lkml.org 
[lkml]   [2022]   [Dec]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[RFC][PATCH 07/12] percpu: Wire up cmpxchg128
In order to replace cmpxchg_double() with the newly minted
cmpxchg128() family of functions, wire it up in this_cpu_cmpxchg().

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
arch/arm64/include/asm/percpu.h | 24 ++++++++++++++++++
arch/s390/include/asm/percpu.h | 20 +++++++++++++++
arch/x86/include/asm/percpu.h | 52 ++++++++++++++++++++++++++++++++++++++++
include/asm-generic/percpu.h | 8 ++++++
include/linux/percpu-defs.h | 20 +++++++++++++--
5 files changed, 122 insertions(+), 2 deletions(-)

--- a/arch/arm64/include/asm/percpu.h
+++ b/arch/arm64/include/asm/percpu.h
@@ -140,6 +140,10 @@ PERCPU_RET_OP(add, add, ldadd)
* re-enabling preemption for preemptible kernels, but doing that in a way
* which builds inside a module would mean messing directly with the preempt
* count. If you do this, peterz and tglx will hunt you down.
+ *
+ * Not to mention it'll break the actual preemption model for missing a
+ * preemption point when TIF_NEED_RESCHED gets set while preemption is
+ * disabled.
*/
#define this_cpu_cmpxchg_double_8(ptr1, ptr2, o1, o2, n1, n2) \
({ \
@@ -240,6 +244,26 @@ PERCPU_RET_OP(add, add, ldadd)
#define this_cpu_cmpxchg_8(pcp, o, n) \
_pcp_protect_return(cmpxchg_relaxed, pcp, o, n)

+#define __pcpu_cast_128(_exp, _val) \
+ _Generic((_exp), \
+ u128: (_val), \
+ s128: (_val), \
+ default: (unsigned long)(_val))
+
+#define this_cpu_cmpxchg_16(pcp, o, n) \
+({ \
+ u128 old__ = __pcpu_cast_128((o), (o)); \
+ u128 new__ = __pcpu_cast_128((n), (n)); \
+ typedef typeof(pcp) pcp_op_T__; \
+ pcp_op_T__ *ptr__; \
+ u128 ret__; \
+ preempt_disable_notrace(); \
+ ptr__ = raw_cpu_ptr(&(pcp)); \
+ ret__ = cmpxchg128_local((void *)ptr__, old__, new__); \
+ preempt_enable_notrace(); \
+ (typeof(pcp))__pcpu_cast_128(*ptr__, ret__); \
+})
+
#ifdef __KVM_NVHE_HYPERVISOR__
extern unsigned long __hyp_per_cpu_offset(unsigned int cpu);
#define __per_cpu_offset
--- a/arch/s390/include/asm/percpu.h
+++ b/arch/s390/include/asm/percpu.h
@@ -148,6 +148,26 @@
#define this_cpu_cmpxchg_4(pcp, oval, nval) arch_this_cpu_cmpxchg(pcp, oval, nval)
#define this_cpu_cmpxchg_8(pcp, oval, nval) arch_this_cpu_cmpxchg(pcp, oval, nval)

+#define __pcpu_cast_128(_exp, _val) \
+ _Generic((_exp), \
+ u128: (_val), \
+ s128: (_val), \
+ default: (unsigned long)(_val))
+
+#define this_cpu_cmpxchg_16(pcp, oval, nval) \
+({ \
+ u128 old__ = __pcpu_cast_128((nval), (nval)); \
+ u128 new__ = __pcpu_cast_128((oval), (oval)); \
+ typedef typeof(pcp) pcp_op_T__; \
+ pcp_op_T__ *ptr__; \
+ u128 ret__; \
+ preempt_disable_notrace(); \
+ ptr__ = raw_cpu_ptr(&(pcp)); \
+ ret__ = cmpxchg128((void *)ptr__, old__, new__); \
+ preempt_enable_notrace(); \
+ (typeof(pcp))__pcpu_cast_128(*ptr__, ret__); \
+})
+
#define arch_this_cpu_xchg(pcp, nval) \
({ \
typeof(pcp) *ptr__; \
--- a/arch/x86/include/asm/percpu.h
+++ b/arch/x86/include/asm/percpu.h
@@ -210,6 +210,58 @@ do { \
(typeof(_var))(unsigned long) pco_old__; \
})

+#if defined(CONFIG_X86_32) && defined(CONFIG_X86_CMPXCHG64)
+#define __pcpu_cast_64(_exp, _val) \
+ _Generic((_exp), \
+ u64: (_val), \
+ s64: (_val), \
+ default: (unsigned long)(_val))
+
+#define percpu_cmpxchg64_op(size, qual, _var, _oval, _nval) \
+({ \
+ __pcpu_type_##size pco_old__ = __pcpu_cast_64((_oval), (_oval));\
+ __pcpu_type_##size pco_new__ = __pcpu_cast_64((_nval), (_nval));\
+ asm qual ("cmpxchg8b " __percpu_arg([var]) \
+ : [var] "+m" (_var), \
+ "+A" (pco_old__) \
+ : "b" ((u32)pco_new__), "c" ((u32)(pco_new__ >> 32)) \
+ : "memory"); \
+ (typeof(_var))__pcpu_cast_64(_var, pco_old__); \
+})
+
+#define raw_cpu_cmpxchg_8(pcp, oval, nval) percpu_cmpxchg64_op(8, , pcp, oval, nval)
+#define this_cpu_cmpxchg_8(pcp, oval, nval) percpu_cmpxchg64_op(8, volatile, pcp, oval, nval)
+#endif
+
+#ifdef CONFIG_X86_64
+#define __pcpu_cast_128(_exp, _val) \
+ _Generic((_exp), \
+ u128: (_val), \
+ s128: (_val), \
+ default: (unsigned long)(_val))
+
+#define percpu_cmpxchg128_op(size, qual, _var, _oval, _nval) \
+({ \
+ union __u128_halves pco_old__ = { \
+ .full = __pcpu_cast_128((_oval), (_oval)) \
+ }; \
+ union __u128_halves pco_new__ = { \
+ .full = __pcpu_cast_128((_nval), (_nval)) \
+ }; \
+ asm qual ("cmpxchg16b " __percpu_arg([var]) \
+ : [var] "+m" (_var), \
+ "+a" (pco_old__.low), \
+ "+d" (pco_old__.high) \
+ : "b" (pco_new__.low), \
+ "c" (pco_new__.high) \
+ : "memory"); \
+ (typeof(_var))__pcpu_cast_128(_var, pco_old__.full); \
+})
+
+#define raw_cpu_cmpxchg_16(pcp, oval, nval) percpu_cmpxchg128_op(16, , pcp, oval, nval)
+#define this_cpu_cmpxchg_16(pcp, oval, nval) percpu_cmpxchg128_op(16, volatile, pcp, oval, nval)
+#endif
+
/*
* this_cpu_read() makes gcc load the percpu variable every time it is
* accessed while this_cpu_read_stable() allows the value to be cached.
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -298,6 +298,10 @@ do { \
#define raw_cpu_cmpxchg_8(pcp, oval, nval) \
raw_cpu_generic_cmpxchg(pcp, oval, nval)
#endif
+#ifndef raw_cpu_cmpxchg_16
+#define raw_cpu_cmpxchg_16(pcp, oval, nval) \
+ raw_cpu_generic_cmpxchg(pcp, oval, nval)
+#endif

#ifndef raw_cpu_cmpxchg_double_1
#define raw_cpu_cmpxchg_double_1(pcp1, pcp2, oval1, oval2, nval1, nval2) \
@@ -423,6 +427,10 @@ do { \
#define this_cpu_cmpxchg_8(pcp, oval, nval) \
this_cpu_generic_cmpxchg(pcp, oval, nval)
#endif
+#ifndef this_cpu_cmpxchg_16
+#define this_cpu_cmpxchg_16(pcp, oval, nval) \
+ this_cpu_generic_cmpxchg(pcp, oval, nval)
+#endif

#ifndef this_cpu_cmpxchg_double_1
#define this_cpu_cmpxchg_double_1(pcp1, pcp2, oval1, oval2, nval1, nval2) \
--- a/include/linux/percpu-defs.h
+++ b/include/linux/percpu-defs.h
@@ -343,6 +343,22 @@ static inline void __this_cpu_preempt_ch
pscr2_ret__; \
})

+#define __pcpu_size16_call_return2(stem, variable, ...) \
+({ \
+ typeof(variable) pscr2_ret__; \
+ __verify_pcpu_ptr(&(variable)); \
+ switch(sizeof(variable)) { \
+ case 1: pscr2_ret__ = stem##1(variable, __VA_ARGS__); break; \
+ case 2: pscr2_ret__ = stem##2(variable, __VA_ARGS__); break; \
+ case 4: pscr2_ret__ = stem##4(variable, __VA_ARGS__); break; \
+ case 8: pscr2_ret__ = stem##8(variable, __VA_ARGS__); break; \
+ case 16: pscr2_ret__ = stem##16(variable, __VA_ARGS__); break; \
+ default: \
+ __bad_size_call_parameter(); break; \
+ } \
+ pscr2_ret__; \
+})
+
/*
* Special handling for cmpxchg_double. cmpxchg_double is passed two
* percpu variables. The first has to be aligned to a double word
@@ -425,7 +441,7 @@ do { \
#define raw_cpu_add_return(pcp, val) __pcpu_size_call_return2(raw_cpu_add_return_, pcp, val)
#define raw_cpu_xchg(pcp, nval) __pcpu_size_call_return2(raw_cpu_xchg_, pcp, nval)
#define raw_cpu_cmpxchg(pcp, oval, nval) \
- __pcpu_size_call_return2(raw_cpu_cmpxchg_, pcp, oval, nval)
+ __pcpu_size16_call_return2(raw_cpu_cmpxchg_, pcp, oval, nval)
#define raw_cpu_cmpxchg_double(pcp1, pcp2, oval1, oval2, nval1, nval2) \
__pcpu_double_call_return_bool(raw_cpu_cmpxchg_double_, pcp1, pcp2, oval1, oval2, nval1, nval2)

@@ -512,7 +528,7 @@ do { \
#define this_cpu_add_return(pcp, val) __pcpu_size_call_return2(this_cpu_add_return_, pcp, val)
#define this_cpu_xchg(pcp, nval) __pcpu_size_call_return2(this_cpu_xchg_, pcp, nval)
#define this_cpu_cmpxchg(pcp, oval, nval) \
- __pcpu_size_call_return2(this_cpu_cmpxchg_, pcp, oval, nval)
+ __pcpu_size16_call_return2(this_cpu_cmpxchg_, pcp, oval, nval)
#define this_cpu_cmpxchg_double(pcp1, pcp2, oval1, oval2, nval1, nval2) \
__pcpu_double_call_return_bool(this_cpu_cmpxchg_double_, pcp1, pcp2, oval1, oval2, nval1, nval2)


\
 
 \ /
  Last update: 2022-12-19 16:47    [W:0.272 / U:0.224 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site