diff mbox series

[V10,05/19] riscv: qspinlock: Introduce combo spinlock

Message ID 20230802164701.192791-6-guoren@kernel.org (mailing list archive)
State Handled Elsewhere
Headers show
Series riscv: Add Native/Paravirt/CNA qspinlock support | expand

Checks

Context Check Description
conchuod/tree_selection fail Failed to apply to next/pending-fixes, riscv/for-next or riscv/master

Commit Message

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

Combo spinlock could support queued and ticket in one Linux Image and
select them during boot time via errata mechanism. Here is the func
size (Bytes) comparison table below:

TYPE			: COMBO | TICKET | QUEUED
arch_spin_lock		: 106	| 60     | 50
arch_spin_unlock	: 54    | 36     | 26
arch_spin_trylock	: 110   | 72     | 54
arch_spin_is_locked	: 48    | 34     | 20
arch_spin_is_contended	: 56    | 40     | 24
rch_spin_value_unlocked	: 48    | 34     | 24

One example of disassemble combo arch_spin_unlock:
   0xffffffff8000409c <+14>:    nop                # detour slot
   0xffffffff800040a0 <+18>:    fence   rw,w       # queued spinlock start
   0xffffffff800040a4 <+22>:    sb      zero,0(a4) # queued spinlock end
   0xffffffff800040a8 <+26>:    ld      s0,8(sp)
   0xffffffff800040aa <+28>:    addi    sp,sp,16
   0xffffffff800040ac <+30>:    ret
   0xffffffff800040ae <+32>:    lw      a5,0(a4)   # ticket spinlock start
   0xffffffff800040b0 <+34>:    sext.w  a5,a5
   0xffffffff800040b2 <+36>:    fence   rw,w
   0xffffffff800040b6 <+40>:    addiw   a5,a5,1
   0xffffffff800040b8 <+42>:    slli    a5,a5,0x30
   0xffffffff800040ba <+44>:    srli    a5,a5,0x30
   0xffffffff800040bc <+46>:    sh      a5,0(a4)   # ticket spinlock end
   0xffffffff800040c0 <+50>:    ld      s0,8(sp)
   0xffffffff800040c2 <+52>:    addi    sp,sp,16
   0xffffffff800040c4 <+54>:    ret

The qspinlock is smaller and faster than ticket-lock when all are in
fast-path, and combo spinlock could provide a compatible Linux Image
for different micro-arch design (weak/strict fwd guarantee) processors.

Signed-off-by: Guo Ren <guoren@kernel.org>
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
---
 arch/riscv/Kconfig                |  9 +++-
 arch/riscv/include/asm/hwcap.h    |  1 +
 arch/riscv/include/asm/spinlock.h | 87 ++++++++++++++++++++++++++++++-
 arch/riscv/kernel/cpufeature.c    | 10 ++++
 4 files changed, 104 insertions(+), 3 deletions(-)

Comments

Waiman Long Aug. 11, 2023, 7:51 p.m. UTC | #1
On 8/2/23 12:46, guoren@kernel.org wrote:
> From: Guo Ren <guoren@linux.alibaba.com>
>
> Combo spinlock could support queued and ticket in one Linux Image and
> select them during boot time via errata mechanism. Here is the func
> size (Bytes) comparison table below:
>
> TYPE			: COMBO | TICKET | QUEUED
> arch_spin_lock		: 106	| 60     | 50
> arch_spin_unlock	: 54    | 36     | 26
> arch_spin_trylock	: 110   | 72     | 54
> arch_spin_is_locked	: 48    | 34     | 20
> arch_spin_is_contended	: 56    | 40     | 24
> rch_spin_value_unlocked	: 48    | 34     | 24
>
> One example of disassemble combo arch_spin_unlock:
>     0xffffffff8000409c <+14>:    nop                # detour slot
>     0xffffffff800040a0 <+18>:    fence   rw,w       # queued spinlock start
>     0xffffffff800040a4 <+22>:    sb      zero,0(a4) # queued spinlock end
>     0xffffffff800040a8 <+26>:    ld      s0,8(sp)
>     0xffffffff800040aa <+28>:    addi    sp,sp,16
>     0xffffffff800040ac <+30>:    ret
>     0xffffffff800040ae <+32>:    lw      a5,0(a4)   # ticket spinlock start
>     0xffffffff800040b0 <+34>:    sext.w  a5,a5
>     0xffffffff800040b2 <+36>:    fence   rw,w
>     0xffffffff800040b6 <+40>:    addiw   a5,a5,1
>     0xffffffff800040b8 <+42>:    slli    a5,a5,0x30
>     0xffffffff800040ba <+44>:    srli    a5,a5,0x30
>     0xffffffff800040bc <+46>:    sh      a5,0(a4)   # ticket spinlock end
>     0xffffffff800040c0 <+50>:    ld      s0,8(sp)
>     0xffffffff800040c2 <+52>:    addi    sp,sp,16
>     0xffffffff800040c4 <+54>:    ret
>
> The qspinlock is smaller and faster than ticket-lock when all are in
> fast-path, and combo spinlock could provide a compatible Linux Image
> for different micro-arch design (weak/strict fwd guarantee) processors.
>
> Signed-off-by: Guo Ren <guoren@kernel.org>
> Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
> ---
>   arch/riscv/Kconfig                |  9 +++-
>   arch/riscv/include/asm/hwcap.h    |  1 +
>   arch/riscv/include/asm/spinlock.h | 87 ++++++++++++++++++++++++++++++-
>   arch/riscv/kernel/cpufeature.c    | 10 ++++
>   4 files changed, 104 insertions(+), 3 deletions(-)
>
> diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
> index e89a3bea3dc1..119e774a3dcf 100644
> --- a/arch/riscv/Kconfig
> +++ b/arch/riscv/Kconfig
> @@ -440,7 +440,7 @@ config NODES_SHIFT
>   
>   choice
>   	prompt "RISC-V spinlock type"
> -	default RISCV_TICKET_SPINLOCKS
> +	default RISCV_COMBO_SPINLOCKS
>   
>   config RISCV_TICKET_SPINLOCKS
>   	bool "Using ticket spinlock"
> @@ -452,6 +452,13 @@ config RISCV_QUEUED_SPINLOCKS
>   	help
>   	  Make sure your micro arch LL/SC has a strong forward progress guarantee.
>   	  Otherwise, stay at ticket-lock.
> +
> +config RISCV_COMBO_SPINLOCKS
> +	bool "Using combo spinlock"
> +	depends on SMP && MMU
> +	select ARCH_USE_QUEUED_SPINLOCKS
> +	help
> +	  Select queued spinlock or ticket-lock via errata.
>   endchoice
>   
>   config RISCV_ALTERNATIVE
> diff --git a/arch/riscv/include/asm/hwcap.h b/arch/riscv/include/asm/hwcap.h
> index f041bfa7f6a0..08ae75a694c2 100644
> --- a/arch/riscv/include/asm/hwcap.h
> +++ b/arch/riscv/include/asm/hwcap.h
> @@ -54,6 +54,7 @@
>   #define RISCV_ISA_EXT_ZIFENCEI		41
>   #define RISCV_ISA_EXT_ZIHPM		42
>   
> +#define RISCV_ISA_EXT_XTICKETLOCK	63
>   #define RISCV_ISA_EXT_MAX		64
>   #define RISCV_ISA_EXT_NAME_LEN_MAX	32
>   
> diff --git a/arch/riscv/include/asm/spinlock.h b/arch/riscv/include/asm/spinlock.h
> index c644a92d4548..9eb3ad31e564 100644
> --- a/arch/riscv/include/asm/spinlock.h
> +++ b/arch/riscv/include/asm/spinlock.h
> @@ -7,11 +7,94 @@
>   #define _Q_PENDING_LOOPS	(1 << 9)
>   #endif
>   

I see why you separated the _Q_PENDING_LOOPS out.


> +#ifdef CONFIG_RISCV_COMBO_SPINLOCKS
> +#include <asm-generic/ticket_spinlock.h>
> +
> +#undef arch_spin_is_locked
> +#undef arch_spin_is_contended
> +#undef arch_spin_value_unlocked
> +#undef arch_spin_lock
> +#undef arch_spin_trylock
> +#undef arch_spin_unlock
> +
> +#include <asm-generic/qspinlock.h>
> +#include <asm/hwcap.h>
> +
> +#undef arch_spin_is_locked
> +#undef arch_spin_is_contended
> +#undef arch_spin_value_unlocked
> +#undef arch_spin_lock
> +#undef arch_spin_trylock
> +#undef arch_spin_unlock
Perhaps you can add a macro like __no_arch_spinlock_redefine to disable 
the various arch_spin_* definition in qspinlock.h and ticket_spinlock.h.
> +
> +#define COMBO_DETOUR				\
> +	asm_volatile_goto(ALTERNATIVE(		\
> +		"nop",				\
> +		"j %l[ticket_spin_lock]",	\
> +		0,				\
> +		RISCV_ISA_EXT_XTICKETLOCK,	\
> +		CONFIG_RISCV_COMBO_SPINLOCKS)	\
> +		: : : : ticket_spin_lock);
> +
> +static __always_inline void arch_spin_lock(arch_spinlock_t *lock)
> +{
> +	COMBO_DETOUR
> +	queued_spin_lock(lock);
> +	return;
> +ticket_spin_lock:
> +	ticket_spin_lock(lock);
> +}
> +
> +static __always_inline bool arch_spin_trylock(arch_spinlock_t *lock)
> +{
> +	COMBO_DETOUR
> +	return queued_spin_trylock(lock);
> +ticket_spin_lock:
> +	return ticket_spin_trylock(lock);
> +}
> +
> +static __always_inline void arch_spin_unlock(arch_spinlock_t *lock)
> +{
> +	COMBO_DETOUR
> +	queued_spin_unlock(lock);
> +	return;
> +ticket_spin_lock:
> +	ticket_spin_unlock(lock);
> +}
> +
> +static __always_inline int arch_spin_value_unlocked(arch_spinlock_t lock)
> +{
> +	COMBO_DETOUR
> +	return queued_spin_value_unlocked(lock);
> +ticket_spin_lock:
> +	return ticket_spin_value_unlocked(lock);
> +}
> +
> +static __always_inline int arch_spin_is_locked(arch_spinlock_t *lock)
> +{
> +	COMBO_DETOUR
> +	return queued_spin_is_locked(lock);
> +ticket_spin_lock:
> +	return ticket_spin_is_locked(lock);
> +}
> +
> +static __always_inline int arch_spin_is_contended(arch_spinlock_t *lock)
> +{
> +	COMBO_DETOUR
> +	return queued_spin_is_contended(lock);
> +ticket_spin_lock:
> +	return ticket_spin_is_contended(lock);
> +}
> +#else /* CONFIG_RISCV_COMBO_SPINLOCKS */
> +
>   #ifdef CONFIG_QUEUED_SPINLOCKS
>   #include <asm/qspinlock.h>
> -#include <asm/qrwlock.h>
>   #else
> -#include <asm-generic/spinlock.h>
> +#include <asm-generic/ticket_spinlock.h>
>   #endif
>   
> +#endif /* CONFIG_RISCV_COMBO_SPINLOCKS */
> +
> +#include <asm/qrwlock.h>
> +
>   #endif /* __ASM_RISCV_SPINLOCK_H */
> diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
> index bdcf460ea53d..e65b0e54152d 100644
> --- a/arch/riscv/kernel/cpufeature.c
> +++ b/arch/riscv/kernel/cpufeature.c
> @@ -324,6 +324,16 @@ void __init riscv_fill_hwcap(void)
>   		set_bit(RISCV_ISA_EXT_ZICSR, isainfo->isa);
>   		set_bit(RISCV_ISA_EXT_ZIFENCEI, isainfo->isa);
>   
> +#ifdef CONFIG_RISCV_COMBO_SPINLOCKS
> +		/*
> +		 * The RISC-V Linux used queued spinlock at first; then, we used ticket_lock
> +		 * as default or queued spinlock by choice. Because ticket_lock would dirty
> +		 * spinlock value, the only way is to change from queued_spinlock to
> +		 * ticket_spinlock, but can not be vice.

The phrase "but can not be vice" is confusing. I think you mean "but not 
vice versa". Right?

Cheers,
Longman
Guo Ren Aug. 12, 2023, 12:22 a.m. UTC | #2
On Sat, Aug 12, 2023 at 3:51 AM Waiman Long <longman@redhat.com> wrote:
>
> On 8/2/23 12:46, guoren@kernel.org wrote:
> > From: Guo Ren <guoren@linux.alibaba.com>
> >
> > Combo spinlock could support queued and ticket in one Linux Image and
> > select them during boot time via errata mechanism. Here is the func
> > size (Bytes) comparison table below:
> >
> > TYPE                  : COMBO | TICKET | QUEUED
> > arch_spin_lock                : 106   | 60     | 50
> > arch_spin_unlock      : 54    | 36     | 26
> > arch_spin_trylock     : 110   | 72     | 54
> > arch_spin_is_locked   : 48    | 34     | 20
> > arch_spin_is_contended        : 56    | 40     | 24
> > rch_spin_value_unlocked       : 48    | 34     | 24
> >
> > One example of disassemble combo arch_spin_unlock:
> >     0xffffffff8000409c <+14>:    nop                # detour slot
> >     0xffffffff800040a0 <+18>:    fence   rw,w       # queued spinlock start
> >     0xffffffff800040a4 <+22>:    sb      zero,0(a4) # queued spinlock end
> >     0xffffffff800040a8 <+26>:    ld      s0,8(sp)
> >     0xffffffff800040aa <+28>:    addi    sp,sp,16
> >     0xffffffff800040ac <+30>:    ret
> >     0xffffffff800040ae <+32>:    lw      a5,0(a4)   # ticket spinlock start
> >     0xffffffff800040b0 <+34>:    sext.w  a5,a5
> >     0xffffffff800040b2 <+36>:    fence   rw,w
> >     0xffffffff800040b6 <+40>:    addiw   a5,a5,1
> >     0xffffffff800040b8 <+42>:    slli    a5,a5,0x30
> >     0xffffffff800040ba <+44>:    srli    a5,a5,0x30
> >     0xffffffff800040bc <+46>:    sh      a5,0(a4)   # ticket spinlock end
> >     0xffffffff800040c0 <+50>:    ld      s0,8(sp)
> >     0xffffffff800040c2 <+52>:    addi    sp,sp,16
> >     0xffffffff800040c4 <+54>:    ret
> >
> > The qspinlock is smaller and faster than ticket-lock when all are in
> > fast-path, and combo spinlock could provide a compatible Linux Image
> > for different micro-arch design (weak/strict fwd guarantee) processors.
> >
> > Signed-off-by: Guo Ren <guoren@kernel.org>
> > Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
> > ---
> >   arch/riscv/Kconfig                |  9 +++-
> >   arch/riscv/include/asm/hwcap.h    |  1 +
> >   arch/riscv/include/asm/spinlock.h | 87 ++++++++++++++++++++++++++++++-
> >   arch/riscv/kernel/cpufeature.c    | 10 ++++
> >   4 files changed, 104 insertions(+), 3 deletions(-)
> >
> > diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
> > index e89a3bea3dc1..119e774a3dcf 100644
> > --- a/arch/riscv/Kconfig
> > +++ b/arch/riscv/Kconfig
> > @@ -440,7 +440,7 @@ config NODES_SHIFT
> >
> >   choice
> >       prompt "RISC-V spinlock type"
> > -     default RISCV_TICKET_SPINLOCKS
> > +     default RISCV_COMBO_SPINLOCKS
> >
> >   config RISCV_TICKET_SPINLOCKS
> >       bool "Using ticket spinlock"
> > @@ -452,6 +452,13 @@ config RISCV_QUEUED_SPINLOCKS
> >       help
> >         Make sure your micro arch LL/SC has a strong forward progress guarantee.
> >         Otherwise, stay at ticket-lock.
> > +
> > +config RISCV_COMBO_SPINLOCKS
> > +     bool "Using combo spinlock"
> > +     depends on SMP && MMU
> > +     select ARCH_USE_QUEUED_SPINLOCKS
> > +     help
> > +       Select queued spinlock or ticket-lock via errata.
> >   endchoice
> >
> >   config RISCV_ALTERNATIVE
> > diff --git a/arch/riscv/include/asm/hwcap.h b/arch/riscv/include/asm/hwcap.h
> > index f041bfa7f6a0..08ae75a694c2 100644
> > --- a/arch/riscv/include/asm/hwcap.h
> > +++ b/arch/riscv/include/asm/hwcap.h
> > @@ -54,6 +54,7 @@
> >   #define RISCV_ISA_EXT_ZIFENCEI              41
> >   #define RISCV_ISA_EXT_ZIHPM         42
> >
> > +#define RISCV_ISA_EXT_XTICKETLOCK    63
> >   #define RISCV_ISA_EXT_MAX           64
> >   #define RISCV_ISA_EXT_NAME_LEN_MAX  32
> >
> > diff --git a/arch/riscv/include/asm/spinlock.h b/arch/riscv/include/asm/spinlock.h
> > index c644a92d4548..9eb3ad31e564 100644
> > --- a/arch/riscv/include/asm/spinlock.h
> > +++ b/arch/riscv/include/asm/spinlock.h
> > @@ -7,11 +7,94 @@
> >   #define _Q_PENDING_LOOPS    (1 << 9)
> >   #endif
> >
>
> I see why you separated the _Q_PENDING_LOOPS out.
haha, yes, I even forget this, :).

>
>
> > +#ifdef CONFIG_RISCV_COMBO_SPINLOCKS
> > +#include <asm-generic/ticket_spinlock.h>
> > +
> > +#undef arch_spin_is_locked
> > +#undef arch_spin_is_contended
> > +#undef arch_spin_value_unlocked
> > +#undef arch_spin_lock
> > +#undef arch_spin_trylock
> > +#undef arch_spin_unlock
> > +
> > +#include <asm-generic/qspinlock.h>
> > +#include <asm/hwcap.h>
> > +
> > +#undef arch_spin_is_locked
> > +#undef arch_spin_is_contended
> > +#undef arch_spin_value_unlocked
> > +#undef arch_spin_lock
> > +#undef arch_spin_trylock
> > +#undef arch_spin_unlock
> Perhaps you can add a macro like __no_arch_spinlock_redefine to disable
> the various arch_spin_* definition in qspinlock.h and ticket_spinlock.h.
That's great; I will try it in the next version.

> > +
> > +#define COMBO_DETOUR                         \
> > +     asm_volatile_goto(ALTERNATIVE(          \
> > +             "nop",                          \
> > +             "j %l[ticket_spin_lock]",       \
> > +             0,                              \
> > +             RISCV_ISA_EXT_XTICKETLOCK,      \
> > +             CONFIG_RISCV_COMBO_SPINLOCKS)   \
> > +             : : : : ticket_spin_lock);
> > +
> > +static __always_inline void arch_spin_lock(arch_spinlock_t *lock)
> > +{
> > +     COMBO_DETOUR
> > +     queued_spin_lock(lock);
> > +     return;
> > +ticket_spin_lock:
> > +     ticket_spin_lock(lock);
> > +}
> > +
> > +static __always_inline bool arch_spin_trylock(arch_spinlock_t *lock)
> > +{
> > +     COMBO_DETOUR
> > +     return queued_spin_trylock(lock);
> > +ticket_spin_lock:
> > +     return ticket_spin_trylock(lock);
> > +}
> > +
> > +static __always_inline void arch_spin_unlock(arch_spinlock_t *lock)
> > +{
> > +     COMBO_DETOUR
> > +     queued_spin_unlock(lock);
> > +     return;
> > +ticket_spin_lock:
> > +     ticket_spin_unlock(lock);
> > +}
> > +
> > +static __always_inline int arch_spin_value_unlocked(arch_spinlock_t lock)
> > +{
> > +     COMBO_DETOUR
> > +     return queued_spin_value_unlocked(lock);
> > +ticket_spin_lock:
> > +     return ticket_spin_value_unlocked(lock);
> > +}
> > +
> > +static __always_inline int arch_spin_is_locked(arch_spinlock_t *lock)
> > +{
> > +     COMBO_DETOUR
> > +     return queued_spin_is_locked(lock);
> > +ticket_spin_lock:
> > +     return ticket_spin_is_locked(lock);
> > +}
> > +
> > +static __always_inline int arch_spin_is_contended(arch_spinlock_t *lock)
> > +{
> > +     COMBO_DETOUR
> > +     return queued_spin_is_contended(lock);
> > +ticket_spin_lock:
> > +     return ticket_spin_is_contended(lock);
> > +}
> > +#else /* CONFIG_RISCV_COMBO_SPINLOCKS */
> > +
> >   #ifdef CONFIG_QUEUED_SPINLOCKS
> >   #include <asm/qspinlock.h>
> > -#include <asm/qrwlock.h>
> >   #else
> > -#include <asm-generic/spinlock.h>
> > +#include <asm-generic/ticket_spinlock.h>
> >   #endif
> >
> > +#endif /* CONFIG_RISCV_COMBO_SPINLOCKS */
> > +
> > +#include <asm/qrwlock.h>
> > +
> >   #endif /* __ASM_RISCV_SPINLOCK_H */
> > diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
> > index bdcf460ea53d..e65b0e54152d 100644
> > --- a/arch/riscv/kernel/cpufeature.c
> > +++ b/arch/riscv/kernel/cpufeature.c
> > @@ -324,6 +324,16 @@ void __init riscv_fill_hwcap(void)
> >               set_bit(RISCV_ISA_EXT_ZICSR, isainfo->isa);
> >               set_bit(RISCV_ISA_EXT_ZIFENCEI, isainfo->isa);
> >
> > +#ifdef CONFIG_RISCV_COMBO_SPINLOCKS
> > +             /*
> > +              * The RISC-V Linux used queued spinlock at first; then, we used ticket_lock
> > +              * as default or queued spinlock by choice. Because ticket_lock would dirty
> > +              * spinlock value, the only way is to change from queued_spinlock to
> > +              * ticket_spinlock, but can not be vice.
>
> The phrase "but can not be vice" is confusing. I think you mean "but not
> vice versa". Right?
Yes, thx for the grammar correction.

>
> Cheers,
> Longman
>
diff mbox series

Patch

diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index e89a3bea3dc1..119e774a3dcf 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -440,7 +440,7 @@  config NODES_SHIFT
 
 choice
 	prompt "RISC-V spinlock type"
-	default RISCV_TICKET_SPINLOCKS
+	default RISCV_COMBO_SPINLOCKS
 
 config RISCV_TICKET_SPINLOCKS
 	bool "Using ticket spinlock"
@@ -452,6 +452,13 @@  config RISCV_QUEUED_SPINLOCKS
 	help
 	  Make sure your micro arch LL/SC has a strong forward progress guarantee.
 	  Otherwise, stay at ticket-lock.
+
+config RISCV_COMBO_SPINLOCKS
+	bool "Using combo spinlock"
+	depends on SMP && MMU
+	select ARCH_USE_QUEUED_SPINLOCKS
+	help
+	  Select queued spinlock or ticket-lock via errata.
 endchoice
 
 config RISCV_ALTERNATIVE
diff --git a/arch/riscv/include/asm/hwcap.h b/arch/riscv/include/asm/hwcap.h
index f041bfa7f6a0..08ae75a694c2 100644
--- a/arch/riscv/include/asm/hwcap.h
+++ b/arch/riscv/include/asm/hwcap.h
@@ -54,6 +54,7 @@ 
 #define RISCV_ISA_EXT_ZIFENCEI		41
 #define RISCV_ISA_EXT_ZIHPM		42
 
+#define RISCV_ISA_EXT_XTICKETLOCK	63
 #define RISCV_ISA_EXT_MAX		64
 #define RISCV_ISA_EXT_NAME_LEN_MAX	32
 
diff --git a/arch/riscv/include/asm/spinlock.h b/arch/riscv/include/asm/spinlock.h
index c644a92d4548..9eb3ad31e564 100644
--- a/arch/riscv/include/asm/spinlock.h
+++ b/arch/riscv/include/asm/spinlock.h
@@ -7,11 +7,94 @@ 
 #define _Q_PENDING_LOOPS	(1 << 9)
 #endif
 
+#ifdef CONFIG_RISCV_COMBO_SPINLOCKS
+#include <asm-generic/ticket_spinlock.h>
+
+#undef arch_spin_is_locked
+#undef arch_spin_is_contended
+#undef arch_spin_value_unlocked
+#undef arch_spin_lock
+#undef arch_spin_trylock
+#undef arch_spin_unlock
+
+#include <asm-generic/qspinlock.h>
+#include <asm/hwcap.h>
+
+#undef arch_spin_is_locked
+#undef arch_spin_is_contended
+#undef arch_spin_value_unlocked
+#undef arch_spin_lock
+#undef arch_spin_trylock
+#undef arch_spin_unlock
+
+#define COMBO_DETOUR				\
+	asm_volatile_goto(ALTERNATIVE(		\
+		"nop",				\
+		"j %l[ticket_spin_lock]",	\
+		0,				\
+		RISCV_ISA_EXT_XTICKETLOCK,	\
+		CONFIG_RISCV_COMBO_SPINLOCKS)	\
+		: : : : ticket_spin_lock);
+
+static __always_inline void arch_spin_lock(arch_spinlock_t *lock)
+{
+	COMBO_DETOUR
+	queued_spin_lock(lock);
+	return;
+ticket_spin_lock:
+	ticket_spin_lock(lock);
+}
+
+static __always_inline bool arch_spin_trylock(arch_spinlock_t *lock)
+{
+	COMBO_DETOUR
+	return queued_spin_trylock(lock);
+ticket_spin_lock:
+	return ticket_spin_trylock(lock);
+}
+
+static __always_inline void arch_spin_unlock(arch_spinlock_t *lock)
+{
+	COMBO_DETOUR
+	queued_spin_unlock(lock);
+	return;
+ticket_spin_lock:
+	ticket_spin_unlock(lock);
+}
+
+static __always_inline int arch_spin_value_unlocked(arch_spinlock_t lock)
+{
+	COMBO_DETOUR
+	return queued_spin_value_unlocked(lock);
+ticket_spin_lock:
+	return ticket_spin_value_unlocked(lock);
+}
+
+static __always_inline int arch_spin_is_locked(arch_spinlock_t *lock)
+{
+	COMBO_DETOUR
+	return queued_spin_is_locked(lock);
+ticket_spin_lock:
+	return ticket_spin_is_locked(lock);
+}
+
+static __always_inline int arch_spin_is_contended(arch_spinlock_t *lock)
+{
+	COMBO_DETOUR
+	return queued_spin_is_contended(lock);
+ticket_spin_lock:
+	return ticket_spin_is_contended(lock);
+}
+#else /* CONFIG_RISCV_COMBO_SPINLOCKS */
+
 #ifdef CONFIG_QUEUED_SPINLOCKS
 #include <asm/qspinlock.h>
-#include <asm/qrwlock.h>
 #else
-#include <asm-generic/spinlock.h>
+#include <asm-generic/ticket_spinlock.h>
 #endif
 
+#endif /* CONFIG_RISCV_COMBO_SPINLOCKS */
+
+#include <asm/qrwlock.h>
+
 #endif /* __ASM_RISCV_SPINLOCK_H */
diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
index bdcf460ea53d..e65b0e54152d 100644
--- a/arch/riscv/kernel/cpufeature.c
+++ b/arch/riscv/kernel/cpufeature.c
@@ -324,6 +324,16 @@  void __init riscv_fill_hwcap(void)
 		set_bit(RISCV_ISA_EXT_ZICSR, isainfo->isa);
 		set_bit(RISCV_ISA_EXT_ZIFENCEI, isainfo->isa);
 
+#ifdef CONFIG_RISCV_COMBO_SPINLOCKS
+		/*
+		 * The RISC-V Linux used queued spinlock at first; then, we used ticket_lock
+		 * as default or queued spinlock by choice. Because ticket_lock would dirty
+		 * spinlock value, the only way is to change from queued_spinlock to
+		 * ticket_spinlock, but can not be vice.
+		 */
+		set_bit(RISCV_ISA_EXT_XTICKETLOCK, isainfo->isa);
+#endif
+
 		/*
 		 * These ones were as they were part of the base ISA when the
 		 * port & dt-bindings were upstreamed, and so can be set