We have developed IPv6 in Qemu -net user mode.
These patches add ICMPv6, NDP, and make UDP and TCP compatible with
IPv6. We have made some refactoring to make current code compatible
with IPv6.
Some patches, like 2 and 13, can be reviewed using
interdiff -w /dev/null patchfile
to get rid of
This patch adds SCALE_S, timer_new_s(), and qemu_clock_get_s in qemu/timer.h to
manage second-scale timers.
Signed-off-by: Guillaume Subiron maet...@subiron.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
---
include/qemu/timer.h | 32
1 file
This patch adds an IPv6 address to the DNS relay. in6_equal_dns() is
developed using this Slirp attribute.
sotranslate_in/out() are also updated to manage the IPv6 case so the
guest can be able to join the host using one of the Slirp addresses.
Signed-off-by: Guillaume Subiron maet...@subiron.org
This patch simply adds a sa_family_t argument to remove the hardcoded
AF_INET in the call of qemu_socket().
Signed-off-by: Guillaume Subiron maet...@subiron.org
---
slirp/slirp.h | 2 +-
slirp/tcp_input.c | 3 ++-
slirp/tcp_subr.c | 5 +++--
3 files changed, 6 insertions(+), 4 deletions(-)
Before this patch, if sosendto fails, udp_input is executed as if the
packet was sent. This could cause memory leak.
This patch adds a goto bad to cut the execution of this function.
Signed-off-by: Guillaume Subiron maet...@subiron.org
---
slirp/udp.c | 1 +
1 file changed, 1 insertion(+)
diff
This patch makes solookup() compatible with all address family. Also,
this function was only compatible with TCP. Having the socket list in
argument, it is now compatible with UDP too. Finally, some optimization
code is factorized inside the function (the function look at the last
returned result
Disambiguation : icmp_error is renamed into icmp_send_error, since it
doesn't manage errors, but only sends ICMP Error messages.
Adding icmp6_send_error to send ICMPv6 Error messages. This function is
simpler than the v4 version.
Adding some calls in various functions to send ICMP errors, when a
A sa_family_t is now passed in argument to udp_attach instead of using a
hardcoded AF_INET to call qemu_socket().
Signed-off-by: Guillaume Subiron maet...@subiron.org
---
slirp/ip_icmp.c | 2 +-
slirp/udp.c | 7 ---
slirp/udp.h | 2 +-
3 files changed, 6 insertions(+), 5 deletions(-)
This patch adds udp6_input() and udp6_output().
It also adds the IPv6 case in sorecvfrom().
Finally, udp_input() is called by ip6_input().
Signed-off-by: Guillaume Subiron maet...@subiron.org
---
slirp/Makefile.objs | 2 +-
slirp/ip6_input.c | 3 +-
slirp/socket.c | 7 ++-
Basically, this patch adds some switch in various TCP functions to
prepare them for the IPv6 case.
To have something to switch in tcp_input() and tcp_respond(), a new
argument is used to give them the sa_family of the addresses they are
working on.
Signed-off-by: Guillaume Subiron
compatible with the new tcpiphdr structure,
whose size will be bigger than sizeof(struct tcphdr) + sizeof(struct ip)
Signed-off-by: Guillaume Subiron maet...@subiron.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
---
slirp/if.h | 4 ++--
slirp/mbuf.c | 3 ++-
slirp
adds ip6_cksum() to compute ICMPv6 checksums using IPv6
pseudo-header.
Signed-off-by: Guillaume Subiron maet...@subiron.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
---
slirp/Makefile.objs | 4 +-
slirp/cksum.c | 23
slirp/ip6.h | 139
This patch replaces foreign and local address/port couples in Socket
structure by 2 sockaddr_storage which can be casted in sockaddr_in or
sockaddr_in6.
Direct access to address and port is still possible thanks to some
\#define, so retrocompatibility of the existing code is assured.
The
This patch adds IPv6 case in TCP functions refactored by the last
patches.
This also adds IPv6 pseudo-header in tcpiphdr structure.
Finally, tcp_input() is called by ip6_input().
Signed-off-by: Guillaume Subiron maet...@subiron.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
is factorized.
Some #define ETH_* are moved upper in slirp.h to make them accessible to
other slirp/*.h
Signed-off-by: Guillaume Subiron maet...@subiron.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
---
slirp/if.c| 2 +-
slirp/mbuf.c | 2 +-
slirp/mbuf.h | 2 +-
slirp/slirp.c
This patchs adds parameters to manage some new options in the qemu -net
command.
Slirp IPv6 address, network prefix, and DNS IPv6 address can be given in
argument to the qemu command.
Defaults parameters are respectively fc00::1, fc00::, /64 and fc00::2.
Signed-off-by: Yann Bordenave
This patch factorizes some duplicate code into a new function,
sotranslate_out(). This function perform the address translation when a
packet is transmitted to the host network. If the paquet is destinated
to the host, the loopback address is used, and if the paquet is
destinated to the virtual
Hello,
Eric Blake, le Mon 21 Oct 2013 22:04:23 +0100, a écrit :
+'*ip6_prefix': 'str',
Why is this a str instead of an integer?
prefix would be e.g. fc00::1/64.
Samuel
Hello,
Eric Blake, le Tue 22 Oct 2013 11:27:27 +0100, a écrit :
On 10/22/2013 11:22 AM, Samuel Thibault wrote:
Eric Blake, le Mon 21 Oct 2013 22:04:23 +0100, a écrit :
+'*ip6_prefix': 'str',
Why is this a str instead of an integer?
prefix would be e.g. fc00::1/64
Eric Blake, le Tue 22 Oct 2013 11:33:52 +0100, a écrit :
On 10/22/2013 11:31 AM, Samuel Thibault wrote:
Hello,
Eric Blake, le Tue 22 Oct 2013 11:27:27 +0100, a écrit :
On 10/22/2013 11:22 AM, Samuel Thibault wrote:
Eric Blake, le Mon 21 Oct 2013 22:04:23 +0100, a écrit
Eric Blake, le Tue 22 Oct 2013 11:52:11 +0100, a écrit :
On 10/22/2013 11:48 AM, Eric Blake wrote:
HMP can let the user abbreviate to net=10.0.2.2/24,
The command line, just like HMP, can use shorthand for convenience.
How are these usually defined? A quick search didn't provide me an
I see. So it would be something like this?
commit 1807466d691f281f430fbf8c0bbff6bf8073247d
Author: Samuel Thibault samuel.thiba...@ens-lyon.org
Date: Tue Oct 22 21:11:46 2013 +0200
qapi-schema, qemu-options slirp: Adding Qemu options for IPv6 addresses
This patchs adds
Paolo Bonzini, le Wed 23 Oct 2013 08:51:21 +0100, a écrit :
+void icmp6_init(Slirp *slirp)
+{
+srand(time(NULL));
+ra_timer = timer_new_s(QEMU_CLOCK_VIRTUAL, ra_timer_handler, slirp);
+timer_mod(ra_timer, qemu_clock_get_s(QEMU_CLOCK_VIRTUAL) +
NDP_Interval);
+}
Should
This patch replaces foreign and local address/port couples in Socket
structure by 2 sockaddr_storage which can be casted in sockaddr_in or
sockaddr_in6.
Direct access to address and port is still possible thanks to some
\#define, so retrocompatibility of the existing code is assured.
The
A sa_family_t is now passed in argument to udp_attach instead of using a
hardcoded AF_INET to call qemu_socket().
Signed-off-by: Guillaume Subiron maet...@subiron.org
---
slirp/ip_icmp.c | 2 +-
slirp/udp.c | 7 ---
slirp/udp.h | 2 +-
3 files changed, 6 insertions(+), 5 deletions(-)
This patch adds SCALE_S, timer_new_s(), and qemu_clock_get_s in qemu/timer.h to
manage second-scale timers.
Signed-off-by: Guillaume Subiron maet...@subiron.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
---
include/qemu/timer.h | 32
1 file
is factorized.
Some #define ETH_* are moved upper in slirp.h to make them accessible to
other slirp/*.h
Signed-off-by: Guillaume Subiron maet...@subiron.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
---
slirp/if.c| 2 +-
slirp/mbuf.c | 2 +-
slirp/mbuf.h | 2 +-
slirp/slirp.c
adds ip6_cksum() to compute ICMPv6 checksums using IPv6
pseudo-header.
Signed-off-by: Guillaume Subiron maet...@subiron.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
---
slirp/Makefile.objs | 4 +-
slirp/cksum.c | 23
slirp/ip6.h | 139
This patch makes solookup() compatible with all address family. Also,
this function was only compatible with TCP. Having the socket list in
argument, it is now compatible with UDP too. Finally, some optimization
code is factorized inside the function (the function look at the last
returned result
This patch simply adds a sa_family_t argument to remove the hardcoded
AF_INET in the call of qemu_socket().
Signed-off-by: Guillaume Subiron maet...@subiron.org
---
slirp/slirp.h | 2 +-
slirp/tcp_input.c | 3 ++-
slirp/tcp_subr.c | 5 +++--
3 files changed, 6 insertions(+), 4 deletions(-)
Before this patch, if sosendto fails, udp_input is executed as if the
packet was sent. This could cause memory leak.
This patch adds a goto bad to cut the execution of this function.
Signed-off-by: Guillaume Subiron maet...@subiron.org
---
slirp/udp.c | 1 +
1 file changed, 1 insertion(+)
diff
This patch adds IPv6 case in TCP functions refactored by the last
patches.
This also adds IPv6 pseudo-header in tcpiphdr structure.
Finally, tcp_input() is called by ip6_input().
Signed-off-by: Guillaume Subiron maet...@subiron.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
This patch adds udp6_input() and udp6_output().
It also adds the IPv6 case in sorecvfrom().
Finally, udp_input() is called by ip6_input().
Signed-off-by: Guillaume Subiron maet...@subiron.org
---
slirp/Makefile.objs | 2 +-
slirp/ip6_input.c | 3 +-
slirp/socket.c | 7 ++-
Disambiguation : icmp_error is renamed into icmp_send_error, since it
doesn't manage errors, but only sends ICMP Error messages.
Adding icmp6_send_error to send ICMPv6 Error messages. This function is
simpler than the v4 version.
Adding some calls in various functions to send ICMP errors, when a
This patch factorizes some duplicate code into a new function,
sotranslate_out(). This function perform the address translation when a
packet is transmitted to the host network. If the paquet is destinated
to the host, the loopback address is used, and if the paquet is
destinated to the virtual
...@meowstars.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
---
net/net.c| 30 ++
net/slirp.c | 46 --
qapi-schema.json | 40 ++--
qemu-options.hx | 18
This patch adds an IPv6 address to the DNS relay. in6_equal_dns() is
developed using this Slirp attribute.
sotranslate_in/out() are also updated to manage the IPv6 case so the
guest can be able to join the host using one of the Slirp addresses.
Signed-off-by: Guillaume Subiron maet...@subiron.org
Basically, this patch adds some switch in various TCP functions to
prepare them for the IPv6 case.
To have something to switch in tcp_input() and tcp_respond(), a new
argument is used to give them the sa_family of the addresses they are
working on.
Signed-off-by: Guillaume Subiron
compatible with the new tcpiphdr structure,
whose size will be bigger than sizeof(struct tcphdr) + sizeof(struct ip)
Signed-off-by: Guillaume Subiron maet...@subiron.org
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
---
slirp/if.h | 4 ++--
slirp/mbuf.c | 3 ++-
slirp
Hello,
This is a respin of IPv6 in Qemu -net user mode.
These patches add ICMPv6, NDP, and make UDP and TCP compatible with
IPv6. We have made some refactoring to make current code compatible with
IPv6.
Some patches, like 2 and 13, can be reviewed using
interdiff -w /dev/null patchfile
to get
Maryyam Muhammad Din, le Mon 09 Dec 2013 12:25:23 +0500, a écrit :
I am unable to figure-out when someone presses control-c on the keyboard,
how host OS routes it to QEMU and how QEMU then
routes it to the guest?
It depends which QEMU UI you are using.
Please suggest me how can I extract
Maryyam Muhammad Din, le Mon 09 Dec 2013 12:44:20 +0500, a écrit :
I have gone through the source code of QEMU but could not figure out how
external interrupts are handled by QEMU for guest.
They aren't. It's the UI that manages keyboard etc. So it depends on the
UI you are using. Unless you
Hi,
Could merging the curses driver patch be considered? It would make qemu
accessible to blind people, and as such, be a considerable help for
working on distribution installers, for instance.
Samuel
___
Qemu-devel mailing list
Qemu-devel@nongnu.org
Hi,
The patch that qemu applies to bochs bios is a bit bugged:
pci_pro_fail:
pop edi
pop esi
- sti
popf
stc
+ and dword ptr[esp+4],0xfffc ;; reset CS.RPL for kqemu
retf
The and operation clears carry... So that pci functions always leave
carry clear, and hence tell
Hi,
I've just thought about implementing a virtual braille device in qemu.
The idea would be that qemu use BrlAPI for accessing the real braille
device (see http://brl.thefreecat.org), and then emulate a virtual USB
braille device for the guest system.
The question I'd have for now is: how to
Hello,
This apparently has been neither applied, nor commented on. Could
either be done?
Samuel
contro...@[\]^_} shouldn't get the 'a' - 'A' offset for correct
translation. ESC is better simulated as escape key.
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
diff --git a/ui
contro...@[\]^_} shouldn't get the 'a' - 'A' offset for correct
translation. ESC is better simulated as escape key.
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
diff --git a/ui/curses.c b/ui/curses.c
index ed3165e..5d949d6 100644
--- a/ui/curses.c
+++ b/ui/curses.c
@@ -238,9
Jes Sorensen, le Wed 01 Sep 2010 11:25:48 +0200, a écrit :
On 08/29/10 18:39, Samuel Thibault wrote:
Linux 2.6.35 hangs at boot when giving -no-acpi to qemu, for instance
the Debian kernel:
qemu -no-acpi -kernel /boot/vmlinuz-2.6.35-trunk-686
If you want anyone to look
Public bug reported:
Linux 2.6.35 hangs at boot when giving -no-acpi to qemu, for instance
the Debian kernel:
qemu -no-acpi -kernel /boot/vmlinuz-2.6.35-trunk-686
There is no output except just Booting the kernel
** Affects: qemu
Importance: Undecided
Status: New
--
linux
Johannes Schindelin, le Tue 26 Feb 2008 12:57:25 +, a écrit :
On Tue, 26 Feb 2008, Samuel Thibault wrote:
I asked on the SDL mailing list, and they answered that qemu should
indeed not use SDL_GetRelativeMouseState(), since that only provides the
latest mouse position
Hello,
I was having a look at the mouse events that the guest receives, and was
surprised to get
pos x:452 y:220 z:0
pos x:452 y:220 z:0
pos x:452 y:220 z:0
Thomas Petazzoni, le Tue 26 Feb 2008 14:51:21 +0100, a écrit :
I'm using Qemu 0.9.0 on Kubuntu, and I have issues with the keymap
selection:
How is the guest configured?
Samuel
Hello,
While using a virtual usbtablet, I noticed that I could not reach the
extreme bottom and right pixels. That is because of the conversion from
screen coordinates to 0..0x7FFF, below is a patch that fixes it.
Samuel
Index: cocoa.m
Johannes Schindelin, le Wed 27 Feb 2008 15:35:01 +, a écrit :
Index: cocoa.m
===
RCS file: /sources/qemu/qemu/cocoa.m,v
retrieving revision 1.15
diff -u -p -r1.15 cocoa.m
--- cocoa.m 22 Jan 2008 23:25:15 -
Hello,
No news about this patch, could someone apply it?
Samuel
Samuel Thibault, le Wed 27 Feb 2008 14:45:55 +, a écrit :
While using a virtual usbtablet, I noticed that I could not reach the
extreme bottom and right pixels. That is because of the conversion from
screen coordinates to 0
Hello,
The patch below fixes SDL mouse events processing:
- GetRelativeMouseState always returns the last position, so when the
polling loop gets several mouse events in one go, we would send
useless 'no move' events.
- So as to make sure we don't miss any mouse click / double click, we
Johannes Schindelin, le Wed 05 Mar 2008 14:09:10 +0100, a écrit :
What is this good for? (I imagine that it would make sense to add a
comment to document why this is here, for clueless people like me.)
Maybe it is to initialise the state of the mouse buttons?
That's it.
This means that
Here is a revamped patch:
This fixes SDL mouse events processing:
- GetRelativeMouseState() always returns the last position, so when the
polling loop gets several mouse events in one go, we would send
useless 'no move' events, let's avoid that.
- So as to make sure we don't miss any mouse
BTW, it would be good for power consumption to provide a virtual HPET
to the guest, and reprogram the host timer as appropriate. 10ms sleeps
kill kitten ;)
Samuel
When SDL is invisible/minimized, there is no need to keep calling the
VGA refresh 33 times per second. This patch reduces in that case the
rate to 2 times per second, which should be responsive enough for the
un-minimizing event.
Index: console.h
Anders, le Wed 12 Mar 2008 00:56:42 +0100, a écrit :
Samuel Thibault wrote:
When SDL is invisible/minimized, there is no need to keep calling the
VGA refresh 33 times per second. This patch reduces in that case the
rate to 2 times per second, which should be responsive enough for the
un
Aurelien Jarno, le Thu 13 Mar 2008 20:50:51 +0100, a écrit :
On Wed, Mar 05, 2008 at 01:54:53PM +, Samuel Thibault wrote:
This fixes SDL mouse events processing:
- GetRelativeMouseState() always returns the last position, so when the
polling loop gets several mouse events in one go
Hello,
There is a small incoherency in curses_keys.h, which makes it fail to
emit \n when using e.g. -k fr: curses2keysym transforms \r and 0x157
into \n, but name2keysym binds \r with Return, not \n. The patch below
fixes that.
Samuel
diff --git a/curses_keys.h b/curses_keys.h
index
Samuel Thibault, le Sun 28 Feb 2010 15:35:19 +0100, a écrit :
There is a small incoherency in curses_keys.h, which makes it fail to
emit \n when using e.g. -k fr: curses2keysym transforms \r and 0x157
into \n, but name2keysym binds \r with Return, not \n. The patch below
fixes that.
Signed
Hello,
curses_keys.h is using obscure constant values while the curses.h header
provides fine defines, let's use the latter.
To be applied on top of my previous patch.
Samuel
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
diff --git a/curses_keys.h b/curses_keys.h
index a6e41cf
Thibault samuel.thiba...@ens-lyon.org
commit 36f0635cb65e1735a7e231d609da98efcda756c5
Author: Samuel Thibault samuel.thiba...@ens-lyon.org
Date: Sun Feb 28 20:48:39 2010 +0100
Fix curses with -k option
diff --git a/curses.c b/curses.c
index 3ce12b9..4b5beac 100644
--- a/curses.c
+++ b
Samuel Thibault, le Sun 28 Feb 2010 21:03:00 +0100, a écrit :
The combination of keymap support (-k option) and curses is currently
very broken. The patch below fixes it by first extending keymap support
to interpret the shift, ctrl, altgr and addupper keywords in keymaps,
and to fix curses
Alexandre Muller, le Fri 26 Mar 2010 21:26:52 -0300, a écrit :
1 - The is a known discussion about it?
Yes. The basic answer is: this is hard. You'd need to keep the memory
coherency etc.
2 - There is a patch or a dirty and quick way to do it? (I not worried about
safety)
Use kvm. In that
Hello,
This adds a destroy hook for the baum character device, to properly
close the BrlAPI connection and free resources.
Signed-off-by: Samuel Thibault samuel.thiba...@ens-lyon.org
commit 447c41758cfda0022ea6e09aaf81137b2b27b915
Author: Samuel Thibault samuel.thiba...@ens-lyon.org
Date
-baum.c
new file mode 100644
index 000..568b8a1
--- /dev/null
+++ b/hw/bt-baum.c
@@ -0,0 +1,158 @@
+/*
+ * QEMU Bluetooth Baum driver.
+ *
+ * Copyright (C) 2010 Samuel Thibault samuel.thiba...@ens-lyon.org
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under
Here is a trivial patch to make only use of posix types in drivers, not
types from the softfloat header.
Signed-off-by: Samuel Thibault [EMAIL PROTECTED]
Index: hw/rtl8139.c
===
RCS file: /sources/qemu/qemu/hw/rtl8139.c,v
retrieving
block-raw.c currently doesn't cope with partial reads and writes, here
is a patch.
Signed-off-by: Samuel Thibault [EMAIL PROTECTED]
Index: block-raw.c
===
RCS file: /sources/qemu/qemu/block-raw.c,v
retrieving revision 1.25
diff -u
Hi,
Qemu currently uses 6 65k tables of pointers for handling ioports, which
makes 3MB on 64bit machines. There's a comment that says XXX: use a two
level table to limit memory usage. But wouldn't it be more simple and
effective to just allocate them through mmap() and when a NULL pointer
is
Samuel Thibault, le Mon 19 Nov 2007 15:20:16 +, a écrit :
Qemu currently uses 6 65k tables of pointers for handling ioports, which
makes 3MB on 64bit machines. There's a comment that says XXX: use a two
level table to limit memory usage. But wouldn't it be more simple and
effective to just
The attached patch drops a variable that became useless.
Samuel
Index: vl.c
===
RCS file: /sources/qemu/qemu/vl.c,v
retrieving revision 1.369
diff -u -p -r1.369 vl.c
--- vl.c19 Nov 2007 01:05:22 - 1.369
+++ vl.c
Paul Brook, le Mon 19 Nov 2007 16:17:26 +, a écrit :
On Monday 19 November 2007, Samuel Thibault wrote:
Samuel Thibault, le Mon 19 Nov 2007 15:20:16 +, a écrit :
Qemu currently uses 6 65k tables of pointers for handling ioports, which
makes 3MB on 64bit machines. There's a comment
Johannes Schindelin, le Wed 28 Nov 2007 12:49:34 +, a écrit :
On Tue, 27 Nov 2007, Hotmail wrote:
The following is another patch against the head that lets me build it on
Windows, as well as makes the adlib.c file debuggable.
Index: Makefile
Johannes Schindelin, le Wed 28 Nov 2007 14:01:21 +, a écrit :
On Wed, 28 Nov 2007, Samuel Thibault wrote:
Johannes Schindelin, le Wed 28 Nov 2007 12:49:34 +, a ?crit :
On Tue, 27 Nov 2007, Hotmail wrote:
The following is another patch against the head that lets me build
Hi,
Laurent Vivier, le Wed 28 Nov 2007 15:02:50 +0100, a écrit :
+ret = posix_memalign((void**)buf, 0x200, 512);
For making this more easily portable, maybe it should be a new
qemu_memalign() function? Also, the alignment may probably be better as
a global macro, since the alignment
Daniel P. Berrange, le Wed 28 Nov 2007 14:27:39 +, a écrit :
On Wed, Nov 28, 2007 at 03:02:50PM +0100, Laurent Vivier wrote:
These patches allow to open file using O_DIRECT and bypass the host I/O
cache.
[PATCH 1/2] Add directio parameter to -drive
Using directio=on with
Laurent Vivier, le Wed 28 Nov 2007 16:00:17 +0100, a écrit :
Le mercredi 28 novembre 2007 à 14:24 +, Samuel Thibault a écrit :
Laurent Vivier, le Wed 28 Nov 2007 15:02:50 +0100, a écrit :
+ret = posix_memalign((void**)buf, 0x200, 512);
For making this more easily portable, maybe
Daniel P. Berrange, le Wed 28 Nov 2007 15:08:20 +, a écrit :
sometimes it is not because the guest is not so i/o friendly
and hence using the host page/buffer cache is useful.
I don't buy that - all OS already do I/O caching because its useful even
on baremetal.
Well, I can see the
Andre Przywara, le Fri 30 Nov 2007 15:10:07 +0100, a écrit :
A more portable and clean solution would be this:
-void *args[MAX_ARGS];
+union
+{
+void* ptr;
+int i;
+} args[MAX_ARGS];
It's not more portable: C99 doesn't asserts that if you write ptr and
read
andrzej zaborowski, le Fri 30 Nov 2007 15:30:40 +0100, a écrit :
I'm not sure why you get a warning here and I'm unable to run a build
at the moment. A void * should be able to store some (unknown size)
integer regardless of the platform.
No exactly. On 32bit platforms, void * are 32bits and
Anthony Liguori, le Mon 03 Dec 2007 09:54:47 -0600, a écrit :
Have you done any performance testing? Buffered IO should absolutely
beat direct IO simply because buffered IO allows writes to complete
before they actually hit disk.
Since qemu can use the aio interface, that shouldn't matter.
Paul Brook, le Mon 03 Dec 2007 15:39:48 +, a écrit :
I think host caching is still useful enough to be enabled by default, and
provides a significant performance increase in several cases.
- The guest typically has a relatively small quantity of RAM, compared to a
modern machine.
Noam Taich, le Mon 31 Dec 2007 09:02:08 -0800, a écrit :
Qemu monitor had many missing keys that could not be (easily) sent,
And there was no mechanism for sending strings properly.
This patch fixes the problem and enhances the monitor's functionality.
Mmm, isn't there a keyboard layout
Andreas Färber, le Fri 04 Jan 2008 14:41:29 +0100, a écrit :
Am 04.01.2008 um 14:20 schrieb Thiemo Seufer:
Carlo Marcelo Arenas Belon wrote:
Trivial fix that ensures that all buffers used for bdrv_read or
bdrv_write
are from an array of the uint8_t type
Do we have a host where this
Hello,
I would like to implement support for braille devices, and for this I'd
need to first implement a USB serial device (FTDI chip). Has anybody
worked on that already?
Samuel
Hello,
Samuel Thibault, le Fri 11 Jan 2008 00:23:12 +, a écrit :
I would like to implement support for braille devices, and for this I'd
need to first implement a USB serial device (FTDI chip). Has anybody
worked on that already?
Ok, was easier than expected, Here is a patch. The serial
Hello,
Samuel Thibault, le Fri 11 Jan 2008 11:09:23 +, a écrit :
Samuel Thibault, le Fri 11 Jan 2008 00:23:12 +, a écrit :
I would like to implement support for braille devices, and for this I'd
need to first implement a USB serial device (FTDI chip). Has anybody
worked
Hello,
Samuel Thibault, le Sun 13 Jan 2008 01:55:56 +, a écrit :
Samuel Thibault, le Fri 11 Jan 2008 11:09:23 +, a écrit :
Samuel Thibault, le Fri 11 Jan 2008 00:23:12 +, a écrit :
I would like to implement support for braille devices, and for this I'd
need to first implement
andrzej zaborowski, le Thu 17 Jan 2008 15:09:54 +0100, a écrit :
On 13/01/2008, Samuel Thibault [EMAIL PROTECTED] wrote:
Samuel Thibault, le Sun 13 Jan 2008 01:55:56 +, a écrit :
Here is an updated version, that takes parameters, so as to be able to
notably provide the product ID
andrzej zaborowski, le Thu 17 Jan 2008 23:25:04 +0100, a écrit :
Thanks, committed although I hoped for something that lets easily test
that the adapter works, e.g. so that after a usb_add serial:...:stdio
you can do cat /dev/ttyXXXN in the guest and see what's being typed on
qemu's stdin. I
andrzej zaborowski, le Sat 19 Jan 2008 14:05:20 +0100, a écrit :
I applied the patch even though the old values worked ok too.
The old values weren't in linux 2.6.12 for instance, and it looks like
the new ones are the default ones for the FTDI builderr, so it should be
fine now.
I also added
Ronan Keryell, le Mon 28 Jan 2008 13:24:27 +0100, a écrit :
But is it possible to use higher-level queue constructions rather than
inlining the queue behaviour in the code?
There is QEMUFIFO code in console.c which could be shared for instance,
yes.
Samuel
Marius Groeger, le Wed 30 Jan 2008 14:41:17 +0100, a écrit :
On Wed, 30 Jan 2008, Hervé Poussineau wrote:
I didn't know how to do a patch which removes the file
hw/mips_pica61.c, so this file will have to be removed manually by
the committer.
One work-around for that is to do:
$
Hello,
We were wondering whether it could be useful to have e.g. a git
repository holding the patches from Xen and KVM: while they are being
merged upstream, they could at least be shared by both projects...
Samuel
Mmm, actually, shouldn't qemu use a more private network like a
RFC1918 172.16.0.0/12 network?
(see http://www.ucam.org/cam-grin/)
Samuel
Andreas Schwab, le Tue 05 Feb 2008 11:32:30 +0100, a écrit :
Samuel Thibault [EMAIL PROTECTED] writes:
Mmm, actually, shouldn't qemu use a more private network like a
RFC1918 172.16.0.0/12 network?
In which way is 172.16.0.0/12 more private than 10.0.0.0/8?
Precisely thanks
1 - 100 of 1560 matches
Mail list logo
