Ideally, we could force gcc to implement switch statements as indirect
jumps with jump tables inline with the code. However, this may not be
possible.
I think Nathaniel was just saying that gcc is likely generating
several hundred sequential if-else blocks for large switch statements.
This gives
The problem with table lookups (I'm assuming you're talking about
function pointer vectors) is that they *destroy* spatial locality of
reference that you could otherwise attain by having series of
if-then-else instructions and some clever instruction prefetching
mechanism on modern processors... No
On Sun, 2005-04-17 at 18:36 -0700, Nathaniel G H wrote:
> I was up until 3:00am studying Qemu, and I came to the conclusion that
> it doesn't make sense to try speeding up the output code, at least not
> yet. A peephole optimizer or hand-coded sequences made to handle common
> combinations of inst
On 17 Apr 2005, at 12:27, Paul Brook wrote:
> Unfortunately it's not that simple. The push instruction may
> cause an exception. Whatever optimizations you apply you've
> got to make sure that the guest state is still consistent when
> the exception occurs.
I brought this up because I want to spee
The attached patch implements single-step mode for arm targets. It doesn't
stop when the instruction condition fails, but I'll fix that later.
Paul
Index: translate.c
===
RCS file: /cvsroot/qemu/qemu/target-arm/translate.c,v
retrievi
> Can it be run as simple user, or power user, or administrator?
> I expect you must be admin on XP to be able to launch some service.
> I'll try tomorrow anyway :)
The kqemu kernel module for Windows is not yet available, but it will
come soon.
Fabrice.
__
On Sun, Apr 17, 2005 at 09:31:32PM +0200, Fabrice Bellard wrote:
>In fact the real fix is to delete the S3VGA stuff - it has never worked
>and I have no plan to support it. Moveover, a correct implementation
>would be made in another driver as the cirrus_vga driver.
Also remove the DEBUG_TB_CHEC
Can it be run as simple user, or power user, or administrator?
I expect you must be admin on XP to be able to launch some service.
I'll try tomorrow anyway :)
--
Christian
On 4/17/05, André Braga <[EMAIL PROTECTED]> wrote:
> WOOHOO!
>
> And I thought that wouldn't be possible without some seriou
The attached patch adds support for ARM "Angel" semihosting syscalls.
This is a simple syscall layer which provides basic OS services to embedded
applications. It is implemented by several free and proprietary toolchains/
debugging tools, including newlib and the GDB simulator. It's what you get
CVSROOT:/cvsroot/qemu
Module name:qemu
Branch:
Changes by: Fabrice Bellard <[EMAIL PROTECTED]> 05/04/17 19:49:03
Modified files:
target-i386: translate.c
Log message:
verr and verw eflags opt fix
CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewc
CVSROOT:/cvsroot/qemu
Module name:qemu
Branch:
Changes by: Fabrice Bellard <[EMAIL PROTECTED]> 05/04/17 19:50:21
Modified files:
target-i386: helper.c
Log message:
make lsl, lar verr and verw exception safe
CVSWeb URLs:
http://savannah.gnu.org/cg
The patch below fixes a thinko in the arm cpu state dumping so the FPSCR is
only displayed once.
Paul
Index: target-arm/translate.c
===
RCS file: /cvsroot/qemu/qemu/target-arm/translate.c,v
retrieving revision 1.20
diff -u -p -r1.20
The patch below adds nominal cygwin host support. It basically just makes it
pretend to be mingw.
Paul
Index: configure
===
RCS file: /cvsroot/qemu/qemu/configure,v
retrieving revision 1.61
diff -u -p -r1.61 configure
--- configure
WOOHOO!
And I thought that wouldn't be possible without some serious rewriting :D
Thumbs up for KKEMU/QEMU's design!
2005/4/17, Fabrice Bellard <[EMAIL PROTECTED]>:
> CVSROOT:/cvsroot/qemu
> Module name:qemu
> Branch:
> Changes by: Fabrice Bellard <[EMAIL PROTECTED]> 05/04/
CVSROOT:/cvsroot/qemu
Module name:qemu
Branch:
Changes by: Fabrice Bellard <[EMAIL PROTECTED]> 05/04/17 19:16:13
Modified files:
. : Makefile.target exec.c gdbstub.c vl.h
linux-user : main.c qemu.h signal.c
target-arm : cp
In fact the real fix is to delete the S3VGA stuff - it has never worked
and I have no plan to support it. Moveover, a correct implementation
would be made in another driver as the cirrus_vga driver.
Fabrice.
Bernhard Fischer wrote:
Hi,
qemu/hw/vga.c: typo: s/uinr32_t/uint32_t/
Thank you,
---
Hi,
qemu/hw/vga.c: typo: s/uinr32_t/uint32_t/
Thank you,
diff -X excl -rduNp qemu.oorig/hw/vga.c qemu/hw/vga.c
--- qemu.oorig/hw/vga.c 2005-02-10 23:00:27.0 +0100
+++ qemu/hw/vga.c 2005-04-17 02:47:10.0 +0200
@@ -956,7 +956,7 @@ static void vga_get_offsets(VGAState *s,
CVSROOT:/cvsroot/qemu
Module name:qemu
Branch:
Changes by: Fabrice Bellard <[EMAIL PROTECTED]> 05/04/17 18:32:14
Modified files:
. : exec-all.h
Log message:
removed unused stuff
CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewcvs/qemu/q
CVSROOT:/cvsroot/qemu
Module name:qemu
Branch:
Changes by: Fabrice Bellard <[EMAIL PROTECTED]> 05/04/17 18:33:47
Modified files:
. : osdep.c kqemu.c
Log message:
windows support for kqemu (Filip Navara)
CVSWeb URLs:
http://savannah.gnu.o
CVSROOT:/cvsroot/qemu
Module name:qemu
Branch:
Changes by: Fabrice Bellard <[EMAIL PROTECTED]> 05/04/17 17:56:18
Modified files:
hw : cirrus_vga_rop2.h cirrus_vga.c
Log message:
destination write mask support, fixed banked memory access, r
On Sunday 17 April 2005 06:58, Joe Luser wrote:
> If you know of areas that will make a big difference in speed, please
> let me know.
IMHO the most effective speedup for most users is to write custom guest device
drivers for IDE/network/video that talk directly to qemu using some low
overhead m
On 17 Apr 2005, at 12:27, Paul Brook wrote:
Unfortunately it's not that simple. The push instruction may cause an
exception. Whatever optimizations you apply you've got to make sure
that the
guest state is still consistent when the exception occurs.
If we just concatenate the C code of the two pro
On Sunday 17 April 2005 09:59, Jonas Maebe wrote:
> On 17 Apr 2005, at 10:21, John R. Hogerhuis wrote:
> > One thought would be to have a peephole optimizer that looks back over
> > the just translated basic block (or a state machine that matches such
> > sequences as an on-line algorithm) and matc
On 17 Apr 2005, at 10:21, John R. Hogerhuis wrote:
One thought would be to have a peephole optimizer that looks back over
the just translated basic block (or a state machine that matches such
sequences as an on-line algorithm) and match against common, known
primitive sequences, and replaces them w
On Sat, 2005-04-16 at 22:58 -0700, Joe Luser wrote:
> Hi folks,
>
> I've been using Qemu on the Mac for a few days now; Several OSes
> running (including Windoze), and I'm impressed. The source looks pretty
> clean, too.
>
> Has anyone done any profiling work to see where Qemu spends most of its
25 matches
Mail list logo
