From: Wei Liu <wei.liu2@citrix.com>
To: xen-devel@lists.xen.org
Cc: Wei Liu <wei.liu2@citrix.com>,
keir@xen.org, ian.campbell@citrix.com, jbeulich@suse.com,
david.vrabel@citrix.com
Subject: [RFC PATCH V4 17/18] Only allow extended event channel on Dom0 and driver domains
Date: Tue, 5 Mar 2013 12:30:39 +0000 [thread overview]
Message-ID: <1362486640-14707-18-git-send-email-wei.liu2@citrix.com> (raw)
In-Reply-To: <1362486640-14707-1-git-send-email-wei.liu2@citrix.com>
For non-Dom0 domains, add a flag to indicate whether it can use any extended
event channel ABIs. Admins can specify this flag when creating a driver
domain.
The rationale behind this option is, extended event channel ABIs will consume
global mapping space in Xen, Admin should have control over these features.
Signed-off-by: Wei Liu <wei.liu2@citrix.com>
---
xen/common/domain.c | 3 +++
xen/common/domctl.c | 6 +++++-
xen/common/event_channel.c | 9 ++++++++-
xen/include/public/domctl.h | 3 +++
xen/include/xen/sched.h | 5 +++++
5 files changed, 24 insertions(+), 2 deletions(-)
diff --git a/xen/common/domain.c b/xen/common/domain.c
index b360de1..f648601 100644
--- a/xen/common/domain.c
+++ b/xen/common/domain.c
@@ -250,6 +250,9 @@ struct domain *domain_create(
if ( domcr_flags & DOMCRF_dummy )
return d;
+ if ( domcr_flags & DOMCRF_evtchn_extended_allowed )
+ d->evtchn_extended_allowed = 1;
+
if ( !is_idle_domain(d) )
{
if ( (err = xsm_domain_create(XSM_HOOK, d, ssidref)) != 0 )
diff --git a/xen/common/domctl.c b/xen/common/domctl.c
index b7f6619..bb15da4 100644
--- a/xen/common/domctl.c
+++ b/xen/common/domctl.c
@@ -369,7 +369,8 @@ long do_domctl(XEN_GUEST_HANDLE_PARAM(xen_domctl_t) u_domctl)
if ( supervisor_mode_kernel ||
(op->u.createdomain.flags &
~(XEN_DOMCTL_CDF_hvm_guest | XEN_DOMCTL_CDF_hap |
- XEN_DOMCTL_CDF_s3_integrity | XEN_DOMCTL_CDF_oos_off)) )
+ XEN_DOMCTL_CDF_s3_integrity | XEN_DOMCTL_CDF_oos_off |
+ XEN_DOMCTL_CDF_evtchn_extended_allowed)) )
break;
dom = op->domain;
@@ -405,6 +406,9 @@ long do_domctl(XEN_GUEST_HANDLE_PARAM(xen_domctl_t) u_domctl)
domcr_flags |= DOMCRF_s3_integrity;
if ( op->u.createdomain.flags & XEN_DOMCTL_CDF_oos_off )
domcr_flags |= DOMCRF_oos_off;
+ if ( op->u.createdomain.flags & XEN_DOMCTL_CDF_evtchn_extended_allowed )
+ domcr_flags |= DOMCRF_evtchn_extended_allowed;
+
d = domain_create(dom, domcr_flags, op->u.createdomain.ssidref);
if ( IS_ERR(d) )
diff --git a/xen/common/event_channel.c b/xen/common/event_channel.c
index 504d769..a49fe3b 100644
--- a/xen/common/event_channel.c
+++ b/xen/common/event_channel.c
@@ -1409,7 +1409,11 @@ long do_event_channel_op(int cmd, XEN_GUEST_HANDLE_PARAM(void) arg)
case EVTCHNOP_query_extended_abis: {
struct evtchn_query_extended_abis query;
- query.abis = extended_event_channel;
+ struct domain *d = current->domain;
+ if ( d->domain_id == 0 || d->evtchn_extended_allowed )
+ query.abis = extended_event_channel;
+ else
+ query.abis = 0;
rc = 0;
if ( __copy_to_guest(arg, &query, 1) )
rc = -EFAULT;
@@ -1418,6 +1422,9 @@ long do_event_channel_op(int cmd, XEN_GUEST_HANDLE_PARAM(void) arg)
case EVTCHNOP_register_3level: {
struct evtchn_register_3level reg;
+ struct domain *d = current->domain;
+ if ( d->domain_id != 0 && !d->evtchn_extended_allowed )
+ return -EPERM;
if ( copy_from_guest(®, arg, 1) != 0 )
return -EFAULT;
rc = evtchn_register_3level(®);
diff --git a/xen/include/public/domctl.h b/xen/include/public/domctl.h
index deb19db..40e9486 100644
--- a/xen/include/public/domctl.h
+++ b/xen/include/public/domctl.h
@@ -59,6 +59,9 @@ struct xen_domctl_createdomain {
/* Disable out-of-sync shadow page tables? */
#define _XEN_DOMCTL_CDF_oos_off 3
#define XEN_DOMCTL_CDF_oos_off (1U<<_XEN_DOMCTL_CDF_oos_off)
+ /* Can this domain use any extended event channel ABIs? */
+#define _XEN_DOMCTL_CDF_evtchn_extended_allowed 4
+#define XEN_DOMCTL_CDF_evtchn_extended_allowed (1U<<_XEN_DOMCTL_CDF_evtchn_extended_allowed)
uint32_t flags;
};
typedef struct xen_domctl_createdomain xen_domctl_createdomain_t;
diff --git a/xen/include/xen/sched.h b/xen/include/xen/sched.h
index cca5e7f..b190fd0 100644
--- a/xen/include/xen/sched.h
+++ b/xen/include/xen/sched.h
@@ -256,6 +256,8 @@ struct domain
bool_t is_paused_by_controller;
/* Domain's VCPUs are pinned 1:1 to physical CPUs? */
bool_t is_pinned;
+ /* Can this domain use any extended event channel ABIs? */
+ bool_t evtchn_extended_allowed;
/* Are any VCPUs polling event channels (SCHEDOP_poll)? */
#if MAX_VIRT_CPUS <= BITS_PER_LONG
@@ -411,6 +413,9 @@ struct domain *domain_create(
/* DOMCRF_oos_off: dont use out-of-sync optimization for shadow page tables */
#define _DOMCRF_oos_off 4
#define DOMCRF_oos_off (1U<<_DOMCRF_oos_off)
+/* DOMCRF_evtchn_extended_allowed: this domain can use extended evtchn ABIs */
+#define _DOMCRF_evtchn_extended_allowed 5
+#define DOMCRF_evtchn_extended_allowed (1U<<_DOMCRF_evtchn_extended_allowed)
/*
* rcu_lock_domain_by_id() is more efficient than get_domain_by_id().
--
1.7.10.4
next prev parent reply other threads:[~2013-03-05 12:30 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-05 12:30 [RFC PATCH V4] Implement 3-level event channel ABI in Xen Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 01/18] xen: correct BITS_PER_EVTCHN_WORD on arm Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 02/18] Clean up trailing whitespaces Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 03/18] Dynamically allocate d->evtchn Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 04/18] Move event channel macros / struct definition to proper place Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 05/18] Add d->max_evtchns Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 06/18] Add evtchn_is_{pending, masked} and evtchn_clear_pending Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 07/18] Implement extended event channel ABIs query Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 08/18] Define 3-level event channel registration interface Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 09/18] Add evtchn_extended in struct domain Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 10/18] Calculate max event channels for EVTCHN_EXTENDED_L3 Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 11/18] Bump EVTCHNS_PER_BUCKET to 512 Wei Liu
2013-03-15 18:15 ` David Vrabel
2013-03-15 18:37 ` Wei Liu
2013-03-15 18:43 ` David Vrabel
2013-03-05 12:30 ` [RFC PATCH V4 12/18] Update Xen public header xen.h Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 13/18] Genneralized event channel operations Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 14/18] Introduce EVTCHN_WORD_BITORDER macro Wei Liu
2013-03-05 12:30 ` [RFC PATCH V4 15/18] Infrastructure to manipulate 3-level event channel pages Wei Liu
2013-03-05 14:22 ` Jan Beulich
2013-03-05 12:30 ` [RFC PATCH V4 16/18] Implement 3-level event channel routines Wei Liu
2013-03-05 14:28 ` Jan Beulich
2013-03-05 16:07 ` Wei Liu
2013-03-05 16:13 ` Jan Beulich
2013-03-05 12:30 ` Wei Liu [this message]
2013-03-05 12:30 ` [RFC PATCH V4 18/18] libxl: add evtchn_extended_allowed flag Wei Liu
2013-03-05 13:48 ` Ian Jackson
2013-03-05 17:11 ` Wei Liu
2013-03-05 17:38 ` Ian Jackson
2013-03-05 17:51 ` Wei Liu
2013-03-05 17:56 ` David Vrabel
2013-03-05 18:08 ` Wei Liu
2013-03-06 17:16 ` Wei Liu
2013-03-07 11:23 ` David Vrabel
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=1362486640-14707-18-git-send-email-wei.liu2@citrix.com \
--to=wei.liu2@citrix.com \
--cc=david.vrabel@citrix.com \
--cc=ian.campbell@citrix.com \
--cc=jbeulich@suse.com \
--cc=keir@xen.org \
--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).