From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Xen-devel <xen-devel@lists.xen.org>
Cc: "Wei Liu" <wei.liu2@citrix.com>,
"Andrew Cooper" <andrew.cooper3@citrix.com>,
"David Wang" <davidwang@zhaoxin.com>,
"Jan Beulich" <JBeulich@suse.com>, "Pu Wen" <puwen@hygon.cn>,
"Roger Pau Monné" <roger.pau@citrix.com>
Subject: [PATCH 3/5] x86/cpu: Renumber X86_VENDOR_* to form a bitmap
Date: Thu, 4 Apr 2019 21:26:30 +0100 [thread overview]
Message-ID: <1554409592-28572-4-git-send-email-andrew.cooper3@citrix.com> (raw)
In-Reply-To: <1554409592-28572-1-git-send-email-andrew.cooper3@citrix.com>
CPUs from different vendors sometimes share characteristics. All users of
X86_VENDOR_* are now direct equal/not-equal comparisons. By expressing the
X86_VENDOR_* constants in a bitmap fashon, we can more concicely and
efficiently test whether a vendor is one of a group.
Update all parts of the code which can already benefit from this improvement.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
---
CC: Jan Beulich <JBeulich@suse.com>
CC: Wei Liu <wei.liu2@citrix.com>
CC: Roger Pau Monné <roger.pau@citrix.com>
CC: David Wang <davidwang@zhaoxin.com>
CC: Pu Wen <puwen@hygon.cn>
The 3 locations which are INTEL | CENTAUR are all to do with SYSENTER MSR
handling, which I expect should include SHANGHAI. I expect this is also true
of !hvm_long_mode_active() handling of X86_FEATURE_SYSCALL.
---
xen/arch/x86/acpi/cpufreq/cpufreq.c | 3 +--
xen/arch/x86/acpi/suspend.c | 6 ++----
xen/arch/x86/pv/emul-priv-op.c | 3 +--
xen/arch/x86/x86_64/traps.c | 3 +--
xen/include/asm-x86/x86-vendors.h | 13 +++++++------
5 files changed, 12 insertions(+), 16 deletions(-)
diff --git a/xen/arch/x86/acpi/cpufreq/cpufreq.c b/xen/arch/x86/acpi/cpufreq/cpufreq.c
index 844ab85..f4e13e1 100644
--- a/xen/arch/x86/acpi/cpufreq/cpufreq.c
+++ b/xen/arch/x86/acpi/cpufreq/cpufreq.c
@@ -661,8 +661,7 @@ int cpufreq_cpu_init(unsigned int cpuid)
int ret;
/* Currently we only handle Intel and AMD processor */
- if ( (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL ) ||
- (boot_cpu_data.x86_vendor == X86_VENDOR_AMD ) )
+ if ( boot_cpu_data.x86_vendor & (X86_VENDOR_INTEL | X86_VENDOR_AMD) )
ret = cpufreq_add_cpu(cpuid);
else
ret = -EFAULT;
diff --git a/xen/arch/x86/acpi/suspend.c b/xen/arch/x86/acpi/suspend.c
index 00e6012..9e69bf2 100644
--- a/xen/arch/x86/acpi/suspend.c
+++ b/xen/arch/x86/acpi/suspend.c
@@ -27,8 +27,7 @@ void save_rest_processor_state(void)
rdmsrl(MSR_SHADOW_GS_BASE, saved_kernel_gs_base);
rdmsrl(MSR_CSTAR, saved_cstar);
rdmsrl(MSR_LSTAR, saved_lstar);
- if ( boot_cpu_data.x86_vendor == X86_VENDOR_INTEL ||
- boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR )
+ if ( boot_cpu_data.x86_vendor & (X86_VENDOR_INTEL | X86_VENDOR_CENTAUR) )
{
rdmsrl(MSR_IA32_SYSENTER_ESP, saved_sysenter_esp);
rdmsrl(MSR_IA32_SYSENTER_EIP, saved_sysenter_eip);
@@ -52,8 +51,7 @@ void restore_rest_processor_state(void)
wrgsbase(saved_gs_base);
wrmsrl(MSR_SHADOW_GS_BASE, saved_kernel_gs_base);
- if ( boot_cpu_data.x86_vendor == X86_VENDOR_INTEL ||
- boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR )
+ if ( boot_cpu_data.x86_vendor & (X86_VENDOR_INTEL | X86_VENDOR_CENTAUR) )
{
/* Recover sysenter MSRs */
wrmsrl(MSR_IA32_SYSENTER_ESP, saved_sysenter_esp);
diff --git a/xen/arch/x86/pv/emul-priv-op.c b/xen/arch/x86/pv/emul-priv-op.c
index 84ce67c..f594065 100644
--- a/xen/arch/x86/pv/emul-priv-op.c
+++ b/xen/arch/x86/pv/emul-priv-op.c
@@ -1066,8 +1066,7 @@ static int write_msr(unsigned int reg, uint64_t val,
case MSR_IA32_MPERF:
case MSR_IA32_APERF:
- if ( (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) &&
- (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) )
+ if ( !(boot_cpu_data.x86_vendor & (X86_VENDOR_INTEL | X86_VENDOR_AMD)) )
break;
if ( likely(!is_cpufreq_controller(currd)) ||
wrmsr_safe(reg, val) == 0 )
diff --git a/xen/arch/x86/x86_64/traps.c b/xen/arch/x86/x86_64/traps.c
index bf7870e..44af765 100644
--- a/xen/arch/x86/x86_64/traps.c
+++ b/xen/arch/x86/x86_64/traps.c
@@ -334,8 +334,7 @@ void subarch_percpu_traps_init(void)
(unsigned long)lstar_enter);
stub_va += offset;
- if ( boot_cpu_data.x86_vendor == X86_VENDOR_INTEL ||
- boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR )
+ if ( boot_cpu_data.x86_vendor & (X86_VENDOR_INTEL | X86_VENDOR_CENTAUR) )
{
/* SYSENTER entry. */
wrmsrl(MSR_IA32_SYSENTER_ESP, stack_bottom);
diff --git a/xen/include/asm-x86/x86-vendors.h b/xen/include/asm-x86/x86-vendors.h
index fca7396..1ecb934 100644
--- a/xen/include/asm-x86/x86-vendors.h
+++ b/xen/include/asm-x86/x86-vendors.h
@@ -4,28 +4,29 @@
/*
* CPU vendor IDs
*
- * - X86_VENDOR_* are Xen-internal identifiers. Values and order are
- * arbitrary.
+ * - X86_VENDOR_* are Xen-internal identifiers. The order is arbitrary, but
+ * values form a bitmap so vendor checks can be made against multiple
+ * vendors at once.
* - X86_VENDOR_*_E?X are architectural information from CPUID leaf 0
*/
#define X86_VENDOR_UNKNOWN 0
-#define X86_VENDOR_INTEL 1
+#define X86_VENDOR_INTEL (1 << 1)
#define X86_VENDOR_INTEL_EBX 0x756e6547U /* "GenuineIntel" */
#define X86_VENDOR_INTEL_ECX 0x6c65746eU
#define X86_VENDOR_INTEL_EDX 0x49656e69U
-#define X86_VENDOR_AMD 2
+#define X86_VENDOR_AMD (1 << 2)
#define X86_VENDOR_AMD_EBX 0x68747541U /* "AuthenticAMD" */
#define X86_VENDOR_AMD_ECX 0x444d4163U
#define X86_VENDOR_AMD_EDX 0x69746e65U
-#define X86_VENDOR_CENTAUR 3
+#define X86_VENDOR_CENTAUR (1 << 3)
#define X86_VENDOR_CENTAUR_EBX 0x746e6543U /* "CentaurHauls" */
#define X86_VENDOR_CENTAUR_ECX 0x736c7561U
#define X86_VENDOR_CENTAUR_EDX 0x48727561U
-#define X86_VENDOR_SHANGHAI 4
+#define X86_VENDOR_SHANGHAI (1 << 4)
#define X86_VENDOR_SHANGHAI_EBX 0x68532020U /* " Shanghai " */
#define X86_VENDOR_SHANGHAI_ECX 0x20206961U
#define X86_VENDOR_SHANGHAI_EDX 0x68676e61U
--
2.1.4
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
next prev parent reply other threads:[~2019-04-04 20:26 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-04 20:26 [PATCH 0/5] x86/cpu: Rework of X86_VENDOR_* constants Andrew Cooper
2019-04-04 20:26 ` [PATCH 1/5] x86/cpu: Drop cpu_devs[] and $VENDOR_init_cpu() hooks Andrew Cooper
2019-04-05 8:57 ` Jan Beulich
2019-04-05 9:27 ` Andrew Cooper
2019-04-04 20:26 ` [PATCH 2/5] x86/cpu: Introduce x86_cpuid_vendor_to_str() and drop cpu_dev.c_vendor[] Andrew Cooper
2019-04-05 8:59 ` Jan Beulich
2019-04-04 20:26 ` Andrew Cooper [this message]
2019-04-05 9:03 ` [PATCH 3/5] x86/cpu: Renumber X86_VENDOR_* to form a bitmap Jan Beulich
2019-04-05 9:28 ` Andrew Cooper
2019-04-04 20:26 ` [PATCH 4/5] x86/cpu: Create Hygon Dhyana architecture support file Andrew Cooper
2019-04-05 9:17 ` Jan Beulich
2019-04-05 15:30 ` Pu Wen
2019-04-05 16:10 ` Jan Beulich
2019-04-04 20:26 ` [PATCH 5/5] x86/msr: Fix handling of MSR_AMD_PATCHLEVEL/MSR_IA32_UCODE_REV Andrew Cooper
2019-04-05 9:20 ` Jan Beulich
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=1554409592-28572-4-git-send-email-andrew.cooper3@citrix.com \
--to=andrew.cooper3@citrix.com \
--cc=JBeulich@suse.com \
--cc=davidwang@zhaoxin.com \
--cc=puwen@hygon.cn \
--cc=roger.pau@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 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).