wed, Dec 16, 2009 at 04:15:16PM +0200, Michael S. Tsirkin wrote:
> > > On Wed, Dec 16, 2009 at 04:17:11PM +0200, Gleb Natapov wrote:
> > > > On Wed, Dec 16, 200
On Wed, Dec 16, 2009 at 04:18:49PM +0200, Gleb Natapov wrote:
> On Wed, Dec 16, 2009 at 04:15:16PM +0200, Michael S. Tsirkin wrote:
> > On Wed, Dec 16, 2009 at 04:17:11PM +0200, Gleb Natapov wrote:
> > > On Wed, Dec 16, 2009 at 09:12:21AM -0500, Kevin O'Connor wrote:
> > > > On Wed, Dec 16, 2009 at
Anthony Liguori wrote:
Gleb Natapov wrote:
I thought KVM emulates the syscall instruction? I swear I've seen
patches for that.
It is. Starting from 2.6.32.
Okay, so this is a performance vs. migration compatibility issue then?
BTW, couldn't we just not advertise syscall in cpuid?
We really need compile _all_ sources for user target with -fpie when
use --enable-user-pie.
It's regression introduced by commit add16157d72454.
Signed-off-by: Kirill A. Shutemov
Cc: Blue Swirl
Signed-off-by: Juan Quintela
---
Makefile.user |7 ++-
configure |3 +++
2 files ch
On Wed, Dec 16, 2009 at 04:15:16PM +0200, Michael S. Tsirkin wrote:
> On Wed, Dec 16, 2009 at 04:17:11PM +0200, Gleb Natapov wrote:
> > On Wed, Dec 16, 2009 at 09:12:21AM -0500, Kevin O'Connor wrote:
> > > On Wed, Dec 16, 2009 at 03:52:34PM +0200, Michael S. Tsirkin wrote:
> > > > I am mostly conce
On Wed, Dec 16, 2009 at 04:17:11PM +0200, Gleb Natapov wrote:
> On Wed, Dec 16, 2009 at 09:12:21AM -0500, Kevin O'Connor wrote:
> > On Wed, Dec 16, 2009 at 03:52:34PM +0200, Michael S. Tsirkin wrote:
> > > I am mostly concerned with migrating between qemu versions with
> > > different roms, while g
On Wed, Dec 16, 2009 at 09:12:21AM -0500, Kevin O'Connor wrote:
> On Wed, Dec 16, 2009 at 03:52:34PM +0200, Michael S. Tsirkin wrote:
> > I am mostly concerned with migrating between qemu versions with
> > different roms, while guest was in the middle of running ROM.
> > This might be solved if we
On Wed, Dec 16, 2009 at 09:12:21AM -0500, Kevin O'Connor wrote:
> On Wed, Dec 16, 2009 at 03:52:34PM +0200, Michael S. Tsirkin wrote:
> > I am mostly concerned with migrating between qemu versions with
> > different roms, while guest was in the middle of running ROM.
> > This might be solved if we
Signed-off-by: Gerd Hoffmann
---
hw/cirrus_vga.c |6 +-
hw/e1000.c |5 +
hw/pci.c| 17 +++--
hw/pci.h|5 -
hw/rtl8139.c|5 +
hw/virtio-pci.c |5 +
6 files changed, 19 insertions(+), 24 deletions(-)
diff --git a/hw/ci
From: Anthony Liguori
Currently, we preload option roms into the option rom space in memory. This
prevents DDIM from functioning correctly which severely limits the number
of roms we can support.
This patch introduces a pci_add_option_rom() which registers the
PCI_ROM_ADDRESS bar which points t
Signed-off-by: Gerd Hoffmann
---
hw/cirrus_vga.c |4 +++-
hw/e1000.c |8 ++--
hw/pci.c| 19 ---
hw/pci.h|3 +--
hw/rtl8139.c|8 ++--
hw/virtio-pci.c |8 ++--
6 files changed, 26 insertions(+), 24 deletions(-)
diff --git
Hi,
Current state of the rom loading patches for testing and discussions.
What is in here:
#1 new seabios binary for testing convinience, git head as of yesterday
plus anthonys patches
#2 anthonys initial patch, unmodified.
#3 romfile property patch, was already on the list but deep in a
On Wednesday 16 December 2009, Leonid Grossman wrote:
> > > 3. Doing the bridging in the NIC using macvlan in passthrough
> > > mode. This lowers the CPU utilization further compared to 2,
> > > at the expense of limiting throughput by the performance of
> > > the PCIe interconnect to the adapter.
On Wed, Dec 16, 2009 at 03:52:34PM +0200, Michael S. Tsirkin wrote:
> I am mostly concerned with migrating between qemu versions with
> different roms, while guest was in the middle of running ROM.
> This might be solved if we migrated ROM content together with
> the device and put some padding in
On Wed, Dec 16, 2009 at 02:42:51PM +0100, Gerd Hoffmann wrote:
> Hi,
>
>> What will happen when we find a bug in one of ROMs
>> I wonder? I think we will need to keep the old ROM
>> around, and put it in as part of compat machine type?
>
> What do you want?
> Old way to load the rom?
> Old rom bi
Hi,
What will happen when we find a bug in one of ROMs
I wonder? I think we will need to keep the old ROM
around, and put it in as part of compat machine type?
What do you want?
Old way to load the rom?
Old rom binaries (i.e. etherboot)?
Both?
The new way to load the roms is a guest-visible
On 12/16/09 11:03, Markus Armbruster wrote:
This fixes "-global isa-fdc,driveB=DRIVE-ID" for me, but problems
remain:
* -drive if=none,id=fd,file=fd.img -global isa-fdc.driveA=fd
dev: isa-fdc, id ""
dev-prop: driveA = floppy0
dev-prop: driveB =
Split default_drive into default_{floppy,cdrom,sdcard}.
Also add QEMUMachine flags to disable them per machine.
Signed-off-by: Gerd Hoffmann
---
hw/boards.h |5 -
vl.c| 23 ---
2 files changed, 24 insertions(+), 4 deletions(-)
diff --git a/hw/boards.h b/hw/
Add isa-fdc (disables default_floppy).
Add ide-drive (disables default_cdrom).
Also walk the -global QemuOpts, so we'll catch
-global isa-fdc.drive{A,B}= too.
Signed-off-by: Gerd Hoffmann
---
qemu-config.h |1 +
vl.c |3 +++
2 files changed, 4 insertions(+), 0 deletions(-)
dif
Add a error message in case we fail to parse a qdev property.
Also make qemu not abort() in case setting a global property can't be
set. This used to be a clear programming error. The introduction of
the -global switch changed that though, so better exit instead (after
printing the new error mes
Gerd Hoffmann writes:
> markus has changed vga names recently and forgot to adapt default_list.
I git-grepped all the changed names then. Either I screwed that up, or
the default list went in after I grepped.
Anyway, I grepped again, and found no additional omissions.
markus has changed vga names recently and forgot to adapt default_list.
Cc: Markus Armbruster
Signed-off-by: Gerd Hoffmann
---
vl.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/vl.c b/vl.c
index 7099231..0318c56 100644
--- a/vl.c
+++ b/vl.c
@@ -288,8 +288,8 @@ sta
Hi,
Basically, if there is some way for SeaBIOS to walk a list of "files"
in the "qemu cfg" space, then it should be straight forward to enhance
the existing code in seabios to extract and deploy roms in addition to
those found in the PCI bar.
I think using fw_config is the only sane way to
Kevin Wolf writes:
> Am 16.12.2009 10:51, schrieb Markus Armbruster:
>> Kevin Wolf writes:
>>
>>> The multiboot implementation assumed that there is only one program header
>>> (which contains the entry point) and that the entry point is at the start of
>>> the code. This doesn't hold true gene
On 12/16/09 05:29, Kevin O'Connor wrote:
On Tue, Dec 15, 2009 at 10:24:29PM +0100, Sebastian Herbszt wrote:
Keep loading custom roms (e.g. from -option-rom) with rom_add_file
starting at 0xc8000. Modify SeaBIOS to scan the memory range for
pre-deployed option roms before deploying PCI roms. Sea
On 12/15/09 18:30, Wolfgang Mauerer wrote:
Hi Gerd,
in commit 5b684b5a56e81f6f, you introduced an explicit check
to prevent adding SCSI controllers without attached disks
to the system.
There was a patch from Daniel removing that check, isn't that one merged
meanwhile? Hmm, looks like it isn
Am 15.12.2009 19:33, schrieb Jamie Lokier:
> Shared backing disks aren't safe after "commit" anyway. Other VMs may
> not be running at the time "commit" renders their image corrupt, so
> locks don't offer adequate protection against the backing disk being changed.
>
> One strategy that would offe
Am 16.12.2009 10:51, schrieb Markus Armbruster:
> Kevin Wolf writes:
>
>> The multiboot implementation assumed that there is only one program header
>> (which contains the entry point) and that the entry point is at the start of
>> the code. This doesn't hold true generally and caused too little
On 12/15/2009 11:26 PM, Jamie Lokier wrote:
When is REX_X(s) not zero, and does it break the index != 4 test?
Some rex bits do affect these tests, and some don't. It's the usual
'consistently inconsistent' rules of x86.
--
error compiling committee.c: too many arguments to function
"Kirill A. Shutemov" writes:
> On Wed, Dec 16, 2009 at 11:15 AM, Paolo Bonzini wrote:
>> On 12/16/2009 09:45 AM, Kirill A. Shutemov wrote:
>>
>> >>> -VPATH=$(SRC_PATH)
>> >>> +vpath %.c $(SRC_PATH)
>
> This change is needed?
>>>
>>> Yes. We dont want to take %.o fro
Gerd Hoffmann writes:
> Fix the floppy controller init wrappers to set the drive properties
> only in case the DriveInfo pointers passed in are non NULL. This allows
> to set the properties using -global.
>
> Signed-off-by: Gerd Hoffmann
This fixes "-global isa-fdc,driveB=DRIVE-ID" for me, but
Is there anything I can provide to help testing possible patches?
A good first step is to write those possible patches. It shouldn't be
difficult, start in target-i386/translate.c:disas_insn().
And see Jamie's suggestion at
http://permalink.gmane.org/gmane.comp.emulators.qemu/59522
which i
Kevin Wolf writes:
> The multiboot implementation assumed that there is only one program header
> (which contains the entry point) and that the entry point is at the start of
> the code. This doesn't hold true generally and caused too little data to be
> loaded.
Out of curiosity: does this affec
On Wed, Dec 16, 2009 at 11:15 AM, Paolo Bonzini wrote:
> On 12/16/2009 09:45 AM, Kirill A. Shutemov wrote:
>
> >>> -VPATH=$(SRC_PATH)
> >>> +vpath %.c $(SRC_PATH)
>>>
>>> >
>>> > This change is needed?
>>
>> Yes. We dont want to take %.o from $(SRC_PATH) since they built without
>>
On 12/16/2009 09:45 AM, Kirill A. Shutemov wrote:
>>> -VPATH=$(SRC_PATH)
>>> +vpath %.c $(SRC_PATH)
>
> This change is needed?
Yes. We dont want to take %.o from $(SRC_PATH) since they built without -fpie.
I think that's fine, it's kind of the purpose of vpath to only look for
sources in
On 12/16/2009 10:56 AM, Clemens Kolbitsch wrote:
On Tuesday 15 December 2009 08:54:04 pm Avi Kivity wrote:
On 12/15/2009 08:48 PM, Clemens Kolbitsch wrote:
Hi list,
I'm experiencing a strange emulation bug with the op-code below. The
instruction raises a segfault in the application (
On Tuesday 15 December 2009 08:54:04 pm Avi Kivity wrote:
> On 12/15/2009 08:48 PM, Clemens Kolbitsch wrote:
> > Hi list,
> >
> > I'm experiencing a strange emulation bug with the op-code below. The
> > instruction raises a segfault in the application (running on the guest),
> > however, if I enabl
On Tue, Dec 15, 2009 at 10:55 PM, Juan Quintela wrote:
> "Kirill A. Shutemov" wrote:
>> Please, apply it.
>>
>> On Tue, Dec 15, 2009 at 3:10 AM, Kirill A. Shutemov
>> wrote:
>>> We really need compile _all_ sources for user target with -fpie when
>>> use --enable-user-pie.
>>>
>>> It's regressio
66 matches
Mail list logo
