[PATCH 0/3] x86/mce: Add Zhaoxin MCE support

[PATCH 0/3] x86/mce: Add Zhaoxin MCE support

[Tony W Wang-oc]

From: Lyle Li <LyleLi@zhaoxin.com>

Zhaoxin consists of two vendors, X86_VENDOR_ZHAOXIN and
X86_VENDOR_CENTAUR, so add the centaur vendor to support
Zhaoxin MCA in mce/core.c and mce/intel.c.

For the sake of code standardization, add zhaoxin.c to
override the Zhaoxin MCA code.

Zhaoxin CPUs support CMCI compatible with Intel, because 
Zhaoxin's UCR error is not reported through CMCI, and in
order to be compatible with intel's CMCI code, so add Zhaoxin
CMCI storm toggle to support the new CMCI storm switching
in mce/intel.c, mce/zhaoxin.c, mce/threshold.c, and mce/internal.h.

Lyle Li (3):
  x86/mce: Add centaur vendor to support Zhaoxin MCA
  x86/mce: Add zhaoxin.c to support Zhaoxin MCA
  x86/mce: Add CMCI storm switching support for Zhaoxin

 arch/x86/Kconfig                    |  8 ++++
 arch/x86/kernel/cpu/mce/Makefile    |  2 +-
 arch/x86/kernel/cpu/mce/core.c      | 70 +++++++++--------------------
 arch/x86/kernel/cpu/mce/intel.c     |  8 ++--
 arch/x86/kernel/cpu/mce/internal.h  | 14 +++++-
 arch/x86/kernel/cpu/mce/threshold.c |  4 ++
 arch/x86/kernel/cpu/mce/zhaoxin.c   | 53 ++++++++++++++++++++++
 7 files changed, 104 insertions(+), 55 deletions(-)
 create mode 100644 arch/x86/kernel/cpu/mce/zhaoxin.c

-- 
2.34.1

[PATCH v1 1/3] x86/mce: Add centaur vendor to support Zhaoxin MCA

[Tony W Wang-oc]

From: Lyle Li <LyleLi@zhaoxin.com>

Zhaoxin consists of two vendors, X86_VENDOR_ZHAOXIN and
X86_VENDOR_CENTAUR, so add the centaur vendor to support
Zhaoxin MCA in mce/core.c and mce/intel.c.

Signed-off-by: Lyle Li <LyleLi@zhaoxin.com>
Reviewed-by: Tony W Wang-oc <TonyWWang-oc@zhaoxin.com>
---
 arch/x86/kernel/cpu/mce/core.c  | 42 ++++++++++++++++-----------------
 arch/x86/kernel/cpu/mce/intel.c |  3 ++-
 2 files changed, 23 insertions(+), 22 deletions(-)

diff --git a/arch/x86/kernel/cpu/mce/core.c b/arch/x86/kernel/cpu/mce/core.c
index ad0623b65..b7b98c33a 100644
--- a/arch/x86/kernel/cpu/mce/core.c
+++ b/arch/x86/kernel/cpu/mce/core.c
@@ -496,6 +496,7 @@ bool mce_usable_address(struct mce *m)
 
 	case X86_VENDOR_INTEL:
 	case X86_VENDOR_ZHAOXIN:
+	case X86_VENDOR_CENTAUR:
 		return intel_mce_usable_address(m);
 
 	default:
@@ -513,6 +514,7 @@ bool mce_is_memory_error(struct mce *m)
 
 	case X86_VENDOR_INTEL:
 	case X86_VENDOR_ZHAOXIN:
+	case X86_VENDOR_CENTAUR:
 		/*
 		 * Intel SDM Volume 3B - 15.9.2 Compound Error Codes
 		 *
@@ -1247,7 +1249,8 @@ static noinstr bool mce_check_crashing_cpu(void)
 
 		mcgstatus = __rdmsr(MSR_IA32_MCG_STATUS);
 
-		if (boot_cpu_data.x86_vendor == X86_VENDOR_ZHAOXIN) {
+		if (boot_cpu_data.x86_vendor == X86_VENDOR_ZHAOXIN ||
+		    boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR) {
 			if (mcgstatus & MCG_STATUS_LMCES)
 				return false;
 		}
@@ -1521,7 +1524,8 @@ noinstr void do_machine_check(struct pt_regs *regs)
 	 * on Intel, Zhaoxin only.
 	 */
 	if (m.cpuvendor == X86_VENDOR_INTEL ||
-	    m.cpuvendor == X86_VENDOR_ZHAOXIN)
+	    m.cpuvendor == X86_VENDOR_ZHAOXIN ||
+	    m.cpuvendor == X86_VENDOR_CENTAUR)
 		lmce = m.mcgstatus & MCG_STATUS_LMCES;
 
 	/*
@@ -1970,6 +1974,18 @@ static int __mcheck_cpu_apply_quirks(struct cpuinfo_x86 *c)
 		}
 	}
 
+	if (c->x86_vendor == X86_VENDOR_CENTAUR) {
+		/*
+		 * All newer Centaur CPUs support MCE broadcasting. Enable
+		 * synchronization with a one second timeout.
+		 */
+		if ((c->x86 == 6 && c->x86_model == 0xf && c->x86_stepping >= 0xe) ||
+		     c->x86 > 6) {
+			if (cfg->monarch_timeout < 0)
+				cfg->monarch_timeout = USEC_PER_SEC;
+		}
+	}
+
 	if (cfg->monarch_timeout < 0)
 		cfg->monarch_timeout = 0;
 	if (cfg->bootlog != 0)
@@ -2012,21 +2028,6 @@ static void __mcheck_cpu_init_early(struct cpuinfo_x86 *c)
 	}
 }
 
-static void mce_centaur_feature_init(struct cpuinfo_x86 *c)
-{
-	struct mca_config *cfg = &mca_cfg;
-
-	 /*
-	  * All newer Centaur CPUs support MCE broadcasting. Enable
-	  * synchronization with a one second timeout.
-	  */
-	if ((c->x86 == 6 && c->x86_model == 0xf && c->x86_stepping >= 0xe) ||
-	     c->x86 > 6) {
-		if (cfg->monarch_timeout < 0)
-			cfg->monarch_timeout = USEC_PER_SEC;
-	}
-}
-
 static void mce_zhaoxin_feature_init(struct cpuinfo_x86 *c)
 {
 	struct mce_bank *mce_banks = this_cpu_ptr(mce_banks_array);
@@ -2072,9 +2073,6 @@ static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)
 		break;
 
 	case X86_VENDOR_CENTAUR:
-		mce_centaur_feature_init(c);
-		break;
-
 	case X86_VENDOR_ZHAOXIN:
 		mce_zhaoxin_feature_init(c);
 		break;
@@ -2092,6 +2090,7 @@ static void __mcheck_cpu_clear_vendor(struct cpuinfo_x86 *c)
 		break;
 
 	case X86_VENDOR_ZHAOXIN:
+	case X86_VENDOR_CENTAUR:
 		mce_zhaoxin_feature_clear(c);
 		break;
 
@@ -2401,7 +2400,8 @@ static void vendor_disable_error_reporting(void)
 	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL ||
 	    boot_cpu_data.x86_vendor == X86_VENDOR_HYGON ||
 	    boot_cpu_data.x86_vendor == X86_VENDOR_AMD ||
-	    boot_cpu_data.x86_vendor == X86_VENDOR_ZHAOXIN)
+	    boot_cpu_data.x86_vendor == X86_VENDOR_ZHAOXIN ||
+	    boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR)
 		return;
 
 	mce_disable_error_reporting();
diff --git a/arch/x86/kernel/cpu/mce/intel.c b/arch/x86/kernel/cpu/mce/intel.c
index f6103e6bf..b7e67f4f7 100644
--- a/arch/x86/kernel/cpu/mce/intel.c
+++ b/arch/x86/kernel/cpu/mce/intel.c
@@ -88,7 +88,8 @@ static int cmci_supported(int *banks)
 	 * makes sure none of the backdoors are entered otherwise.
 	 */
 	if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL &&
-	    boot_cpu_data.x86_vendor != X86_VENDOR_ZHAOXIN)
+	    boot_cpu_data.x86_vendor != X86_VENDOR_ZHAOXIN &&
+	    boot_cpu_data.x86_vendor != X86_VENDOR_CENTAUR)
 		return 0;
 
 	if (!boot_cpu_has(X86_FEATURE_APIC) || lapic_get_maxlvt() < 6)
-- 
2.34.1

[PATCH v1 2/3] x86/mce: Add zhaoxin.c to support Zhaoxin MCA

[Tony W Wang-oc]

From: Lyle Li <LyleLi@zhaoxin.com>

For the sake of code standardization, add zhaoxin.c to
override the Zhaoxin MCA code.

Signed-off-by: Lyle Li <LyleLi@zhaoxin.com>
Reviewed-by: Tony W Wang-oc <TonyWWang-oc@zhaoxin.com>
---
 arch/x86/Kconfig                   |  8 +++++++
 arch/x86/kernel/cpu/mce/Makefile   |  2 +-
 arch/x86/kernel/cpu/mce/core.c     | 28 ------------------------
 arch/x86/kernel/cpu/mce/internal.h |  7 ++++++
 arch/x86/kernel/cpu/mce/zhaoxin.c  | 35 ++++++++++++++++++++++++++++++
 5 files changed, 51 insertions(+), 29 deletions(-)
 create mode 100644 arch/x86/kernel/cpu/mce/zhaoxin.c

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 1d7122a18..b908cdfb9 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1171,6 +1171,14 @@ config X86_MCE_INTEL
 	  Additional support for intel specific MCE features such as
 	  the thermal monitor.
 
+config X86_MCE_ZHAOXIN
+	def_bool y
+	prompt "Zhaoxin MCE features"
+	depends on X86_MCE_INTEL
+	help
+	  Additional support for zhaoxin specific MCE features such as
+	  the corrected machine check interrupt.
+
 config X86_MCE_AMD
 	def_bool y
 	prompt "AMD MCE features"
diff --git a/arch/x86/kernel/cpu/mce/Makefile b/arch/x86/kernel/cpu/mce/Makefile
index 015856abd..2e863e78d 100644
--- a/arch/x86/kernel/cpu/mce/Makefile
+++ b/arch/x86/kernel/cpu/mce/Makefile
@@ -5,7 +5,7 @@ obj-$(CONFIG_X86_ANCIENT_MCE)	+= winchip.o p5.o
 obj-$(CONFIG_X86_MCE_INTEL)	+= intel.o
 obj-$(CONFIG_X86_MCE_AMD)	+= amd.o
 obj-$(CONFIG_X86_MCE_THRESHOLD) += threshold.o
-
+obj-$(CONFIG_X86_MCE_ZHAOXIN)   += zhaoxin.o
 mce-inject-y			:= inject.o
 obj-$(CONFIG_X86_MCE_INJECT)	+= mce-inject.o
 
diff --git a/arch/x86/kernel/cpu/mce/core.c b/arch/x86/kernel/cpu/mce/core.c
index b7b98c33a..b32bfd9f3 100644
--- a/arch/x86/kernel/cpu/mce/core.c
+++ b/arch/x86/kernel/cpu/mce/core.c
@@ -2028,34 +2028,6 @@ static void __mcheck_cpu_init_early(struct cpuinfo_x86 *c)
 	}
 }
 
-static void mce_zhaoxin_feature_init(struct cpuinfo_x86 *c)
-{
-	struct mce_bank *mce_banks = this_cpu_ptr(mce_banks_array);
-
-	/*
-	 * These CPUs have MCA bank 8 which reports only one error type called
-	 * SVAD (System View Address Decoder). The reporting of that error is
-	 * controlled by IA32_MC8.CTL.0.
-	 *
-	 * If enabled, prefetching on these CPUs will cause SVAD MCE when
-	 * virtual machines start and result in a system  panic. Always disable
-	 * bank 8 SVAD error by default.
-	 */
-	if ((c->x86 == 7 && c->x86_model == 0x1b) ||
-	    (c->x86_model == 0x19 || c->x86_model == 0x1f)) {
-		if (this_cpu_read(mce_num_banks) > 8)
-			mce_banks[8].ctl = 0;
-	}
-
-	intel_init_cmci();
-	intel_init_lmce();
-}
-
-static void mce_zhaoxin_feature_clear(struct cpuinfo_x86 *c)
-{
-	intel_clear_lmce();
-}
-
 static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)
 {
 	switch (c->x86_vendor) {
diff --git a/arch/x86/kernel/cpu/mce/internal.h b/arch/x86/kernel/cpu/mce/internal.h
index 01f8f0396..e9b06b825 100644
--- a/arch/x86/kernel/cpu/mce/internal.h
+++ b/arch/x86/kernel/cpu/mce/internal.h
@@ -334,4 +334,11 @@ static __always_inline u32 mca_msr_reg(int bank, enum mca_msr reg)
 }
 
 extern void (*mc_poll_banks)(void);
+#ifdef CONFIG_X86_MCE_ZHAOXIN
+void mce_zhaoxin_feature_init(struct cpuinfo_x86 *c);
+void mce_zhaoxin_feature_clear(struct cpuinfo_x86 *c);
+#else
+void mce_zhaoxin_feature_init(struct cpuinfo_x86 *c) {}
+void mce_zhaoxin_feature_clear(struct cpuinfo_x86 *c) {}
+#endif
 #endif /* __X86_MCE_INTERNAL_H__ */
diff --git a/arch/x86/kernel/cpu/mce/zhaoxin.c b/arch/x86/kernel/cpu/mce/zhaoxin.c
new file mode 100644
index 000000000..97d12ce0c
--- /dev/null
+++ b/arch/x86/kernel/cpu/mce/zhaoxin.c
@@ -0,0 +1,35 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Zhaoxin specific MCE features
+ * Author: Lyle Li
+ */
+#include <asm/msr.h>
+#include "internal.h"
+
+void mce_zhaoxin_feature_init(struct cpuinfo_x86 *c)
+{
+	struct mce_bank *mce_banks = this_cpu_ptr(mce_banks_array);
+
+	/*
+	 * These CPUs have MCA bank 8 which reports only one error type called
+	 * SVAD (System View Address Decoder). The reporting of that error is
+	 * controlled by IA32_MC8.CTL.0.
+	 *
+	 * If enabled, prefetching on these CPUs will cause SVAD MCE when
+	 * virtual machines start and result in a system  panic. Always disable
+	 * bank 8 SVAD error by default.
+	 */
+	if ((c->x86 == 7 && c->x86_model == 0x1b) ||
+	    (c->x86_model == 0x19 || c->x86_model == 0x1f)) {
+		if (this_cpu_read(mce_num_banks) > 8)
+			mce_banks[8].ctl = 0;
+	}
+
+	intel_init_cmci();
+	intel_init_lmce();
+}
+
+void mce_zhaoxin_feature_clear(struct cpuinfo_x86 *c)
+{
+	intel_clear_lmce();
+}
-- 
2.34.1

[PATCH v1 3/3] x86/mce: Add CMCI storm switching support for Zhaoxin

[Tony W Wang-oc]

From: Lyle Li <LyleLi@zhaoxin.com>

Zhaoxin CPUs support CMCI compatible with Intel, because
Zhaoxin's UCR error is not reported through CMCI, and in
order to be compatible with intel's CMCI code, so add Zhaoxin
CMCI storm toggle to support the new CMCI storm switching
in mce/intel.c, mce/zhaoxin.c, mce/threshold.c, and mce/internal.h.

Signed-off-by: Lyle Li <LyleLi@zhaoxin.com>
Reviewed-by: Tony W Wang-oc <TonyWWang-oc@zhaoxin.com>
---
 arch/x86/kernel/cpu/mce/intel.c     |  5 ++---
 arch/x86/kernel/cpu/mce/internal.h  |  7 ++++++-
 arch/x86/kernel/cpu/mce/threshold.c |  4 ++++
 arch/x86/kernel/cpu/mce/zhaoxin.c   | 18 ++++++++++++++++++
 4 files changed, 30 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kernel/cpu/mce/intel.c b/arch/x86/kernel/cpu/mce/intel.c
index b7e67f4f7..aa75e2848 100644
--- a/arch/x86/kernel/cpu/mce/intel.c
+++ b/arch/x86/kernel/cpu/mce/intel.c
@@ -45,7 +45,7 @@ static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned);
  * cmci_discover_lock protects against parallel discovery attempts
  * which could race against each other.
  */
-static DEFINE_RAW_SPINLOCK(cmci_discover_lock);
+DEFINE_RAW_SPINLOCK(cmci_discover_lock);
 
 /*
  * On systems that do support CMCI but it's disabled, polling for MCEs can
@@ -61,7 +61,7 @@ static DEFINE_SPINLOCK(cmci_poll_lock);
  * MCi_CTL2 threshold for each bank when there is no storm.
  * Default value for each bank may have been set by BIOS.
  */
-static u16 cmci_threshold[MAX_NR_BANKS];
+u16 cmci_threshold[MAX_NR_BANKS];
 
 /*
  * High threshold to limit CMCI rate during storms. Max supported is
@@ -73,7 +73,6 @@ static u16 cmci_threshold[MAX_NR_BANKS];
  * to corrected errors, so keeping CMCI enabled means that uncorrected
  * errors will still be processed in a timely fashion.
  */
-#define CMCI_STORM_THRESHOLD	32749
 
 static int cmci_supported(int *banks)
 {
diff --git a/arch/x86/kernel/cpu/mce/internal.h b/arch/x86/kernel/cpu/mce/internal.h
index e9b06b825..8fa1f590f 100644
--- a/arch/x86/kernel/cpu/mce/internal.h
+++ b/arch/x86/kernel/cpu/mce/internal.h
@@ -7,7 +7,7 @@
 
 #include <linux/device.h>
 #include <asm/mce.h>
-
+#include <linux/spinlock.h>
 enum severity_level {
 	MCE_NO_SEVERITY,
 	MCE_DEFERRED_SEVERITY,
@@ -334,11 +334,16 @@ static __always_inline u32 mca_msr_reg(int bank, enum mca_msr reg)
 }
 
 extern void (*mc_poll_banks)(void);
+#define CMCI_STORM_THRESHOLD    32749
+extern raw_spinlock_t cmci_discover_lock;
+extern u16 cmci_threshold[MAX_NR_BANKS];
 #ifdef CONFIG_X86_MCE_ZHAOXIN
 void mce_zhaoxin_feature_init(struct cpuinfo_x86 *c);
 void mce_zhaoxin_feature_clear(struct cpuinfo_x86 *c);
+void mce_zhaoxin_handle_storm(int bank, bool on);
 #else
 void mce_zhaoxin_feature_init(struct cpuinfo_x86 *c) {}
 void mce_zhaoxin_feature_clear(struct cpuinfo_x86 *c) {}
+void mce_zhaoxin_handle_storm(int bank, bool on) {}
 #endif
 #endif /* __X86_MCE_INTERNAL_H__ */
diff --git a/arch/x86/kernel/cpu/mce/threshold.c b/arch/x86/kernel/cpu/mce/threshold.c
index 89e31e1e5..200280387 100644
--- a/arch/x86/kernel/cpu/mce/threshold.c
+++ b/arch/x86/kernel/cpu/mce/threshold.c
@@ -63,6 +63,10 @@ static void mce_handle_storm(unsigned int bank, bool on)
 	case X86_VENDOR_INTEL:
 		mce_intel_handle_storm(bank, on);
 		break;
+	case X86_VENDOR_ZHAOXIN:
+	case X86_VENDOR_CENTAUR:
+		mce_zhaoxin_handle_storm(bank, on);
+		break;
 	}
 }
 
diff --git a/arch/x86/kernel/cpu/mce/zhaoxin.c b/arch/x86/kernel/cpu/mce/zhaoxin.c
index 97d12ce0c..38beca449 100644
--- a/arch/x86/kernel/cpu/mce/zhaoxin.c
+++ b/arch/x86/kernel/cpu/mce/zhaoxin.c
@@ -33,3 +33,21 @@ void mce_zhaoxin_feature_clear(struct cpuinfo_x86 *c)
 {
 	intel_clear_lmce();
 }
+
+void mce_zhaoxin_handle_storm(int bank, bool on)
+{
+	unsigned long flags;
+	u64 val;
+
+	raw_spin_lock_irqsave(&cmci_discover_lock, flags);
+	rdmsrl(MSR_IA32_MCx_CTL2(bank), val);
+	if (on) {
+		val &= ~(MCI_CTL2_CMCI_EN | MCI_CTL2_CMCI_THRESHOLD_MASK);
+		val |= CMCI_STORM_THRESHOLD;
+	} else {
+		val &= ~MCI_CTL2_CMCI_THRESHOLD_MASK;
+		val |= (MCI_CTL2_CMCI_EN | cmci_threshold[bank]);
+	}
+	wrmsrl(MSR_IA32_MCx_CTL2(bank), val);
+	raw_spin_unlock_irqrestore(&cmci_discover_lock, flags);
+}
-- 
2.34.1

Re: [PATCH v1 2/3] x86/mce: Add zhaoxin.c to support Zhaoxin MCA

[kernel test robot]

Hi Tony,

kernel test robot noticed the following build errors:

[auto build test ERROR on tip/x86/core]
[also build test ERROR on tip/master linus/master v6.11-rc7 next-20240909]
[cannot apply to tip/auto-latest]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Tony-W-Wang-oc/x86-mce-Add-centaur-vendor-to-support-Zhaoxin-MCA/20240909-192507
base:   tip/x86/core
patch link:    https://lore.kernel.org/r/20240909104349.3349-3-TonyWWang-oc%40zhaoxin.com
patch subject: [PATCH v1 2/3] x86/mce: Add zhaoxin.c to support Zhaoxin MCA
config: x86_64-buildonly-randconfig-001-20240910 (https://download.01.org/0day-ci/archive/20240910/202409100925.oZtxKGQi-lkp@intel.com/config)
compiler: gcc-12 (Debian 12.2.0-14) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240910/202409100925.oZtxKGQi-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202409100925.oZtxKGQi-lkp@intel.com/

All errors (new ones prefixed by >>):

   ld: arch/x86/kernel/cpu/mce/severity.o: in function `mce_zhaoxin_feature_init':
>> severity.c:(.text+0x2c0): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/severity.o: in function `mce_zhaoxin_feature_clear':
>> severity.c:(.text+0x2f0): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/genpool.o: in function `mce_zhaoxin_feature_init':
   genpool.c:(.text+0x10): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/genpool.o: in function `mce_zhaoxin_feature_clear':
   genpool.c:(.text+0x40): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/amd.o: in function `mce_zhaoxin_feature_init':
   amd.c:(.text+0x1730): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/amd.o: in function `mce_zhaoxin_feature_clear':
   amd.c:(.text+0x1760): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/threshold.o: in function `mce_zhaoxin_feature_init':
   threshold.c:(.text+0x70): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/threshold.o: in function `mce_zhaoxin_feature_clear':
   threshold.c:(.text+0xa0): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/inject.o: in function `mce_zhaoxin_feature_init':
   inject.c:(.text+0xd80): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/inject.o: in function `mce_zhaoxin_feature_clear':
   inject.c:(.text+0xdb0): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/apei.o: in function `mce_zhaoxin_feature_init':
   apei.c:(.text+0xf0): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/apei.o: in function `mce_zhaoxin_feature_clear':
   apei.c:(.text+0x120): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

Re: [PATCH v1 3/3] x86/mce: Add CMCI storm switching support for Zhaoxin

[kernel test robot]

Hi Tony,

kernel test robot noticed the following build errors:

[auto build test ERROR on tip/x86/core]
[also build test ERROR on tip/master linus/master v6.11-rc7 next-20240909]
[cannot apply to tip/auto-latest]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Tony-W-Wang-oc/x86-mce-Add-centaur-vendor-to-support-Zhaoxin-MCA/20240909-192507
base:   tip/x86/core
patch link:    https://lore.kernel.org/r/20240909104349.3349-4-TonyWWang-oc%40zhaoxin.com
patch subject: [PATCH v1 3/3] x86/mce: Add CMCI storm switching support for Zhaoxin
config: x86_64-buildonly-randconfig-001-20240910 (https://download.01.org/0day-ci/archive/20240910/202409101353.K4jjCjRN-lkp@intel.com/config)
compiler: gcc-12 (Debian 12.2.0-14) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240910/202409101353.K4jjCjRN-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202409101353.K4jjCjRN-lkp@intel.com/

All errors (new ones prefixed by >>):

   ld: arch/x86/kernel/cpu/mce/severity.o: in function `mce_zhaoxin_feature_init':
   severity.c:(.text+0x2c0): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/severity.o: in function `mce_zhaoxin_feature_clear':
   severity.c:(.text+0x2f0): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/severity.o: in function `mce_zhaoxin_handle_storm':
>> severity.c:(.text+0x320): multiple definition of `mce_zhaoxin_handle_storm'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x19b0): first defined here
   ld: arch/x86/kernel/cpu/mce/genpool.o: in function `mce_zhaoxin_feature_init':
   genpool.c:(.text+0x10): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/genpool.o: in function `mce_zhaoxin_feature_clear':
   genpool.c:(.text+0x40): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/genpool.o: in function `mce_zhaoxin_handle_storm':
   genpool.c:(.text+0x70): multiple definition of `mce_zhaoxin_handle_storm'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x19b0): first defined here
   ld: arch/x86/kernel/cpu/mce/amd.o: in function `mce_zhaoxin_feature_init':
   amd.c:(.text+0x1730): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/amd.o: in function `mce_zhaoxin_feature_clear':
   amd.c:(.text+0x1760): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/amd.o: in function `mce_zhaoxin_handle_storm':
   amd.c:(.text+0x1790): multiple definition of `mce_zhaoxin_handle_storm'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x19b0): first defined here
   ld: arch/x86/kernel/cpu/mce/threshold.o: in function `mce_zhaoxin_feature_init':
   threshold.c:(.text+0x70): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/threshold.o: in function `mce_zhaoxin_feature_clear':
   threshold.c:(.text+0xa0): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/threshold.o: in function `mce_zhaoxin_handle_storm':
   threshold.c:(.text+0xd0): multiple definition of `mce_zhaoxin_handle_storm'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x19b0): first defined here
   ld: arch/x86/kernel/cpu/mce/inject.o: in function `mce_zhaoxin_feature_init':
   inject.c:(.text+0xd80): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/inject.o: in function `mce_zhaoxin_feature_clear':
   inject.c:(.text+0xdb0): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/inject.o: in function `mce_zhaoxin_handle_storm':
   inject.c:(.text+0xde0): multiple definition of `mce_zhaoxin_handle_storm'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x19b0): first defined here
   ld: arch/x86/kernel/cpu/mce/apei.o: in function `mce_zhaoxin_feature_init':
   apei.c:(.text+0xf0): multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1950): first defined here
   ld: arch/x86/kernel/cpu/mce/apei.o: in function `mce_zhaoxin_feature_clear':
   apei.c:(.text+0x120): multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x1980): first defined here
   ld: arch/x86/kernel/cpu/mce/apei.o: in function `mce_zhaoxin_handle_storm':
   apei.c:(.text+0x150): multiple definition of `mce_zhaoxin_handle_storm'; arch/x86/kernel/cpu/mce/core.o:core.c:(.text+0x19b0): first defined here

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

Re: [PATCH v1 2/3] x86/mce: Add zhaoxin.c to support Zhaoxin MCA

[kernel test robot]

Hi Tony,

kernel test robot noticed the following build errors:

[auto build test ERROR on tip/x86/core]
[also build test ERROR on tip/master linus/master v6.11-rc7 next-20240909]
[cannot apply to tip/auto-latest]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Tony-W-Wang-oc/x86-mce-Add-centaur-vendor-to-support-Zhaoxin-MCA/20240909-192507
base:   tip/x86/core
patch link:    https://lore.kernel.org/r/20240909104349.3349-3-TonyWWang-oc%40zhaoxin.com
patch subject: [PATCH v1 2/3] x86/mce: Add zhaoxin.c to support Zhaoxin MCA
config: x86_64-randconfig-123-20240910 (https://download.01.org/0day-ci/archive/20240910/202409101604.VYS2JlHA-lkp@intel.com/config)
compiler: gcc-12 (Debian 12.2.0-14) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240910/202409101604.VYS2JlHA-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202409101604.VYS2JlHA-lkp@intel.com/

All errors (new ones prefixed by >>):

   ld: arch/x86/kernel/cpu/mce/severity.o: in function `mce_zhaoxin_feature_init':
>> arch/x86/kernel/cpu/mce/internal.h:341: multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:arch/x86/kernel/cpu/mce/internal.h:341: first defined here
   ld: arch/x86/kernel/cpu/mce/severity.o: in function `mce_zhaoxin_feature_clear':
>> arch/x86/kernel/cpu/mce/internal.h:342: multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:arch/x86/kernel/cpu/mce/internal.h:342: first defined here
   ld: arch/x86/kernel/cpu/mce/genpool.o: in function `mce_zhaoxin_feature_init':
>> arch/x86/kernel/cpu/mce/internal.h:341: multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:arch/x86/kernel/cpu/mce/internal.h:341: first defined here
   ld: arch/x86/kernel/cpu/mce/genpool.o: in function `mce_zhaoxin_feature_clear':
>> arch/x86/kernel/cpu/mce/internal.h:342: multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:arch/x86/kernel/cpu/mce/internal.h:342: first defined here
   ld: arch/x86/kernel/cpu/mce/intel.o: in function `mce_zhaoxin_feature_init':
>> arch/x86/kernel/cpu/mce/internal.h:341: multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:arch/x86/kernel/cpu/mce/internal.h:341: first defined here
   ld: arch/x86/kernel/cpu/mce/intel.o: in function `mce_zhaoxin_feature_clear':
>> arch/x86/kernel/cpu/mce/internal.h:342: multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:arch/x86/kernel/cpu/mce/internal.h:342: first defined here
   ld: arch/x86/kernel/cpu/mce/threshold.o: in function `mce_zhaoxin_feature_init':
>> arch/x86/kernel/cpu/mce/internal.h:341: multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:arch/x86/kernel/cpu/mce/internal.h:341: first defined here
   ld: arch/x86/kernel/cpu/mce/threshold.o: in function `mce_zhaoxin_feature_clear':
>> arch/x86/kernel/cpu/mce/internal.h:342: multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:arch/x86/kernel/cpu/mce/internal.h:342: first defined here
   ld: arch/x86/kernel/cpu/mce/dev-mcelog.o: in function `mce_zhaoxin_feature_init':
>> arch/x86/kernel/cpu/mce/internal.h:341: multiple definition of `mce_zhaoxin_feature_init'; arch/x86/kernel/cpu/mce/core.o:arch/x86/kernel/cpu/mce/internal.h:341: first defined here
   ld: arch/x86/kernel/cpu/mce/dev-mcelog.o: in function `mce_zhaoxin_feature_clear':
>> arch/x86/kernel/cpu/mce/internal.h:342: multiple definition of `mce_zhaoxin_feature_clear'; arch/x86/kernel/cpu/mce/core.o:arch/x86/kernel/cpu/mce/internal.h:342: first defined here


vim +341 arch/x86/kernel/cpu/mce/internal.h

   335	
   336	extern void (*mc_poll_banks)(void);
   337	#ifdef CONFIG_X86_MCE_ZHAOXIN
   338	void mce_zhaoxin_feature_init(struct cpuinfo_x86 *c);
   339	void mce_zhaoxin_feature_clear(struct cpuinfo_x86 *c);
   340	#else
 > 341	void mce_zhaoxin_feature_init(struct cpuinfo_x86 *c) {}
 > 342	void mce_zhaoxin_feature_clear(struct cpuinfo_x86 *c) {}

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

Re: [PATCH v1 1/3] x86/mce: Add centaur vendor to support Zhaoxin MCA

[Borislav Petkov]

On Mon, Sep 09, 2024 at 06:43:47PM +0800, Tony W Wang-oc wrote:
> @@ -1970,6 +1974,18 @@ static int __mcheck_cpu_apply_quirks(struct cpuinfo_x86 *c)
>  		}
>  	}
>  
> +	if (c->x86_vendor == X86_VENDOR_CENTAUR) {
> +		/*
> +		 * All newer Centaur CPUs support MCE broadcasting. Enable
> +		 * synchronization with a one second timeout.
> +		 */
> +		if ((c->x86 == 6 && c->x86_model == 0xf && c->x86_stepping >= 0xe) ||
> +		     c->x86 > 6) {
> +			if (cfg->monarch_timeout < 0)
> +				cfg->monarch_timeout = USEC_PER_SEC;
> +		}
> +	}

So if centaur == zhaoxin, why aren't you moving this hunk to
mce_zhaoxin_feature_init() instead?


-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCH v1 1/3] x86/mce: Add centaur vendor to support Zhaoxin MCA

[Borislav Petkov]

On Mon, Sep 09, 2024 at 06:43:47PM +0800, Tony W Wang-oc wrote:
> From: Lyle Li <LyleLi@zhaoxin.com>
> 
> Zhaoxin consists of two vendors, X86_VENDOR_ZHAOXIN and
> X86_VENDOR_CENTAUR, so add the centaur vendor to support
> Zhaoxin MCA in mce/core.c and mce/intel.c.
> 
> Signed-off-by: Lyle Li <LyleLi@zhaoxin.com>
> Reviewed-by: Tony W Wang-oc <TonyWWang-oc@zhaoxin.com>

If you're sending someone else's patch, then you need to add your SOB too.

Please educate yourself on the process:

https://kernel.org/doc/html/latest/process/submitting-patches.html

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCH v1 1/3] x86/mce: Add centaur vendor to support Zhaoxin MCA

[Dave Hansen]

On 9/13/24 07:27, Borislav Petkov wrote:
>> +	if (c->x86_vendor == X86_VENDOR_CENTAUR) {
>> +		/*
>> +		 * All newer Centaur CPUs support MCE broadcasting. Enable
>> +		 * synchronization with a one second timeout.
>> +		 */
>> +		if ((c->x86 == 6 && c->x86_model == 0xf && c->x86_stepping >= 0xe) ||
>> +		     c->x86 > 6) {
>> +			if (cfg->monarch_timeout < 0)
>> +				cfg->monarch_timeout = USEC_PER_SEC;
>> +		}
>> +	}
> So if centaur == zhaoxin, why aren't you moving this hunk to
> mce_zhaoxin_feature_init() instead?

The centaur and zhaoxin logic is also _really_ close here:

>                 if (c->x86 > 6 || (c->x86_model == 0x19 || c->x86_model == 0x1f)) {
>                         if (cfg->monarch_timeout < 0)
>                                 cfg->monarch_timeout = USEC_PER_SEC;
>                 }

vs

>         if ((c->x86 == 6 && c->x86_model == 0xf && c->x86_stepping >= 0xe) ||
>              c->x86 > 6) {
>                 if (cfg->monarch_timeout < 0)
>                         cfg->monarch_timeout = USEC_PER_SEC;
>         }

I'd just randomly guess that the zhaoxin version is buggy because it
doesn't do a c->x86 check before the "(c->x86_model == 0x19 ||
c->x86_model == 0x1f)".

So instead of copying and pasting the same block over and over, can we
consolidate it a bit?

foo()
{
	/* Older CPUs do not do MCE broadcast: */
	if (c->x86 < 6)
		return;
	/* All newer ones do: */
	if (c->x86 > 6)
		goto mce_broadcast;

	/* Family 6 is mixed: */
	if (c->x86_vendor == X86_VENDOR_CENTAUR) {
		if (c->x86_model == 0xf &&
		    c->x86_stepping >= 0xe)
			goto mce_broadcast;
	} else if (c->x86_vendor == X86_VENDOR_ZHAOXIN) {
		if (c->x86_model == 0x19 ||
		    c->x86_model == 0x1f))
			goto mce_broadcast;
	}

	return;

mce_broadcast:
	if (cfg->monarch_timeout < 0)
		cfg->monarch_timeout = USEC_PER_SEC;
}


Heck, the Intel code can even go in there I think.  Wouldn't that tell
the story a bit better?

Re: [PATCH v1 1/3] x86/mce: Add centaur vendor to support Zhaoxin MCA

[Tony W Wang-oc]

On 2024/9/13 22:27, Borislav Petkov wrote:
> 
> 
> [这封邮件来自外部发件人 谨防风险]
> 
> On Mon, Sep 09, 2024 at 06:43:47PM +0800, Tony W Wang-oc wrote:
>> @@ -1970,6 +1974,18 @@ static int __mcheck_cpu_apply_quirks(struct cpuinfo_x86 *c)
>>                }
>>        }
>>
>> +     if (c->x86_vendor == X86_VENDOR_CENTAUR) {
>> +             /*
>> +              * All newer Centaur CPUs support MCE broadcasting. Enable
>> +              * synchronization with a one second timeout.
>> +              */
>> +             if ((c->x86 == 6 && c->x86_model == 0xf && c->x86_stepping >= 0xe) ||
>> +                  c->x86 > 6) {
>> +                     if (cfg->monarch_timeout < 0)
>> +                             cfg->monarch_timeout = USEC_PER_SEC;
>> +             }
>> +     }
> 
> So if centaur == zhaoxin, why aren't you moving this hunk to
> mce_zhaoxin_feature_init() instead?
> 
>
Ok, will adjust the patch in the next version.

Sincerely!
TonyWWang-oc

Re: [PATCH v1 1/3] x86/mce: Add centaur vendor to support Zhaoxin MCA

[Tony W Wang-oc]

On 2024/9/13 23:47, Dave Hansen wrote:
> 
> 
> [这封邮件来自外部发件人 谨防风险]
> 
> On 9/13/24 07:27, Borislav Petkov wrote:
>>> +    if (c->x86_vendor == X86_VENDOR_CENTAUR) {
>>> +            /*
>>> +             * All newer Centaur CPUs support MCE broadcasting. Enable
>>> +             * synchronization with a one second timeout.
>>> +             */
>>> +            if ((c->x86 == 6 && c->x86_model == 0xf && c->x86_stepping >= 0xe) ||
>>> +                 c->x86 > 6) {
>>> +                    if (cfg->monarch_timeout < 0)
>>> +                            cfg->monarch_timeout = USEC_PER_SEC;
>>> +            }
>>> +    }
>> So if centaur == zhaoxin, why aren't you moving this hunk to
>> mce_zhaoxin_feature_init() instead?
> 
> The centaur and zhaoxin logic is also _really_ close here:
> 
>>                  if (c->x86 > 6 || (c->x86_model == 0x19 || c->x86_model == 0x1f)) {
>>                          if (cfg->monarch_timeout < 0)
>>                                  cfg->monarch_timeout = USEC_PER_SEC;
>>                  }
> 
> vs
> 
>>          if ((c->x86 == 6 && c->x86_model == 0xf && c->x86_stepping >= 0xe) ||
>>               c->x86 > 6) {
>>                  if (cfg->monarch_timeout < 0)
>>                          cfg->monarch_timeout = USEC_PER_SEC;
>>          }
> 
> I'd just randomly guess that the zhaoxin version is buggy because it
> doesn't do a c->x86 check before the "(c->x86_model == 0x19 ||
> c->x86_model == 0x1f)".
> 

Yes, the check for c->x86 == 6 is omitted in the zhaoxin version.

> So instead of copying and pasting the same block over and over, can we
> consolidate it a bit?
> 
> foo()
> {
>          /* Older CPUs do not do MCE broadcast: */
>          if (c->x86 < 6)
>                  return;
>          /* All newer ones do: */
>          if (c->x86 > 6)
>                  goto mce_broadcast;
> 
>          /* Family 6 is mixed: */
>          if (c->x86_vendor == X86_VENDOR_CENTAUR) {
>                  if (c->x86_model == 0xf &&
>                      c->x86_stepping >= 0xe)
>                          goto mce_broadcast;
>          } else if (c->x86_vendor == X86_VENDOR_ZHAOXIN) {
>                  if (c->x86_model == 0x19 ||
>                      c->x86_model == 0x1f))
>                          goto mce_broadcast;
>          }
> 
>          return;
> 
> mce_broadcast:
>          if (cfg->monarch_timeout < 0)
>                  cfg->monarch_timeout = USEC_PER_SEC;
> }
>

Thank you! That makes more sense, and will adopt it into zhaoxin.c

Sincerely!
TonyWWang-oc