Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <944d6a0a-7e38-e362-5bb2-0c838967e1c2@xen0n.name>
Date:   Sun, 15 May 2022 17:47:04 +0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.0a1
Subject: Re: [PATCH V10 13/22] LoongArch: Add system call support
Content-Language: en-US
To:     Huacai Chen <chenhuacai@gmail.com>, WANG Xuerui <kernel@xen0n.name>
Cc:     Huacai Chen <chenhuacai@loongson.cn>,
        Arnd Bergmann <arnd@arndb.de>,
        Andy Lutomirski <luto@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Peter Zijlstra <peterz@infradead.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        David Airlie <airlied@linux.ie>,
        Jonathan Corbet <corbet@lwn.net>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        linux-arch <linux-arch@vger.kernel.org>,
        "open list:DOCUMENTATION" <linux-doc@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Xuefeng Li <lixuefeng@loongson.cn>,
        Yanteng Si <siyanteng@loongson.cn>,
        Guo Ren <guoren@kernel.org>,
        Jiaxun Yang <jiaxun.yang@flygoat.com>,
        Stephen Rothwell <sfr@canb.auug.org.au>
References: <20220514080402.2650181-1-chenhuacai@loongson.cn>
 <20220514080402.2650181-14-chenhuacai@loongson.cn>
 <1b5eb372-0e62-bdc8-ebdc-f49eab220d1b@xen0n.name>
 <CAAhV-H44ESukAX5qhj8TQttOar7Kg5qU0_1grfLdLDAHbFAoQA@mail.gmail.com>
From:   WANG Xuerui <kernel@xen0n.name>
In-Reply-To: <CAAhV-H44ESukAX5qhj8TQttOar7Kg5qU0_1grfLdLDAHbFAoQA@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 5/14/22 22:39, Huacai Chen wrote:

> Hi, Xuerui,
>
> On Sat, May 14, 2022 at 9:53 PM WANG Xuerui <kernel@xen0n.name> wrote:
>> Hi,
>>
>> On 5/14/22 16:03, Huacai Chen wrote:
>>> Add system call support and related uaccess.h for LoongArch.
>>>
>>> Q: Why keep _ARCH_WANT_SYS_CLONE definition while there is clone3:
>>> A: The latest glibc release (2.35) has some basic support for clone3 but
>>>      it isn't complete. E.g., pthread_create() and spawni() have converted
>>>      to use clone3 but fork() will still use clone. Moreover, some seccomp
>>>      related applications can still not work perfectly with clone3. E.g.,
>>>      Chromium sandbox cannot work at all and there is no solution for it,
>>>      which is more terrible than the fork() story [1].
>>>
>>> [1] https://chromium-review.googlesource.com/c/chromium/src/+/2936184
>>>
>>> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
>>> ---
>>>    arch/loongarch/include/asm/seccomp.h     |  20 ++
>>>    arch/loongarch/include/asm/syscall.h     |  74 +++++++
>>>    arch/loongarch/include/asm/uaccess.h     | 269 +++++++++++++++++++++++
>>>    arch/loongarch/include/asm/unistd.h      |  11 +
>>>    arch/loongarch/include/uapi/asm/unistd.h |   5 +
>>>    arch/loongarch/kernel/entry.S            |  89 ++++++++
>>>    arch/loongarch/kernel/syscall.c          |  63 ++++++
>>>    7 files changed, 531 insertions(+)
>>>    create mode 100644 arch/loongarch/include/asm/seccomp.h
>>>    create mode 100644 arch/loongarch/include/asm/syscall.h
>>>    create mode 100644 arch/loongarch/include/asm/uaccess.h
>>>    create mode 100644 arch/loongarch/include/asm/unistd.h
>>>    create mode 100644 arch/loongarch/include/uapi/asm/unistd.h
>>>    create mode 100644 arch/loongarch/kernel/entry.S
>>>    create mode 100644 arch/loongarch/kernel/syscall.c
>>>
>>> diff --git a/arch/loongarch/include/asm/seccomp.h b/arch/loongarch/include/asm/seccomp.h
>>> new file mode 100644
>>> index 000000000000..31d6ab42e43e
>>> --- /dev/null
>>> +++ b/arch/loongarch/include/asm/seccomp.h
>>> @@ -0,0 +1,20 @@
>>> +/* SPDX-License-Identifier: GPL-2.0 */
>>> +
>>> +#ifndef _ASM_SECCOMP_H
>>> +#define _ASM_SECCOMP_H
>>> +
>>> +#include <asm/unistd.h>
>>> +
>>> +#include <asm-generic/seccomp.h>
>>> +
>>> +#ifdef CONFIG_32BIT
>>> +# define SECCOMP_ARCH_NATIVE         AUDIT_ARCH_LOONGARCH32
>>> +# define SECCOMP_ARCH_NATIVE_NR              NR_syscalls
>>> +# define SECCOMP_ARCH_NATIVE_NAME    "loongarch32"
>>> +#else
>>> +# define SECCOMP_ARCH_NATIVE         AUDIT_ARCH_LOONGARCH64
>>> +# define SECCOMP_ARCH_NATIVE_NR              NR_syscalls
>>> +# define SECCOMP_ARCH_NATIVE_NAME    "loongarch64"
>>> +#endif
>>> +
>>> +#endif /* _ASM_SECCOMP_H */
>>> diff --git a/arch/loongarch/include/asm/syscall.h b/arch/loongarch/include/asm/syscall.h
>>> new file mode 100644
>>> index 000000000000..e286dc58476e
>>> --- /dev/null
>>> +++ b/arch/loongarch/include/asm/syscall.h
>>> @@ -0,0 +1,74 @@
>>> +/* SPDX-License-Identifier: GPL-2.0+ */
>>> +/*
>>> + * Author: Hanlu Li <lihanlu@loongson.cn>
>>> + *         Huacai Chen <chenhuacai@loongson.cn>
>>> + *
>>> + * Copyright (C) 2020-2022 Loongson Technology Corporation Limited
>>> + */
>>> +
>>> +#ifndef __ASM_LOONGARCH_SYSCALL_H
>>> +#define __ASM_LOONGARCH_SYSCALL_H
>>> +
>>> +#include <linux/compiler.h>
>>> +#include <uapi/linux/audit.h>
>>> +#include <linux/elf-em.h>
>>> +#include <linux/kernel.h>
>>> +#include <linux/sched.h>
>>> +#include <linux/uaccess.h>
>>> +#include <asm/ptrace.h>
>>> +#include <asm/unistd.h>
>>> +
>>> +extern void *sys_call_table[];
>>> +
>>> +static inline long syscall_get_nr(struct task_struct *task,
>>> +                               struct pt_regs *regs)
>>> +{
>>> +     return regs->regs[11];
>> It may be better readability-wise to somehow change this 11 to something
>> resembling A7, but it's okay as-is.
> OK, keep it as is.
>
>>> +}
>>> +
>>> +static inline void syscall_rollback(struct task_struct *task,
>>> +                                 struct pt_regs *regs)
>>> +{
>>> +        regs->regs[4] = regs->orig_a0;
>>> +}
>>> +
>>> +static inline long syscall_get_error(struct task_struct *task,
>>> +                                  struct pt_regs *regs)
>>> +{
>>> +     unsigned long error = regs->regs[4];
>>> +
>>> +     return IS_ERR_VALUE(error) ? error : 0;
>>> +}
>>> +
>>> +static inline long syscall_get_return_value(struct task_struct *task,
>>> +                                         struct pt_regs *regs)
>>> +{
>>> +     return regs->regs[4];
>>> +}
>>> +
>>> +static inline void syscall_set_return_value(struct task_struct *task,
>>> +                                         struct pt_regs *regs,
>>> +                                         int error, long val)
>>> +{
>>> +     regs->regs[4] = (long) error ? error : val;
>>> +}
>>> +
>>> +static inline void syscall_get_arguments(struct task_struct *task,
>>> +                                      struct pt_regs *regs,
>>> +                                      unsigned long *args)
>>> +{
>>> +     args[0] = regs->orig_a0;
>>> +     memcpy(&args[1], &regs->regs[5], 5 * sizeof(long));
>>> +}
>>> +
>>> +static inline int syscall_get_arch(struct task_struct *task)
>>> +{
>>> +     return AUDIT_ARCH_LOONGARCH64;
>>> +}
>>> +
>>> +static inline bool arch_syscall_is_vdso_sigreturn(struct pt_regs *regs)
>>> +{
>>> +     return false;
>>> +}
>>> +
>>> +#endif       /* __ASM_LOONGARCH_SYSCALL_H */
>>> diff --git a/arch/loongarch/include/asm/uaccess.h b/arch/loongarch/include/asm/uaccess.h
>>> new file mode 100644
>>> index 000000000000..217c6a3727b1
>>> --- /dev/null
>>> +++ b/arch/loongarch/include/asm/uaccess.h
>>> @@ -0,0 +1,269 @@
>>> +/* SPDX-License-Identifier: GPL-2.0 */
>>> +/*
>>> + * Copyright (C) 2020-2022 Loongson Technology Corporation Limited
>>> + *
>>> + * Derived from MIPS:
>>> + * Copyright (C) 1996, 1997, 1998, 1999, 2000, 03, 04 by Ralf Baechle
>>> + * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
>>> + * Copyright (C) 2007  Maciej W. Rozycki
>>> + * Copyright (C) 2014, Imagination Technologies Ltd.
>>> + */
>>> +#ifndef _ASM_UACCESS_H
>>> +#define _ASM_UACCESS_H
>>> +
>>> +#include <linux/kernel.h>
>>> +#include <linux/string.h>
>>> +#include <linux/extable.h>
>>> +#include <asm/pgtable.h>
>>> +#include <asm-generic/extable.h>
>>> +#include <asm-generic/access_ok.h>
>>> +
>>> +extern u64 __ua_limit;
>>> +
>>> +#define __UA_ADDR    ".dword"
>>> +#define __UA_LA              "la.abs"
>>> +#define __UA_LIMIT   __ua_limit
>>> +
>>> +/*
>>> + * get_user: - Get a simple variable from user space.
>>> + * @x:        Variable to store result.
>>> + * @ptr: Source address, in user space.
>>> + *
>>> + * Context: User context only. This function may sleep if pagefaults are
>>> + *          enabled.
>>> + *
>>> + * This macro copies a single simple variable from user space to kernel
>>> + * space.  It supports simple types like char and int, but not larger
>>> + * data types like structures or arrays.
>>> + *
>>> + * @ptr must have pointer-to-simple-variable type, and the result of
>>> + * dereferencing @ptr must be assignable to @x without a cast.
>>> + *
>>> + * Returns zero on success, or -EFAULT on error.
>>> + * On error, the variable @x is set to zero.
>>> + */
>>> +#define get_user(x, ptr) \
>>> +({                                                                   \
>>> +     const __typeof__(*(ptr)) __user *__p = (ptr);                   \
>>> +                                                                     \
>>> +     might_fault();                                                  \
>>> +     access_ok(__p, sizeof(*__p)) ? __get_user((x), __p) :           \
>>> +                                    ((x) = 0, -EFAULT);              \
>>> +})
>>> +
>>> +/*
>>> + * put_user: - Write a simple value into user space.
>>> + * @x:        Value to copy to user space.
>>> + * @ptr: Destination address, in user space.
>>> + *
>>> + * Context: User context only. This function may sleep if pagefaults are
>>> + *          enabled.
>>> + *
>>> + * This macro copies a single simple value from kernel space to user
>>> + * space.  It supports simple types like char and int, but not larger
>>> + * data types like structures or arrays.
>>> + *
>>> + * @ptr must have pointer-to-simple-variable type, and @x must be assignable
>>> + * to the result of dereferencing @ptr.
>>> + *
>>> + * Returns zero on success, or -EFAULT on error.
>>> + */
>>> +#define put_user(x, ptr) \
>>> +({                                                                   \
>>> +     __typeof__(*(ptr)) __user *__p = (ptr);                         \
>>> +                                                                     \
>>> +     might_fault();                                                  \
>>> +     access_ok(__p, sizeof(*__p)) ? __put_user((x), __p) : -EFAULT;  \
>>> +})
>>> +
>>> +/*
>>> + * __get_user: - Get a simple variable from user space, with less checking.
>>> + * @x:        Variable to store result.
>>> + * @ptr: Source address, in user space.
>>> + *
>>> + * Context: User context only. This function may sleep if pagefaults are
>>> + *          enabled.
>>> + *
>>> + * This macro copies a single simple variable from user space to kernel
>>> + * space.  It supports simple types like char and int, but not larger
>>> + * data types like structures or arrays.
>>> + *
>>> + * @ptr must have pointer-to-simple-variable type, and the result of
>>> + * dereferencing @ptr must be assignable to @x without a cast.
>>> + *
>>> + * Caller must check the pointer with access_ok() before calling this
>>> + * function.
>>> + *
>>> + * Returns zero on success, or -EFAULT on error.
>>> + * On error, the variable @x is set to zero.
>>> + */
>>> +#define __get_user(x, ptr) \
>>> +({                                                                   \
>>> +     int __gu_err = 0;                                               \
>>> +                                                                     \
>>> +     __chk_user_ptr(ptr);                                            \
>>> +     __get_user_common((x), sizeof(*(ptr)), ptr);                    \
>>> +     __gu_err;                                                       \
>>> +})
>>> +
>>> +/*
>>> + * __put_user: - Write a simple value into user space, with less checking.
>>> + * @x:        Value to copy to user space.
>>> + * @ptr: Destination address, in user space.
>>> + *
>>> + * Context: User context only. This function may sleep if pagefaults are
>>> + *          enabled.
>>> + *
>>> + * This macro copies a single simple value from kernel space to user
>>> + * space.  It supports simple types like char and int, but not larger
>>> + * data types like structures or arrays.
>>> + *
>>> + * @ptr must have pointer-to-simple-variable type, and @x must be assignable
>>> + * to the result of dereferencing @ptr.
>>> + *
>>> + * Caller must check the pointer with access_ok() before calling this
>>> + * function.
>>> + *
>>> + * Returns zero on success, or -EFAULT on error.
>>> + */
>>> +#define __put_user(x, ptr) \
>>> +({                                                                   \
>>> +     int __pu_err = 0;                                               \
>>> +     __typeof__(*(ptr)) __pu_val;                                    \
>>> +                                                                     \
>>> +     __pu_val = (x);                                                 \
>>> +     __chk_user_ptr(ptr);                                            \
>>> +     __put_user_common(ptr, sizeof(*(ptr)));                         \
>>> +     __pu_err;                                                       \
>>> +})
>>> +
>>> +struct __large_struct { unsigned long buf[100]; };
>>> +#define __m(x) (*(struct __large_struct __user *)(x))
>>> +
>>> +#define __get_user_common(val, size, ptr)                            \
>>> +do {                                                                 \
>>> +     switch (size) {                                                 \
>>> +     case 1: __get_data_asm(val, "ld.b", ptr); break;                \
>>> +     case 2: __get_data_asm(val, "ld.h", ptr); break;                \
>>> +     case 4: __get_data_asm(val, "ld.w", ptr); break;                \
>>> +     case 8: __get_data_asm(val, "ld.d", ptr); break;                \
>>> +     default: BUILD_BUG(); break;                                    \
>>> +     }                                                               \
>>> +} while (0)
>>> +
>>> +#define __get_kernel_common(val, size, ptr) __get_user_common(val, size, ptr)
>>> +
>>> +#define __get_data_asm(val, insn, ptr)                                       \
>>> +{                                                                    \
>>> +     long __gu_tmp;                                                  \
>>> +                                                                     \
>>> +     __asm__ __volatile__(                                           \
>>> +     "1:     " insn "        %1, %2                          \n"     \
>>> +     "2:                                                     \n"     \
>>> +     "       .section .fixup,\"ax\"                          \n"     \
>>> +     "3:     li.w    %0, %3                                  \n"     \
>>> +     "       or      %1, $r0, $r0                            \n"     \
>>> +     "       b       2b                                      \n"     \
>>> +     "       .previous                                       \n"     \
>>> +     "       .section __ex_table,\"a\"                       \n"     \
>>> +     "       "__UA_ADDR "\t1b, 3b                            \n"     \
>>> +     "       .previous                                       \n"     \
>>> +     : "+r" (__gu_err), "=r" (__gu_tmp)                              \
>>> +     : "m" (__m(ptr)), "i" (-EFAULT));                               \
>>> +                                                                     \
>>> +     (val) = (__typeof__(*(ptr))) __gu_tmp;                          \
>>> +}
>>> +
>>> +#define __put_user_common(ptr, size)                                 \
>>> +do {                                                                 \
>>> +     switch (size) {                                                 \
>>> +     case 1: __put_data_asm("st.b", ptr); break;                     \
>>> +     case 2: __put_data_asm("st.h", ptr); break;                     \
>>> +     case 4: __put_data_asm("st.w", ptr); break;                     \
>>> +     case 8: __put_data_asm("st.d", ptr); break;                     \
>>> +     default: BUILD_BUG(); break;                                    \
>>> +     }                                                               \
>>> +} while (0)
>>> +
>>> +#define __put_kernel_common(ptr, size) __put_user_common(ptr, size)
>>> +
>>> +#define __put_data_asm(insn, ptr)                                    \
>>> +{                                                                    \
>>> +     __asm__ __volatile__(                                           \
>>> +     "1:     " insn "        %z2, %1         # __put_user_asm\n"     \
>>> +     "2:                                                     \n"     \
>>> +     "       .section        .fixup,\"ax\"                   \n"     \
>>> +     "3:     li.w    %0, %3                                  \n"     \
>>> +     "       b       2b                                      \n"     \
>>> +     "       .previous                                       \n"     \
>>> +     "       .section        __ex_table,\"a\"                \n"     \
>>> +     "       " __UA_ADDR "   1b, 3b                          \n"     \
>>> +     "       .previous                                       \n"     \
>>> +     : "+r" (__pu_err), "=m" (__m(ptr))                              \
>>> +     : "Jr" (__pu_val), "i" (-EFAULT));                              \
>>> +}
>>> +
>>> +#define __get_kernel_nofault(dst, src, type, err_label)                      \
>>> +do {                                                                 \
>>> +     int __gu_err = 0;                                               \
>>> +                                                                     \
>>> +     __get_kernel_common(*((type *)(dst)), sizeof(type),             \
>>> +                         (__force type *)(src));                     \
>>> +     if (unlikely(__gu_err))                                         \
>>> +             goto err_label;                                         \
>>> +} while (0)
>>> +
>>> +#define __put_kernel_nofault(dst, src, type, err_label)                      \
>>> +do {                                                                 \
>>> +     type __pu_val;                                                  \
>>> +     int __pu_err = 0;                                               \
>>> +                                                                     \
>>> +     __pu_val = *(__force type *)(src);                              \
>>> +     __put_kernel_common(((type *)(dst)), sizeof(type));             \
>>> +     if (unlikely(__pu_err))                                         \
>>> +             goto err_label;                                         \
>>> +} while (0)
>>> +
>>> +extern unsigned long __copy_user(void *to, const void *from, __kernel_size_t n);
>>> +
>>> +static inline unsigned long __must_check
>>> +raw_copy_from_user(void *to, const void __user *from, unsigned long n)
>>> +{
>>> +     return __copy_user(to, from, n);
>>> +}
>>> +
>>> +static inline unsigned long __must_check
>>> +raw_copy_to_user(void __user *to, const void *from, unsigned long n)
>>> +{
>>> +     return __copy_user(to, from, n);
>>> +}
>>> +
>>> +#define INLINE_COPY_FROM_USER
>>> +#define INLINE_COPY_TO_USER
>>> +
>>> +/*
>>> + * __clear_user: - Zero a block of memory in user space, with less checking.
>>> + * @addr: Destination address, in user space.
>>> + * @size: Number of bytes to zero.
>>> + *
>>> + * Zero a block of memory in user space.  Caller must check
>>> + * the specified block with access_ok() before calling this function.
>>> + *
>>> + * Returns number of bytes that could not be cleared.
>>> + * On success, this will be zero.
>>> + */
>>> +extern unsigned long __clear_user(void __user *addr, __kernel_size_t size);
>>> +
>>> +#define clear_user(addr, n)                                          \
>>> +({                                                                   \
>>> +     void __user *__cl_addr = (addr);                                \
>>> +     unsigned long __cl_size = (n);                                  \
>>> +     if (__cl_size && access_ok(__cl_addr, __cl_size))               \
>>> +             __cl_size = __clear_user(__cl_addr, __cl_size);         \
>>> +     __cl_size;                                                      \
>>> +})
>>> +
>>> +extern long strncpy_from_user(char *to, const char __user *from, long n);
>>> +extern long strnlen_user(const char __user *str, long n);
>>> +
>>> +#endif /* _ASM_UACCESS_H */
>>> diff --git a/arch/loongarch/include/asm/unistd.h b/arch/loongarch/include/asm/unistd.h
>>> new file mode 100644
>>> index 000000000000..cfddb0116a8c
>>> --- /dev/null
>>> +++ b/arch/loongarch/include/asm/unistd.h
>>> @@ -0,0 +1,11 @@
>>> +/* SPDX-License-Identifier: GPL-2.0+ */
>>> +/*
>>> + * Author: Hanlu Li <lihanlu@loongson.cn>
>>> + *         Huacai Chen <chenhuacai@loongson.cn>
>>> + *
>>> + * Copyright (C) 2020-2022 Loongson Technology Corporation Limited
>>> + */
>>> +
>>> +#include <uapi/asm/unistd.h>
>>> +
>>> +#define NR_syscalls (__NR_syscalls)
>>> diff --git a/arch/loongarch/include/uapi/asm/unistd.h b/arch/loongarch/include/uapi/asm/unistd.h
>>> new file mode 100644
>>> index 000000000000..fcb668984f03
>>> --- /dev/null
>>> +++ b/arch/loongarch/include/uapi/asm/unistd.h
>>> @@ -0,0 +1,5 @@
>>> +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
>>> +#define __ARCH_WANT_SYS_CLONE
>>> +#define __ARCH_WANT_SYS_CLONE3
>> Having read the discussion in v9 regarding the sandbox scenario, I agree
>> leaving clone alone would probably be the better choice. It's relieving
>> seeing the stat cruft dropped... World rebuilds of new-world distros
>> pending on the horizon. :-D
> clone3() is not just a beautiful version of clone(), it has some new
> features such as pidfd, so we need both clone() and clone3().
Apologizes if I didn't make it clear, but to clear any confusion: I'm 
not saying we don't want clone3, and we should keep both clone (for not 
breaking end user workflows involving sandboxed processes such as 
browsers) and clone3 (the path forward).
>
>>> +
>>> +#include <asm-generic/unistd.h>
>>> diff --git a/arch/loongarch/kernel/entry.S b/arch/loongarch/kernel/entry.S
>>> new file mode 100644
>>> index 000000000000..d5b3dbcf5425
>>> --- /dev/null
>>> +++ b/arch/loongarch/kernel/entry.S
>>> @@ -0,0 +1,89 @@
>>> +/* SPDX-License-Identifier: GPL-2.0 */
>>> +/*
>>> + * Copyright (C) 2020-2022 Loongson Technology Corporation Limited
>>> + *
>>> + * Derived from MIPS:
>>> + * Copyright (C) 1994 - 2000, 2001, 2003 Ralf Baechle
>>> + * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
>>> + * Copyright (C) 2001 MIPS Technologies, Inc.
>>> + */
>>> +
>>> +#include <asm/asm.h>
>>> +#include <asm/asmmacro.h>
>>> +#include <asm/loongarch.h>
>>> +#include <asm/regdef.h>
>>> +#include <asm/stackframe.h>
>>> +#include <asm/thread_info.h>
>>> +
>>> +     .text
>>> +     .cfi_sections   .debug_frame
>>> +     .align  5
>>> +SYM_FUNC_START(handle_syscall)
>>> +     csrrd   t0, PERCPU_BASE_KS
>>> +     la.abs  t1, kernelsp
>>> +     add.d   t1, t1, t0
>>> +     move    t2, sp
>>> +     ld.d    sp, t1, 0
>>> +
>>> +     addi.d  sp, sp, -PT_SIZE
>>> +     cfi_st  t2, PT_R3
>>> +     cfi_rel_offset  sp, PT_R3
>>> +     st.d    zero, sp, PT_R0
>>> +     csrrd   t2, LOONGARCH_CSR_PRMD
>>> +     st.d    t2, sp, PT_PRMD
>>> +     csrrd   t2, LOONGARCH_CSR_CRMD
>>> +     st.d    t2, sp, PT_CRMD
>>> +     csrrd   t2, LOONGARCH_CSR_EUEN
>>> +     st.d    t2, sp, PT_EUEN
>>> +     csrrd   t2, LOONGARCH_CSR_ECFG
>>> +     st.d    t2, sp, PT_ECFG
>>> +     csrrd   t2, LOONGARCH_CSR_ESTAT
>>> +     st.d    t2, sp, PT_ESTAT
>>> +     cfi_st  ra, PT_R1
>>> +     cfi_st  a0, PT_R4
>>> +     cfi_st  a1, PT_R5
>>> +     cfi_st  a2, PT_R6
>>> +     cfi_st  a3, PT_R7
>>> +     cfi_st  a4, PT_R8
>>> +     cfi_st  a5, PT_R9
>>> +     cfi_st  a6, PT_R10
>>> +     cfi_st  a7, PT_R11
>>> +     csrrd   ra, LOONGARCH_CSR_ERA
>>> +     st.d    ra, sp, PT_ERA
>>> +     cfi_rel_offset ra, PT_ERA
>>> +
>>> +     cfi_st  tp, PT_R2
>>> +     cfi_st  u0, PT_R21
>>> +     cfi_st  fp, PT_R22
>>> +
>>> +     SAVE_STATIC
>>> +
>>> +     move    u0, t0
>>> +     li.d    tp, ~_THREAD_MASK
>>> +     and     tp, tp, sp
>>> +
>>> +     move    a0, sp
>>> +     bl      do_syscall
>>> +
>>> +     RESTORE_ALL_AND_RET
>>> +SYM_FUNC_END(handle_syscall)
>>> +
>>> +SYM_CODE_START(ret_from_fork)
>>> +     bl      schedule_tail           # a0 = struct task_struct *prev
>>> +     move    a0, sp
>>> +     bl      syscall_exit_to_user_mode
>>> +     RESTORE_STATIC
>>> +     RESTORE_SOME
>>> +     RESTORE_SP_AND_RET
>>> +SYM_CODE_END(ret_from_fork)
>>> +
>>> +SYM_CODE_START(ret_from_kernel_thread)
>>> +     bl      schedule_tail           # a0 = struct task_struct *prev
>>> +     move    a0, s1
>>> +     jirl    ra, s0, 0
>>> +     move    a0, sp
>>> +     bl      syscall_exit_to_user_mode
>>> +     RESTORE_STATIC
>>> +     RESTORE_SOME
>>> +     RESTORE_SP_AND_RET
>>> +SYM_CODE_END(ret_from_kernel_thread)
>>> diff --git a/arch/loongarch/kernel/syscall.c b/arch/loongarch/kernel/syscall.c
>>> new file mode 100644
>>> index 000000000000..3fc4211db989
>>> --- /dev/null
>>> +++ b/arch/loongarch/kernel/syscall.c
>>> @@ -0,0 +1,63 @@
>>> +// SPDX-License-Identifier: GPL-2.0+
>>> +/*
>>> + * Author: Hanlu Li <lihanlu@loongson.cn>
>>> + *         Huacai Chen <chenhuacai@loongson.cn>
>>> + *
>>> + * Copyright (C) 2020-2022 Loongson Technology Corporation Limited
>>> + */
>>> +#include <linux/capability.h>
>>> +#include <linux/entry-common.h>
>>> +#include <linux/errno.h>
>>> +#include <linux/linkage.h>
>>> +#include <linux/syscalls.h>
>>> +#include <linux/unistd.h>
>>> +
>>> +#include <asm/asm.h>
>>> +#include <asm/signal.h>
>>> +#include <asm/switch_to.h>
>>> +#include <asm-generic/syscalls.h>
>>> +
>>> +#undef __SYSCALL
>>> +#define __SYSCALL(nr, call)  [nr] = (call),
>>> +
>>> +SYSCALL_DEFINE6(mmap, unsigned long, addr, unsigned long, len, unsigned long,
>>> +             prot, unsigned long, flags, unsigned long, fd, off_t, offset)
>>> +{
>>> +     if (offset & ~PAGE_MASK)
>>> +             return -EINVAL;
>>> +
>>> +     return ksys_mmap_pgoff(addr, len, prot, flags, fd, offset >> PAGE_SHIFT);
>>> +}
>>> +
>>> +void *sys_call_table[__NR_syscalls] = {
>>> +     [0 ... __NR_syscalls - 1] = sys_ni_syscall,
>>> +#include <asm/unistd.h>
>>> +};
>>> +
>>> +typedef long (*sys_call_fn)(unsigned long, unsigned long,
>>> +     unsigned long, unsigned long, unsigned long, unsigned long);
>>> +
>>> +void noinstr do_syscall(struct pt_regs *regs)
>>> +{
>>> +     unsigned long nr;
>>> +     sys_call_fn syscall_fn;
>>> +
>>> +     nr = regs->regs[11];
>>> +     /* Set for syscall restarting */
>>> +     if (nr < NR_syscalls)
>>> +             regs->regs[0] = nr + 1;
>>> +
>>> +     regs->csr_era += 4;
>>> +     regs->orig_a0 = regs->regs[4];
>>> +     regs->regs[4] = -ENOSYS;
>>> +
>>> +     nr = syscall_enter_from_user_mode(regs, nr);
>>> +
>>> +     if (nr < NR_syscalls) {
>> I saw in the manual that there's the BADI (出错指令) CSR that records the
>> faulting instruction word for exceptions other than INT, GCHC and MERR;
>> this implies we can read the trapping syscall insn word from this CSR,
>> to inspect its immediate field, so we can reserve non-zero immediates on
>> the syscall insn for future use.
>>
>> This would take 3 insns to complete: csrrd, bitfield extraction, and
>> bail if non-zero; this should be an acceptable tradeoff for a lot of
>> extensibility and scalability?
> This need much efforts, we can optimize it in future.

Fair point. No one has used syscall insns with non-zero immediates, at 
least from what I've seen during the year porting software and reviewing 
others' patches, so we could add the check later and breakage should 
only be theoretical.

>
> Huacai
>>> +             syscall_fn = sys_call_table[nr];
>>> +             regs->regs[4] = syscall_fn(regs->orig_a0, regs->regs[5], regs->regs[6],
>>> +                                        regs->regs[7], regs->regs[8], regs->regs[9]);
>>> +     }
>>> +
>>> +     syscall_exit_to_user_mode(regs);
>>> +}

Now that we've gone through every point:

Reviewed-by: WANG Xuerui <git@xen0n.name>