From: "Jan Beulich" <JBeulich@suse.com>
To: Feng Wu <feng.wu@intel.com>
Cc: george.dunlap@eu.citrix.com, andrew.cooper3@citrix.com,
dario.faggioli@citrix.com, kevin.tian@intel.com,
xen-devel@lists.xen.org
Subject: Re: [PATCH v5 4/7] VMX: Make sure PI is in proper state before install the hooks
Date: Wed, 12 Oct 2016 07:45:16 -0600 [thread overview]
Message-ID: <57FE5A8C0200007800116C2F@prv-mh.provo.novell.com> (raw)
In-Reply-To: <1476147473-30970-5-git-send-email-feng.wu@intel.com>
>>> On 11.10.16 at 02:57, <feng.wu@intel.com> wrote:
> static void pi_desc_init(struct vcpu *v)
> {
> - uint32_t dest;
> -
> v->arch.hvm_vmx.pi_desc.nv = posted_intr_vector;
>
> - dest = cpu_physical_id(v->processor);
> -
> - if ( x2apic_enabled )
> - v->arch.hvm_vmx.pi_desc.ndst = dest;
> - else
> - v->arch.hvm_vmx.pi_desc.ndst = MASK_INSR(dest, PI_xAPIC_NDST_MASK);
> + /*
> + * Mark NDST as invalid, then we can use this invalid value as a
> + * marker to whether update NDST or not in vmx_pi_hooks_assign().
> + */
> + v->arch.hvm_vmx.pi_desc.ndst = 0xffffffff;
I think I had at the same time asked to make this a #define, so the
two (currently) instance can be connected together.
> --- a/xen/arch/x86/hvm/vmx/vmx.c
> +++ b/xen/arch/x86/hvm/vmx/vmx.c
> @@ -211,14 +211,39 @@ static void vmx_pi_do_resume(struct vcpu *v)
> /* This function is called when pcidevs_lock is held */
> void vmx_pi_hooks_assign(struct domain *d)
> {
> + struct vcpu *v;
> +
> if ( !iommu_intpost || !has_hvm_container_domain(d) )
> return;
>
> ASSERT(!d->arch.hvm_domain.vmx.vcpu_block);
>
> - d->arch.hvm_domain.vmx.vcpu_block = vmx_vcpu_block;
> + /*
> + * We carefully handle the timing here:
> + * - Install the context switch first
> + * - Then set the NDST field
> + * - Install the block and resume hooks in the end
> + *
> + * This can make sure the PI (especially the NDST feild) is
> + * in proper state when we call vmx_vcpu_block().
> + */
> d->arch.hvm_domain.vmx.pi_switch_from = vmx_pi_switch_from;
> d->arch.hvm_domain.vmx.pi_switch_to = vmx_pi_switch_to;
> +
> + for_each_vcpu ( d, v )
> + {
> + unsigned int dest = cpu_physical_id(v->processor);
> + struct pi_desc *pi_desc = &v->arch.hvm_vmx.pi_desc;
> +
> + /*
> + * We don't need to update NDST if 'arch.hvm_domain.vmx.pi_switch_to'
> + * already gets called
I think you mean "got" or "has got". Also I think you'd better refer to
the actual VMX function instead of the hook pointer field.
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
next prev parent reply other threads:[~2016-10-12 13:45 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-10-11 0:57 [PATCH v5 0/7] VMX: Properly handle pi descriptor and per-cpu blocking list Feng Wu
2016-10-11 0:57 ` [PATCH v5 1/7] VMX: Statically assign two PI hooks Feng Wu
2016-10-11 8:11 ` Tian, Kevin
2016-10-12 13:25 ` Jan Beulich
2016-10-11 0:57 ` [PATCH v5 2/7] VMX: Properly handle pi when all the assigned devices are removed Feng Wu
2016-10-11 8:20 ` Tian, Kevin
2016-10-11 0:57 ` [PATCH v5 3/7] VMX: Cleanup PI per-cpu blocking list when vcpu is destroyed Feng Wu
2016-10-11 0:57 ` [PATCH v5 4/7] VMX: Make sure PI is in proper state before install the hooks Feng Wu
2016-10-12 13:45 ` Jan Beulich [this message]
2016-10-17 6:26 ` Wu, Feng
2016-10-24 7:22 ` Jan Beulich
2016-10-24 7:45 ` Wu, Feng
2016-10-24 7:59 ` Jan Beulich
2016-10-11 0:57 ` [PATCH v5 5/7] VT-d: No need to set irq affinity for posted format IRTE Feng Wu
2016-10-12 13:56 ` Jan Beulich
2016-10-17 7:02 ` Wu, Feng
2016-10-24 7:27 ` Jan Beulich
2016-10-24 8:57 ` Wu, Feng
2016-10-24 9:53 ` Jan Beulich
2016-10-24 10:18 ` Wu, Feng
2016-10-24 10:57 ` Jan Beulich
2016-10-24 11:10 ` Wu, Feng
2016-10-24 11:31 ` Jan Beulich
2016-10-25 1:04 ` Wu, Feng
2016-10-25 5:57 ` Tian, Kevin
2016-10-25 8:09 ` Jan Beulich
2016-10-26 9:12 ` Wu, Feng
2016-10-26 9:51 ` Jan Beulich
2016-10-11 0:57 ` [PATCH v5 6/7] VT-d: Some cleanups Feng Wu
2016-10-12 14:01 ` Jan Beulich
2016-10-11 0:57 ` [PATCH v5 7/7] VMX: Fixup PI descriptor when cpu is offline Feng Wu
2016-10-11 8:38 ` Tian, Kevin
2016-10-11 11:46 ` Wu, Feng
2016-10-11 8:08 ` [PATCH v5 0/7] VMX: Properly handle pi descriptor and per-cpu blocking list Tian, Kevin
2016-10-11 8:11 ` Wu, Feng
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=57FE5A8C0200007800116C2F@prv-mh.provo.novell.com \
--to=jbeulich@suse.com \
--cc=andrew.cooper3@citrix.com \
--cc=dario.faggioli@citrix.com \
--cc=feng.wu@intel.com \
--cc=george.dunlap@eu.citrix.com \
--cc=kevin.tian@intel.com \
--cc=xen-devel@lists.xen.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).