Skip to contentSkip to conversation

LoongArch: Add 128-bit atomic cmpxchg support (v3)

PATCH Superseded George Guo 3 participants 9 messages Nov 26, 2025 – Nov 27, 2025 LoongArch

v3 of 10 Superseded Latest is v10 →

lore
Share & tools
lore

George Guo

author
#0 lore raw

v3 cover LoongArch: Add 128-bit atomic cmpxchg support (v3)

This patch series adds 128-bit atomic compare-and-exchange support for LoongArch architecture, which fixes BPF scheduler test failures caused by missing 128-bit atomics support.

The series consists of two patches:

1. "LoongArch: Add 128-bit atomic cmpxchg support"
- Implements 128-bit atomic compare-and-exchange using LoongArch's LL.D/SC.Q instructions
- Fixes BPF scheduler test failures (scx_central scx_qmap) where kmalloc_nolock_noprof returns NULL due to missing 128-bit atomics, leading to -ENOMEM errors during scheduler initialization

2. "LoongArch: Enable 128-bit atomics cmpxchg support"
- Adds select HAVE_CMPXCHG_DOUBLE and select HAVE_ALIGNED_STRUCT_PAGE in Kconfig to enable 128-bit atomic cmpxchg support

The issue was identified through BPF scheduler test failures where scx_central and scx_qmap schedulers would fail to initialize. Testing was performed using the scx_qmap scheduler from tools/sched_ext/, confirming that the patches resolve the initialization failures.

---
Changes in v3:
- dbar 0 -> __WEAK_LLSC_MB
- =ZB" (__ptr[0]) -> "r" (__ptr)
- Link to v2: https://lore.kernel.org/r/[email protected]

Changes in v2:
- Use a normal ld.d for the high word instead of ll.d to avoid race condition
- Insert a dbar between ll.d and ld.d to prevent reordering
- Simply __cmpxchg128_asm("ll.d", "sc.q", ptr, o, n) to __cmpxchg128_asm(ptr, o, n)
- Fix address operand constraints after testing different approaches:
* ld.d with "m"
* ll.d with "ZC",
* sc.q with "ZB"(alternative constraints caused issues:
- "r" caused system hang
- "ZC" caused compiler error:
{standard input}: Assembler messages:
{standard input}:10037: Fatal error: Immediate overflow.
format: u0:0 )
- Link to v1: https://lore.kernel.org/r/[email protected]

---
George Guo (2):
LoongArch: Add 128-bit atomic cmpxchg support
LoongArch: Enable 128-bit atomics cmpxchg support

 arch/loongarch/Kconfig               |  2 ++
 arch/loongarch/include/asm/cmpxchg.h | 47 ++++++++++++++++++++++++++++++++++++
 2 files changed, 49 insertions(+)

---
base-commit: d5ae5ac32615e4af729f0610fdc11ff4f4798aef
change-id: 20251120-2-d03862b2cf6d

Best regards,

Signature · 1 line
George Guo <[email protected]>
author
#1 lore raw

PATCH v3 1/2 LoongArch: Add 128-bit atomic cmpxchg support

From: George Guo <[email protected]>

Implement 128-bit atomic compare-and-exchange using LoongArch's LL.D/SC.Q instructions.

At the same time, fix BPF scheduler test failures (scx_central scx_qmap) caused by kmalloc_nolock_noprof returning NULL due to missing 128-bit atomics. The NULL returns led to -ENOMEM errors during scheduler initialization, causing test cases to fail.

Verified by testing with the scx_qmap scheduler (located in tools/sched_ext/). Building with make and running ./tools/sched_ext/build/bin/scx_qmap.

---
 arch/loongarch/include/asm/cmpxchg.h | 47 ++++++++++++++++++++++++++++++++++++
 1 file changed, 47 insertions(+)
arch/loongarch/include/asm/cmpxchg.h +47 -0
index 979fde61bba8a42cb4f019f13ded2a3119d4aaf4..88553ae63e178ebff16ebbc2e01b0e23831b0918 100644
@@ -111,6 +111,44 @@ __arch_xchg(volatile void *ptr, unsigned long x, int size)
 	__ret;								\
 })
 
+union __u128_halves {
+	u128 full;
+	struct {
+		u64 low;
+		u64 high;
+	};
+};
+
+#define __cmpxchg128_asm(ptr, old, new)					\
+({									\
+	union __u128_halves __old, __new, __ret;			\
+	volatile u64 *__ptr = (volatile u64 *)(ptr);			\
+									\
+	__old.full = (old);                                             \
+	__new.full = (new);						\
+									\
+	__asm__ __volatile__(						\
+	"1:   ll.d    %0, %3		# 128-bit cmpxchg low	\n"	\
+	__WEAK_LLSC_MB							\
+	"     ld.d    %1, %4		# 128-bit cmpxchg high	\n"	\
+	"     bne     %0, %z5, 2f				\n"	\
+	"     bne     %1, %z6, 2f				\n"	\
+	"     move    $t0, %z7					\n"	\
+	"     move    $t1, %z8					\n"	\
+	"     sc.q    $t0, $t1, %2				\n"	\
+	"     beqz    $t0, 1b					\n"	\
+	"2:							\n"	\
+	__WEAK_LLSC_MB							\
+	: "=&r" (__ret.low), "=&r" (__ret.high)				\
+	: "r" (__ptr),							\
+	  "ZC" (__ptr[0]), "m" (__ptr[1]),				\
+	  "Jr" (__old.low), "Jr" (__old.high),				\
+	  "Jr" (__new.low), "Jr" (__new.high)				\
+	: "t0", "t1", "memory");					\
+									\
+	__ret.full;							\
+})
+
 static inline unsigned int __cmpxchg_small(volatile void *ptr, unsigned int old,
 					   unsigned int new, unsigned int size)
 {
@@ -198,6 +236,15 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, unsigned int
 	__res;								\
 })
 
+/* cmpxchg128 */
+#define system_has_cmpxchg128()		1
+
+#define arch_cmpxchg128(ptr, o, n)					\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 16);				\
+	__cmpxchg128_asm(ptr, o, n);			\
+})
+
 #ifdef CONFIG_64BIT
 #define arch_cmpxchg64_local(ptr, o, n)					\
   ({									\
Signature · 1 line
2.48.1
author
#2 lore raw

PATCH v3 2/2 LoongArch: Enable 128-bit atomics cmpxchg support

In reply to George Guo

From: George Guo <[email protected]>

Add select HAVE_CMPXCHG_DOUBLE and select HAVE_ALIGNED_STRUCT_PAGE in Kconfig to enable 128-bit atomic cmpxchg support on LoongArch.

---
 arch/loongarch/Kconfig | 2 ++
 1 file changed, 2 insertions(+)
arch/loongarch/Kconfig +2 -0
index 5b1116733d881bc2b1b43fb93f20367add4dbc54..6fb2c253969f9ddece5478920423d7326c3ec046 100644
@@ -114,6 +114,7 @@ config LOONGARCH
 	select GENERIC_TIME_VSYSCALL
 	select GPIOLIB
 	select HAS_IOPORT
+	select HAVE_ALIGNED_STRUCT_PAGE
 	select HAVE_ARCH_AUDITSYSCALL
 	select HAVE_ARCH_JUMP_LABEL
 	select HAVE_ARCH_JUMP_LABEL_RELATIVE
@@ -140,6 +141,7 @@ config LOONGARCH
 	select HAVE_DYNAMIC_FTRACE_WITH_DIRECT_CALLS
 	select HAVE_DYNAMIC_FTRACE_WITH_REGS
 	select HAVE_EBPF_JIT
+	select HAVE_CMPXCHG_DOUBLE
 	select HAVE_EFFICIENT_UNALIGNED_ACCESS if !ARCH_STRICT_ALIGN
 	select HAVE_EXIT_THREAD
 	select HAVE_GENERIC_TIF_BITS
Signature · 1 line
2.48.1
maintainer
#3 lore raw
In reply to George Guo

Hi, George,

Quoted 4 lines · George Guo

This patch series adds 128-bit atomic compare-and-exchange support for LoongArch architecture, which fixes BPF scheduler test failures caused by missing 128-bit atomics support.

Have you tested your code on Loongson-3A5000/3C5000?

Huacai

Quoted 58 lines

The series consists of two patches:

1. "LoongArch: Add 128-bit atomic cmpxchg support"
- Implements 128-bit atomic compare-and-exchange using LoongArch's LL.D/SC.Q instructions
- Fixes BPF scheduler test failures (scx_central scx_qmap) where kmalloc_nolock_noprof returns NULL due to missing 128-bit atomics, leading to -ENOMEM errors during scheduler initialization

2. "LoongArch: Enable 128-bit atomics cmpxchg support"
- Adds select HAVE_CMPXCHG_DOUBLE and select HAVE_ALIGNED_STRUCT_PAGE in Kconfig to enable 128-bit atomic cmpxchg support

The issue was identified through BPF scheduler test failures where scx_central and scx_qmap schedulers would fail to initialize. Testing was performed using the scx_qmap scheduler from tools/sched_ext/, confirming that the patches resolve the initialization failures.

---
Changes in v3:
- dbar 0 -> __WEAK_LLSC_MB
- =ZB" (__ptr[0]) -> "r" (__ptr)
- Link to v2: https://lore.kernel.org/r/[email protected]

Changes in v2:
- Use a normal ld.d for the high word instead of ll.d to avoid race condition
- Insert a dbar between ll.d and ld.d to prevent reordering
- Simply __cmpxchg128_asm("ll.d", "sc.q", ptr, o, n) to __cmpxchg128_asm(ptr, o, n)
- Fix address operand constraints after testing different approaches:
* ld.d with "m"
* ll.d with "ZC",
* sc.q with "ZB"(alternative constraints caused issues:
- "r" caused system hang
- "ZC" caused compiler error:
{standard input}: Assembler messages:
{standard input}:10037: Fatal error: Immediate overflow.
format: u0:0 )
- Link to v1: https://lore.kernel.org/r/[email protected]

---
George Guo (2):
LoongArch: Add 128-bit atomic cmpxchg support
LoongArch: Enable 128-bit atomics cmpxchg support

 arch/loongarch/Kconfig               |  2 ++
 arch/loongarch/include/asm/cmpxchg.h | 47 ++++++++++++++++++++++++++++++++++++
 2 files changed, 49 insertions(+)

---
base-commit: d5ae5ac32615e4af729f0610fdc11ff4f4798aef
change-id: 20251120-2-d03862b2cf6d

Best regards,
--
George Guo <[email protected]>

author
#4 lore raw
Quoted 5 lines · On Wed, 26 Nov 2025 12:44:44 +0800 Huacai Chen <[email protected]> wrote:

Hi, George,

Quoted 1 line · George Guo
Quoted 7 lines
Quoted 3 lines

This patch series adds 128-bit atomic compare-and-exchange support for LoongArch architecture, which fixes BPF scheduler test failures caused by missing 128-bit atomics support.

Have you tested your code on Loongson-3A5000/3C5000?

Huacai

Hi Huacai,

I have tested it on a virtual machine with fedora-42.

Quoted 64 lines
Quoted 64 lines

The series consists of two patches:

1. "LoongArch: Add 128-bit atomic cmpxchg support"
- Implements 128-bit atomic compare-and-exchange using
LoongArch's LL.D/SC.Q instructions
- Fixes BPF scheduler test failures (scx_central scx_qmap) where kmalloc_nolock_noprof returns NULL due to missing 128-bit
atomics, leading to -ENOMEM errors during scheduler initialization

2. "LoongArch: Enable 128-bit atomics cmpxchg support"
- Adds select HAVE_CMPXCHG_DOUBLE and select
HAVE_ALIGNED_STRUCT_PAGE in Kconfig to enable 128-bit atomic cmpxchg support

The issue was identified through BPF scheduler test failures where scx_central and scx_qmap schedulers would fail to initialize. Testing was performed using the scx_qmap scheduler from tools/sched_ext/, confirming that the patches resolve the initialization failures.

---
Changes in v3:
- dbar 0 -> __WEAK_LLSC_MB
- =ZB" (__ptr[0]) -> "r" (__ptr)
- Link to v2:
https://lore.kernel.org/r/[email protected]

Changes in v2:
- Use a normal ld.d for the high word instead of ll.d to avoid race condition
- Insert a dbar between ll.d and ld.d to prevent reordering
- Simply __cmpxchg128_asm("ll.d", "sc.q", ptr, o, n) to __cmpxchg128_asm(ptr, o, n)
- Fix address operand constraints after testing different approaches:
* ld.d with "m"
* ll.d with "ZC",
* sc.q with "ZB"(alternative constraints caused issues:
- "r" caused system hang
- "ZC" caused compiler error:
{standard input}: Assembler messages:
{standard input}:10037: Fatal error: Immediate overflow.
format: u0:0 )
- Link to v1:
https://lore.kernel.org/r/[email protected]

---
George Guo (2):
LoongArch: Add 128-bit atomic cmpxchg support
LoongArch: Enable 128-bit atomics cmpxchg support

 arch/loongarch/Kconfig               |  2 ++
 arch/loongarch/include/asm/cmpxchg.h | 47

++++++++++++++++++++++++++++++++++++ 2 files changed, 49 insertions(+) ---
base-commit: d5ae5ac32615e4af729f0610fdc11ff4f4798aef
change-id: 20251120-2-d03862b2cf6d

Best regards,
--
George Guo <[email protected]>

#5 lore raw
In reply to George Guo
Quoted 5 lines · George Guo

This patch series adds 128-bit atomic compare-and-exchange support for LoongArch architecture, which fixes BPF scheduler test failures caused by missing 128-bit atomics support.

Quoted 9 lines

The series consists of two patches:

1. "LoongArch: Add 128-bit atomic cmpxchg support"
- Implements 128-bit atomic compare-and-exchange using LoongArch's LL.D/SC.Q instructions
- Fixes BPF scheduler test failures (scx_central scx_qmap) where kmalloc_nolock_noprof returns NULL due to missing 128-bit atomics, leading to -ENOMEM errors during scheduler initialization

This kmalloc_nolock_noprof() was introduced in v6.18-rc1 and has no caller for now.
Why is this related to the sched_ext failure ?

Quoted 49 lines

2. "LoongArch: Enable 128-bit atomics cmpxchg support"
- Adds select HAVE_CMPXCHG_DOUBLE and select HAVE_ALIGNED_STRUCT_PAGE in Kconfig to enable 128-bit atomic cmpxchg support

The issue was identified through BPF scheduler test failures where scx_central and scx_qmap schedulers would fail to initialize. Testing was performed using the scx_qmap scheduler from tools/sched_ext/, confirming that the patches resolve the initialization failures.

---
Changes in v3:
- dbar 0 -> __WEAK_LLSC_MB
- =ZB" (__ptr[0]) -> "r" (__ptr)
- Link to v2: https://lore.kernel.org/r/[email protected]

Changes in v2:
- Use a normal ld.d for the high word instead of ll.d to avoid race condition
- Insert a dbar between ll.d and ld.d to prevent reordering
- Simply __cmpxchg128_asm("ll.d", "sc.q", ptr, o, n) to __cmpxchg128_asm(ptr, o, n)
- Fix address operand constraints after testing different approaches:
* ld.d with "m"
* ll.d with "ZC",
* sc.q with "ZB"(alternative constraints caused issues:
- "r" caused system hang
- "ZC" caused compiler error:
{standard input}: Assembler messages:
{standard input}:10037: Fatal error: Immediate overflow.
format: u0:0 )
- Link to v1: https://lore.kernel.org/r/[email protected]

---
George Guo (2):
LoongArch: Add 128-bit atomic cmpxchg support
LoongArch: Enable 128-bit atomics cmpxchg support

 arch/loongarch/Kconfig               |  2 ++
 arch/loongarch/include/asm/cmpxchg.h | 47 ++++++++++++++++++++++++++++++++++++
 2 files changed, 49 insertions(+)

---
base-commit: d5ae5ac32615e4af729f0610fdc11ff4f4798aef
change-id: 20251120-2-d03862b2cf6d

Best regards,
--
George Guo <[email protected]>

author
#6 lore raw
Quoted 17 lines · On Wed, 26 Nov 2025 13:23:57 +0800 Hengqi Chen <[email protected]> wrote:
Quoted 5 lines · George Guo

This patch series adds 128-bit atomic compare-and-exchange support for LoongArch architecture, which fixes BPF scheduler test failures caused by missing 128-bit atomics support.

Quoted 9 lines

The series consists of two patches:

1. "LoongArch: Add 128-bit atomic cmpxchg support"
- Implements 128-bit atomic compare-and-exchange using
LoongArch's LL.D/SC.Q instructions
- Fixes BPF scheduler test failures (scx_central scx_qmap) where kmalloc_nolock_noprof returns NULL due to missing 128-bit
atomics, leading to -ENOMEM errors during scheduler initialization

Quoted 4 lines

This kmalloc_nolock_noprof() was introduced in v6.18-rc1 and has no caller for now.
Why is this related to the sched_ext failure ?

Hi Hengqi,

When running scx_central, function call chain as below:
central_init->bpf_timer_init->__bpf_async_init->bpf_map_kmalloc_nolock->kmalloc_nolock ->kmalloc_nolock_noprof

The function kmalloc_nolock_noprof returns NULL due to the following condition:

if (!(s->flags & __CMPXCHG_DOUBLE) && !kmem_cache_debug(s))
/*
* kmalloc_nolock() is not supported on architectures that
* don't implement cmpxchg16b, but debug caches don't use
* per-cpu slab and per-cpu partial slabs. They rely on
* kmem_cache_node->list_lock, so kmalloc_nolock() can
* attempt to allocate from debug caches by
* spin_trylock_irqsave(&n->list_lock, ...)
*/
return NULL;

The NULL return occurs because kmalloc_nolock is not supported on Loongarch, which don't implement cmpxchg16b. So I am giving the patch.

Also I tried with debug caches(CONFIG_SLUB_DEBUG_ON=y), it works, but not a good idea.

Quoted 55 lines
Quoted 55 lines

2. "LoongArch: Enable 128-bit atomics cmpxchg support"
- Adds select HAVE_CMPXCHG_DOUBLE and select
HAVE_ALIGNED_STRUCT_PAGE in Kconfig to enable 128-bit atomic cmpxchg support

The issue was identified through BPF scheduler test failures where scx_central and scx_qmap schedulers would fail to initialize. Testing was performed using the scx_qmap scheduler from tools/sched_ext/, confirming that the patches resolve the initialization failures.

---
Changes in v3:
- dbar 0 -> __WEAK_LLSC_MB
- =ZB" (__ptr[0]) -> "r" (__ptr)
- Link to v2:
https://lore.kernel.org/r/[email protected]

Changes in v2:
- Use a normal ld.d for the high word instead of ll.d to avoid race condition
- Insert a dbar between ll.d and ld.d to prevent reordering
- Simply __cmpxchg128_asm("ll.d", "sc.q", ptr, o, n) to __cmpxchg128_asm(ptr, o, n)
- Fix address operand constraints after testing different approaches:
* ld.d with "m"
* ll.d with "ZC",
* sc.q with "ZB"(alternative constraints caused issues:
- "r" caused system hang
- "ZC" caused compiler error:
{standard input}: Assembler messages:
{standard input}:10037: Fatal error: Immediate overflow.
format: u0:0 )
- Link to v1:
https://lore.kernel.org/r/[email protected]

---
George Guo (2):
LoongArch: Add 128-bit atomic cmpxchg support
LoongArch: Enable 128-bit atomics cmpxchg support

 arch/loongarch/Kconfig               |  2 ++
 arch/loongarch/include/asm/cmpxchg.h | 47

++++++++++++++++++++++++++++++++++++ 2 files changed, 49 insertions(+) ---
base-commit: d5ae5ac32615e4af729f0610fdc11ff4f4798aef
change-id: 20251120-2-d03862b2cf6d

Best regards,
--
George Guo <[email protected]>

#7 lore raw
Quoted 21 lines · George Guo
Quoted 17 lines · On Wed, 26 Nov 2025 13:23:57 +0800 Hengqi Chen <[email protected]> wrote:
Quoted 14 lines · George Guo

This patch series adds 128-bit atomic compare-and-exchange support for LoongArch architecture, which fixes BPF scheduler test failures caused by missing 128-bit atomics support.

The series consists of two patches:

1. "LoongArch: Add 128-bit atomic cmpxchg support"
- Implements 128-bit atomic compare-and-exchange using
LoongArch's LL.D/SC.Q instructions
- Fixes BPF scheduler test failures (scx_central scx_qmap) where kmalloc_nolock_noprof returns NULL due to missing 128-bit
atomics, leading to -ENOMEM errors during scheduler initialization

Quoted 2 lines
Quoted 2 lines

This kmalloc_nolock_noprof() was introduced in v6.18-rc1 and has no caller for now.

OK, it does have a caller in rc-2 [1].

[1]: https://lore.kernel.org/bpf/[email protected]/

Quoted 8 lines
Quoted 2 lines

Why is this related to the sched_ext failure ?

Hi Hengqi,

When running scx_central, function call chain as below:
central_init->bpf_timer_init->__bpf_async_init->bpf_map_kmalloc_nolock->kmalloc_nolock ->kmalloc_nolock_noprof

Thanks, will test this series.

Quoted 76 lines

The function kmalloc_nolock_noprof returns NULL due to the following condition:

if (!(s->flags & __CMPXCHG_DOUBLE) && !kmem_cache_debug(s))
/*
* kmalloc_nolock() is not supported on architectures that
* don't implement cmpxchg16b, but debug caches don't use
* per-cpu slab and per-cpu partial slabs. They rely on
* kmem_cache_node->list_lock, so kmalloc_nolock() can
* attempt to allocate from debug caches by
* spin_trylock_irqsave(&n->list_lock, ...)
*/
return NULL;

The NULL return occurs because kmalloc_nolock is not supported on Loongarch, which don't implement cmpxchg16b. So I am giving the patch.

Also I tried with debug caches(CONFIG_SLUB_DEBUG_ON=y), it works, but not a good idea.

Quoted 55 lines
Quoted 55 lines

2. "LoongArch: Enable 128-bit atomics cmpxchg support"
- Adds select HAVE_CMPXCHG_DOUBLE and select
HAVE_ALIGNED_STRUCT_PAGE in Kconfig to enable 128-bit atomic cmpxchg support

The issue was identified through BPF scheduler test failures where scx_central and scx_qmap schedulers would fail to initialize. Testing was performed using the scx_qmap scheduler from tools/sched_ext/, confirming that the patches resolve the initialization failures.

---
Changes in v3:
- dbar 0 -> __WEAK_LLSC_MB
- =ZB" (__ptr[0]) -> "r" (__ptr)
- Link to v2:
https://lore.kernel.org/r/[email protected]

Changes in v2:
- Use a normal ld.d for the high word instead of ll.d to avoid race condition
- Insert a dbar between ll.d and ld.d to prevent reordering
- Simply __cmpxchg128_asm("ll.d", "sc.q", ptr, o, n) to __cmpxchg128_asm(ptr, o, n)
- Fix address operand constraints after testing different approaches:
* ld.d with "m"
* ll.d with "ZC",
* sc.q with "ZB"(alternative constraints caused issues:
- "r" caused system hang
- "ZC" caused compiler error:
{standard input}: Assembler messages:
{standard input}:10037: Fatal error: Immediate overflow.
format: u0:0 )
- Link to v1:
https://lore.kernel.org/r/[email protected]

---
George Guo (2):
LoongArch: Add 128-bit atomic cmpxchg support
LoongArch: Enable 128-bit atomics cmpxchg support

 arch/loongarch/Kconfig               |  2 ++
 arch/loongarch/include/asm/cmpxchg.h | 47

++++++++++++++++++++++++++++++++++++ 2 files changed, 49 insertions(+) ---
base-commit: d5ae5ac32615e4af729f0610fdc11ff4f4798aef
change-id: 20251120-2-d03862b2cf6d

Best regards,
--
George Guo <[email protected]>

#8 lore raw
Quoted 32 lines · Hengqi Chen
Quoted 21 lines · George Guo
Quoted 17 lines · On Wed, 26 Nov 2025 13:23:57 +0800 Hengqi Chen <[email protected]> wrote:
Quoted 14 lines · George Guo

This patch series adds 128-bit atomic compare-and-exchange support for LoongArch architecture, which fixes BPF scheduler test failures caused by missing 128-bit atomics support.

The series consists of two patches:

1. "LoongArch: Add 128-bit atomic cmpxchg support"
- Implements 128-bit atomic compare-and-exchange using
LoongArch's LL.D/SC.Q instructions
- Fixes BPF scheduler test failures (scx_central scx_qmap) where kmalloc_nolock_noprof returns NULL due to missing 128-bit
atomics, leading to -ENOMEM errors during scheduler initialization

Quoted 2 lines
Quoted 2 lines

This kmalloc_nolock_noprof() was introduced in v6.18-rc1 and has no caller for now.

OK, it does have a caller in rc-2 [1].

[1]: https://lore.kernel.org/bpf/[email protected]/

Quoted 2 lines
Quoted 2 lines

Why is this related to the sched_ext failure ?

Quoted 8 lines
Quoted 6 lines

Hi Hengqi,

When running scx_central, function call chain as below:
central_init->bpf_timer_init->__bpf_async_init->bpf_map_kmalloc_nolock->kmalloc_nolock ->kmalloc_nolock_noprof

Thanks, will test this series.

I tried with qemu, but it seems the kernel can't even boot.

Quoted 77 lines
Quoted 77 lines

The function kmalloc_nolock_noprof returns NULL due to the following condition:

if (!(s->flags & __CMPXCHG_DOUBLE) && !kmem_cache_debug(s))
/*
* kmalloc_nolock() is not supported on architectures that
* don't implement cmpxchg16b, but debug caches don't use
* per-cpu slab and per-cpu partial slabs. They rely on
* kmem_cache_node->list_lock, so kmalloc_nolock() can
* attempt to allocate from debug caches by
* spin_trylock_irqsave(&n->list_lock, ...)
*/
return NULL;

The NULL return occurs because kmalloc_nolock is not supported on Loongarch, which don't implement cmpxchg16b. So I am giving the patch.

Also I tried with debug caches(CONFIG_SLUB_DEBUG_ON=y), it works, but not a good idea.

Quoted 55 lines
Quoted 55 lines

2. "LoongArch: Enable 128-bit atomics cmpxchg support"
- Adds select HAVE_CMPXCHG_DOUBLE and select
HAVE_ALIGNED_STRUCT_PAGE in Kconfig to enable 128-bit atomic cmpxchg support

The issue was identified through BPF scheduler test failures where scx_central and scx_qmap schedulers would fail to initialize. Testing was performed using the scx_qmap scheduler from tools/sched_ext/, confirming that the patches resolve the initialization failures.

---
Changes in v3:
- dbar 0 -> __WEAK_LLSC_MB
- =ZB" (__ptr[0]) -> "r" (__ptr)
- Link to v2:
https://lore.kernel.org/r/[email protected]

Changes in v2:
- Use a normal ld.d for the high word instead of ll.d to avoid race condition
- Insert a dbar between ll.d and ld.d to prevent reordering
- Simply __cmpxchg128_asm("ll.d", "sc.q", ptr, o, n) to __cmpxchg128_asm(ptr, o, n)
- Fix address operand constraints after testing different approaches:
* ld.d with "m"
* ll.d with "ZC",
* sc.q with "ZB"(alternative constraints caused issues:
- "r" caused system hang
- "ZC" caused compiler error:
{standard input}: Assembler messages:
{standard input}:10037: Fatal error: Immediate overflow.
format: u0:0 )
- Link to v1:
https://lore.kernel.org/r/[email protected]

---
George Guo (2):
LoongArch: Add 128-bit atomic cmpxchg support
LoongArch: Enable 128-bit atomics cmpxchg support

 arch/loongarch/Kconfig               |  2 ++
 arch/loongarch/include/asm/cmpxchg.h | 47

++++++++++++++++++++++++++++++++++++ 2 files changed, 49 insertions(+) ---
base-commit: d5ae5ac32615e4af729f0610fdc11ff4f4798aef
change-id: 20251120-2-d03862b2cf6d

Best regards,
--
George Guo <[email protected]>