From: Paolo Bonzini <pbonzini@redhat.com>
To: Xiaoyao Li <xiaoyao.li@intel.com>,
Eduardo Habkost <ehabkost@redhat.com>,
Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>, qemu-devel@nongnu.org
Subject: Re: [PATCH v2] accel/kvm: Make "kernel_irqchip" default on
Date: Tue, 7 Jan 2020 11:32:42 +0100 [thread overview]
Message-ID: <05532242-3a70-b547-3765-0ab29583c4b0@redhat.com> (raw)
In-Reply-To: <20191228104326.21732-1-xiaoyao.li@intel.com>
On 28/12/19 11:43, Xiaoyao Li wrote:
> Commit 11bc4a13d1f4 ("kvm: convert "-machine kernel_irqchip" to an
> accelerator property") moves kernel_irqchip property from "-machine" to
> "-accel kvm", but it forgets to set the default value of
> kernel_irqchip_allowed and kernel_irqchip_split.
>
> Also cleaning up the three useless members (kernel_irqchip_allowed,
> kernel_irqchip_required, kernel_irqchip_split) in struct MachineState.
>
> Fixes: 11bc4a13d1f4 ("kvm: convert "-machine kernel_irqchip" to an accelerator property")
> Reported-by: Vitaly Kuznetsov <vkuznets@redhat.com>
> Signed-off-by: Xiaoyao Li <xiaoyao.li@intel.com>
> ---
> Changes in v2:
> - Add Reported-by tag;
> - Initialize kernel_irqchip_split in init_machine();
Now that I am actually reviewing the patch on something other than a
phone, I think this would break "-machine kernel_irqchip=split". I'll
test, and squash if it works, something like this:
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index afbbe0a1af..ea35433170 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -98,7 +98,7 @@ struct KVMState
int kvm_shadow_mem;
bool kernel_irqchip_allowed;
bool kernel_irqchip_required;
- bool kernel_irqchip_split;
+ OnOffAuto kernel_irqchip_split;
bool sync_mmu;
bool manual_dirty_log_protect;
/* The man page (and posix) say ioctl numbers are signed int, but
@@ -1799,7 +1799,7 @@ static void kvm_irqchip_create(KVMState *s)
* in-kernel irqchip for us */
ret = kvm_arch_irqchip_create(s);
if (ret == 0) {
- if (s->kernel_irqchip_split) {
+ if (s->kernel_irqchip_split == ON_OFF_AUTO_ON) {
perror("Split IRQ chip mode not supported.");
exit(1);
} else {
@@ -2070,7 +2070,9 @@ static int kvm_init(MachineState *ms)
goto err;
}
- s->kernel_irqchip_split = mc->default_kernel_irqchip_split;
+ if (s->kernel_irqchip_split == ON_OFF_AUTO_AUTO) {
+ s->kernel_irqchip_split = mc->default_kernel_irqchip_split ? ON_OFF_AUTO_ON : ON_OFF_AUTO_OFF;
+ }
if (s->kernel_irqchip_allowed) {
kvm_irqchip_create(s);
@@ -3007,17 +3009,17 @@ static void kvm_set_kernel_irqchip(Object *obj, Visitor *v,
case ON_OFF_SPLIT_ON:
s->kernel_irqchip_allowed = true;
s->kernel_irqchip_required = true;
- s->kernel_irqchip_split = false;
+ s->kernel_irqchip_split = ON_OFF_AUTO_OFF;
break;
case ON_OFF_SPLIT_OFF:
s->kernel_irqchip_allowed = false;
s->kernel_irqchip_required = false;
- s->kernel_irqchip_split = false;
+ s->kernel_irqchip_split = ON_OFF_AUTO_OFF;
break;
case ON_OFF_SPLIT_SPLIT:
s->kernel_irqchip_allowed = true;
s->kernel_irqchip_required = true;
- s->kernel_irqchip_split = true;
+ s->kernel_irqchip_split = ON_OFF_AUTO_ON;
break;
default:
/* The value was checked in visit_type_OnOffSplit() above. If
@@ -3040,7 +3042,7 @@ bool kvm_kernel_irqchip_required(void)
bool kvm_kernel_irqchip_split(void)
{
- return kvm_state->kernel_irqchip_split;
+ return kvm_state->kernel_irqchip_split == ON_OFF_AUTO_ON;
}
static void kvm_accel_instance_init(Object *obj)
@@ -3049,6 +3051,7 @@ static void kvm_accel_instance_init(Object *obj)
s->kvm_shadow_mem = -1;
s->kernel_irqchip_allowed = true;
+ s->kernel_irqchip_split = ON_OFF_AUTO_AUTO;
}
static void kvm_accel_class_init(ObjectClass *oc, void *data)
As a follow up, kernel_irqchip_allowed and kernel_irqchip_required could also be changed to a single OnOffAuto field, I think.
Paolo
next prev parent reply other threads:[~2020-01-07 10:52 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-28 10:43 [PATCH v2] accel/kvm: Make "kernel_irqchip" default on Xiaoyao Li
2019-12-30 12:01 ` Michal Prívozník
2020-01-07 10:32 ` Paolo Bonzini [this message]
2020-01-08 10:17 ` Cornelia Huck
2020-01-08 11:38 ` Xiaoyao Li
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=05532242-3a70-b547-3765-0ab29583c4b0@redhat.com \
--to=pbonzini@redhat.com \
--cc=ehabkost@redhat.com \
--cc=marcel.apfelbaum@gmail.com \
--cc=qemu-devel@nongnu.org \
--cc=vkuznets@redhat.com \
--cc=xiaoyao.li@intel.com \
/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).