From: Gleb Natapov <gleb@redhat.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: "avi@redhat.com" <avi@redhat.com>, kvm-devel <kvm@vger.kernel.org>
Subject: Re: [PATCH] call kvm_cpu_synchronize_state() on target vcpu
Date: Wed, 9 Sep 2009 19:07:34 +0300 [thread overview]
Message-ID: <20090909160734.GG22885@redhat.com> (raw)
In-Reply-To: <4AA7D074.5010207@siemens.com>
On Wed, Sep 09, 2009 at 05:57:40PM +0200, Jan Kiszka wrote:
> Gleb Natapov wrote:
> > On Wed, Sep 09, 2009 at 05:47:13PM +0200, Jan Kiszka wrote:
> >> Gleb Natapov wrote:
> >>> regs_modified logic doesn't work if io thread calls
> >>> kvm_cpu_synchronize_state() since kvm_arch_get_registers()
> >>> returns only after vcpu thread is back to kernel. Setting
> >>> regs_modified to 1 at this stage causes loading of wrong vcpu
> >>> state on the next vcpu_run().
> >> We need this upstream too, right? Could you file the corresponding patch?
> >>
> > Upstream is single threaded. It shouldn't suffer from this bug.
>
> Not if you enable iothread support (though I don't remember if that
It can't work with kvm since all vcpu ioctls are called on the thread
that issues them.
> works now for kvm) + you are also touching shared code here. So qemu-kvm
> would benefit from keeping the diff small.
>
The patch doesn't touch shared code. (it is almost impossible to tell
what code is shared and what's not nowadays)
> Jan
>
> >
> >> Thanks,
> >> Jan
> >>
> >>> Signed-off-by: Gleb Natapov <gleb@redhat.com>
> >>> diff --git a/qemu-kvm.c b/qemu-kvm.c
> >>> index 06efd41..9ab0cec 100644
> >>> --- a/qemu-kvm.c
> >>> +++ b/qemu-kvm.c
> >>> @@ -874,14 +874,6 @@ int kvm_set_mpstate(kvm_vcpu_context_t vcpu, struct kvm_mp_state *mp_state)
> >>> }
> >>> #endif
> >>>
> >>> -void kvm_cpu_synchronize_state(CPUState *env)
> >>> -{
> >>> - if (!env->kvm_cpu_state.regs_modified) {
> >>> - kvm_arch_get_registers(env);
> >>> - env->kvm_cpu_state.regs_modified = 1;
> >>> - }
> >>> -}
> >>> -
> >>> static int handle_mmio(kvm_vcpu_context_t vcpu)
> >>> {
> >>> unsigned long addr = vcpu->run->mmio.phys_addr;
> >>> @@ -1539,6 +1531,22 @@ static void on_vcpu(CPUState *env, void (*func)(void *data), void *data)
> >>> qemu_cond_wait(&qemu_work_cond);
> >>> }
> >>>
> >>> +static void do_kvm_cpu_synchronize_state(void *_env)
> >>> +{
> >>> + CPUState *env = _env;
> >>> + if (!env->kvm_cpu_state.regs_modified) {
> >>> + kvm_arch_save_regs(env);
> >>> + kvm_arch_load_mpstate(env);
> >>> + env->kvm_cpu_state.regs_modified = 1;
> >>> + }
> >>> +}
> >>> +
> >>> +void kvm_cpu_synchronize_state(CPUState *env)
> >>> +{
> >>> + if (!env->kvm_cpu_state.regs_modified)
> >>> + on_vcpu(env, do_kvm_cpu_synchronize_state, env);
> >>> +}
> >>> +
> >>> static void inject_interrupt(void *data)
> >>> {
> >>> cpu_interrupt(current_env, (long) data);
> >>> diff --git a/qemu-kvm.h b/qemu-kvm.h
> >>> index 2c1730b..32f74b3 100644
> >>> --- a/qemu-kvm.h
> >>> +++ b/qemu-kvm.h
> >>> @@ -1153,12 +1153,6 @@ static inline int kvm_sync_vcpus(void)
> >>> return 0;
> >>> }
> >>>
> >>> -static inline void kvm_arch_get_registers(CPUState *env)
> >>> -{
> >>> - kvm_save_registers(env);
> >>> - kvm_save_mpstate(env);
> >>> -}
> >>> -
> >>> static inline void kvm_arch_put_registers(CPUState *env)
> >>> {
> >>> kvm_load_registers(env);
> >>> diff --git a/target-i386/helper.c b/target-i386/helper.c
> >>> index 4a16887..57c74a2 100644
> >>> --- a/target-i386/helper.c
> >>> +++ b/target-i386/helper.c
> >>> @@ -746,7 +746,7 @@ void cpu_dump_state(CPUState *env, FILE *f,
> >>> static const char *seg_name[6] = { "ES", "CS", "SS", "DS", "FS", "GS" };
> >>>
> >>> if (kvm_enabled())
> >>> - kvm_arch_get_registers(env);
> >>> + kvm_cpu_synchronize_state(env);
> >>>
> >>> eflags = env->eflags;
> >>> #ifdef TARGET_X86_64
> >>> --
> >>> Gleb.
> >> --
> >> Siemens AG, Corporate Technology, CT SE 2
> >> Corporate Competence Center Embedded Linux
> >
> > --
> > Gleb.
>
> --
> Siemens AG, Corporate Technology, CT SE 2
> Corporate Competence Center Embedded Linux
--
Gleb.
next prev parent reply other threads:[~2009-09-09 16:07 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-09-09 15:33 [PATCH] call kvm_cpu_synchronize_state() on target vcpu Gleb Natapov
2009-09-09 15:41 ` Avi Kivity
2009-09-09 15:47 ` Jan Kiszka
2009-09-09 15:49 ` Gleb Natapov
2009-09-09 15:57 ` Jan Kiszka
2009-09-09 16:07 ` Gleb Natapov [this message]
2009-09-09 16:21 ` Jan Kiszka
2009-09-09 16:27 ` Gleb Natapov
2009-09-09 16:32 ` Jan Kiszka
2009-09-09 16:36 ` Gleb Natapov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20090909160734.GG22885@redhat.com \
--to=gleb@redhat.com \
--cc=avi@redhat.com \
--cc=jan.kiszka@siemens.com \
--cc=kvm@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).