[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [CFT PATCH 00/12] Tricky parts of my iothread-for-win32
From: |
Aurelien Jarno |
Subject: |
Re: [Qemu-devel] [CFT PATCH 00/12] Tricky parts of my iothread-for-win32 stuff |
Date: |
Tue, 8 Feb 2011 20:31:19 +0100 |
User-agent: |
Mutt/1.5.18 (2008-05-17) |
On Tue, Feb 08, 2011 at 06:18:17PM +0100, Paolo Bonzini wrote:
> Ok, so let's start the dance...
>
> These are the generic parts of the iothread-for-win32 work I made
> last year, rebased to today's qemu (qemu-kvm.git/uq/master merged
> with qemu.git/master, because it conflicts with some patches from Jan).
> I never tested them too much, and that's one of the reasons why I never
> submitted the stuff. Perhaps with some help they can make it this time.
>
> Testing I made now is still a bit limited, but the patches can boot
> RHEL and Fedora under both Windows iothread and Linux iothread. FWIW,
> under Wine I need iothread to boot Linux.
>
> The main idea is to replace timedwaits with waits. First because the
> timedwaits were such only out of laziness; second because timedwait
> is a pain to implement for Win32 where we have to manually implement
> condvars.
Thanks a lot for those patches, if we don't have anybody picking up the
work with this basis, we can definitely consider Windows support dead.
> Anthony Liguori (1):
> io-thread: make sure to initialize qemu_work_cond and qemu_cpu_cond
>
> Paolo Bonzini (11):
> cris, microblaze: use cpu_has_work
> inline cpu_halted into sole caller
> change qemu_thread_equal API to always compare with current thread
> always qemu_cpu_kick after unhalting a cpu
> exit round-robin vcpu loop if cpu->stopped is true
> always signal pause_cond after stopping a VCPU
> do not use timedwait on qemu_halt_cond
> do not use timedwait on qemu_system_cond
> do not use timedwait on qemu_pause_cond
> do not use timedwait on qemu_cpu_cond
> iothread stops the vcpu thread via IPI
>
> cpu-defs.h | 1 +
> cpu-exec.c | 9 +++++++--
> cpus.c | 38 +++++++++++++++++++-------------------
> hw/apic.c | 1 +
> hw/ppc.c | 2 ++
> hw/sun4m.c | 11 +++++++++--
> hw/sun4u.c | 4 ++--
> qemu-thread.c | 4 ++--
> qemu-thread.h | 2 +-
> target-alpha/exec.h | 11 -----------
> target-arm/exec.h | 13 -------------
> target-cris/exec.h | 13 -------------
> target-i386/exec.h | 12 ------------
> target-m68k/exec.h | 10 ----------
> target-microblaze/exec.h | 13 -------------
> target-mips/exec.h | 11 -----------
> target-ppc/exec.h | 11 -----------
> target-s390x/exec.h | 12 ------------
> target-s390x/kvm.c | 1 +
> target-sh4/cpu.h | 1 -
> target-sh4/exec.h | 11 -----------
> target-sparc/exec.h | 10 ----------
> 22 files changed, 45 insertions(+), 156 deletions(-)
>
> --
> 1.7.3.5
>
>
>
--
Aurelien Jarno GPG: 1024D/F1BCDB73
address@hidden http://www.aurel32.net
- [Qemu-devel] Re: [CFT PATCH 03/12] inline cpu_halted into sole caller, (continued)
- Re: [Qemu-devel] [CFT PATCH 00/12] Tricky parts of my iothread-for-win32 stuff,
Aurelien Jarno <=
- [Qemu-devel] [CFT PATCH 10/12] do not use timedwait on qemu_pause_cond, Paolo Bonzini, 2011/02/08
- [Qemu-devel] [CFT PATCH 04/12] change qemu_thread_equal API to always compare with current thread, Paolo Bonzini, 2011/02/08
- [Qemu-devel] [CFT PATCH 05/12] always qemu_cpu_kick after unhalting a cpu, Paolo Bonzini, 2011/02/08
- [Qemu-devel] Re: [CFT PATCH 00/12] Tricky parts of my iothread-for-win32 stuff, Jan Kiszka, 2011/02/08