Stop reinventing the wheel. Use the pixman library for raster ops.
Signed-off-by: Gerd Hoffmann
---
configure | 12
ui/fbdev.c | 172 +++
2 files changed, 114 insertions(+), 70 deletions(-)
diff --git a/configure b/configure
inde
Add support for scaling the guest display.
Ctrl-Alt-S hotkey toggles scaling.
Signed-off-by: Gerd Hoffmann
---
ui/fbdev.c | 61 ++-
1 files changed, 51 insertions(+), 10 deletions(-)
diff --git a/ui/fbdev.c b/ui/fbdev.c
index 6835fef..55
Display works, requires truecolor framebuffer with 16 or 32 bpp on the
host. 32bpp is recommended. The framebuffer is used as-is, qemu
doesn't try to switch modes. With LCD displays mode switching is pretty
pointless IMHO, also it wouldn't work anyway with the most common
fbdev drivers (vesafb,
Also change the way the gui_timer is initialized: each time a
displaychangelistener is registered or unregistered we'll check
whenever we need a timer (due to dpy_refresh callback being present)
and if so setup a timer, otherwise zap it. This way the gui timer works
correctly with displaychangelis
When adding DisplayChangeListeners the set_mouse and cursor_define
callbacks have been left in DisplayState for some reason. Fix it.
Signed-off-by: Gerd Hoffmann
---
console.c |2 +-
console.h | 39 +++
hw/jazz_led.c |2 +-
hw
Add CONFIG_FBDEV, add --enable-fbdev and --disable-fbdev
configure switches so fbdev can be enabled/disabled at
compile time.
Signed-off-by: Gerd Hoffmann
---
configure| 12
qmp.c|2 +-
ui/Makefile.objs |2 +-
vl.c |4 ++--
4 files chang
This patch adds a fbdev monitor command to enable/disable
the fbdev display at runtime to both qmp and hmp.
qmp: fbdev enable=on|off
hmp: fbdev on|off
Signed-off-by: Gerd Hoffmann
---
hmp-commands.hx | 15 +++
hmp.c|9 +
hmp.h|1 +
qapi-sch
Signed-off-by: Gerd Hoffmann
---
console.h | 72 +++
hw/xenfb.c |2 +-
vl.c |9 ++-
3 files changed, 42 insertions(+), 41 deletions(-)
diff --git a/console.h b/console.h
index f990684..646ad4b 100644
--- a/console.h
+++
Add mouse_set and cursor_define DisplayChangeListener callbacks
and mouse pointer rendering support.
Signed-off-by: Gerd Hoffmann
---
ui/fbdev.c | 95
1 files changed, 95 insertions(+), 0 deletions(-)
diff --git a/ui/fbdev.c b/ui/fb
Hi,
Third round of the framebuffer display driver patches, including git
tree for pull as I think it's ready now.
Changes: Addressed review comments from Markus. Catch a few more fatal
signals, especially SIGABRT, so fbdev restores your console when qemu
runs into an assert(). Misc little tw
Il 18/09/2012 08:27, Alex Barcelo ha scritto:
>
> I saw some things[1] about multiple vcpu, smp and things like that. It
> seemed to me that --enable-io-thread enables it.
iothread means that the QEMU main thread only services an event loop
(I/O, bottom halves, timers, etc.). Running CPUs is off
Am 17.09.2012 18:43, schrieb Paolo Bonzini:
> Il 07/09/2012 18:11, Kevin Wolf ha scritto:
>> I was planning to review it in more detail next week, but I just had a
>> quick look. I'm not sure if automatically shutting down the NBD server
>> when the guest stops using it is always right (for removab
Libqblock was placed in new directory ./libqblock, libtool will build
dynamic library there, source files of block layer remains in ./block.
So block related source code will generate 3 sets of binary, first is old
ones used in qemu, second and third are non PIC and PIC ones in ./libqblock.
GCC
This patch introduce libqblock API, make libqblock.la and make check-libqblock
could build this library.
Functionalities:
1 create a new image.
2 sync access of an image.
3 basic image information retrieving such as backing file.
4 detect if a sector is allocated in an image.
Supported Format
> > Luckily removable media are usually not too interesting, so a
> > slightly suboptimal behavior is okay as long as it does not break the
> > important use cases---mostly migration without shared storage, where also
> > uninteresting images have to be mirrored or exposed via NBD. Those
> > shou
On 2012-09-18 03:52, Wen Congyang wrote:
> At 09/18/2012 01:56 AM, Luiz Capitulino Wrote:
>> Hi Wen,
>>
>> We've re-reviewed the dump-guest-memory command and found some
>> possible issues with the -p option.
>>
>> The main issue is that it seems possible for a malicious guest to set
>> page tables
This patch contains type and defines used in APIs, one file for public usage
by user, one for libqblock internal usage.
Signed-off-by: Wenchao Xia
---
libqblock/libqblock-internal.h | 56 +
libqblock/libqblock-types.h| 268
2 files change
Hi,
On 09/17/2012 06:22 PM, Jan Kiszka wrote:
If that does not work, add the debug parameter to the usb-redir device,
set it
to 4, collect logs of trying to redirect the device and send me the logs
please, ie:
-device usb-redir,chardev=usbredirchardev1,id=usbredirdev1,debug=4
Also be aware that
Created a new directory in tests, make chekc-libqblock will build an
executable binrary, make clean will delete it.
Signed-off-by: Wenchao Xia
---
.gitignore |1 +
Makefile |1 +
tests/Makefile |3 +++
tests/libqblock/
This patch contains the major APIs in the library.
Important APIs:
1 QBroker. These structure was used to retrieve errors, every thread must
create one first, later maybe thread related staff could be added into it.
2 QBlockState. It stands for an block image object.
3 QBlockStaticInfo. It
In this example, user first create two qcow2 images, and then get the
backing file relationship information of them. Then does write and read
sync IO on them.
Signed-off-by: Wenchao Xia
---
tests/libqblock/libqblock-test.c | 233 ++
1 files changed, 233 ins
Am 18.09.2012 11:09, schrieb Paolo Bonzini:
>>> Luckily removable media are usually not too interesting, so a
>>> slightly suboptimal behavior is okay as long as it does not break the
>>> important use cases---mostly migration without shared storage, where also
>>> uninteresting images have to be m
Il 18/09/2012 11:40, Kevin Wolf ha scritto:
>>> >> Note that after completing the refactoring, we'll only have one combined
>>> >> bdrv_close/delete function and so there won't be BlockDriverStates
>>> >> that are closed. In this case, I think it's quite obvious that not
>>> >> closing
>>> >> the
This patch adds a mmio bar to the qemu standard vga which allows to
access the standard vga registers and bochs dispi interface registers
via mmio.
Cc: Benjamin Herrenschmidt
Signed-off-by: Gerd Hoffmann
---
hw/vga-pci.c | 97 ++
hw/vga.
Am 18.09.2012 11:48, schrieb Paolo Bonzini:
> Il 18/09/2012 11:40, Kevin Wolf ha scritto:
>> Note that after completing the refactoring, we'll only have one combined
>> bdrv_close/delete function and so there won't be BlockDriverStates
>> that are closed. In this case, I think it's quit
Il 18/09/2012 11:01, Wenchao Xia ha scritto:
> This patch contains type and defines used in APIs, one file for public usage
> by user, one for libqblock internal usage.
>
> Signed-off-by: Wenchao Xia
> ---
> libqblock/libqblock-internal.h | 56 +
> libqblock/libqblock-types.h| 2
Il 18/09/2012 11:01, Wenchao Xia ha scritto:
> Libqblock was placed in new directory ./libqblock, libtool will build
> dynamic library there, source files of block layer remains in ./block.
> So block related source code will generate 3 sets of binary, first is old
> ones used in qemu, second and
Il 18/09/2012 11:01, Wenchao Xia ha scritto:
> Created a new directory in tests, make chekc-libqblock will build an
> executable binrary, make clean will delete it.
>
> Signed-off-by: Wenchao Xia
> ---
> .gitignore |1 +
> Makefile |1 +
>
Il 18/09/2012 11:01, Wenchao Xia ha scritto:
> In this example, user first create two qcow2 images, and then get the
> backing file relationship information of them. Then does write and read
> sync IO on them.
Please use gtest so that this can be easily extensible.
Paolo
> Signed-off-by: Wench
On Tue, 18 Sep 2012, Xu, Dongxiao wrote:
> Hi Stefano,
>
> Is these patches merged with Xen 4.2? I didn't see them in the upstream.
> The uint/int fix is critical to fix the nested guest boot issue.
They are not. Ian decided that he wanted to merge a different version of
them.
On 09/17/12 22:12, Peter Lieven wrote:
On 09/17/12 10:41, Kevin Wolf wrote:
Am 16.09.2012 12:13, schrieb Peter Lieven:
Hi,
when trying to block migrate a VM from one node to another, the source
VM crashed with the following assertion:
block.c:3829: bdrv_set_in_use: Assertion `bs->in_use != in_
Am 18.09.2012 12:28, schrieb Peter Lieven:
> On 09/17/12 22:12, Peter Lieven wrote:
>> On 09/17/12 10:41, Kevin Wolf wrote:
>>> Am 16.09.2012 12:13, schrieb Peter Lieven:
Hi,
when trying to block migrate a VM from one node to another, the source
VM crashed with the following ass
Il 18/09/2012 11:01, Wenchao Xia ha scritto:
> Libqblock was placed in new directory ./libqblock, libtool will build
> dynamic library there, source files of block layer remains in ./block.
> So block related source code will generate 3 sets of binary, first is old
> ones used in qemu, second and
On Tue, 2012-09-18 at 11:51 +0200, Gerd Hoffmann wrote:
> This patch adds a mmio bar to the qemu standard vga which allows to
> access the standard vga registers and bochs dispi interface registers
> via mmio.
I had a patch like that somewhere (or is that it ? :-)
I dropped it in favor of a more
On Tue, 18 Sep 2012, Gerd Hoffmann wrote:
> Also change the way the gui_timer is initialized: each time a
> displaychangelistener is registered or unregistered we'll check
> whenever we need a timer (due to dpy_refresh callback being present)
> and if so setup a timer, otherwise zap it. This way t
On Tue, Sep 18, 2012 at 09:14:55AM +0100, Richard Davies wrote:
> Hi Mel,
>
> Thanks for your latest patch, I attach a perf report below with this on top
> of all previous patches. There is still lock contention, though in a
> different place.
>
> 59.97% qemu-kvm [kernel.kallsyms]
Am 18.09.2012 02:08, schrieb David Gibson:
> On Mon, Sep 17, 2012 at 01:24:51PM -0500, Anthony Liguori wrote:
>> David Gibson writes:
>>
>>> tcp_chr_connect(), unlike for example udp_chr_update_read_handler() does
>>> not check if the fd it is using is valid (>= 0) before passing it to
>>> qemu_se
On Tue, 18 Sep 2012, Max Filippov wrote:
> On Tue, Sep 18, 2012 at 12:15 AM, Eduardo Habkost wrote:
> > On Mon, Sep 17, 2012 at 11:54:42PM +0400, malc wrote:
> >> On Mon, 17 Sep 2012, Anthony Liguori wrote:
> >>
> >> > malc writes:
> >> >
> >> > > Some(thing|one) broke compilation with pcspk ena
On 17 September 2012 21:07, Francesco Lavra wrote:
> In the A series memory map (implemented in the Cortex A15 CoreTile), the
> first NOR flash bank (flash 0) is mapped to address 0x0800, while
> address 0x can be configured as alias to either the first or the
> second flash bank. This
Even for writes to already allocated clusters, an l2meta is allocated,
though it stays effectively unused. After this patch, only allocating
requests still have one. Each l2meta now describes an in-flight request
that writes to clusters that are not yet hooked up in the L2 table.
Signed-off-by: Ke
Not exactly bisectable, but one large patch isn't much better either :-(
m->error is used to allow bdrv_drain() to stop with l2meta in error
state rather than go into an endless loop.
Signed-off-by: Kevin Wolf
---
block/qcow2.c | 44
block/qcow2.h
Handling overlapping allocations aren't just a detail of cluster
allocation. They are rather one of three ways to get the host cluster
offset for a write request:
1. If a request overlaps an in-flight allocations, the cluster offset
can be taken from there (this is what handle_dependencies will
Signed-off-by: Kevin Wolf
---
block/qcow2-cluster.c | 29 +
block/qcow2.c | 31 ---
block/qcow2.h | 10 ++
3 files changed, 67 insertions(+), 3 deletions(-)
diff --git a/block/qcow2-cluster.c b/block/qcow2-clus
There's no reason for run_dependent_requests() to hold s->lock, and a
later patch will require that in fact the lock is not held.
Also, before this patch, run_dependent_requests() not only does what its
name suggests, but also removes the l2meta from the list of in-flight
requests. Change this, wh
On 09/18/12 12:32, Benjamin Herrenschmidt wrote:
> On Tue, 2012-09-18 at 11:51 +0200, Gerd Hoffmann wrote:
>> This patch adds a mmio bar to the qemu standard vga which allows to
>> access the standard vga registers and bochs dispi interface registers
>> via mmio.
>
> I had a patch like that somewh
Signed-off-by: Kevin Wolf
---
block/qcow2-cluster.c | 70 +---
1 files changed, 42 insertions(+), 28 deletions(-)
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index 7a038ac..468ef1b 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-c
This makes it easier to address the areas for which a COW must be
performed. As a nice side effect, the COW code in
qcow2_alloc_cluster_link_l2 becomes really trivial.
Signed-off-by: Kevin Wolf
---
block/qcow2-cluster.c | 85 +++--
block/qcow2.h
Signed-off-by: Kevin Wolf
---
block/qcow2-cluster.c | 26 --
block/qcow2.h | 11 +++
2 files changed, 27 insertions(+), 10 deletions(-)
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index 4d5c3da..440fdbf 100644
--- a/block/qcow2-cluster.c
This is the first part of an optimisation to improve the performance of
sequential cluster allocations.
Typically, write requests aren't aligned to cluster boundaries, so
sequential allocation means that every other request has to wait for the
COW of the previous request to complete. We can do bet
This patch adds SPARC ASI mappings that are used by the LEON processor.It also
corrects the MMU context register and context table pointer mask of the LEON3.
Signed-off-by: Ronald Hecht
---
target-sparc/cpu.c |4 ++--
target-sparc/ldst_helper.c |6 ++
2 files changed, 8 inser
This patch adds SPARC ASI mappings that are used by the LEON processor.It also
corrects the MMU context register and context table pointer mask of the LEON3.
Signed-off-by: Ronald Hecht
---
target-sparc/cpu.c |4 ++--
target-sparc/ldst_helper.c |6 ++
2 files changed, 8 inser
Jan Kiszka writes:
> On 2012-09-18 03:52, Wen Congyang wrote:
>> At 09/18/2012 01:56 AM, Luiz Capitulino Wrote:
>>> Hi Wen,
>>>
>>> We've re-reviewed the dump-guest-memory command and found some
>>> possible issues with the -p option.
>>>
>>> The main issue is that it seems possible for a malicio
Cancelling COW when it's overwritten by a subsequent write request of
the guest was a good start, however in practice we don't gain
performance yet. The second write request is split in two, the first one
containing the overwritten COW area, and the second one allocating
another cluster.
We can do
In preparation of delayed COW (i.e. completing the guest write request
before the associated COWs have completed) we must make sure that after
the guest data has written the new data is read back, even if the COW
hasn't completed and the new cluster isn't linked in the L2 table yet.
Signed-off-by:
Has it been applied to anything? I don't think so.
Is it still needed?
Thanks,
/mjt
On 07.06.2012 20:23, Andreas Färber wrote:
> config-devices.mak.d is included from Makefile.target, i.e. from inside
> the *-softmmu/ directory. It included the directory path, so never
> applied to the actual c
During the last few releases we have got rid of most of the overhead of
metadata writes during cluster allocation. What's left is the COW for
unaligned allocating write requests, and it's quite expensive.
In the general case, this cost cannot be avoided. However, if we're
lucky enough that before
This creates a separate coroutine for processing the COW and the L2
table update of allocating requests. The request itself can then
complete while the second part is still being processed.
We need a qemu_aio_flush() hook in order to ensure that these
coroutines for the second part aren't still ru
The offset within the cluster is already present as n_start and this is
what the code uses. QCowL2Meta.offset is only needed at a cluster
granularity.
Signed-off-by: Kevin Wolf
---
block/qcow2-cluster.c |4 ++--
block/qcow2.h | 22 ++
2 files changed, 24 inserti
There's no real reason to have an l2meta for normal requests that don't
allocate anything. Before we can get rid of it, we must return the host
cluster offset in a different way.
Signed-off-by: Kevin Wolf
---
block/qcow2-cluster.c | 10 ++
block/qcow2.c | 14 +++---
b
On 2012-09-18 14:23, Markus Armbruster wrote:
> Jan Kiszka writes:
>
>> On 2012-09-18 03:52, Wen Congyang wrote:
>>> At 09/18/2012 01:56 AM, Luiz Capitulino Wrote:
Hi Wen,
We've re-reviewed the dump-guest-memory command and found some
possible issues with the -p option.
>
As soon as delayed COW is introduced, the l2meta struct is needed even
after completion of the request, so it can't live on the stack.
Signed-off-by: Kevin Wolf
---
block/qcow2.c | 26 +++---
1 files changed, 15 insertions(+), 11 deletions(-)
diff --git a/block/qcow2.c b/b
On Tue, 18 Sep 2012 09:17:10 +0200
Gerd Hoffmann wrote:
> This patch adds a fbdev monitor command to enable/disable
> the fbdev display at runtime to both qmp and hmp.
>
> qmp: fbdev enable=on|off
> hmp: fbdev on|off
>
> Signed-off-by: Gerd Hoffmann
> ---
> hmp-commands.hx | 15 +++
Am 17.09.2012 17:22, schrieb Bharata B Rao:
> sockets: Make inet_parse() non static.
>
> From: Bharata B Rao
>
> Make inet_parse() non-static so that other subsystems like gluster
> can use it to parse inet addresses. As a pre-requisite, define and
> globalize the qemu_inet_opts.
>
> Signed-off
This is closer to where the dirty flag is really needed, and it avoids
having checks for special cases related to cluster allocation directly
in the writev loop.
Signed-off-by: Kevin Wolf
---
block/qcow2-cluster.c |5 -
block/qcow2.c |7 +--
block/qcow2.h |2 +
On 09/18/2012 01:47 PM, Ronald Hecht wrote:
> This patch adds SPARC ASI mappings that are used by the LEON processor.It also
> corrects the MMU context register and context table pointer mask of the LEON3.
>
Reviewed-by: Fabien Chouteau
--
Fabien Chouteau
Il 18/09/2012 14:47, Kevin Wolf ha scritto:
>> > Makefile.objs |2 +-
>> > qemu-config.c | 31 +++
>> > qemu-sockets.c | 42 ++
>> > qemu_socket.h |1 +
>> > 4 files changed, 39 insertions(+), 37 deletions(-)
> Maki
Signed-off-by: Ronald Hecht
---
target-sparc/cpu.c |2 +-
target-sparc/cpu.h |1 +
target-sparc/helper.h |1 +
target-sparc/ldst_helper.c |9 +
target-sparc/translate.c |5 +
5 files changed, 17 insertions(+), 1 deletions(-)
diff --git a/tar
On 2012-09-18 14:50, GaoYi wrote:
> Hi Jan,
>
>I have followed a previous thread about ELI proposed by Abel Gordon,
> http://www.spinics.net/lists/kvm/msg73907.html.
>I wonder whether this mechanism will be incorporated in KVM someday.
Likely not. Both Intel and AMD will soon ship hardwa
Hello all,
In order to test and debug a linux-based real-time system, I'd like to
hook it up to a software simulator, which sadly does not run fast
enough to meet the real-time constraints.
To workaround this problem, I had the idea of running the real-time
system in guest whose clock would be co
Am 17.09.2012 17:23, schrieb Bharata B Rao:
> sockets: Change inet_parse() to accept address specification without port
>
> From: Bharata B Rao
>
> inet_parse() expects address:port. Change it to work without explicit port
> specification. In addition, don't depend solely on the return value of
Juan Quintela wrote:
> Hi
>
> Please send in any agenda items you are interested in covering.
there are no agenda so . call is cancelled.
Happy hacking, Juan.
Signed-off-by: Ronald Hecht
---
target-sparc/helper.h |1 +
target-sparc/ldst_helper.c |6 ++
target-sparc/translate.c | 10 +++---
3 files changed, 10 insertions(+), 7 deletions(-)
diff --git a/target-sparc/helper.h b/target-sparc/helper.h
index 74ecad1..23a2ad4 100644
On Tue, 18 Sep 2012 14:41:53 +0200
Jan Kiszka wrote:
> On 2012-09-18 14:23, Markus Armbruster wrote:
> > Jan Kiszka writes:
> >
> >> On 2012-09-18 03:52, Wen Congyang wrote:
> >>> At 09/18/2012 01:56 AM, Luiz Capitulino Wrote:
> Hi Wen,
>
> We've re-reviewed the dump-guest-memory
Jan Kiszka writes:
> On 2012-09-18 14:23, Markus Armbruster wrote:
>> Jan Kiszka writes:
>>
>>> On 2012-09-18 03:52, Wen Congyang wrote:
At 09/18/2012 01:56 AM, Luiz Capitulino Wrote:
> Hi Wen,
>
> We've re-reviewed the dump-guest-memory command and found some
> possible is
Il 18/09/2012 15:22, Kevin Wolf ha scritto:
> Am 17.09.2012 17:23, schrieb Bharata B Rao:
>> sockets: Change inet_parse() to accept address specification without port
>>
>> From: Bharata B Rao
>>
>> inet_parse() expects address:port. Change it to work without explicit port
>> specification. In add
On 17 September 2012 21:08, Francesco Lavra wrote:
> This patch adds modelling of the two NOR flash banks found on the
> Versatile Express motherboard. Tested with U-Boot running on an emulated
> Versatile Express, with either A9 or A15 CoreTile.
>
> Signed-off-by: Francesco Lavra
> ---
> Changes
Drop the private reimplementation of ctz32() from pflash_cfi0[12]
in favour of using the standard version from host-utils.h.
Signed-off-by: Peter Maydell
---
hw/pflash_cfi01.c | 37 +
hw/pflash_cfi02.c | 37 +
2 files ch
Am 17.09.2012 17:26, schrieb Bharata B Rao:
> block: Support GlusterFS as a QEMU block backend.
>
> From: Bharata B Rao
>
> This patch adds gluster as the new block backend in QEMU. This gives
> QEMU the ability to boot VM images from gluster volumes. Its already
> possible to boot from VM image
Am 18.09.2012 15:31, schrieb Paolo Bonzini:
> Il 18/09/2012 15:22, Kevin Wolf ha scritto:
>> Am 17.09.2012 17:23, schrieb Bharata B Rao:
>>> sockets: Change inet_parse() to accept address specification without port
>>>
>>> From: Bharata B Rao
>>>
>>> inet_parse() expects address:port. Change it to
On Tue, 18 Sep 2012, Gerd Hoffmann wrote:
> When adding DisplayChangeListeners the set_mouse and cursor_define
> callbacks have been left in DisplayState for some reason. Fix it.
>
> Signed-off-by: Gerd Hoffmann
This patch is good.
The one thing I don't like is dpy_cursor_define_supported, beca
Hi,
Am 18.09.2012 14:58, schrieb Ronald Hecht:
> Signed-off-by: Ronald Hecht
That's not a valid SoB, please fix. :)
Also please use a "target-sparc: " prefix in your subjects to make clear
what the patch is about (applies to all three patches). If the patches
are otherwise okay, hopefully Blue
On 09/18/2012 03:29 PM, Ronald Hecht wrote:
> Signed-off-by: Ronald Hecht
Reviewed-by: Fabien Chouteau
--
Fabien Chouteau
While swapping constants to the second operand, swap
sources matching destinations to the first operand.
Signed-off-by: Richard Henderson
---
tcg/optimize.c | 22 +++---
1 file changed, 19 insertions(+), 3 deletions(-)
diff --git a/tcg/optimize.c b/tcg/optimize.c
index 5b0a8ce..
Il 18/09/2012 13:40, Kevin Wolf ha scritto:
> +again:
> +QLIST_FOREACH(m, &s->cluster_allocs, next_in_flight) {
> +if (m->sleeping) {
> +qemu_coroutine_enter(m->co, NULL);
> +/* next_in_flight link could have become invalid */
> +goto again;
> +
Il 18/09/2012 13:40, Kevin Wolf ha scritto:
> Not exactly bisectable, but one large patch isn't much better either
For better bisectability you could add the co_sleep_ns in a separate
patch, later in the series.
Paolo
Am 18.09.2012 14:29, schrieb Michael Tokarev:
> Has it been applied to anything? I don't think so.
> Is it still needed?
Not in qemu.git yet, still applicable AFAICT. CC'ing Paolo.
/-F
>
> Thanks,
>
> /mjt
>
> On 07.06.2012 20:23, Andreas Färber wrote:
>> config-devices.mak.d is included fro
Il 18/09/2012 13:40, Kevin Wolf ha scritto:
> static void run_dependent_requests(BDRVQcowState *s, QCowL2Meta *m)
> {
> -/* Take the request off the list of running requests */
> -if (m->nb_clusters != 0) {
> -QLIST_REMOVE(m, next_in_flight);
> -}
> -
> /* Restart all dep
Implemented with setcond if the target does not provide
the optional opcode.
Signed-off-by: Richard Henderson
---
tcg/README | 6 ++
tcg/arm/tcg-target.h | 1 +
tcg/hppa/tcg-target.h | 1 +
tcg/i386/tcg-target.h | 2 ++
tcg/ia64/tcg-target.h | 2 ++
tcg/mips/tcg-target.
Signed-off-by: Ronald Hecht
---
target-sparc/cpu.c |2 +-
target-sparc/cpu.h |1 +
target-sparc/helper.h |1 +
target-sparc/ldst_helper.c |9 +
target-sparc/translate.c |5 +
5 files changed, 17 insertions(+), 1 deletions(-)
diff --git a/ta
Signed-off-by: Richard Henderson
---
tcg/optimize.c | 29 +
1 file changed, 29 insertions(+)
diff --git a/tcg/optimize.c b/tcg/optimize.c
index fba0ed9..5b0a8ce 100644
--- a/tcg/optimize.c
+++ b/tcg/optimize.c
@@ -397,6 +397,14 @@ static TCGArg *tcg_constant_folding(T
Am 18.09.2012 16:24, schrieb Paolo Bonzini:
> Il 18/09/2012 13:40, Kevin Wolf ha scritto:
>> +qemu_co_mutex_unlock(&s->lock);
>> +qemu_co_rwlock_rdlock(&s->l2meta_flush);
>
> Should this lock be taken in process_l2meta? It's a bit easier to follow.
I'm pretty sure there w
Il 18/09/2012 13:40, Kevin Wolf ha scritto:
> +qemu_co_mutex_unlock(&s->lock);
> +qemu_co_rwlock_wrlock(&m->l2_writeback_lock);
Can anybody else take the lock as reader again at this point? If not, I
wonder if this is more clear if you write it as a CoQueue.
Paolo
> +has_wr_lock = t
For proper cmov insns, as well as the non-goto-tb case
of conditional branch.
Signed-off-by: Richard Henderson
---
target-alpha/translate.c | 102 ++-
1 file changed, 48 insertions(+), 54 deletions(-)
diff --git a/target-alpha/translate.c b/target-alp
Am 18.09.2012 16:27, schrieb Paolo Bonzini:
> Il 18/09/2012 13:40, Kevin Wolf ha scritto:
>> +again:
>> +QLIST_FOREACH(m, &s->cluster_allocs, next_in_flight) {
>> +if (m->sleeping) {
>> +qemu_coroutine_enter(m->co, NULL);
>> +/* next_in_flight link could have bec
Looking at http://lkml.indiana.edu/hypermail/linux/kernel/1205.0/00100.html
The new value for EAX is 0x4001.
This depends on
http://lists.gnu.org/archive/html/qemu-devel/2012-09/msg02497.html
As far as I known it is #5. It depends on (1), (2), (3) and (4).
Based on cpu-queue[1] branch.
(Fr
>From http://lkml.indiana.edu/hypermail/linux/kernel/1205.0/00100.html
EAX should be KVM_CPUID_FEATURES (0x4001) not 0.
---
target-i386/kvm.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index 761a9b1..0c9f5dd 100644
--- a/targ
Il 18/09/2012 13:40, Kevin Wolf ha scritto:
> +qemu_co_mutex_unlock(&s->lock);
> +qemu_co_rwlock_rdlock(&s->l2meta_flush);
Should this lock be taken in process_l2meta? It's a bit easier to follow.
Paolo
> +l2meta->is_written = true;
> +co = qemu_c
>From http://lkml.indiana.edu/hypermail/linux/kernel/1205.0/00100.html
EAX should be KVM_CPUID_FEATURES (0x4001) not 0.
If kvm is not configured, the additional option of hypervisor-level=1
(or hypervisor-level=0x4001) needs to be specified to get this.
---
target-i386/cpu.c | 12 ++
Am 18.09.2012 16:33, schrieb Paolo Bonzini:
> Il 18/09/2012 13:40, Kevin Wolf ha scritto:
>> static void run_dependent_requests(BDRVQcowState *s, QCowL2Meta *m)
>> {
>> -/* Take the request off the list of running requests */
>> -if (m->nb_clusters != 0) {
>> -QLIST_REMOVE(m, next
As recently discussed, with the optional fallback to setcond.
I include a patch to target-alpha to test correctness both
before and after implementing the opcode in the i386 backend,
as well as in the optimizations.
r~
Richard Henderson (5):
tcg: Introduce movcond
target-alpha: Use movcond
1 - 100 of 261 matches
Mail list logo
