From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Xen-devel <xen-devel@lists.xenproject.org>
Cc: "Andrew Cooper" <andrew.cooper3@citrix.com>,
"Ian Jackson" <Ian.Jackson@citrix.com>, "Wei Liu" <wl@xen.org>,
"Jan Beulich" <JBeulich@suse.com>,
"Roger Pau Monné" <roger.pau@citrix.com>
Subject: [Xen-devel] [PATCH v2 07/10] tools/libxc: Rework xc_cpuid_set() to use {get, set}_cpu_policy()
Date: Fri, 13 Sep 2019 20:27:56 +0100 [thread overview]
Message-ID: <20190913192759.10795-8-andrew.cooper3@citrix.com> (raw)
In-Reply-To: <20190913192759.10795-1-andrew.cooper3@citrix.com>
The purpose of this change is to stop using xc_cpuid_do_domctl(), and to stop
basing decisions on a local CPUID instruction. This is not an appropriate way
to construct policy information for other domains.
Obtain the host and domain-max policies from Xen, and mix the results as
before. Provide rather more error logging than before.
No semantics changes to xc_cpuid_set(). There are conceptual problems with
how the function works, which will be addressed in future toolstack work.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
---
CC: Jan Beulich <JBeulich@suse.com>
CC: Wei Liu <wl@xen.org>
CC: Roger Pau Monné <roger.pau@citrix.com>
CC: Ian Jackson <Ian.Jackson@citrix.com>
---
tools/libxc/xc_cpuid_x86.c | 95 ++++++++++++++++++++++++++++++++++++++++------
1 file changed, 84 insertions(+), 11 deletions(-)
diff --git a/tools/libxc/xc_cpuid_x86.c b/tools/libxc/xc_cpuid_x86.c
index 8785cae329..77f96a4ea6 100644
--- a/tools/libxc/xc_cpuid_x86.c
+++ b/tools/libxc/xc_cpuid_x86.c
@@ -902,20 +902,80 @@ int xc_cpuid_set(
const char **config, char **config_transformed)
{
int rc;
- unsigned int i, j, regs[4], polregs[4];
- struct cpuid_domain_info info = {};
+ unsigned int i, j, regs[4] = {}, polregs[4] = {};
+ xc_dominfo_t di;
+ xen_cpuid_leaf_t *leaves = NULL;
+ unsigned int nr_leaves, policy_leaves, nr_msrs;
+ uint32_t err_leaf = -1, err_subleaf = -1, err_msr = -1;
for ( i = 0; i < 4; ++i )
config_transformed[i] = NULL;
- rc = get_cpuid_domain_info(xch, domid, &info, NULL, 0);
+ if ( xc_domain_getinfo(xch, domid, 1, &di) != 1 ||
+ di.domid != domid )
+ {
+ ERROR("Failed to obtain d%d info", domid);
+ rc = -ESRCH;
+ goto fail;
+ }
+
+ rc = xc_get_cpu_policy_size(xch, &nr_leaves, &nr_msrs);
if ( rc )
- goto out;
+ {
+ PERROR("Failed to obtain policy info size");
+ rc = -errno;
+ goto fail;
+ }
- cpuid(input, regs);
+ rc = -ENOMEM;
+ if ( (leaves = calloc(nr_leaves, sizeof(*leaves))) == NULL )
+ {
+ ERROR("Unable to allocate memory for %u CPUID leaves", nr_leaves);
+ goto fail;
+ }
- memcpy(polregs, regs, sizeof(regs));
- xc_cpuid_policy(&info, input, polregs);
+ /* Get the domain's max policy. */
+ nr_msrs = 0;
+ policy_leaves = nr_leaves;
+ rc = xc_get_system_cpu_policy(xch, di.hvm ? XEN_SYSCTL_cpu_policy_hvm_max
+ : XEN_SYSCTL_cpu_policy_pv_max,
+ &policy_leaves, leaves, &nr_msrs, NULL);
+ if ( rc )
+ {
+ PERROR("Failed to obtain %s max policy", di.hvm ? "hvm" : "pv");
+ rc = -errno;
+ goto fail;
+ }
+ for ( i = 0; i < policy_leaves; ++i )
+ if ( leaves[i].leaf == input[0] && leaves[i].subleaf == input[1] )
+ {
+ polregs[0] = leaves[i].a;
+ polregs[1] = leaves[i].b;
+ polregs[2] = leaves[i].c;
+ polregs[3] = leaves[i].d;
+ break;
+ }
+
+ /* Get the host policy. */
+ nr_msrs = 0;
+ policy_leaves = nr_leaves;
+ rc = xc_get_system_cpu_policy(xch, XEN_SYSCTL_cpu_policy_host,
+ &policy_leaves, leaves, &nr_msrs, NULL);
+ if ( rc )
+ {
+ PERROR("Failed to obtain host policy");
+ rc = -errno;
+ goto fail;
+ }
+ for ( i = 0; i < policy_leaves; ++i )
+ if ( leaves[i].leaf == input[0] && leaves[i].subleaf == input[1] )
+ {
+ regs[0] = leaves[i].a;
+ regs[1] = leaves[i].b;
+ regs[2] = leaves[i].c;
+ regs[3] = leaves[i].d;
+ break;
+ }
for ( i = 0; i < 4; i++ )
{
@@ -966,9 +1026,21 @@ int xc_cpuid_set(
}
}
- rc = xc_cpuid_do_domctl(xch, domid, input, regs);
- if ( rc == 0 )
- goto out;
+ /* Feed the transformed leaf back up to Xen. */
+ leaves[0] = (xen_cpuid_leaf_t){ input[0], input[1],
+ regs[0], regs[1], regs[2], regs[3] };
+ rc = xc_set_domain_cpu_policy(xch, domid, 1, leaves, 0, NULL,
+ &err_leaf, &err_subleaf, &err_msr);
+ if ( rc )
+ {
+ PERROR("Failed to set d%d's policy (err leaf %#x, subleaf %#x, msr %#x)",
+ domid, err_leaf, err_subleaf, err_msr);
+ rc = -errno;
+ goto fail;
+ }
+
+ /* Success! */
+ goto out;
fail:
for ( i = 0; i < 4; i++ )
@@ -978,6 +1050,7 @@ int xc_cpuid_set(
}
out:
- free_cpuid_domain_info(&info);
+ free(leaves);
+
return rc;
}
--
2.11.0
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
next prev parent reply other threads:[~2019-09-13 19:29 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-13 19:27 [Xen-devel] [PATCH v2 00/10] x86/cpuid: Switch to using XEN_DOMCTL_set_cpumsr_policy Andrew Cooper
2019-09-13 19:27 ` [Xen-devel] [PATCH v2 01/10] x86/msr: Offer CPUID Faulting to PVH control domains Andrew Cooper
2019-09-16 10:53 ` Jan Beulich
2019-09-13 19:27 ` [Xen-devel] [PATCH v2 02/10] libx86: Proactively initialise error pointers Andrew Cooper
[not found] ` <527f33ad-3de1-15c7-eb4b-603eaf65f3c5@suse.com>
[not found] ` <65f18521-15c5-72a9-29f6-cd5d621e1283@citrix.com>
2019-09-16 15:46 ` Jan Beulich
2019-09-13 19:27 ` [Xen-devel] [PATCH v2 03/10] libx86: Introduce x86_cpu_policies_are_compatible() Andrew Cooper
2019-09-16 10:59 ` Jan Beulich
2019-09-16 15:31 ` Andrew Cooper
2019-09-13 19:27 ` [Xen-devel] [PATCH v2 04/10] x86/cpuid: Split update_domain_cpuid_info() in half Andrew Cooper
2019-09-13 19:27 ` [Xen-devel] [PATCH v2 05/10] x86/domctl: Implement XEN_DOMCTL_set_cpumsr_policy Andrew Cooper
2019-09-16 11:04 ` Jan Beulich
2019-09-16 15:40 ` Andrew Cooper
2019-09-13 19:27 ` [Xen-devel] [PATCH v2 06/10] tools/libxc: Pre-cleanup for xc_cpuid_{set, apply_policy}() Andrew Cooper
2019-09-16 11:09 ` Jan Beulich
2019-09-16 15:42 ` Andrew Cooper
2019-09-13 19:27 ` Andrew Cooper [this message]
2019-09-13 19:27 ` [Xen-devel] [PATCH v2 08/10] tools/libxc: Rework xc_cpuid_apply_policy() to use {get, set}_cpu_policy() Andrew Cooper
2019-09-16 11:17 ` Jan Beulich
2019-09-16 13:41 ` Wei Liu
2019-09-16 15:49 ` Andrew Cooper
2019-09-16 16:05 ` Jan Beulich
2019-09-18 16:09 ` Jan Beulich
2019-09-19 8:48 ` Andrew Cooper
2019-09-25 18:11 ` [Xen-devel] [PATCH v3 " Andrew Cooper
2019-09-26 8:04 ` Jan Beulich
2019-09-26 12:25 ` Andrew Cooper
2019-09-13 19:27 ` [Xen-devel] [PATCH v2 09/10] x86/domctl: Drop XEN_DOMCTL_set_cpuid Andrew Cooper
2019-09-13 19:27 ` [Xen-devel] [PATCH v2 10/10] x86/cpuid: Enable CPUID Faulting for PV control domains by default Andrew Cooper
2019-09-16 11:22 ` Jan Beulich
2019-09-16 15:52 ` Andrew Cooper
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=20190913192759.10795-8-andrew.cooper3@citrix.com \
--to=andrew.cooper3@citrix.com \
--cc=Ian.Jackson@citrix.com \
--cc=JBeulich@suse.com \
--cc=roger.pau@citrix.com \
--cc=wl@xen.org \
--cc=xen-devel@lists.xenproject.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).