* [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode()
@ 2025-03-20 17:29 Charlie Jenkins
2025-03-20 17:29 ` [PATCH v6 1/4] riscv: entry: Convert ret_from_fork() to C Charlie Jenkins
` (4 more replies)
0 siblings, 5 replies; 14+ messages in thread
From: Charlie Jenkins @ 2025-03-20 17:29 UTC (permalink / raw)
To: Paul Walmsley, Palmer Dabbelt, Huacai Chen, WANG Xuerui,
Thomas Gleixner, Peter Zijlstra, Andy Lutomirski, Alexandre Ghiti,
Arnd Bergmann, Albert Ou, Alexandre Ghiti
Cc: linux-riscv, linux-kernel, loongarch, Charlie Jenkins
Similar to commit 221a164035fd ("entry: Move
syscall_enter_from_user_mode() to header file"), move
syscall_exit_to_user_mode() to the header file as well.
Testing was done with the byte-unixbench [1] syscall benchmark (which
calls getpid) and QEMU. On riscv I measured a 7.09246% improvement, on
x86 a 2.98843% improvement, on loongarch a 6.07954% improvement, and on
s390 a 11.1328% improvement.
The Intel bot also reported "kernel test robot noticed a 1.9%
improvement of stress-ng.seek.ops_per_sec" [2]
Since this is on QEMU, I know these numbers are not perfect, but they
show a trend of general improvement across all architectures that use
the generic entry code.
[1] https://github.com/kdlucas/byte-unixbench
[2] https://lore.kernel.org/linux-riscv/202502051555.85ae6844-lkp@intel.com/
Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
---
Changes in v6:
- Update commit message of patch 4 to contain performance numbers
- Link to v5: https://lore.kernel.org/r/20250305-riscv_optimize_entry-v5-0-6507b5dff3ce@rivosinc.com
Changes in v5:
- Rebase on 6.14-rc5
- Link to v4: https://lore.kernel.org/r/20250127-riscv_optimize_entry-v4-0-868cf7702dc9@rivosinc.com
Changes in v4:
- I had messed up warning for ct_state() on rebase, correct that issue
- Link to v3: https://lore.kernel.org/r/20250124-riscv_optimize_entry-v3-0-869f36b9e43b@rivosinc.com
Changes in v3:
- Fixup comment to properly reflect args (Alex)
- Fix prototypes for loongarch (Huacai)
- Link to v2: https://lore.kernel.org/r/20250123-riscv_optimize_entry-v2-0-7c259492d508@rivosinc.com
Changes in v2:
- Fixup compilation issues for loongarch
- Fixup compilation issues with CONFIG_CONTEXT_TRACKING_USER
- Link to v1: https://lore.kernel.org/r/20250122-riscv_optimize_entry-v1-0-4ee95559cfd0@rivosinc.com
---
Charlie Jenkins (4):
riscv: entry: Convert ret_from_fork() to C
riscv: entry: Split ret_from_fork() into user and kernel
LoongArch: entry: Migrate ret_from_fork() to C
entry: Inline syscall_exit_to_user_mode()
arch/loongarch/include/asm/asm-prototypes.h | 8 +++++
arch/loongarch/kernel/entry.S | 22 ++++++-------
arch/loongarch/kernel/process.c | 33 +++++++++++++++----
arch/riscv/include/asm/asm-prototypes.h | 2 ++
arch/riscv/kernel/entry.S | 20 +++++++-----
arch/riscv/kernel/process.c | 21 +++++++++++--
include/linux/entry-common.h | 43 +++++++++++++++++++++++--
kernel/entry/common.c | 49 +----------------------------
8 files changed, 119 insertions(+), 79 deletions(-)
---
base-commit: 7eb172143d5508b4da468ed59ee857c6e5e01da6
change-id: 20240402-riscv_optimize_entry-583843420325
--
- Charlie
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH v6 1/4] riscv: entry: Convert ret_from_fork() to C
2025-03-20 17:29 [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode() Charlie Jenkins
@ 2025-03-20 17:29 ` Charlie Jenkins
2025-04-29 6:33 ` [tip: core/entry] " tip-bot2 for Charlie Jenkins
2025-03-20 17:29 ` [PATCH v6 2/4] riscv: entry: Split ret_from_fork() into user and kernel Charlie Jenkins
` (3 subsequent siblings)
4 siblings, 1 reply; 14+ messages in thread
From: Charlie Jenkins @ 2025-03-20 17:29 UTC (permalink / raw)
To: Paul Walmsley, Palmer Dabbelt, Huacai Chen, WANG Xuerui,
Thomas Gleixner, Peter Zijlstra, Andy Lutomirski, Alexandre Ghiti,
Arnd Bergmann, Albert Ou, Alexandre Ghiti
Cc: linux-riscv, linux-kernel, loongarch, Charlie Jenkins
Move the main section of ret_from_fork() to C to allow inlining of
syscall_exit_to_user_mode().
Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com>
---
arch/riscv/include/asm/asm-prototypes.h | 1 +
arch/riscv/kernel/entry.S | 15 ++++++---------
arch/riscv/kernel/process.c | 14 ++++++++++++--
3 files changed, 19 insertions(+), 11 deletions(-)
diff --git a/arch/riscv/include/asm/asm-prototypes.h b/arch/riscv/include/asm/asm-prototypes.h
index cd627ec289f163a630b73dd03dd52a6b28692997..733ff609778797001006c33bba9e3cc5b1f15387 100644
--- a/arch/riscv/include/asm/asm-prototypes.h
+++ b/arch/riscv/include/asm/asm-prototypes.h
@@ -52,6 +52,7 @@ DECLARE_DO_ERROR_INFO(do_trap_ecall_s);
DECLARE_DO_ERROR_INFO(do_trap_ecall_m);
DECLARE_DO_ERROR_INFO(do_trap_break);
+asmlinkage void ret_from_fork(void *fn_arg, int (*fn)(void *), struct pt_regs *regs);
asmlinkage void handle_bad_stack(struct pt_regs *regs);
asmlinkage void do_page_fault(struct pt_regs *regs);
asmlinkage void do_irq(struct pt_regs *regs);
diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S
index 33a5a9f2a0d4e1eeccfb3621b9e518b88e1b0704..b2dc5e7c7b3a843fa4aa02eba2a911eb3ce31d1f 100644
--- a/arch/riscv/kernel/entry.S
+++ b/arch/riscv/kernel/entry.S
@@ -319,17 +319,14 @@ SYM_CODE_END(handle_kernel_stack_overflow)
ASM_NOKPROBE(handle_kernel_stack_overflow)
#endif
-SYM_CODE_START(ret_from_fork)
+SYM_CODE_START(ret_from_fork_asm)
call schedule_tail
- beqz s0, 1f /* not from kernel thread */
- /* Call fn(arg) */
- move a0, s1
- jalr s0
-1:
- move a0, sp /* pt_regs */
- call syscall_exit_to_user_mode
+ move a0, s1 /* fn_arg */
+ move a1, s0 /* fn */
+ move a2, sp /* pt_regs */
+ call ret_from_fork
j ret_from_exception
-SYM_CODE_END(ret_from_fork)
+SYM_CODE_END(ret_from_fork_asm)
#ifdef CONFIG_IRQ_STACKS
/*
diff --git a/arch/riscv/kernel/process.c b/arch/riscv/kernel/process.c
index 7c244de7718008947075357ea4502d56419d507c..7b0a0bfe29aec896c2bdd8976d855dd390de88d7 100644
--- a/arch/riscv/kernel/process.c
+++ b/arch/riscv/kernel/process.c
@@ -17,7 +17,9 @@
#include <linux/ptrace.h>
#include <linux/uaccess.h>
#include <linux/personality.h>
+#include <linux/entry-common.h>
+#include <asm/asm-prototypes.h>
#include <asm/unistd.h>
#include <asm/processor.h>
#include <asm/csr.h>
@@ -36,7 +38,7 @@ unsigned long __stack_chk_guard __read_mostly;
EXPORT_SYMBOL(__stack_chk_guard);
#endif
-extern asmlinkage void ret_from_fork(void);
+extern asmlinkage void ret_from_fork_asm(void);
void noinstr arch_cpu_idle(void)
{
@@ -206,6 +208,14 @@ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
return 0;
}
+asmlinkage void ret_from_fork(void *fn_arg, int (*fn)(void *), struct pt_regs *regs)
+{
+ if (unlikely(fn))
+ fn(fn_arg);
+
+ syscall_exit_to_user_mode(regs);
+}
+
int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
{
unsigned long clone_flags = args->flags;
@@ -242,7 +252,7 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
p->thread.riscv_v_flags = 0;
if (has_vector() || has_xtheadvector())
riscv_v_thread_alloc(p);
- p->thread.ra = (unsigned long)ret_from_fork;
+ p->thread.ra = (unsigned long)ret_from_fork_asm;
p->thread.sp = (unsigned long)childregs; /* kernel sp */
return 0;
}
--
2.43.0
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH v6 2/4] riscv: entry: Split ret_from_fork() into user and kernel
2025-03-20 17:29 [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode() Charlie Jenkins
2025-03-20 17:29 ` [PATCH v6 1/4] riscv: entry: Convert ret_from_fork() to C Charlie Jenkins
@ 2025-03-20 17:29 ` Charlie Jenkins
2025-03-21 6:36 ` Alexandre Ghiti
2025-04-29 6:33 ` [tip: core/entry] " tip-bot2 for Charlie Jenkins
2025-03-20 17:29 ` [PATCH v6 3/4] LoongArch: entry: Migrate ret_from_fork() to C Charlie Jenkins
` (2 subsequent siblings)
4 siblings, 2 replies; 14+ messages in thread
From: Charlie Jenkins @ 2025-03-20 17:29 UTC (permalink / raw)
To: Paul Walmsley, Palmer Dabbelt, Huacai Chen, WANG Xuerui,
Thomas Gleixner, Peter Zijlstra, Andy Lutomirski, Alexandre Ghiti,
Arnd Bergmann, Albert Ou, Alexandre Ghiti
Cc: linux-riscv, linux-kernel, loongarch, Charlie Jenkins
This function was unified into a single function in commit ab9164dae273
("riscv: entry: Consolidate ret_from_kernel_thread into ret_from_fork").
However that imposed a performance degradation. Partially reverting this
commit to have ret_from_fork() split again results in a 1% increase on
the number of times fork is able to be called per second.
Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
---
arch/riscv/include/asm/asm-prototypes.h | 3 ++-
arch/riscv/kernel/entry.S | 13 ++++++++++---
arch/riscv/kernel/process.c | 17 +++++++++++------
3 files changed, 23 insertions(+), 10 deletions(-)
diff --git a/arch/riscv/include/asm/asm-prototypes.h b/arch/riscv/include/asm/asm-prototypes.h
index 733ff609778797001006c33bba9e3cc5b1f15387..bfc8ea5f9319b19449ec59493b45b926df888832 100644
--- a/arch/riscv/include/asm/asm-prototypes.h
+++ b/arch/riscv/include/asm/asm-prototypes.h
@@ -52,7 +52,8 @@ DECLARE_DO_ERROR_INFO(do_trap_ecall_s);
DECLARE_DO_ERROR_INFO(do_trap_ecall_m);
DECLARE_DO_ERROR_INFO(do_trap_break);
-asmlinkage void ret_from_fork(void *fn_arg, int (*fn)(void *), struct pt_regs *regs);
+asmlinkage void ret_from_fork_kernel(void *fn_arg, int (*fn)(void *), struct pt_regs *regs);
+asmlinkage void ret_from_fork_user(struct pt_regs *regs);
asmlinkage void handle_bad_stack(struct pt_regs *regs);
asmlinkage void do_page_fault(struct pt_regs *regs);
asmlinkage void do_irq(struct pt_regs *regs);
diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S
index b2dc5e7c7b3a843fa4aa02eba2a911eb3ce31d1f..0fb338000c6dc0358742cd03497fa54b9e9d1aec 100644
--- a/arch/riscv/kernel/entry.S
+++ b/arch/riscv/kernel/entry.S
@@ -319,14 +319,21 @@ SYM_CODE_END(handle_kernel_stack_overflow)
ASM_NOKPROBE(handle_kernel_stack_overflow)
#endif
-SYM_CODE_START(ret_from_fork_asm)
+SYM_CODE_START(ret_from_fork_kernel_asm)
call schedule_tail
move a0, s1 /* fn_arg */
move a1, s0 /* fn */
move a2, sp /* pt_regs */
- call ret_from_fork
+ call ret_from_fork_kernel
j ret_from_exception
-SYM_CODE_END(ret_from_fork_asm)
+SYM_CODE_END(ret_from_fork_kernel_asm)
+
+SYM_CODE_START(ret_from_fork_user_asm)
+ call schedule_tail
+ move a0, sp /* pt_regs */
+ call ret_from_fork_user
+ j ret_from_exception
+SYM_CODE_END(ret_from_fork_user_asm)
#ifdef CONFIG_IRQ_STACKS
/*
diff --git a/arch/riscv/kernel/process.c b/arch/riscv/kernel/process.c
index 7b0a0bfe29aec896c2bdd8976d855dd390de88d7..485ec7a80a56097e8905cd6395af29633846b5c8 100644
--- a/arch/riscv/kernel/process.c
+++ b/arch/riscv/kernel/process.c
@@ -38,7 +38,8 @@ unsigned long __stack_chk_guard __read_mostly;
EXPORT_SYMBOL(__stack_chk_guard);
#endif
-extern asmlinkage void ret_from_fork_asm(void);
+extern asmlinkage void ret_from_fork_kernel_asm(void);
+extern asmlinkage void ret_from_fork_user_asm(void);
void noinstr arch_cpu_idle(void)
{
@@ -208,14 +209,18 @@ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
return 0;
}
-asmlinkage void ret_from_fork(void *fn_arg, int (*fn)(void *), struct pt_regs *regs)
+asmlinkage void ret_from_fork_kernel(void *fn_arg, int (*fn)(void *), struct pt_regs *regs)
{
- if (unlikely(fn))
- fn(fn_arg);
+ fn(fn_arg);
syscall_exit_to_user_mode(regs);
}
+asmlinkage void ret_from_fork_user(struct pt_regs *regs)
+{
+ syscall_exit_to_user_mode(regs);
+}
+
int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
{
unsigned long clone_flags = args->flags;
@@ -238,6 +243,7 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
p->thread.s[0] = (unsigned long)args->fn;
p->thread.s[1] = (unsigned long)args->fn_arg;
+ p->thread.ra = (unsigned long)ret_from_fork_kernel_asm;
} else {
*childregs = *(current_pt_regs());
/* Turn off status.VS */
@@ -247,12 +253,11 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
if (clone_flags & CLONE_SETTLS)
childregs->tp = tls;
childregs->a0 = 0; /* Return value of fork() */
- p->thread.s[0] = 0;
+ p->thread.ra = (unsigned long)ret_from_fork_user_asm;
}
p->thread.riscv_v_flags = 0;
if (has_vector() || has_xtheadvector())
riscv_v_thread_alloc(p);
- p->thread.ra = (unsigned long)ret_from_fork_asm;
p->thread.sp = (unsigned long)childregs; /* kernel sp */
return 0;
}
--
2.43.0
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH v6 3/4] LoongArch: entry: Migrate ret_from_fork() to C
2025-03-20 17:29 [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode() Charlie Jenkins
2025-03-20 17:29 ` [PATCH v6 1/4] riscv: entry: Convert ret_from_fork() to C Charlie Jenkins
2025-03-20 17:29 ` [PATCH v6 2/4] riscv: entry: Split ret_from_fork() into user and kernel Charlie Jenkins
@ 2025-03-20 17:29 ` Charlie Jenkins
2025-04-29 6:33 ` [tip: core/entry] " tip-bot2 for Charlie Jenkins
2025-05-05 4:58 ` [PATCH v6 3/4] " Huacai Chen
2025-03-20 17:29 ` [PATCH v6 4/4] entry: Inline syscall_exit_to_user_mode() Charlie Jenkins
2025-03-21 21:22 ` [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode() Thomas Gleixner
4 siblings, 2 replies; 14+ messages in thread
From: Charlie Jenkins @ 2025-03-20 17:29 UTC (permalink / raw)
To: Paul Walmsley, Palmer Dabbelt, Huacai Chen, WANG Xuerui,
Thomas Gleixner, Peter Zijlstra, Andy Lutomirski, Alexandre Ghiti,
Arnd Bergmann, Albert Ou, Alexandre Ghiti
Cc: linux-riscv, linux-kernel, loongarch, Charlie Jenkins
LoongArch is the only architecture that calls
syscall_exit_to_user_mode() from asm. Move the call into C so that this
function can be inlined across all architectures.
Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
---
arch/loongarch/include/asm/asm-prototypes.h | 8 +++++++
arch/loongarch/kernel/entry.S | 22 +++++++++----------
arch/loongarch/kernel/process.c | 33 +++++++++++++++++++++++------
3 files changed, 45 insertions(+), 18 deletions(-)
diff --git a/arch/loongarch/include/asm/asm-prototypes.h b/arch/loongarch/include/asm/asm-prototypes.h
index 51f224bcfc654228ae423e9a066b25b35102a5b9..704066b4f7368be15be960fadbcd6c2574bbf6c0 100644
--- a/arch/loongarch/include/asm/asm-prototypes.h
+++ b/arch/loongarch/include/asm/asm-prototypes.h
@@ -12,3 +12,11 @@ __int128_t __ashlti3(__int128_t a, int b);
__int128_t __ashrti3(__int128_t a, int b);
__int128_t __lshrti3(__int128_t a, int b);
#endif
+
+asmlinkage void noinstr __no_stack_protector ret_from_fork(struct task_struct *prev,
+ struct pt_regs *regs);
+
+asmlinkage void noinstr __no_stack_protector ret_from_kernel_thread(struct task_struct *prev,
+ struct pt_regs *regs,
+ int (*fn)(void *),
+ void *fn_arg);
diff --git a/arch/loongarch/kernel/entry.S b/arch/loongarch/kernel/entry.S
index 48e7e34e355e83eae8165957ba2eac05a8bf17df..2abc29e573810e000f2fef4646ddca0dbb80eabe 100644
--- a/arch/loongarch/kernel/entry.S
+++ b/arch/loongarch/kernel/entry.S
@@ -77,24 +77,22 @@ SYM_CODE_START(handle_syscall)
SYM_CODE_END(handle_syscall)
_ASM_NOKPROBE(handle_syscall)
-SYM_CODE_START(ret_from_fork)
+SYM_CODE_START(ret_from_fork_asm)
UNWIND_HINT_REGS
- bl schedule_tail # a0 = struct task_struct *prev
- move a0, sp
- bl syscall_exit_to_user_mode
+ move a1, sp
+ bl ret_from_fork
RESTORE_STATIC
RESTORE_SOME
RESTORE_SP_AND_RET
-SYM_CODE_END(ret_from_fork)
+SYM_CODE_END(ret_from_fork_asm)
-SYM_CODE_START(ret_from_kernel_thread)
+SYM_CODE_START(ret_from_kernel_thread_asm)
UNWIND_HINT_REGS
- bl schedule_tail # a0 = struct task_struct *prev
- move a0, s1
- jirl ra, s0, 0
- move a0, sp
- bl syscall_exit_to_user_mode
+ move a1, sp
+ move a2, s0
+ move a3, s1
+ bl ret_from_kernel_thread
RESTORE_STATIC
RESTORE_SOME
RESTORE_SP_AND_RET
-SYM_CODE_END(ret_from_kernel_thread)
+SYM_CODE_END(ret_from_kernel_thread_asm)
diff --git a/arch/loongarch/kernel/process.c b/arch/loongarch/kernel/process.c
index 6e58f65455c7ca3eae2e88ed852c8655a6701e5c..98bc60d7c550fcc0225e8452f81a7d6cd7888015 100644
--- a/arch/loongarch/kernel/process.c
+++ b/arch/loongarch/kernel/process.c
@@ -14,6 +14,7 @@
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/errno.h>
+#include <linux/entry-common.h>
#include <linux/sched.h>
#include <linux/sched/debug.h>
#include <linux/sched/task.h>
@@ -33,6 +34,7 @@
#include <linux/prctl.h>
#include <linux/nmi.h>
+#include <asm/asm-prototypes.h>
#include <asm/asm.h>
#include <asm/bootinfo.h>
#include <asm/cpu.h>
@@ -47,6 +49,7 @@
#include <asm/pgtable.h>
#include <asm/processor.h>
#include <asm/reg.h>
+#include <asm/switch_to.h>
#include <asm/unwind.h>
#include <asm/vdso.h>
@@ -63,8 +66,9 @@ EXPORT_SYMBOL(__stack_chk_guard);
unsigned long boot_option_idle_override = IDLE_NO_OVERRIDE;
EXPORT_SYMBOL(boot_option_idle_override);
-asmlinkage void ret_from_fork(void);
-asmlinkage void ret_from_kernel_thread(void);
+asmlinkage void restore_and_ret(void);
+asmlinkage void ret_from_fork_asm(void);
+asmlinkage void ret_from_kernel_thread_asm(void);
void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long sp)
{
@@ -138,6 +142,23 @@ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
return 0;
}
+asmlinkage void noinstr __no_stack_protector ret_from_fork(struct task_struct *prev,
+ struct pt_regs *regs)
+{
+ schedule_tail(prev);
+ syscall_exit_to_user_mode(regs);
+}
+
+asmlinkage void noinstr __no_stack_protector ret_from_kernel_thread(struct task_struct *prev,
+ struct pt_regs *regs,
+ int (*fn)(void *),
+ void *fn_arg)
+{
+ schedule_tail(prev);
+ fn(fn_arg);
+ syscall_exit_to_user_mode(regs);
+}
+
/*
* Copy architecture-specific thread state
*/
@@ -165,8 +186,8 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
p->thread.reg03 = childksp;
p->thread.reg23 = (unsigned long)args->fn;
p->thread.reg24 = (unsigned long)args->fn_arg;
- p->thread.reg01 = (unsigned long)ret_from_kernel_thread;
- p->thread.sched_ra = (unsigned long)ret_from_kernel_thread;
+ p->thread.reg01 = (unsigned long)ret_from_kernel_thread_asm;
+ p->thread.sched_ra = (unsigned long)ret_from_kernel_thread_asm;
memset(childregs, 0, sizeof(struct pt_regs));
childregs->csr_euen = p->thread.csr_euen;
childregs->csr_crmd = p->thread.csr_crmd;
@@ -182,8 +203,8 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
childregs->regs[3] = usp;
p->thread.reg03 = (unsigned long) childregs;
- p->thread.reg01 = (unsigned long) ret_from_fork;
- p->thread.sched_ra = (unsigned long) ret_from_fork;
+ p->thread.reg01 = (unsigned long) ret_from_fork_asm;
+ p->thread.sched_ra = (unsigned long) ret_from_fork_asm;
/*
* New tasks lose permission to use the fpu. This accelerates context
--
2.43.0
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH v6 4/4] entry: Inline syscall_exit_to_user_mode()
2025-03-20 17:29 [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode() Charlie Jenkins
` (2 preceding siblings ...)
2025-03-20 17:29 ` [PATCH v6 3/4] LoongArch: entry: Migrate ret_from_fork() to C Charlie Jenkins
@ 2025-03-20 17:29 ` Charlie Jenkins
2025-04-29 6:33 ` [tip: core/entry] " tip-bot2 for Charlie Jenkins
2025-03-21 21:22 ` [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode() Thomas Gleixner
4 siblings, 1 reply; 14+ messages in thread
From: Charlie Jenkins @ 2025-03-20 17:29 UTC (permalink / raw)
To: Paul Walmsley, Palmer Dabbelt, Huacai Chen, WANG Xuerui,
Thomas Gleixner, Peter Zijlstra, Andy Lutomirski, Alexandre Ghiti,
Arnd Bergmann, Albert Ou, Alexandre Ghiti
Cc: linux-riscv, linux-kernel, loongarch, Charlie Jenkins
Similar to commit 221a164035fd ("entry: Move
syscall_enter_from_user_mode() to header file"), move
syscall_exit_to_user_mode() to the header file as well.
Testing was done with the byte-unixbench [1] syscall benchmark (which
calls getpid) and QEMU. On riscv I measured a 7.09246% improvement, on
x86 a 2.98843% improvement, on loongarch a 6.07954% improvement, and on
s390 a 11.1328% improvement.
The Intel bot also reported "kernel test robot noticed a 1.9%
improvement of stress-ng.seek.ops_per_sec" [2]
[1] https://github.com/kdlucas/byte-unixbench
[2] https://lore.kernel.org/linux-riscv/202502051555.85ae6844-lkp@intel.com/
Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com>
---
include/linux/entry-common.h | 43 ++++++++++++++++++++++++++++++++++++--
kernel/entry/common.c | 49 +-------------------------------------------
2 files changed, 42 insertions(+), 50 deletions(-)
diff --git a/include/linux/entry-common.h b/include/linux/entry-common.h
index fc61d0205c97084acc89c8e45e088946f5e6d9b2..f94f3fdf15fc0091223cc9f7b823970302e67312 100644
--- a/include/linux/entry-common.h
+++ b/include/linux/entry-common.h
@@ -14,6 +14,7 @@
#include <linux/kmsan.h>
#include <asm/entry-common.h>
+#include <asm/syscall.h>
/*
* Define dummy _TIF work flags if not defined by the architecture or for
@@ -366,6 +367,15 @@ static __always_inline void exit_to_user_mode(void)
lockdep_hardirqs_on(CALLER_ADDR0);
}
+/**
+ * syscall_exit_work - Handle work before returning to user mode
+ * @regs: Pointer to current pt_regs
+ * @work: Current thread syscall work
+ *
+ * Do one-time syscall specific work.
+ */
+void syscall_exit_work(struct pt_regs *regs, unsigned long work);
+
/**
* syscall_exit_to_user_mode_work - Handle work before returning to user mode
* @regs: Pointer to currents pt_regs
@@ -379,7 +389,30 @@ static __always_inline void exit_to_user_mode(void)
* make the final state transitions. Interrupts must stay disabled between
* return from this function and the invocation of exit_to_user_mode().
*/
-void syscall_exit_to_user_mode_work(struct pt_regs *regs);
+static __always_inline void syscall_exit_to_user_mode_work(struct pt_regs *regs)
+{
+ unsigned long work = READ_ONCE(current_thread_info()->syscall_work);
+ unsigned long nr = syscall_get_nr(current, regs);
+
+ CT_WARN_ON(ct_state() != CT_STATE_KERNEL);
+
+ if (IS_ENABLED(CONFIG_PROVE_LOCKING)) {
+ if (WARN(irqs_disabled(), "syscall %lu left IRQs disabled", nr))
+ local_irq_enable();
+ }
+
+ rseq_syscall(regs);
+
+ /*
+ * Do one-time syscall specific work. If these work items are
+ * enabled, we want to run them exactly once per syscall exit with
+ * interrupts enabled.
+ */
+ if (unlikely(work & SYSCALL_WORK_EXIT))
+ syscall_exit_work(regs, work);
+ local_irq_disable_exit_to_user();
+ exit_to_user_mode_prepare(regs);
+}
/**
* syscall_exit_to_user_mode - Handle work before returning to user mode
@@ -410,7 +443,13 @@ void syscall_exit_to_user_mode_work(struct pt_regs *regs);
* exit_to_user_mode(). This function is preferred unless there is a
* compelling architectural reason to use the separate functions.
*/
-void syscall_exit_to_user_mode(struct pt_regs *regs);
+static __always_inline void syscall_exit_to_user_mode(struct pt_regs *regs)
+{
+ instrumentation_begin();
+ syscall_exit_to_user_mode_work(regs);
+ instrumentation_end();
+ exit_to_user_mode();
+}
/**
* irqentry_enter_from_user_mode - Establish state before invoking the irq handler
diff --git a/kernel/entry/common.c b/kernel/entry/common.c
index e33691d5adf7aab4af54cf2bf8e5ef5bd6ad1424..f55e421fb196dd5f9d4e34dd85ae096c774cf879 100644
--- a/kernel/entry/common.c
+++ b/kernel/entry/common.c
@@ -146,7 +146,7 @@ static inline bool report_single_step(unsigned long work)
return work & SYSCALL_WORK_SYSCALL_EXIT_TRAP;
}
-static void syscall_exit_work(struct pt_regs *regs, unsigned long work)
+void syscall_exit_work(struct pt_regs *regs, unsigned long work)
{
bool step;
@@ -173,53 +173,6 @@ static void syscall_exit_work(struct pt_regs *regs, unsigned long work)
ptrace_report_syscall_exit(regs, step);
}
-/*
- * Syscall specific exit to user mode preparation. Runs with interrupts
- * enabled.
- */
-static void syscall_exit_to_user_mode_prepare(struct pt_regs *regs)
-{
- unsigned long work = READ_ONCE(current_thread_info()->syscall_work);
- unsigned long nr = syscall_get_nr(current, regs);
-
- CT_WARN_ON(ct_state() != CT_STATE_KERNEL);
-
- if (IS_ENABLED(CONFIG_PROVE_LOCKING)) {
- if (WARN(irqs_disabled(), "syscall %lu left IRQs disabled", nr))
- local_irq_enable();
- }
-
- rseq_syscall(regs);
-
- /*
- * Do one-time syscall specific work. If these work items are
- * enabled, we want to run them exactly once per syscall exit with
- * interrupts enabled.
- */
- if (unlikely(work & SYSCALL_WORK_EXIT))
- syscall_exit_work(regs, work);
-}
-
-static __always_inline void __syscall_exit_to_user_mode_work(struct pt_regs *regs)
-{
- syscall_exit_to_user_mode_prepare(regs);
- local_irq_disable_exit_to_user();
- exit_to_user_mode_prepare(regs);
-}
-
-void syscall_exit_to_user_mode_work(struct pt_regs *regs)
-{
- __syscall_exit_to_user_mode_work(regs);
-}
-
-__visible noinstr void syscall_exit_to_user_mode(struct pt_regs *regs)
-{
- instrumentation_begin();
- __syscall_exit_to_user_mode_work(regs);
- instrumentation_end();
- exit_to_user_mode();
-}
-
noinstr void irqentry_enter_from_user_mode(struct pt_regs *regs)
{
enter_from_user_mode(regs);
--
2.43.0
^ permalink raw reply related [flat|nested] 14+ messages in thread
* Re: [PATCH v6 2/4] riscv: entry: Split ret_from_fork() into user and kernel
2025-03-20 17:29 ` [PATCH v6 2/4] riscv: entry: Split ret_from_fork() into user and kernel Charlie Jenkins
@ 2025-03-21 6:36 ` Alexandre Ghiti
2025-04-29 6:33 ` [tip: core/entry] " tip-bot2 for Charlie Jenkins
1 sibling, 0 replies; 14+ messages in thread
From: Alexandre Ghiti @ 2025-03-21 6:36 UTC (permalink / raw)
To: Charlie Jenkins, Paul Walmsley, Palmer Dabbelt, Huacai Chen,
WANG Xuerui, Thomas Gleixner, Peter Zijlstra, Andy Lutomirski,
Alexandre Ghiti, Arnd Bergmann, Albert Ou
Cc: linux-riscv, linux-kernel, loongarch
Hi Charlie,
On 20/03/2025 18:29, Charlie Jenkins wrote:
> This function was unified into a single function in commit ab9164dae273
> ("riscv: entry: Consolidate ret_from_kernel_thread into ret_from_fork").
> However that imposed a performance degradation. Partially reverting this
> commit to have ret_from_fork() split again results in a 1% increase on
> the number of times fork is able to be called per second.
>
> Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
> ---
> arch/riscv/include/asm/asm-prototypes.h | 3 ++-
> arch/riscv/kernel/entry.S | 13 ++++++++++---
> arch/riscv/kernel/process.c | 17 +++++++++++------
> 3 files changed, 23 insertions(+), 10 deletions(-)
>
> diff --git a/arch/riscv/include/asm/asm-prototypes.h b/arch/riscv/include/asm/asm-prototypes.h
> index 733ff609778797001006c33bba9e3cc5b1f15387..bfc8ea5f9319b19449ec59493b45b926df888832 100644
> --- a/arch/riscv/include/asm/asm-prototypes.h
> +++ b/arch/riscv/include/asm/asm-prototypes.h
> @@ -52,7 +52,8 @@ DECLARE_DO_ERROR_INFO(do_trap_ecall_s);
> DECLARE_DO_ERROR_INFO(do_trap_ecall_m);
> DECLARE_DO_ERROR_INFO(do_trap_break);
>
> -asmlinkage void ret_from_fork(void *fn_arg, int (*fn)(void *), struct pt_regs *regs);
> +asmlinkage void ret_from_fork_kernel(void *fn_arg, int (*fn)(void *), struct pt_regs *regs);
> +asmlinkage void ret_from_fork_user(struct pt_regs *regs);
> asmlinkage void handle_bad_stack(struct pt_regs *regs);
> asmlinkage void do_page_fault(struct pt_regs *regs);
> asmlinkage void do_irq(struct pt_regs *regs);
> diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S
> index b2dc5e7c7b3a843fa4aa02eba2a911eb3ce31d1f..0fb338000c6dc0358742cd03497fa54b9e9d1aec 100644
> --- a/arch/riscv/kernel/entry.S
> +++ b/arch/riscv/kernel/entry.S
> @@ -319,14 +319,21 @@ SYM_CODE_END(handle_kernel_stack_overflow)
> ASM_NOKPROBE(handle_kernel_stack_overflow)
> #endif
>
> -SYM_CODE_START(ret_from_fork_asm)
> +SYM_CODE_START(ret_from_fork_kernel_asm)
> call schedule_tail
> move a0, s1 /* fn_arg */
> move a1, s0 /* fn */
> move a2, sp /* pt_regs */
> - call ret_from_fork
> + call ret_from_fork_kernel
> j ret_from_exception
> -SYM_CODE_END(ret_from_fork_asm)
> +SYM_CODE_END(ret_from_fork_kernel_asm)
> +
> +SYM_CODE_START(ret_from_fork_user_asm)
> + call schedule_tail
> + move a0, sp /* pt_regs */
> + call ret_from_fork_user
> + j ret_from_exception
> +SYM_CODE_END(ret_from_fork_user_asm)
>
> #ifdef CONFIG_IRQ_STACKS
> /*
> diff --git a/arch/riscv/kernel/process.c b/arch/riscv/kernel/process.c
> index 7b0a0bfe29aec896c2bdd8976d855dd390de88d7..485ec7a80a56097e8905cd6395af29633846b5c8 100644
> --- a/arch/riscv/kernel/process.c
> +++ b/arch/riscv/kernel/process.c
> @@ -38,7 +38,8 @@ unsigned long __stack_chk_guard __read_mostly;
> EXPORT_SYMBOL(__stack_chk_guard);
> #endif
>
> -extern asmlinkage void ret_from_fork_asm(void);
> +extern asmlinkage void ret_from_fork_kernel_asm(void);
> +extern asmlinkage void ret_from_fork_user_asm(void);
>
> void noinstr arch_cpu_idle(void)
> {
> @@ -208,14 +209,18 @@ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
> return 0;
> }
>
> -asmlinkage void ret_from_fork(void *fn_arg, int (*fn)(void *), struct pt_regs *regs)
> +asmlinkage void ret_from_fork_kernel(void *fn_arg, int (*fn)(void *), struct pt_regs *regs)
> {
> - if (unlikely(fn))
> - fn(fn_arg);
> + fn(fn_arg);
>
> syscall_exit_to_user_mode(regs);
> }
>
> +asmlinkage void ret_from_fork_user(struct pt_regs *regs)
> +{
> + syscall_exit_to_user_mode(regs);
> +}
> +
> int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
> {
> unsigned long clone_flags = args->flags;
> @@ -238,6 +243,7 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
>
> p->thread.s[0] = (unsigned long)args->fn;
> p->thread.s[1] = (unsigned long)args->fn_arg;
> + p->thread.ra = (unsigned long)ret_from_fork_kernel_asm;
> } else {
> *childregs = *(current_pt_regs());
> /* Turn off status.VS */
> @@ -247,12 +253,11 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
> if (clone_flags & CLONE_SETTLS)
> childregs->tp = tls;
> childregs->a0 = 0; /* Return value of fork() */
> - p->thread.s[0] = 0;
> + p->thread.ra = (unsigned long)ret_from_fork_user_asm;
> }
> p->thread.riscv_v_flags = 0;
> if (has_vector() || has_xtheadvector())
> riscv_v_thread_alloc(p);
> - p->thread.ra = (unsigned long)ret_from_fork_asm;
> p->thread.sp = (unsigned long)childregs; /* kernel sp */
> return 0;
> }
>
Acked-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Thanks,
Alex
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode()
2025-03-20 17:29 [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode() Charlie Jenkins
` (3 preceding siblings ...)
2025-03-20 17:29 ` [PATCH v6 4/4] entry: Inline syscall_exit_to_user_mode() Charlie Jenkins
@ 2025-03-21 21:22 ` Thomas Gleixner
2025-04-28 21:54 ` Charlie Jenkins
4 siblings, 1 reply; 14+ messages in thread
From: Thomas Gleixner @ 2025-03-21 21:22 UTC (permalink / raw)
To: Charlie Jenkins, Paul Walmsley, Palmer Dabbelt, Huacai Chen,
WANG Xuerui, Peter Zijlstra, Andy Lutomirski, Alexandre Ghiti,
Arnd Bergmann, Albert Ou, Alexandre Ghiti
Cc: linux-riscv, linux-kernel, loongarch, Charlie Jenkins
On Thu, Mar 20 2025 at 10:29, Charlie Jenkins wrote:
> Similar to commit 221a164035fd ("entry: Move
> syscall_enter_from_user_mode() to header file"), move
> syscall_exit_to_user_mode() to the header file as well.
>
> Testing was done with the byte-unixbench [1] syscall benchmark (which
> calls getpid) and QEMU. On riscv I measured a 7.09246% improvement, on
> x86 a 2.98843% improvement, on loongarch a 6.07954% improvement, and on
> s390 a 11.1328% improvement.
>
> The Intel bot also reported "kernel test robot noticed a 1.9%
> improvement of stress-ng.seek.ops_per_sec" [2]
>
> Since this is on QEMU, I know these numbers are not perfect, but they
> show a trend of general improvement across all architectures that use
> the generic entry code.
This looks sane now. I've bookmarked it as post-merge window material.
Thanks,
tglx
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode()
2025-03-21 21:22 ` [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode() Thomas Gleixner
@ 2025-04-28 21:54 ` Charlie Jenkins
2025-04-29 6:22 ` Thomas Gleixner
0 siblings, 1 reply; 14+ messages in thread
From: Charlie Jenkins @ 2025-04-28 21:54 UTC (permalink / raw)
To: Thomas Gleixner
Cc: Paul Walmsley, Palmer Dabbelt, Huacai Chen, WANG Xuerui,
Peter Zijlstra, Andy Lutomirski, Alexandre Ghiti, Arnd Bergmann,
Albert Ou, Alexandre Ghiti, linux-riscv, linux-kernel, loongarch
On Fri, Mar 21, 2025 at 10:22:36PM +0100, Thomas Gleixner wrote:
> On Thu, Mar 20 2025 at 10:29, Charlie Jenkins wrote:
> > Similar to commit 221a164035fd ("entry: Move
> > syscall_enter_from_user_mode() to header file"), move
> > syscall_exit_to_user_mode() to the header file as well.
> >
> > Testing was done with the byte-unixbench [1] syscall benchmark (which
> > calls getpid) and QEMU. On riscv I measured a 7.09246% improvement, on
> > x86 a 2.98843% improvement, on loongarch a 6.07954% improvement, and on
> > s390 a 11.1328% improvement.
> >
> > The Intel bot also reported "kernel test robot noticed a 1.9%
> > improvement of stress-ng.seek.ops_per_sec" [2]
> >
> > Since this is on QEMU, I know these numbers are not perfect, but they
> > show a trend of general improvement across all architectures that use
> > the generic entry code.
>
> This looks sane now. I've bookmarked it as post-merge window material.
Has this been applied somewhere?
- Charlie
>
> Thanks,
>
> tglx
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode()
2025-04-28 21:54 ` Charlie Jenkins
@ 2025-04-29 6:22 ` Thomas Gleixner
0 siblings, 0 replies; 14+ messages in thread
From: Thomas Gleixner @ 2025-04-29 6:22 UTC (permalink / raw)
To: Charlie Jenkins
Cc: Paul Walmsley, Palmer Dabbelt, Huacai Chen, WANG Xuerui,
Peter Zijlstra, Andy Lutomirski, Alexandre Ghiti, Arnd Bergmann,
Albert Ou, Alexandre Ghiti, linux-riscv, linux-kernel, loongarch
On Mon, Apr 28 2025 at 14:54, Charlie Jenkins wrote:
> On Fri, Mar 21, 2025 at 10:22:36PM +0100, Thomas Gleixner wrote:
>>
>> This looks sane now. I've bookmarked it as post-merge window material.
>
> Has this been applied somewhere?
Thanks for the nudge. It indeed fell through the cracks. Tending to it
now.
Thanks,
tglx
^ permalink raw reply [flat|nested] 14+ messages in thread
* [tip: core/entry] entry: Inline syscall_exit_to_user_mode()
2025-03-20 17:29 ` [PATCH v6 4/4] entry: Inline syscall_exit_to_user_mode() Charlie Jenkins
@ 2025-04-29 6:33 ` tip-bot2 for Charlie Jenkins
0 siblings, 0 replies; 14+ messages in thread
From: tip-bot2 for Charlie Jenkins @ 2025-04-29 6:33 UTC (permalink / raw)
To: linux-tip-commits
Cc: Charlie Jenkins, Thomas Gleixner, Alexandre Ghiti, x86,
linux-kernel
The following commit has been merged into the core/entry branch of tip:
Commit-ID: e43b8bb56e537bfc8d9076793091e7679020fc9c
Gitweb: https://git.kernel.org/tip/e43b8bb56e537bfc8d9076793091e7679020fc9c
Author: Charlie Jenkins <charlie@rivosinc.com>
AuthorDate: Thu, 20 Mar 2025 10:29:24 -07:00
Committer: Thomas Gleixner <tglx@linutronix.de>
CommitterDate: Tue, 29 Apr 2025 08:27:10 +02:00
entry: Inline syscall_exit_to_user_mode()
Similar to commit 221a164035fd ("entry: Move syscall_enter_from_user_mode()
to header file"), move syscall_exit_to_user_mode() to the header file as
well.
Testing was done with the byte-unixbench syscall benchmark (which calls
getpid) and QEMU. On riscv I measured a 7.09246% improvement, on x86 a
2.98843% improvement, on loongarch a 6.07954% improvement, and on s390 a
11.1328% improvement.
The Intel bot also reported "kernel test robot noticed a 1.9% improvement
of stress-ng.seek.ops_per_sec".
Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Link: https://lore.kernel.org/all/20250320-riscv_optimize_entry-v6-4-63e187e26041@rivosinc.com
Link: https://lore.kernel.org/linux-riscv/202502051555.85ae6844-lkp@intel.com/
---
include/linux/entry-common.h | 43 +++++++++++++++++++++++++++++--
kernel/entry/common.c | 49 +-----------------------------------
2 files changed, 42 insertions(+), 50 deletions(-)
diff --git a/include/linux/entry-common.h b/include/linux/entry-common.h
index fc61d02..f94f3fd 100644
--- a/include/linux/entry-common.h
+++ b/include/linux/entry-common.h
@@ -14,6 +14,7 @@
#include <linux/kmsan.h>
#include <asm/entry-common.h>
+#include <asm/syscall.h>
/*
* Define dummy _TIF work flags if not defined by the architecture or for
@@ -367,6 +368,15 @@ static __always_inline void exit_to_user_mode(void)
}
/**
+ * syscall_exit_work - Handle work before returning to user mode
+ * @regs: Pointer to current pt_regs
+ * @work: Current thread syscall work
+ *
+ * Do one-time syscall specific work.
+ */
+void syscall_exit_work(struct pt_regs *regs, unsigned long work);
+
+/**
* syscall_exit_to_user_mode_work - Handle work before returning to user mode
* @regs: Pointer to currents pt_regs
*
@@ -379,7 +389,30 @@ static __always_inline void exit_to_user_mode(void)
* make the final state transitions. Interrupts must stay disabled between
* return from this function and the invocation of exit_to_user_mode().
*/
-void syscall_exit_to_user_mode_work(struct pt_regs *regs);
+static __always_inline void syscall_exit_to_user_mode_work(struct pt_regs *regs)
+{
+ unsigned long work = READ_ONCE(current_thread_info()->syscall_work);
+ unsigned long nr = syscall_get_nr(current, regs);
+
+ CT_WARN_ON(ct_state() != CT_STATE_KERNEL);
+
+ if (IS_ENABLED(CONFIG_PROVE_LOCKING)) {
+ if (WARN(irqs_disabled(), "syscall %lu left IRQs disabled", nr))
+ local_irq_enable();
+ }
+
+ rseq_syscall(regs);
+
+ /*
+ * Do one-time syscall specific work. If these work items are
+ * enabled, we want to run them exactly once per syscall exit with
+ * interrupts enabled.
+ */
+ if (unlikely(work & SYSCALL_WORK_EXIT))
+ syscall_exit_work(regs, work);
+ local_irq_disable_exit_to_user();
+ exit_to_user_mode_prepare(regs);
+}
/**
* syscall_exit_to_user_mode - Handle work before returning to user mode
@@ -410,7 +443,13 @@ void syscall_exit_to_user_mode_work(struct pt_regs *regs);
* exit_to_user_mode(). This function is preferred unless there is a
* compelling architectural reason to use the separate functions.
*/
-void syscall_exit_to_user_mode(struct pt_regs *regs);
+static __always_inline void syscall_exit_to_user_mode(struct pt_regs *regs)
+{
+ instrumentation_begin();
+ syscall_exit_to_user_mode_work(regs);
+ instrumentation_end();
+ exit_to_user_mode();
+}
/**
* irqentry_enter_from_user_mode - Establish state before invoking the irq handler
diff --git a/kernel/entry/common.c b/kernel/entry/common.c
index 2015457..a8dd1f2 100644
--- a/kernel/entry/common.c
+++ b/kernel/entry/common.c
@@ -146,7 +146,7 @@ static inline bool report_single_step(unsigned long work)
return work & SYSCALL_WORK_SYSCALL_EXIT_TRAP;
}
-static void syscall_exit_work(struct pt_regs *regs, unsigned long work)
+void syscall_exit_work(struct pt_regs *regs, unsigned long work)
{
bool step;
@@ -173,53 +173,6 @@ static void syscall_exit_work(struct pt_regs *regs, unsigned long work)
ptrace_report_syscall_exit(regs, step);
}
-/*
- * Syscall specific exit to user mode preparation. Runs with interrupts
- * enabled.
- */
-static void syscall_exit_to_user_mode_prepare(struct pt_regs *regs)
-{
- unsigned long work = READ_ONCE(current_thread_info()->syscall_work);
- unsigned long nr = syscall_get_nr(current, regs);
-
- CT_WARN_ON(ct_state() != CT_STATE_KERNEL);
-
- if (IS_ENABLED(CONFIG_PROVE_LOCKING)) {
- if (WARN(irqs_disabled(), "syscall %lu left IRQs disabled", nr))
- local_irq_enable();
- }
-
- rseq_syscall(regs);
-
- /*
- * Do one-time syscall specific work. If these work items are
- * enabled, we want to run them exactly once per syscall exit with
- * interrupts enabled.
- */
- if (unlikely(work & SYSCALL_WORK_EXIT))
- syscall_exit_work(regs, work);
-}
-
-static __always_inline void __syscall_exit_to_user_mode_work(struct pt_regs *regs)
-{
- syscall_exit_to_user_mode_prepare(regs);
- local_irq_disable_exit_to_user();
- exit_to_user_mode_prepare(regs);
-}
-
-void syscall_exit_to_user_mode_work(struct pt_regs *regs)
-{
- __syscall_exit_to_user_mode_work(regs);
-}
-
-__visible noinstr void syscall_exit_to_user_mode(struct pt_regs *regs)
-{
- instrumentation_begin();
- __syscall_exit_to_user_mode_work(regs);
- instrumentation_end();
- exit_to_user_mode();
-}
-
noinstr void irqentry_enter_from_user_mode(struct pt_regs *regs)
{
enter_from_user_mode(regs);
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [tip: core/entry] LoongArch: entry: Migrate ret_from_fork() to C
2025-03-20 17:29 ` [PATCH v6 3/4] LoongArch: entry: Migrate ret_from_fork() to C Charlie Jenkins
@ 2025-04-29 6:33 ` tip-bot2 for Charlie Jenkins
2025-05-05 4:58 ` [PATCH v6 3/4] " Huacai Chen
1 sibling, 0 replies; 14+ messages in thread
From: tip-bot2 for Charlie Jenkins @ 2025-04-29 6:33 UTC (permalink / raw)
To: linux-tip-commits; +Cc: Charlie Jenkins, Thomas Gleixner, x86, linux-kernel
The following commit has been merged into the core/entry branch of tip:
Commit-ID: 7ace1602abf21da505993d77ccbae1df2496b324
Gitweb: https://git.kernel.org/tip/7ace1602abf21da505993d77ccbae1df2496b324
Author: Charlie Jenkins <charlie@rivosinc.com>
AuthorDate: Thu, 20 Mar 2025 10:29:23 -07:00
Committer: Thomas Gleixner <tglx@linutronix.de>
CommitterDate: Tue, 29 Apr 2025 08:27:10 +02:00
LoongArch: entry: Migrate ret_from_fork() to C
LoongArch is the only architecture that calls syscall_exit_to_user_mode()
from assembly.
Move the call into C so that this function can be inlined across all
architectures.
Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/20250320-riscv_optimize_entry-v6-3-63e187e26041@rivosinc.com
---
arch/loongarch/include/asm/asm-prototypes.h | 8 +++++-
arch/loongarch/kernel/entry.S | 22 +++++--------
arch/loongarch/kernel/process.c | 33 ++++++++++++++++----
3 files changed, 45 insertions(+), 18 deletions(-)
diff --git a/arch/loongarch/include/asm/asm-prototypes.h b/arch/loongarch/include/asm/asm-prototypes.h
index 51f224b..704066b 100644
--- a/arch/loongarch/include/asm/asm-prototypes.h
+++ b/arch/loongarch/include/asm/asm-prototypes.h
@@ -12,3 +12,11 @@ __int128_t __ashlti3(__int128_t a, int b);
__int128_t __ashrti3(__int128_t a, int b);
__int128_t __lshrti3(__int128_t a, int b);
#endif
+
+asmlinkage void noinstr __no_stack_protector ret_from_fork(struct task_struct *prev,
+ struct pt_regs *regs);
+
+asmlinkage void noinstr __no_stack_protector ret_from_kernel_thread(struct task_struct *prev,
+ struct pt_regs *regs,
+ int (*fn)(void *),
+ void *fn_arg);
diff --git a/arch/loongarch/kernel/entry.S b/arch/loongarch/kernel/entry.S
index 48e7e34..2abc29e 100644
--- a/arch/loongarch/kernel/entry.S
+++ b/arch/loongarch/kernel/entry.S
@@ -77,24 +77,22 @@ SYM_CODE_START(handle_syscall)
SYM_CODE_END(handle_syscall)
_ASM_NOKPROBE(handle_syscall)
-SYM_CODE_START(ret_from_fork)
+SYM_CODE_START(ret_from_fork_asm)
UNWIND_HINT_REGS
- bl schedule_tail # a0 = struct task_struct *prev
- move a0, sp
- bl syscall_exit_to_user_mode
+ move a1, sp
+ bl ret_from_fork
RESTORE_STATIC
RESTORE_SOME
RESTORE_SP_AND_RET
-SYM_CODE_END(ret_from_fork)
+SYM_CODE_END(ret_from_fork_asm)
-SYM_CODE_START(ret_from_kernel_thread)
+SYM_CODE_START(ret_from_kernel_thread_asm)
UNWIND_HINT_REGS
- bl schedule_tail # a0 = struct task_struct *prev
- move a0, s1
- jirl ra, s0, 0
- move a0, sp
- bl syscall_exit_to_user_mode
+ move a1, sp
+ move a2, s0
+ move a3, s1
+ bl ret_from_kernel_thread
RESTORE_STATIC
RESTORE_SOME
RESTORE_SP_AND_RET
-SYM_CODE_END(ret_from_kernel_thread)
+SYM_CODE_END(ret_from_kernel_thread_asm)
diff --git a/arch/loongarch/kernel/process.c b/arch/loongarch/kernel/process.c
index 6e58f65..98bc60d 100644
--- a/arch/loongarch/kernel/process.c
+++ b/arch/loongarch/kernel/process.c
@@ -14,6 +14,7 @@
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/errno.h>
+#include <linux/entry-common.h>
#include <linux/sched.h>
#include <linux/sched/debug.h>
#include <linux/sched/task.h>
@@ -33,6 +34,7 @@
#include <linux/prctl.h>
#include <linux/nmi.h>
+#include <asm/asm-prototypes.h>
#include <asm/asm.h>
#include <asm/bootinfo.h>
#include <asm/cpu.h>
@@ -47,6 +49,7 @@
#include <asm/pgtable.h>
#include <asm/processor.h>
#include <asm/reg.h>
+#include <asm/switch_to.h>
#include <asm/unwind.h>
#include <asm/vdso.h>
@@ -63,8 +66,9 @@ EXPORT_SYMBOL(__stack_chk_guard);
unsigned long boot_option_idle_override = IDLE_NO_OVERRIDE;
EXPORT_SYMBOL(boot_option_idle_override);
-asmlinkage void ret_from_fork(void);
-asmlinkage void ret_from_kernel_thread(void);
+asmlinkage void restore_and_ret(void);
+asmlinkage void ret_from_fork_asm(void);
+asmlinkage void ret_from_kernel_thread_asm(void);
void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long sp)
{
@@ -138,6 +142,23 @@ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
return 0;
}
+asmlinkage void noinstr __no_stack_protector ret_from_fork(struct task_struct *prev,
+ struct pt_regs *regs)
+{
+ schedule_tail(prev);
+ syscall_exit_to_user_mode(regs);
+}
+
+asmlinkage void noinstr __no_stack_protector ret_from_kernel_thread(struct task_struct *prev,
+ struct pt_regs *regs,
+ int (*fn)(void *),
+ void *fn_arg)
+{
+ schedule_tail(prev);
+ fn(fn_arg);
+ syscall_exit_to_user_mode(regs);
+}
+
/*
* Copy architecture-specific thread state
*/
@@ -165,8 +186,8 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
p->thread.reg03 = childksp;
p->thread.reg23 = (unsigned long)args->fn;
p->thread.reg24 = (unsigned long)args->fn_arg;
- p->thread.reg01 = (unsigned long)ret_from_kernel_thread;
- p->thread.sched_ra = (unsigned long)ret_from_kernel_thread;
+ p->thread.reg01 = (unsigned long)ret_from_kernel_thread_asm;
+ p->thread.sched_ra = (unsigned long)ret_from_kernel_thread_asm;
memset(childregs, 0, sizeof(struct pt_regs));
childregs->csr_euen = p->thread.csr_euen;
childregs->csr_crmd = p->thread.csr_crmd;
@@ -182,8 +203,8 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
childregs->regs[3] = usp;
p->thread.reg03 = (unsigned long) childregs;
- p->thread.reg01 = (unsigned long) ret_from_fork;
- p->thread.sched_ra = (unsigned long) ret_from_fork;
+ p->thread.reg01 = (unsigned long) ret_from_fork_asm;
+ p->thread.sched_ra = (unsigned long) ret_from_fork_asm;
/*
* New tasks lose permission to use the fpu. This accelerates context
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [tip: core/entry] riscv: entry: Split ret_from_fork() into user and kernel
2025-03-20 17:29 ` [PATCH v6 2/4] riscv: entry: Split ret_from_fork() into user and kernel Charlie Jenkins
2025-03-21 6:36 ` Alexandre Ghiti
@ 2025-04-29 6:33 ` tip-bot2 for Charlie Jenkins
1 sibling, 0 replies; 14+ messages in thread
From: tip-bot2 for Charlie Jenkins @ 2025-04-29 6:33 UTC (permalink / raw)
To: linux-tip-commits
Cc: Charlie Jenkins, Thomas Gleixner, Alexandre Ghiti, x86,
linux-kernel
The following commit has been merged into the core/entry branch of tip:
Commit-ID: 5b3d6103b343d59e19bd641e4c31df519f4d250d
Gitweb: https://git.kernel.org/tip/5b3d6103b343d59e19bd641e4c31df519f4d250d
Author: Charlie Jenkins <charlie@rivosinc.com>
AuthorDate: Thu, 20 Mar 2025 10:29:22 -07:00
Committer: Thomas Gleixner <tglx@linutronix.de>
CommitterDate: Tue, 29 Apr 2025 08:27:10 +02:00
riscv: entry: Split ret_from_fork() into user and kernel
This function was unified into a single function in commit ab9164dae273
("riscv: entry: Consolidate ret_from_kernel_thread into ret_from_fork").
However that imposed a performance degradation.
Partially reverting this commit to have ret_from_fork() split again,
results in a 1% increase on the number of times fork is able to be called
per second.
Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Link: https://lore.kernel.org/all/20250320-riscv_optimize_entry-v6-2-63e187e26041@rivosinc.com
---
arch/riscv/include/asm/asm-prototypes.h | 3 ++-
arch/riscv/kernel/entry.S | 13 ++++++++++---
arch/riscv/kernel/process.c | 17 +++++++++++------
3 files changed, 23 insertions(+), 10 deletions(-)
diff --git a/arch/riscv/include/asm/asm-prototypes.h b/arch/riscv/include/asm/asm-prototypes.h
index 733ff60..bfc8ea5 100644
--- a/arch/riscv/include/asm/asm-prototypes.h
+++ b/arch/riscv/include/asm/asm-prototypes.h
@@ -52,7 +52,8 @@ DECLARE_DO_ERROR_INFO(do_trap_ecall_s);
DECLARE_DO_ERROR_INFO(do_trap_ecall_m);
DECLARE_DO_ERROR_INFO(do_trap_break);
-asmlinkage void ret_from_fork(void *fn_arg, int (*fn)(void *), struct pt_regs *regs);
+asmlinkage void ret_from_fork_kernel(void *fn_arg, int (*fn)(void *), struct pt_regs *regs);
+asmlinkage void ret_from_fork_user(struct pt_regs *regs);
asmlinkage void handle_bad_stack(struct pt_regs *regs);
asmlinkage void do_page_fault(struct pt_regs *regs);
asmlinkage void do_irq(struct pt_regs *regs);
diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S
index b2dc5e7..0fb3380 100644
--- a/arch/riscv/kernel/entry.S
+++ b/arch/riscv/kernel/entry.S
@@ -319,14 +319,21 @@ SYM_CODE_END(handle_kernel_stack_overflow)
ASM_NOKPROBE(handle_kernel_stack_overflow)
#endif
-SYM_CODE_START(ret_from_fork_asm)
+SYM_CODE_START(ret_from_fork_kernel_asm)
call schedule_tail
move a0, s1 /* fn_arg */
move a1, s0 /* fn */
move a2, sp /* pt_regs */
- call ret_from_fork
+ call ret_from_fork_kernel
j ret_from_exception
-SYM_CODE_END(ret_from_fork_asm)
+SYM_CODE_END(ret_from_fork_kernel_asm)
+
+SYM_CODE_START(ret_from_fork_user_asm)
+ call schedule_tail
+ move a0, sp /* pt_regs */
+ call ret_from_fork_user
+ j ret_from_exception
+SYM_CODE_END(ret_from_fork_user_asm)
#ifdef CONFIG_IRQ_STACKS
/*
diff --git a/arch/riscv/kernel/process.c b/arch/riscv/kernel/process.c
index 7b0a0bf..485ec7a 100644
--- a/arch/riscv/kernel/process.c
+++ b/arch/riscv/kernel/process.c
@@ -38,7 +38,8 @@ unsigned long __stack_chk_guard __read_mostly;
EXPORT_SYMBOL(__stack_chk_guard);
#endif
-extern asmlinkage void ret_from_fork_asm(void);
+extern asmlinkage void ret_from_fork_kernel_asm(void);
+extern asmlinkage void ret_from_fork_user_asm(void);
void noinstr arch_cpu_idle(void)
{
@@ -208,14 +209,18 @@ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
return 0;
}
-asmlinkage void ret_from_fork(void *fn_arg, int (*fn)(void *), struct pt_regs *regs)
+asmlinkage void ret_from_fork_kernel(void *fn_arg, int (*fn)(void *), struct pt_regs *regs)
{
- if (unlikely(fn))
- fn(fn_arg);
+ fn(fn_arg);
syscall_exit_to_user_mode(regs);
}
+asmlinkage void ret_from_fork_user(struct pt_regs *regs)
+{
+ syscall_exit_to_user_mode(regs);
+}
+
int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
{
unsigned long clone_flags = args->flags;
@@ -238,6 +243,7 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
p->thread.s[0] = (unsigned long)args->fn;
p->thread.s[1] = (unsigned long)args->fn_arg;
+ p->thread.ra = (unsigned long)ret_from_fork_kernel_asm;
} else {
*childregs = *(current_pt_regs());
/* Turn off status.VS */
@@ -247,12 +253,11 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
if (clone_flags & CLONE_SETTLS)
childregs->tp = tls;
childregs->a0 = 0; /* Return value of fork() */
- p->thread.s[0] = 0;
+ p->thread.ra = (unsigned long)ret_from_fork_user_asm;
}
p->thread.riscv_v_flags = 0;
if (has_vector() || has_xtheadvector())
riscv_v_thread_alloc(p);
- p->thread.ra = (unsigned long)ret_from_fork_asm;
p->thread.sp = (unsigned long)childregs; /* kernel sp */
return 0;
}
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [tip: core/entry] riscv: entry: Convert ret_from_fork() to C
2025-03-20 17:29 ` [PATCH v6 1/4] riscv: entry: Convert ret_from_fork() to C Charlie Jenkins
@ 2025-04-29 6:33 ` tip-bot2 for Charlie Jenkins
0 siblings, 0 replies; 14+ messages in thread
From: tip-bot2 for Charlie Jenkins @ 2025-04-29 6:33 UTC (permalink / raw)
To: linux-tip-commits
Cc: Charlie Jenkins, Thomas Gleixner, Alexandre Ghiti, x86,
linux-kernel
The following commit has been merged into the core/entry branch of tip:
Commit-ID: f955aa8723a65759e920d4de8e5d076cef412afc
Gitweb: https://git.kernel.org/tip/f955aa8723a65759e920d4de8e5d076cef412afc
Author: Charlie Jenkins <charlie@rivosinc.com>
AuthorDate: Thu, 20 Mar 2025 10:29:21 -07:00
Committer: Thomas Gleixner <tglx@linutronix.de>
CommitterDate: Tue, 29 Apr 2025 08:27:10 +02:00
riscv: entry: Convert ret_from_fork() to C
Move the main section of ret_from_fork() to C to allow inlining of
syscall_exit_to_user_mode().
Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Link: https://lore.kernel.org/all/20250320-riscv_optimize_entry-v6-1-63e187e26041@rivosinc.com
---
arch/riscv/include/asm/asm-prototypes.h | 1 +
arch/riscv/kernel/entry.S | 15 ++++++---------
arch/riscv/kernel/process.c | 14 ++++++++++++--
3 files changed, 19 insertions(+), 11 deletions(-)
diff --git a/arch/riscv/include/asm/asm-prototypes.h b/arch/riscv/include/asm/asm-prototypes.h
index cd627ec..733ff60 100644
--- a/arch/riscv/include/asm/asm-prototypes.h
+++ b/arch/riscv/include/asm/asm-prototypes.h
@@ -52,6 +52,7 @@ DECLARE_DO_ERROR_INFO(do_trap_ecall_s);
DECLARE_DO_ERROR_INFO(do_trap_ecall_m);
DECLARE_DO_ERROR_INFO(do_trap_break);
+asmlinkage void ret_from_fork(void *fn_arg, int (*fn)(void *), struct pt_regs *regs);
asmlinkage void handle_bad_stack(struct pt_regs *regs);
asmlinkage void do_page_fault(struct pt_regs *regs);
asmlinkage void do_irq(struct pt_regs *regs);
diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S
index 33a5a9f..b2dc5e7 100644
--- a/arch/riscv/kernel/entry.S
+++ b/arch/riscv/kernel/entry.S
@@ -319,17 +319,14 @@ SYM_CODE_END(handle_kernel_stack_overflow)
ASM_NOKPROBE(handle_kernel_stack_overflow)
#endif
-SYM_CODE_START(ret_from_fork)
+SYM_CODE_START(ret_from_fork_asm)
call schedule_tail
- beqz s0, 1f /* not from kernel thread */
- /* Call fn(arg) */
- move a0, s1
- jalr s0
-1:
- move a0, sp /* pt_regs */
- call syscall_exit_to_user_mode
+ move a0, s1 /* fn_arg */
+ move a1, s0 /* fn */
+ move a2, sp /* pt_regs */
+ call ret_from_fork
j ret_from_exception
-SYM_CODE_END(ret_from_fork)
+SYM_CODE_END(ret_from_fork_asm)
#ifdef CONFIG_IRQ_STACKS
/*
diff --git a/arch/riscv/kernel/process.c b/arch/riscv/kernel/process.c
index 7c244de..7b0a0bf 100644
--- a/arch/riscv/kernel/process.c
+++ b/arch/riscv/kernel/process.c
@@ -17,7 +17,9 @@
#include <linux/ptrace.h>
#include <linux/uaccess.h>
#include <linux/personality.h>
+#include <linux/entry-common.h>
+#include <asm/asm-prototypes.h>
#include <asm/unistd.h>
#include <asm/processor.h>
#include <asm/csr.h>
@@ -36,7 +38,7 @@ unsigned long __stack_chk_guard __read_mostly;
EXPORT_SYMBOL(__stack_chk_guard);
#endif
-extern asmlinkage void ret_from_fork(void);
+extern asmlinkage void ret_from_fork_asm(void);
void noinstr arch_cpu_idle(void)
{
@@ -206,6 +208,14 @@ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
return 0;
}
+asmlinkage void ret_from_fork(void *fn_arg, int (*fn)(void *), struct pt_regs *regs)
+{
+ if (unlikely(fn))
+ fn(fn_arg);
+
+ syscall_exit_to_user_mode(regs);
+}
+
int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
{
unsigned long clone_flags = args->flags;
@@ -242,7 +252,7 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
p->thread.riscv_v_flags = 0;
if (has_vector() || has_xtheadvector())
riscv_v_thread_alloc(p);
- p->thread.ra = (unsigned long)ret_from_fork;
+ p->thread.ra = (unsigned long)ret_from_fork_asm;
p->thread.sp = (unsigned long)childregs; /* kernel sp */
return 0;
}
^ permalink raw reply related [flat|nested] 14+ messages in thread
* Re: [PATCH v6 3/4] LoongArch: entry: Migrate ret_from_fork() to C
2025-03-20 17:29 ` [PATCH v6 3/4] LoongArch: entry: Migrate ret_from_fork() to C Charlie Jenkins
2025-04-29 6:33 ` [tip: core/entry] " tip-bot2 for Charlie Jenkins
@ 2025-05-05 4:58 ` Huacai Chen
1 sibling, 0 replies; 14+ messages in thread
From: Huacai Chen @ 2025-05-05 4:58 UTC (permalink / raw)
To: Charlie Jenkins
Cc: Paul Walmsley, Palmer Dabbelt, WANG Xuerui, Thomas Gleixner,
Peter Zijlstra, Andy Lutomirski, Alexandre Ghiti, Arnd Bergmann,
Albert Ou, Alexandre Ghiti, linux-riscv, linux-kernel, loongarch
Hi, Charlie,
There are some small issues.
On Fri, Mar 21, 2025 at 1:29 AM Charlie Jenkins <charlie@rivosinc.com> wrote:
>
> LoongArch is the only architecture that calls
> syscall_exit_to_user_mode() from asm. Move the call into C so that this
> function can be inlined across all architectures.
>
> Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
> ---
> arch/loongarch/include/asm/asm-prototypes.h | 8 +++++++
> arch/loongarch/kernel/entry.S | 22 +++++++++----------
> arch/loongarch/kernel/process.c | 33 +++++++++++++++++++++++------
> 3 files changed, 45 insertions(+), 18 deletions(-)
>
> diff --git a/arch/loongarch/include/asm/asm-prototypes.h b/arch/loongarch/include/asm/asm-prototypes.h
> index 51f224bcfc654228ae423e9a066b25b35102a5b9..704066b4f7368be15be960fadbcd6c2574bbf6c0 100644
> --- a/arch/loongarch/include/asm/asm-prototypes.h
> +++ b/arch/loongarch/include/asm/asm-prototypes.h
> @@ -12,3 +12,11 @@ __int128_t __ashlti3(__int128_t a, int b);
> __int128_t __ashrti3(__int128_t a, int b);
> __int128_t __lshrti3(__int128_t a, int b);
> #endif
> +
> +asmlinkage void noinstr __no_stack_protector ret_from_fork(struct task_struct *prev,
> + struct pt_regs *regs);
> +
> +asmlinkage void noinstr __no_stack_protector ret_from_kernel_thread(struct task_struct *prev,
> + struct pt_regs *regs,
> + int (*fn)(void *),
> + void *fn_arg);
> diff --git a/arch/loongarch/kernel/entry.S b/arch/loongarch/kernel/entry.S
> index 48e7e34e355e83eae8165957ba2eac05a8bf17df..2abc29e573810e000f2fef4646ddca0dbb80eabe 100644
> --- a/arch/loongarch/kernel/entry.S
> +++ b/arch/loongarch/kernel/entry.S
> @@ -77,24 +77,22 @@ SYM_CODE_START(handle_syscall)
> SYM_CODE_END(handle_syscall)
> _ASM_NOKPROBE(handle_syscall)
>
> -SYM_CODE_START(ret_from_fork)
> +SYM_CODE_START(ret_from_fork_asm)
> UNWIND_HINT_REGS
> - bl schedule_tail # a0 = struct task_struct *prev
> - move a0, sp
> - bl syscall_exit_to_user_mode
> + move a1, sp
> + bl ret_from_fork
> RESTORE_STATIC
> RESTORE_SOME
> RESTORE_SP_AND_RET
> -SYM_CODE_END(ret_from_fork)
> +SYM_CODE_END(ret_from_fork_asm)
>
> -SYM_CODE_START(ret_from_kernel_thread)
> +SYM_CODE_START(ret_from_kernel_thread_asm)
> UNWIND_HINT_REGS
> - bl schedule_tail # a0 = struct task_struct *prev
> - move a0, s1
> - jirl ra, s0, 0
> - move a0, sp
> - bl syscall_exit_to_user_mode
> + move a1, sp
> + move a2, s0
> + move a3, s1
> + bl ret_from_kernel_thread
> RESTORE_STATIC
> RESTORE_SOME
> RESTORE_SP_AND_RET
> -SYM_CODE_END(ret_from_kernel_thread)
> +SYM_CODE_END(ret_from_kernel_thread_asm)
> diff --git a/arch/loongarch/kernel/process.c b/arch/loongarch/kernel/process.c
> index 6e58f65455c7ca3eae2e88ed852c8655a6701e5c..98bc60d7c550fcc0225e8452f81a7d6cd7888015 100644
> --- a/arch/loongarch/kernel/process.c
> +++ b/arch/loongarch/kernel/process.c
> @@ -14,6 +14,7 @@
> #include <linux/init.h>
> #include <linux/kernel.h>
> #include <linux/errno.h>
> +#include <linux/entry-common.h>
For alpa-betical order, it should be before errno.h.
> #include <linux/sched.h>
> #include <linux/sched/debug.h>
> #include <linux/sched/task.h>
> @@ -33,6 +34,7 @@
> #include <linux/prctl.h>
> #include <linux/nmi.h>
>
> +#include <asm/asm-prototypes.h>
For alpa-betical order, it should be after asm.h.
Huacai
> #include <asm/asm.h>
> #include <asm/bootinfo.h>
> #include <asm/cpu.h>
> @@ -47,6 +49,7 @@
> #include <asm/pgtable.h>
> #include <asm/processor.h>
> #include <asm/reg.h>
> +#include <asm/switch_to.h>
> #include <asm/unwind.h>
> #include <asm/vdso.h>
>
> @@ -63,8 +66,9 @@ EXPORT_SYMBOL(__stack_chk_guard);
> unsigned long boot_option_idle_override = IDLE_NO_OVERRIDE;
> EXPORT_SYMBOL(boot_option_idle_override);
>
> -asmlinkage void ret_from_fork(void);
> -asmlinkage void ret_from_kernel_thread(void);
> +asmlinkage void restore_and_ret(void);
> +asmlinkage void ret_from_fork_asm(void);
> +asmlinkage void ret_from_kernel_thread_asm(void);
>
> void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long sp)
> {
> @@ -138,6 +142,23 @@ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
> return 0;
> }
>
> +asmlinkage void noinstr __no_stack_protector ret_from_fork(struct task_struct *prev,
> + struct pt_regs *regs)
> +{
> + schedule_tail(prev);
> + syscall_exit_to_user_mode(regs);
> +}
> +
> +asmlinkage void noinstr __no_stack_protector ret_from_kernel_thread(struct task_struct *prev,
> + struct pt_regs *regs,
> + int (*fn)(void *),
> + void *fn_arg)
> +{
> + schedule_tail(prev);
> + fn(fn_arg);
> + syscall_exit_to_user_mode(regs);
> +}
> +
> /*
> * Copy architecture-specific thread state
> */
> @@ -165,8 +186,8 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
> p->thread.reg03 = childksp;
> p->thread.reg23 = (unsigned long)args->fn;
> p->thread.reg24 = (unsigned long)args->fn_arg;
> - p->thread.reg01 = (unsigned long)ret_from_kernel_thread;
> - p->thread.sched_ra = (unsigned long)ret_from_kernel_thread;
> + p->thread.reg01 = (unsigned long)ret_from_kernel_thread_asm;
> + p->thread.sched_ra = (unsigned long)ret_from_kernel_thread_asm;
> memset(childregs, 0, sizeof(struct pt_regs));
> childregs->csr_euen = p->thread.csr_euen;
> childregs->csr_crmd = p->thread.csr_crmd;
> @@ -182,8 +203,8 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
> childregs->regs[3] = usp;
>
> p->thread.reg03 = (unsigned long) childregs;
> - p->thread.reg01 = (unsigned long) ret_from_fork;
> - p->thread.sched_ra = (unsigned long) ret_from_fork;
> + p->thread.reg01 = (unsigned long) ret_from_fork_asm;
> + p->thread.sched_ra = (unsigned long) ret_from_fork_asm;
>
> /*
> * New tasks lose permission to use the fpu. This accelerates context
>
> --
> 2.43.0
>
^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2025-05-05 4:59 UTC | newest]
Thread overview: 14+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-03-20 17:29 [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode() Charlie Jenkins
2025-03-20 17:29 ` [PATCH v6 1/4] riscv: entry: Convert ret_from_fork() to C Charlie Jenkins
2025-04-29 6:33 ` [tip: core/entry] " tip-bot2 for Charlie Jenkins
2025-03-20 17:29 ` [PATCH v6 2/4] riscv: entry: Split ret_from_fork() into user and kernel Charlie Jenkins
2025-03-21 6:36 ` Alexandre Ghiti
2025-04-29 6:33 ` [tip: core/entry] " tip-bot2 for Charlie Jenkins
2025-03-20 17:29 ` [PATCH v6 3/4] LoongArch: entry: Migrate ret_from_fork() to C Charlie Jenkins
2025-04-29 6:33 ` [tip: core/entry] " tip-bot2 for Charlie Jenkins
2025-05-05 4:58 ` [PATCH v6 3/4] " Huacai Chen
2025-03-20 17:29 ` [PATCH v6 4/4] entry: Inline syscall_exit_to_user_mode() Charlie Jenkins
2025-04-29 6:33 ` [tip: core/entry] " tip-bot2 for Charlie Jenkins
2025-03-21 21:22 ` [PATCH v6 0/4] entry: Move ret_from_fork() to C and inline syscall_exit_to_user_mode() Thomas Gleixner
2025-04-28 21:54 ` Charlie Jenkins
2025-04-29 6:22 ` Thomas Gleixner
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox