From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Xen-devel <xen-devel@lists.xen.org>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>,
Ian Jackson <Ian.Jackson@eu.citrix.com>,
Ian Campbell <Ian.Campbell@citrix.com>,
Wei Liu <wei.liu2@citrix.com>
Subject: [PATCH] tools/libxc: Improve efficiency of xc_cpuid_apply_policy()
Date: Mon, 5 Oct 2015 14:12:17 +0100 [thread overview]
Message-ID: <1444050737-8437-1-git-send-email-andrew.cooper3@citrix.com> (raw)
Having the internals of xc_cpuid_policy() make hypercalls to collect domain
information causes xc_cpuid_apply_policy() to be very inefficient.
Re-order operations to collect all information at once at the outermost layer,
and pass a structure in to all cpuid policy generation functions.
This removes several hypercalls (4 from HVM, 3 from PV) for each of the
up-to 108 leaves processed.
No change in the eventual policy provided, although all the information
gathering how has (or has correct) error checking.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
---
CC: Ian Campbell <Ian.Campbell@citrix.com>
CC: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Wei Liu <wei.liu2@citrix.com>
---
tools/libxc/xc_cpuid_x86.c | 245 +++++++++++++++++++++++++-------------------
1 file changed, 141 insertions(+), 104 deletions(-)
diff --git a/tools/libxc/xc_cpuid_x86.c b/tools/libxc/xc_cpuid_x86.c
index e146a3e..715f30c 100644
--- a/tools/libxc/xc_cpuid_x86.c
+++ b/tools/libxc/xc_cpuid_x86.c
@@ -33,6 +33,27 @@
#define DEF_MAX_INTELEXT 0x80000008u
#define DEF_MAX_AMDEXT 0x8000001cu
+struct cpuid_domain_info
+{
+ enum
+ {
+ VENDOR_UNKNOWN,
+ VENDOR_INTEL,
+ VENDOR_AMD,
+ } vendor;
+
+ bool hvm;
+ bool pvh;
+ uint64_t xfeature_mask;
+
+ /* PV-only information. */
+ bool pv64;
+
+ /* HVM-only infomation. */
+ bool pae;
+ bool nestedhvm;
+};
+
static void cpuid(const unsigned int *input, unsigned int *regs)
{
unsigned int count = (input[1] == XEN_CPUID_INPUT_UNUSED) ? 0 : input[1];
@@ -55,24 +76,75 @@ static void cpuid(const unsigned int *input, unsigned int *regs)
#endif
}
-/* Get the manufacturer brand name of the host processor. */
-static void xc_cpuid_brand_get(char *str)
+static int get_cpuid_domain_info(xc_interface *xch, domid_t domid,
+ struct cpuid_domain_info *info)
{
- unsigned int input[2] = { 0, 0 };
- unsigned int regs[4];
+ struct xen_domctl domctl = {};
+ xc_dominfo_t di;
+ unsigned int in[2] = { 0, ~0U }, regs[4];
+ int rc;
- cpuid(input, regs);
+ cpuid(in, regs);
+ if ( regs[1] == 0x756e6547U && /* "GenuineIntel" */
+ regs[2] == 0x6c65746eU &&
+ regs[3] == 0x49656e69U )
+ info->vendor = VENDOR_INTEL;
+ else if ( regs[1] == 0x68747541U && /* "AuthenticAMD" */
+ regs[2] == 0x444d4163U &&
+ regs[3] == 0x69746e65U )
+ info->vendor = VENDOR_AMD;
+ else
+ info->vendor = VENDOR_UNKNOWN;
+
+ if ( xc_domain_getinfo(xch, domid, 1, &di) != 1 ||
+ di.domid != domid )
+ return -ESRCH;
+
+ info->hvm = di.hvm;
+ info->pvh = di.pvh;
+
+ /* Get xstate information. */
+ domctl.cmd = XEN_DOMCTL_getvcpuextstate;
+ domctl.domain = domid;
+ rc = do_domctl(xch, &domctl);
+ if ( rc )
+ return rc;
+
+ info->xfeature_mask = domctl.u.vcpuextstate.xfeature_mask;
+
+ if ( di.hvm )
+ {
+ uint64_t val;
+
+ rc = xc_hvm_param_get(xch, domid, HVM_PARAM_PAE_ENABLED, &val);
+ if ( rc )
+ return rc;
+
+ info->pae = !!val;
+
+ rc = xc_hvm_param_get(xch, domid, HVM_PARAM_NESTEDHVM, &val);
+ if ( rc )
+ return rc;
+
+ info->nestedhvm = !!val;
+ }
+ else
+ {
+ unsigned int width;
+
+ rc = xc_domain_get_guest_width(xch, domid, &width);
+ if ( rc )
+ return rc;
+
+ info->pv64 = (width == 8);
+ }
- *(uint32_t *)(str + 0) = regs[1];
- *(uint32_t *)(str + 4) = regs[3];
- *(uint32_t *)(str + 8) = regs[2];
- str[12] = '\0';
+ return 0;
}
-static void amd_xc_cpuid_policy(
- xc_interface *xch, domid_t domid,
- const unsigned int *input, unsigned int *regs,
- int is_pae, int is_nestedhvm)
+static void amd_xc_cpuid_policy(xc_interface *xch,
+ const struct cpuid_domain_info *info,
+ const unsigned int *input, unsigned int *regs)
{
switch ( input[0] )
{
@@ -87,13 +159,13 @@ static void amd_xc_cpuid_policy(
break;
case 0x80000001: {
- if ( !is_pae )
+ if ( !info->pae )
clear_bit(X86_FEATURE_PAE, regs[3]);
/* Filter all other features according to a whitelist. */
regs[2] &= (bitmaskof(X86_FEATURE_LAHF_LM) |
bitmaskof(X86_FEATURE_CMP_LEGACY) |
- (is_nestedhvm ? bitmaskof(X86_FEATURE_SVM) : 0) |
+ (info->nestedhvm ? bitmaskof(X86_FEATURE_SVM) : 0) |
bitmaskof(X86_FEATURE_CR8_LEGACY) |
bitmaskof(X86_FEATURE_ABM) |
bitmaskof(X86_FEATURE_SSE4A) |
@@ -127,7 +199,8 @@ static void amd_xc_cpuid_policy(
break;
case 0x8000000a: {
- if (!is_nestedhvm) {
+ if ( !info->nestedhvm )
+ {
regs[0] = regs[1] = regs[2] = regs[3] = 0;
break;
}
@@ -157,16 +230,15 @@ static void amd_xc_cpuid_policy(
}
}
-static void intel_xc_cpuid_policy(
- xc_interface *xch, domid_t domid,
- const unsigned int *input, unsigned int *regs,
- int is_pae, int is_nestedhvm)
+static void intel_xc_cpuid_policy(xc_interface *xch,
+ const struct cpuid_domain_info *info,
+ const unsigned int *input, unsigned int *regs)
{
switch ( input[0] )
{
case 0x00000001:
/* ECX[5] is availability of VMX */
- if (is_nestedhvm)
+ if ( info->nestedhvm )
set_bit(X86_FEATURE_VMXE, regs[2]);
break;
@@ -211,11 +283,11 @@ static void intel_xc_cpuid_policy(
#define XSAVEOPT (1 << 0)
/* Configure extended state enumeration leaves (0x0000000D for xsave) */
-static void xc_cpuid_config_xsave(
- xc_interface *xch, domid_t domid, uint64_t xfeature_mask,
- const unsigned int *input, unsigned int *regs)
+static void xc_cpuid_config_xsave(xc_interface *xch,
+ const struct cpuid_domain_info *info,
+ const unsigned int *input, unsigned int *regs)
{
- if ( xfeature_mask == 0 )
+ if ( info->xfeature_mask == 0 )
{
regs[0] = regs[1] = regs[2] = regs[3] = 0;
return;
@@ -225,9 +297,9 @@ static void xc_cpuid_config_xsave(
{
case 0:
/* EAX: low 32bits of xfeature_enabled_mask */
- regs[0] = xfeature_mask & 0xFFFFFFFF;
+ regs[0] = info->xfeature_mask & 0xFFFFFFFF;
/* EDX: high 32bits of xfeature_enabled_mask */
- regs[3] = (xfeature_mask >> 32) & 0xFFFFFFFF;
+ regs[3] = (info->xfeature_mask >> 32) & 0xFFFFFFFF;
/* ECX: max size required by all HW features */
{
unsigned int _input[2] = {0xd, 0x0}, _regs[4];
@@ -250,7 +322,7 @@ static void xc_cpuid_config_xsave(
regs[1] = regs[2] = regs[3] = 0;
break;
case 2 ... 63: /* sub-leaves */
- if ( !(xfeature_mask & (1ULL << input[1])) )
+ if ( !(info->xfeature_mask & (1ULL << input[1])) )
{
regs[0] = regs[1] = regs[2] = regs[3] = 0;
break;
@@ -261,29 +333,10 @@ static void xc_cpuid_config_xsave(
}
}
-static void xc_cpuid_hvm_policy(
- xc_interface *xch, domid_t domid,
- const unsigned int *input, unsigned int *regs)
+static void xc_cpuid_hvm_policy(xc_interface *xch,
+ const struct cpuid_domain_info *info,
+ const unsigned int *input, unsigned int *regs)
{
- DECLARE_DOMCTL;
- char brand[13];
- uint64_t val;
- int is_pae, is_nestedhvm;
- uint64_t xfeature_mask;
-
- xc_hvm_param_get(xch, domid, HVM_PARAM_PAE_ENABLED, &val);
- is_pae = !!val;
-
- /* Detecting Xen's atitude towards XSAVE */
- memset(&domctl, 0, sizeof(domctl));
- domctl.cmd = XEN_DOMCTL_getvcpuextstate;
- domctl.domain = domid;
- do_domctl(xch, &domctl);
- xfeature_mask = domctl.u.vcpuextstate.xfeature_mask;
-
- xc_hvm_param_get(xch, domid, HVM_PARAM_NESTEDHVM, &val);
- is_nestedhvm = !!val;
-
switch ( input[0] )
{
case 0x00000000:
@@ -311,7 +364,7 @@ static void xc_cpuid_hvm_policy(
bitmaskof(X86_FEATURE_AES) |
bitmaskof(X86_FEATURE_F16C) |
bitmaskof(X86_FEATURE_RDRAND) |
- ((xfeature_mask != 0) ?
+ ((info->xfeature_mask != 0) ?
(bitmaskof(X86_FEATURE_AVX) |
bitmaskof(X86_FEATURE_XSAVE)) : 0));
@@ -346,7 +399,8 @@ static void xc_cpuid_hvm_policy(
/* We always support MTRR MSRs. */
regs[3] |= bitmaskof(X86_FEATURE_MTRR);
- if ( !is_pae ) {
+ if ( !info->pae )
+ {
clear_bit(X86_FEATURE_PAE, regs[3]);
clear_bit(X86_FEATURE_PSE36, regs[3]);
}
@@ -373,7 +427,7 @@ static void xc_cpuid_hvm_policy(
break;
case 0x0000000d:
- xc_cpuid_config_xsave(xch, domid, xfeature_mask, input, regs);
+ xc_cpuid_config_xsave(xch, info, input, regs);
break;
case 0x80000000:
@@ -381,7 +435,7 @@ static void xc_cpuid_hvm_policy(
break;
case 0x80000001:
- if ( !is_pae )
+ if ( !info->pae )
{
clear_bit(X86_FEATURE_LAHF_LM, regs[2]);
clear_bit(X86_FEATURE_LM, regs[3]);
@@ -422,40 +476,21 @@ static void xc_cpuid_hvm_policy(
break;
}
- xc_cpuid_brand_get(brand);
- if ( strstr(brand, "AMD") )
- amd_xc_cpuid_policy(xch, domid, input, regs, is_pae, is_nestedhvm);
+ if ( info->vendor == VENDOR_AMD )
+ amd_xc_cpuid_policy(xch, info, input, regs);
else
- intel_xc_cpuid_policy(xch, domid, input, regs, is_pae, is_nestedhvm);
-
+ intel_xc_cpuid_policy(xch, info, input, regs);
}
-static void xc_cpuid_pv_policy(
- xc_interface *xch, domid_t domid,
- const unsigned int *input, unsigned int *regs, bool pvh)
+static void xc_cpuid_pv_policy(xc_interface *xch,
+ const struct cpuid_domain_info *info,
+ const unsigned int *input, unsigned int *regs)
{
- DECLARE_DOMCTL;
- unsigned int guest_width;
- int guest_64bit;
- char brand[13];
- uint64_t xfeature_mask;
-
- xc_cpuid_brand_get(brand);
-
- xc_domain_get_guest_width(xch, domid, &guest_width);
- guest_64bit = (guest_width == 8);
-
- /* Detecting Xen's atitude towards XSAVE */
- memset(&domctl, 0, sizeof(domctl));
- domctl.cmd = XEN_DOMCTL_getvcpuextstate;
- domctl.domain = domid;
- do_domctl(xch, &domctl);
- xfeature_mask = domctl.u.vcpuextstate.xfeature_mask;
-
if ( (input[0] & 0x7fffffff) == 0x00000001 )
{
clear_bit(X86_FEATURE_VME, regs[3]);
- if ( !pvh ) {
+ if ( !info->pvh )
+ {
clear_bit(X86_FEATURE_PSE, regs[3]);
clear_bit(X86_FEATURE_PGE, regs[3]);
}
@@ -468,7 +503,7 @@ static void xc_cpuid_pv_policy(
switch ( input[0] )
{
case 0x00000001:
- if ( strstr(brand, "AMD") )
+ if ( info->vendor == VENDOR_AMD )
clear_bit(X86_FEATURE_SEP, regs[3]);
clear_bit(X86_FEATURE_DS, regs[3]);
clear_bit(X86_FEATURE_ACC, regs[3]);
@@ -481,9 +516,9 @@ static void xc_cpuid_pv_policy(
clear_bit(X86_FEATURE_SMXE, regs[2]);
clear_bit(X86_FEATURE_EST, regs[2]);
clear_bit(X86_FEATURE_TM2, regs[2]);
- if ( !guest_64bit )
+ if ( !info->pv64 )
clear_bit(X86_FEATURE_CX16, regs[2]);
- if ( xfeature_mask == 0 )
+ if ( info->xfeature_mask == 0 )
{
clear_bit(X86_FEATURE_XSAVE, regs[2]);
clear_bit(X86_FEATURE_AVX, regs[2]);
@@ -512,22 +547,22 @@ static void xc_cpuid_pv_policy(
break;
case 0x0000000d:
- xc_cpuid_config_xsave(xch, domid, xfeature_mask, input, regs);
+ xc_cpuid_config_xsave(xch, info, input, regs);
break;
case 0x80000001:
- if ( !guest_64bit )
+ if ( !info->pv64 )
{
clear_bit(X86_FEATURE_LM, regs[3]);
clear_bit(X86_FEATURE_LAHF_LM, regs[2]);
- if ( !strstr(brand, "AMD") )
+ if ( info->vendor != VENDOR_AMD )
clear_bit(X86_FEATURE_SYSCALL, regs[3]);
}
else
{
set_bit(X86_FEATURE_SYSCALL, regs[3]);
}
- if ( !pvh )
+ if ( !info->pvh )
clear_bit(X86_FEATURE_PAGE1GB, regs[3]);
clear_bit(X86_FEATURE_RDTSCP, regs[3]);
@@ -553,12 +588,10 @@ static void xc_cpuid_pv_policy(
}
}
-static int xc_cpuid_policy(
- xc_interface *xch, domid_t domid,
- const unsigned int *input, unsigned int *regs)
+static int xc_cpuid_policy(xc_interface *xch,
+ const struct cpuid_domain_info *info,
+ const unsigned int *input, unsigned int *regs)
{
- xc_dominfo_t info;
-
/*
* For hypervisor leaves (0x4000XXXX) only 0x4000xx00.EAX[7:0] bits (max
* number of leaves) can be set by user. Hypervisor will enforce this so
@@ -570,13 +603,10 @@ static int xc_cpuid_policy(
return 0;
}
- if ( xc_domain_getinfo(xch, domid, 1, &info) == 0 )
- return -EINVAL;
-
- if ( info.hvm )
- xc_cpuid_hvm_policy(xch, domid, input, regs);
+ if ( info->hvm )
+ xc_cpuid_hvm_policy(xch, info, input, regs);
else
- xc_cpuid_pv_policy(xch, domid, input, regs, info.pvh);
+ xc_cpuid_pv_policy(xch, info, input, regs);
return 0;
}
@@ -625,19 +655,21 @@ void xc_cpuid_to_str(const unsigned int *regs, char **strs)
int xc_cpuid_apply_policy(xc_interface *xch, domid_t domid)
{
+ struct cpuid_domain_info info = {};
unsigned int input[2] = { 0, 0 }, regs[4];
unsigned int base_max, ext_max;
- char brand[13];
int rc;
+ rc = get_cpuid_domain_info(xch, domid, &info);
+ if ( rc )
+ return rc;
cpuid(input, regs);
base_max = (regs[0] <= DEF_MAX_BASE) ? regs[0] : DEF_MAX_BASE;
input[0] = 0x80000000;
cpuid(input, regs);
- xc_cpuid_brand_get(brand);
- if ( strstr(brand, "AMD") )
+ if ( info.vendor == VENDOR_AMD )
ext_max = (regs[0] <= DEF_MAX_AMDEXT) ? regs[0] : DEF_MAX_AMDEXT;
else
ext_max = (regs[0] <= DEF_MAX_INTELEXT) ? regs[0] : DEF_MAX_INTELEXT;
@@ -647,7 +679,7 @@ int xc_cpuid_apply_policy(xc_interface *xch, domid_t domid)
for ( ; ; )
{
cpuid(input, regs);
- xc_cpuid_policy(xch, domid, input, regs);
+ xc_cpuid_policy(xch, &info, input, regs);
if ( regs[0] || regs[1] || regs[2] || regs[3] )
{
@@ -766,13 +798,18 @@ int xc_cpuid_set(
{
int rc;
unsigned int i, j, regs[4], polregs[4];
+ struct cpuid_domain_info info = {};
memset(config_transformed, 0, 4 * sizeof(*config_transformed));
+ rc = get_cpuid_domain_info(xch, domid, &info);
+ if ( rc )
+ return rc;
+
cpuid(input, regs);
memcpy(polregs, regs, sizeof(regs));
- xc_cpuid_policy(xch, domid, input, polregs);
+ xc_cpuid_policy(xch, &info, input, polregs);
for ( i = 0; i < 4; i++ )
{
--
1.7.10.4
next reply other threads:[~2015-10-05 13:12 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-05 13:12 Andrew Cooper [this message]
2015-10-06 9:14 ` [PATCH] tools/libxc: Improve efficiency of xc_cpuid_apply_policy() Wei Liu
2015-10-07 11:52 ` Ian Campbell
2015-10-07 11:54 ` 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=1444050737-8437-1-git-send-email-andrew.cooper3@citrix.com \
--to=andrew.cooper3@citrix.com \
--cc=Ian.Campbell@citrix.com \
--cc=Ian.Jackson@eu.citrix.com \
--cc=wei.liu2@citrix.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.