diff mbox series

[V10,08/19] riscv: qspinlock: Use new static key for controlling call of virt_spin_lock()

Message ID 20230802164701.192791-9-guoren@kernel.org (mailing list archive)
State New, archived
Headers show
Series riscv: Add Native/Paravirt/CNA qspinlock support | expand

Commit Message

Guo Ren Aug. 2, 2023, 4:46 p.m. UTC
From: Guo Ren <guoren@linux.alibaba.com>

Add a static key controlling whether virt_spin_lock() should be
called or not. When running on bare metal set the new key to
false.

The KVM guests fall back to a Test-and-Set spinlock, because fair
locks have horrible lock 'holder' preemption issues. The
virt_spin_lock_key would shortcut for the
queued_spin_lock_slowpath() function that allow virt_spin_lock to
hijack it.

Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>
---
 arch/riscv/Kconfig                |  1 +
 arch/riscv/include/asm/sbi.h      |  8 ++++++++
 arch/riscv/include/asm/spinlock.h | 22 ++++++++++++++++++++++
 arch/riscv/kernel/cpufeature.c    |  4 +++-
 arch/riscv/kernel/sbi.c           |  2 +-
 arch/riscv/kernel/setup.c         | 19 +++++++++++++++++++
 6 files changed, 54 insertions(+), 2 deletions(-)
diff mbox series

Patch

diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index 119e774a3dcf..42ae45c42b4d 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -20,6 +20,7 @@  config RISCV
 	select ARCH_ENABLE_THP_MIGRATION if TRANSPARENT_HUGEPAGE
 	select ARCH_HAS_BINFMT_FLAT
 	select ARCH_HAS_CURRENT_STACK_POINTER
+	select ARCH_HAS_CPU_FINALIZE_INIT
 	select ARCH_HAS_DEBUG_VIRTUAL if MMU
 	select ARCH_HAS_DEBUG_VM_PGTABLE
 	select ARCH_HAS_DEBUG_WX
diff --git a/arch/riscv/include/asm/sbi.h b/arch/riscv/include/asm/sbi.h
index e1523c8624cc..b7ced34b79a3 100644
--- a/arch/riscv/include/asm/sbi.h
+++ b/arch/riscv/include/asm/sbi.h
@@ -51,6 +51,13 @@  enum sbi_ext_base_fid {
 	SBI_EXT_BASE_GET_MIMPID,
 };
 
+enum sbi_ext_base_impl_id {
+	SBI_EXT_BASE_IMPL_ID_BBL = 0,
+	SBI_EXT_BASE_IMPL_ID_OPENSBI,
+	SBI_EXT_BASE_IMPL_ID_XVISOR,
+	SBI_EXT_BASE_IMPL_ID_KVM,
+};
+
 enum sbi_ext_time_fid {
 	SBI_EXT_TIME_SET_TIMER = 0,
 };
@@ -286,6 +293,7 @@  int sbi_console_getchar(void);
 long sbi_get_mvendorid(void);
 long sbi_get_marchid(void);
 long sbi_get_mimpid(void);
+long sbi_get_firmware_id(void);
 void sbi_set_timer(uint64_t stime_value);
 void sbi_shutdown(void);
 void sbi_send_ipi(unsigned int cpu);
diff --git a/arch/riscv/include/asm/spinlock.h b/arch/riscv/include/asm/spinlock.h
index 9eb3ad31e564..13f3e14500c0 100644
--- a/arch/riscv/include/asm/spinlock.h
+++ b/arch/riscv/include/asm/spinlock.h
@@ -4,6 +4,28 @@ 
 #define __ASM_RISCV_SPINLOCK_H
 
 #ifdef CONFIG_QUEUED_SPINLOCKS
+/*
+ * The KVM guests fall back to a Test-and-Set spinlock, because fair locks
+ * have horrible lock 'holder' preemption issues. The virt_spin_lock_key
+ * would shortcut for the queued_spin_lock_slowpath() function that allow
+ * virt_spin_lock to hijack it.
+ */
+DECLARE_STATIC_KEY_TRUE(virt_spin_lock_key);
+
+#define virt_spin_lock virt_spin_lock
+static inline bool virt_spin_lock(struct qspinlock *lock)
+{
+	if (!static_branch_likely(&virt_spin_lock_key))
+		return false;
+
+	do {
+		while (atomic_read(&lock->val) != 0)
+			cpu_relax();
+	} while (atomic_cmpxchg(&lock->val, 0, _Q_LOCKED_VAL) != 0);
+
+	return true;
+}
+
 #define _Q_PENDING_LOOPS	(1 << 9)
 #endif
 
diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
index d9694fe40a9a..26826aa590e9 100644
--- a/arch/riscv/kernel/cpufeature.c
+++ b/arch/riscv/kernel/cpufeature.c
@@ -21,6 +21,7 @@ 
 #include <asm/hwcap.h>
 #include <asm/patch.h>
 #include <asm/processor.h>
+#include <asm/sbi.h>
 #include <asm/vector.h>
 
 #define NUM_ALPHA_EXTS ('z' - 'a' + 1)
@@ -343,7 +344,8 @@  void __init riscv_fill_hwcap(void)
 		 * ticket_spinlock, but can not be vice.
 		 */
 		if (!force_qspinlock &&
-		    !riscv_has_errata_thead_qspinlock()) {
+		    !riscv_has_errata_thead_qspinlock() &&
+		    (sbi_get_firmware_id() != SBI_EXT_BASE_IMPL_ID_KVM)) {
 			set_bit(RISCV_ISA_EXT_XTICKETLOCK, isainfo->isa);
 		}
 #endif
diff --git a/arch/riscv/kernel/sbi.c b/arch/riscv/kernel/sbi.c
index c672c8ba9a2a..398b768a02e6 100644
--- a/arch/riscv/kernel/sbi.c
+++ b/arch/riscv/kernel/sbi.c
@@ -555,7 +555,7 @@  static inline long sbi_get_spec_version(void)
 	return __sbi_base_ecall(SBI_EXT_BASE_GET_SPEC_VERSION);
 }
 
-static inline long sbi_get_firmware_id(void)
+long sbi_get_firmware_id(void)
 {
 	return __sbi_base_ecall(SBI_EXT_BASE_GET_IMP_ID);
 }
diff --git a/arch/riscv/kernel/setup.c b/arch/riscv/kernel/setup.c
index 971fe776e2f8..def89fd8ea55 100644
--- a/arch/riscv/kernel/setup.c
+++ b/arch/riscv/kernel/setup.c
@@ -26,6 +26,7 @@ 
 #include <asm/alternative.h>
 #include <asm/cacheflush.h>
 #include <asm/cpu_ops.h>
+#include <asm/cpufeature.h>
 #include <asm/early_ioremap.h>
 #include <asm/pgtable.h>
 #include <asm/setup.h>
@@ -264,6 +265,19 @@  static void __init parse_dtb(void)
 #endif
 }
 
+#ifdef CONFIG_QUEUED_SPINLOCKS
+DEFINE_STATIC_KEY_TRUE(virt_spin_lock_key);
+
+static void __init virt_spin_lock_init(void)
+{
+	if (sbi_get_firmware_id() != SBI_EXT_BASE_IMPL_ID_KVM ||
+	    force_qspinlock)
+		static_branch_disable(&virt_spin_lock_key);
+}
+#else
+static void __init virt_spin_lock_init(void) {}
+#endif
+
 extern void __init init_rt_signal_env(void);
 
 void __init setup_arch(char **cmdline_p)
@@ -313,6 +327,11 @@  void __init setup_arch(char **cmdline_p)
 		riscv_noncoherent_supported();
 }
 
+void __init arch_cpu_finalize_init(void)
+{
+	virt_spin_lock_init();
+}
+
 static int __init topology_init(void)
 {
 	int i, ret;