Documentation/admin-guide/kernel-parameters.txt | 4 + arch/arm/include/asm/uaccess.h | 26 +- arch/arm64/include/asm/uaccess.h | 4 +- arch/arm64/kernel/entry-common.c | 2 +- arch/powerpc/include/asm/uaccess.h | 8 +- arch/riscv/include/asm/uaccess.h | 8 +- arch/s390/include/asm/uaccess.h | 4 +- arch/x86/entry/syscall_32.c | 3 +- arch/x86/include/asm/futex.h | 75 +-- arch/x86/include/asm/ptrace.h | 20 +- arch/x86/include/asm/uaccess.h | 12 +- drivers/hv/mshv_root_main.c | 3 + fs/binfmt_elf.c | 2 +- fs/exec.c | 2 +- fs/select.c | 12 +- include/asm-generic/thread_info_tif.h | 3 + include/linux/bitmap.h | 15 + include/linux/cleanup.h | 30 +- include/linux/cpumask.h | 26 +- include/linux/entry-common.h | 38 +- include/linux/irq-entry-common.h | 75 ++- include/linux/irq_work.h | 9 +- include/linux/irq_work_types.h | 14 + include/linux/mm.h | 25 - include/linux/mm_types.h | 128 +--- include/linux/resume_user_mode.h | 2 +- include/linux/rseq.h | 214 +++--- include/linux/rseq_entry.h | 616 ++++++++++++++++++ include/linux/rseq_types.h | 164 +++++ include/linux/sched.h | 57 +- include/linux/thread_info.h | 5 + include/linux/uaccess.h | 314 ++++++++- include/trace/events/rseq.h | 4 +- include/uapi/linux/rseq.h | 21 +- init/Kconfig | 28 +- init/init_task.c | 3 + kernel/cpu.c | 19 + kernel/entry/common.c | 39 +- kernel/entry/syscall-common.c | 8 - kernel/exit.c | 1 + kernel/fork.c | 7 +- kernel/futex/core.c | 4 +- kernel/futex/futex.h | 58 +- kernel/ptrace.c | 6 +- kernel/rseq.c | 655 ++++++++----------- kernel/sched/core.c | 827 ++++++++++++------------ kernel/sched/membarrier.c | 8 +- kernel/sched/sched.h | 392 +++++------ kernel/signal.c | 2 - lib/bitmap.c | 6 + virt/kvm/kvm_main.c | 7 + 51 files changed, 2458 insertions(+), 1557 deletions(-) create mode 100644 include/linux/irq_work_types.h create mode 100644 include/linux/rseq_entry.h create mode 100644 include/linux/rseq_types.h
Linus,
please pull the latest core/rseq branch from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git core-rseq-2025-11-30
up to: 653fda7ae73d: sched/mmcid: Switch over to the new mechanism
A large overhaul of the restartable sequences and CID management:
The recent enablement of RSEQ in glibc resulted in regressions which are
caused by the related overhead. It turned out that the decision to invoke
the exit to user work was not really a decision. More or less each
context switch caused that. There is a long list of small issues which
sums up nicely and results in a 3-4% regression in I/O benchmarks.
The other detail which caused issues due to extra work in context switch
and task migration is the CID (memory context ID) management. It also
requires to use a task work to consolidate the CID space, which is
executed in the context of an arbitrary task and results in sporadic
uncontrolled exit latencies.
The rewrite addresses this by:
- Removing deprecated and long unsupported functionality
- Moving the related data into dedicated data structures which are
optimized for fast path processing.
- Caching values so actual decisions can be made
- Replacing the current implementation with a optimized inlined variant.
- Separating fast and slow path for architectures which use the generic
entry code, so that only fault and error handling goes into the
TIF_NOTIFY_RESUME handler.
- Rewriting the CID management so that it becomes mostly invisible in the
context switch path. That moves the work of switching modes into the
fork/exit path, which is a reasonable tradeoff. That work is only
required when a process creates more threads than the cpuset it is
allowed to run on or when enough threads exit after that. An artificial
thread pool benchmarks which triggers this did not degrade, it actually
improved significantly.
The main effect in migration heavy scenarios is that runqueue lock held
time and therefore contention goes down significantly.
Note: This contains parts of the scoped uaccess series.
Thanks,
tglx
------------------>
Dan Carpenter (1):
rseq: Delete duplicate if statement in rseq_virt_userspace_exit()
Peter Zijlstra (2):
cleanup: Always inline everything
x86/ptrace: Always inline trivial accessors
Thomas Gleixner (63):
ARM: uaccess: Implement missing __get_user_asm_dword()
uaccess: Provide ASM GOTO safe wrappers for unsafe_*_user()
x86/uaccess: Use unsafe wrappers for ASM GOTO
powerpc/uaccess: Use unsafe wrappers for ASM GOTO
riscv/uaccess: Use unsafe wrappers for ASM GOTO
s390/uaccess: Use unsafe wrappers for ASM GOTO
arm64: uaccess: Use unsafe wrappers for ASM GOTO
uaccess: Provide scoped user access regions
uaccess: Provide put/get_user_inline()
futex: Convert to get/put_user_inline()
x86/futex: Convert to scoped user access
select: Convert to scoped user access
rseq: Avoid pointless evaluation in __rseq_notify_resume()
rseq: Condense the inline stubs
rseq: Move algorithm comment to top
rseq: Remove the ksig argument from rseq_handle_notify_resume()
rseq: Simplify registration
rseq: Simplify the event notification
rseq, virt: Retrigger RSEQ after vcpu_run()
rseq: Avoid CPU/MM CID updates when no event pending
rseq: Introduce struct rseq_data
entry: Clean up header
entry: Remove syscall_enter_from_user_mode_prepare()
entry: Inline irqentry_enter/exit_from/to_user_mode()
sched: Move MM CID related functions to sched.h
rseq: Cache CPU ID and MM CID values
rseq: Record interrupt from user space
rseq: Provide tracepoint wrappers for inline code
rseq: Expose lightweight statistics in debugfs
rseq: Provide static branch for runtime debugging
rseq: Provide and use rseq_update_user_cs()
rseq: Replace the original debug implementation
rseq: Make exit debugging static branch based
rseq: Use static branch for syscall exit debug when GENERIC_IRQ_ENTRY=y
rseq: Provide and use rseq_set_ids()
rseq: Separate the signal delivery path
rseq: Rework the TIF_NOTIFY handler
rseq: Optimize event setting
rseq: Implement fast path for exit to user
rseq: Switch to fast path processing on exit to user
entry: Split up exit_to_user_mode_prepare()
rseq: Split up rseq_exit_to_user_mode()
rseq: Switch to TIF_RSEQ if supported
sched/mmcid: Revert the complex CID management
sched/mmcid: Use proper data structures
sched/mmcid: Cacheline align MM CID storage
sched: Fixup whitespace damage
sched/mmcid: Move scheduler code out of global header
sched/mmcid: Prevent pointless work in mm_update_cpus_allowed()
cpumask: Introduce cpumask_weighted_or()
sched/mmcid: Use cpumask_weighted_or()
cpumask: Cache num_possible_cpus()
sched/mmcid: Convert mm CID mask to a bitmap
signal: Move MMCID exit out of sighand lock
sched/mmcid: Move initialization out of line
sched/mmcid: Provide precomputed maximal value
sched/mmcid: Serialize sched_mm_cid_fork()/exit() with a mutex
sched/mmcid: Introduce per task/CPU ownership infrastructure
sched/mmcid: Provide new scheduler CID mechanism
sched/mmcid: Provide CID ownership mode fixup functions
irqwork: Move data struct to a types header
sched/mmcid: Implement deferred mode change
sched/mmcid: Switch over to the new mechanism
Documentation/admin-guide/kernel-parameters.txt | 4 +
arch/arm/include/asm/uaccess.h | 26 +-
arch/arm64/include/asm/uaccess.h | 4 +-
arch/arm64/kernel/entry-common.c | 2 +-
arch/powerpc/include/asm/uaccess.h | 8 +-
arch/riscv/include/asm/uaccess.h | 8 +-
arch/s390/include/asm/uaccess.h | 4 +-
arch/x86/entry/syscall_32.c | 3 +-
arch/x86/include/asm/futex.h | 75 +--
arch/x86/include/asm/ptrace.h | 20 +-
arch/x86/include/asm/uaccess.h | 12 +-
drivers/hv/mshv_root_main.c | 3 +
fs/binfmt_elf.c | 2 +-
fs/exec.c | 2 +-
fs/select.c | 12 +-
include/asm-generic/thread_info_tif.h | 3 +
include/linux/bitmap.h | 15 +
include/linux/cleanup.h | 30 +-
include/linux/cpumask.h | 26 +-
include/linux/entry-common.h | 38 +-
include/linux/irq-entry-common.h | 75 ++-
include/linux/irq_work.h | 9 +-
include/linux/irq_work_types.h | 14 +
include/linux/mm.h | 25 -
include/linux/mm_types.h | 128 +---
include/linux/resume_user_mode.h | 2 +-
include/linux/rseq.h | 214 +++---
include/linux/rseq_entry.h | 616 ++++++++++++++++++
include/linux/rseq_types.h | 164 +++++
include/linux/sched.h | 57 +-
include/linux/thread_info.h | 5 +
include/linux/uaccess.h | 314 ++++++++-
include/trace/events/rseq.h | 4 +-
include/uapi/linux/rseq.h | 21 +-
init/Kconfig | 28 +-
init/init_task.c | 3 +
kernel/cpu.c | 19 +
kernel/entry/common.c | 39 +-
kernel/entry/syscall-common.c | 8 -
kernel/exit.c | 1 +
kernel/fork.c | 7 +-
kernel/futex/core.c | 4 +-
kernel/futex/futex.h | 58 +-
kernel/ptrace.c | 6 +-
kernel/rseq.c | 655 ++++++++-----------
kernel/sched/core.c | 827 ++++++++++++------------
kernel/sched/membarrier.c | 8 +-
kernel/sched/sched.h | 392 +++++------
kernel/signal.c | 2 -
lib/bitmap.c | 6 +
virt/kvm/kvm_main.c | 7 +
51 files changed, 2458 insertions(+), 1557 deletions(-)
create mode 100644 include/linux/irq_work_types.h
create mode 100644 include/linux/rseq_entry.h
create mode 100644 include/linux/rseq_types.h
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 6c42061ca20e..e63827475792 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -6500,6 +6500,10 @@
Memory area to be used by remote processor image,
managed by CMA.
+ rseq_debug= [KNL] Enable or disable restartable sequence
+ debug mode. Defaults to CONFIG_RSEQ_DEBUG_DEFAULT_ENABLE.
+ Format: <bool>
+
rt_group_sched= [KNL] Enable or disable SCHED_RR/FIFO group scheduling
when CONFIG_RT_GROUP_SCHED=y. Defaults to
!CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED.
diff --git a/arch/arm/include/asm/uaccess.h b/arch/arm/include/asm/uaccess.h
index f90be312418e..d6ae80b5df36 100644
--- a/arch/arm/include/asm/uaccess.h
+++ b/arch/arm/include/asm/uaccess.h
@@ -283,10 +283,17 @@ extern int __put_user_8(void *, unsigned long long);
__gu_err; \
})
+/*
+ * This is a type: either unsigned long, if the argument fits into
+ * that type, or otherwise unsigned long long.
+ */
+#define __long_type(x) \
+ __typeof__(__builtin_choose_expr(sizeof(x) > sizeof(0UL), 0ULL, 0UL))
+
#define __get_user_err(x, ptr, err, __t) \
do { \
unsigned long __gu_addr = (unsigned long)(ptr); \
- unsigned long __gu_val; \
+ __long_type(x) __gu_val; \
unsigned int __ua_flags; \
__chk_user_ptr(ptr); \
might_fault(); \
@@ -295,6 +302,7 @@ do { \
case 1: __get_user_asm_byte(__gu_val, __gu_addr, err, __t); break; \
case 2: __get_user_asm_half(__gu_val, __gu_addr, err, __t); break; \
case 4: __get_user_asm_word(__gu_val, __gu_addr, err, __t); break; \
+ case 8: __get_user_asm_dword(__gu_val, __gu_addr, err, __t); break; \
default: (__gu_val) = __get_user_bad(); \
} \
uaccess_restore(__ua_flags); \
@@ -353,6 +361,22 @@ do { \
#define __get_user_asm_word(x, addr, err, __t) \
__get_user_asm(x, addr, err, "ldr" __t)
+#ifdef __ARMEB__
+#define __WORD0_OFFS 4
+#define __WORD1_OFFS 0
+#else
+#define __WORD0_OFFS 0
+#define __WORD1_OFFS 4
+#endif
+
+#define __get_user_asm_dword(x, addr, err, __t) \
+ ({ \
+ unsigned long __w0, __w1; \
+ __get_user_asm(__w0, addr + __WORD0_OFFS, err, "ldr" __t); \
+ __get_user_asm(__w1, addr + __WORD1_OFFS, err, "ldr" __t); \
+ (x) = ((u64)__w1 << 32) | (u64) __w0; \
+})
+
#define __put_user_switch(x, ptr, __err, __fn) \
do { \
const __typeof__(*(ptr)) __user *__pu_ptr = (ptr); \
diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uaccess.h
index 1aa4ecb73429..6490930deef8 100644
--- a/arch/arm64/include/asm/uaccess.h
+++ b/arch/arm64/include/asm/uaccess.h
@@ -422,9 +422,9 @@ static __must_check __always_inline bool user_access_begin(const void __user *pt
}
#define user_access_begin(a,b) user_access_begin(a,b)
#define user_access_end() uaccess_ttbr0_disable()
-#define unsafe_put_user(x, ptr, label) \
+#define arch_unsafe_put_user(x, ptr, label) \
__raw_put_mem("sttr", x, uaccess_mask_ptr(ptr), label, U)
-#define unsafe_get_user(x, ptr, label) \
+#define arch_unsafe_get_user(x, ptr, label) \
__raw_get_mem("ldtr", x, uaccess_mask_ptr(ptr), label, U)
/*
diff --git a/arch/arm64/kernel/entry-common.c b/arch/arm64/kernel/entry-common.c
index a9c81715ce59..0a97e2621f60 100644
--- a/arch/arm64/kernel/entry-common.c
+++ b/arch/arm64/kernel/entry-common.c
@@ -100,7 +100,7 @@ static __always_inline void arm64_enter_from_user_mode(struct pt_regs *regs)
static __always_inline void arm64_exit_to_user_mode(struct pt_regs *regs)
{
local_irq_disable();
- exit_to_user_mode_prepare(regs);
+ exit_to_user_mode_prepare_legacy(regs);
local_daif_mask();
mte_check_tfsr_exit();
exit_to_user_mode();
diff --git a/arch/powerpc/include/asm/uaccess.h b/arch/powerpc/include/asm/uaccess.h
index 4f5a46a77fa2..784a00e681fa 100644
--- a/arch/powerpc/include/asm/uaccess.h
+++ b/arch/powerpc/include/asm/uaccess.h
@@ -451,7 +451,7 @@ user_write_access_begin(const void __user *ptr, size_t len)
#define user_write_access_begin user_write_access_begin
#define user_write_access_end prevent_current_write_to_user
-#define unsafe_get_user(x, p, e) do { \
+#define arch_unsafe_get_user(x, p, e) do { \
__long_type(*(p)) __gu_val; \
__typeof__(*(p)) __user *__gu_addr = (p); \
\
@@ -459,7 +459,7 @@ user_write_access_begin(const void __user *ptr, size_t len)
(x) = (__typeof__(*(p)))__gu_val; \
} while (0)
-#define unsafe_put_user(x, p, e) \
+#define arch_unsafe_put_user(x, p, e) \
__put_user_size_goto((__typeof__(*(p)))(x), (p), sizeof(*(p)), e)
#define unsafe_copy_from_user(d, s, l, e) \
@@ -504,11 +504,11 @@ do { \
unsafe_put_user(*(u8*)(_src + _i), (u8 __user *)(_dst + _i), e); \
} while (0)
-#define __get_kernel_nofault(dst, src, type, err_label) \
+#define arch_get_kernel_nofault(dst, src, type, err_label) \
__get_user_size_goto(*((type *)(dst)), \
(__force type __user *)(src), sizeof(type), err_label)
-#define __put_kernel_nofault(dst, src, type, err_label) \
+#define arch_put_kernel_nofault(dst, src, type, err_label) \
__put_user_size_goto(*((type *)(src)), \
(__force type __user *)(dst), sizeof(type), err_label)
diff --git a/arch/riscv/include/asm/uaccess.h b/arch/riscv/include/asm/uaccess.h
index f5f4f7f85543..36bba6720c26 100644
--- a/arch/riscv/include/asm/uaccess.h
+++ b/arch/riscv/include/asm/uaccess.h
@@ -437,10 +437,10 @@ unsigned long __must_check clear_user(void __user *to, unsigned long n)
__clear_user(untagged_addr(to), n) : n;
}
-#define __get_kernel_nofault(dst, src, type, err_label) \
+#define arch_get_kernel_nofault(dst, src, type, err_label) \
__get_user_nocheck(*((type *)(dst)), (__force __user type *)(src), err_label)
-#define __put_kernel_nofault(dst, src, type, err_label) \
+#define arch_put_kernel_nofault(dst, src, type, err_label) \
__put_user_nocheck(*((type *)(src)), (__force __user type *)(dst), err_label)
static __must_check __always_inline bool user_access_begin(const void __user *ptr, size_t len)
@@ -460,10 +460,10 @@ static inline void user_access_restore(unsigned long enabled) { }
* We want the unsafe accessors to always be inlined and use
* the error labels - thus the macro games.
*/
-#define unsafe_put_user(x, ptr, label) \
+#define arch_unsafe_put_user(x, ptr, label) \
__put_user_nocheck(x, (ptr), label)
-#define unsafe_get_user(x, ptr, label) do { \
+#define arch_unsafe_get_user(x, ptr, label) do { \
__inttype(*(ptr)) __gu_val; \
__get_user_nocheck(__gu_val, (ptr), label); \
(x) = (__force __typeof__(*(ptr)))__gu_val; \
diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h
index 3e5b8b677057..c5e02addcd67 100644
--- a/arch/s390/include/asm/uaccess.h
+++ b/arch/s390/include/asm/uaccess.h
@@ -468,8 +468,8 @@ do { \
#endif /* CONFIG_CC_HAS_ASM_GOTO_OUTPUT && CONFIG_CC_HAS_ASM_AOR_FORMAT_FLAGS */
-#define __get_kernel_nofault __mvc_kernel_nofault
-#define __put_kernel_nofault __mvc_kernel_nofault
+#define arch_get_kernel_nofault __mvc_kernel_nofault
+#define arch_put_kernel_nofault __mvc_kernel_nofault
void __cmpxchg_user_key_called_with_bad_pointer(void);
diff --git a/arch/x86/entry/syscall_32.c b/arch/x86/entry/syscall_32.c
index 2b15ea17bb7c..a67a644d0cfe 100644
--- a/arch/x86/entry/syscall_32.c
+++ b/arch/x86/entry/syscall_32.c
@@ -274,9 +274,10 @@ static noinstr bool __do_fast_syscall_32(struct pt_regs *regs)
* fetch EBP before invoking any of the syscall entry work
* functions.
*/
- syscall_enter_from_user_mode_prepare(regs);
+ enter_from_user_mode(regs);
instrumentation_begin();
+ local_irq_enable();
/* Fetch EBP from where the vDSO stashed it. */
if (IS_ENABLED(CONFIG_X86_64)) {
/*
diff --git a/arch/x86/include/asm/futex.h b/arch/x86/include/asm/futex.h
index 6e2458088800..fe5d9a10d900 100644
--- a/arch/x86/include/asm/futex.h
+++ b/arch/x86/include/asm/futex.h
@@ -46,38 +46,31 @@ do { \
} while(0)
static __always_inline int arch_futex_atomic_op_inuser(int op, int oparg, int *oval,
- u32 __user *uaddr)
+ u32 __user *uaddr)
{
- if (can_do_masked_user_access())
- uaddr = masked_user_access_begin(uaddr);
- else if (!user_access_begin(uaddr, sizeof(u32)))
- return -EFAULT;
-
- switch (op) {
- case FUTEX_OP_SET:
- unsafe_atomic_op1("xchgl %0, %2", oval, uaddr, oparg, Efault);
- break;
- case FUTEX_OP_ADD:
- unsafe_atomic_op1(LOCK_PREFIX "xaddl %0, %2", oval,
- uaddr, oparg, Efault);
- break;
- case FUTEX_OP_OR:
- unsafe_atomic_op2("orl %4, %3", oval, uaddr, oparg, Efault);
- break;
- case FUTEX_OP_ANDN:
- unsafe_atomic_op2("andl %4, %3", oval, uaddr, ~oparg, Efault);
- break;
- case FUTEX_OP_XOR:
- unsafe_atomic_op2("xorl %4, %3", oval, uaddr, oparg, Efault);
- break;
- default:
- user_access_end();
- return -ENOSYS;
+ scoped_user_rw_access(uaddr, Efault) {
+ switch (op) {
+ case FUTEX_OP_SET:
+ unsafe_atomic_op1("xchgl %0, %2", oval, uaddr, oparg, Efault);
+ break;
+ case FUTEX_OP_ADD:
+ unsafe_atomic_op1(LOCK_PREFIX "xaddl %0, %2", oval, uaddr, oparg, Efault);
+ break;
+ case FUTEX_OP_OR:
+ unsafe_atomic_op2("orl %4, %3", oval, uaddr, oparg, Efault);
+ break;
+ case FUTEX_OP_ANDN:
+ unsafe_atomic_op2("andl %4, %3", oval, uaddr, ~oparg, Efault);
+ break;
+ case FUTEX_OP_XOR:
+ unsafe_atomic_op2("xorl %4, %3", oval, uaddr, oparg, Efault);
+ break;
+ default:
+ return -ENOSYS;
+ }
}
- user_access_end();
return 0;
Efault:
- user_access_end();
return -EFAULT;
}
@@ -86,21 +79,19 @@ static inline int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
{
int ret = 0;
- if (can_do_masked_user_access())
- uaddr = masked_user_access_begin(uaddr);
- else if (!user_access_begin(uaddr, sizeof(u32)))
- return -EFAULT;
- asm volatile("\n"
- "1:\t" LOCK_PREFIX "cmpxchgl %3, %2\n"
- "2:\n"
- _ASM_EXTABLE_TYPE_REG(1b, 2b, EX_TYPE_EFAULT_REG, %0) \
- : "+r" (ret), "=a" (oldval), "+m" (*uaddr)
- : "r" (newval), "1" (oldval)
- : "memory"
- );
- user_access_end();
- *uval = oldval;
+ scoped_user_rw_access(uaddr, Efault) {
+ asm_inline volatile("\n"
+ "1:\t" LOCK_PREFIX "cmpxchgl %3, %2\n"
+ "2:\n"
+ _ASM_EXTABLE_TYPE_REG(1b, 2b, EX_TYPE_EFAULT_REG, %0)
+ : "+r" (ret), "=a" (oldval), "+m" (*uaddr)
+ : "r" (newval), "1" (oldval)
+ : "memory");
+ *uval = oldval;
+ }
return ret;
+Efault:
+ return -EFAULT;
}
#endif
diff --git a/arch/x86/include/asm/ptrace.h b/arch/x86/include/asm/ptrace.h
index 50f75467f73d..b5dec859bc75 100644
--- a/arch/x86/include/asm/ptrace.h
+++ b/arch/x86/include/asm/ptrace.h
@@ -187,12 +187,12 @@ convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs);
extern void send_sigtrap(struct pt_regs *regs, int error_code, int si_code);
-static inline unsigned long regs_return_value(struct pt_regs *regs)
+static __always_inline unsigned long regs_return_value(struct pt_regs *regs)
{
return regs->ax;
}
-static inline void regs_set_return_value(struct pt_regs *regs, unsigned long rc)
+static __always_inline void regs_set_return_value(struct pt_regs *regs, unsigned long rc)
{
regs->ax = rc;
}
@@ -277,34 +277,34 @@ static __always_inline bool ip_within_syscall_gap(struct pt_regs *regs)
}
#endif
-static inline unsigned long kernel_stack_pointer(struct pt_regs *regs)
+static __always_inline unsigned long kernel_stack_pointer(struct pt_regs *regs)
{
return regs->sp;
}
-static inline unsigned long instruction_pointer(struct pt_regs *regs)
+static __always_inline unsigned long instruction_pointer(struct pt_regs *regs)
{
return regs->ip;
}
-static inline void instruction_pointer_set(struct pt_regs *regs,
- unsigned long val)
+static __always_inline
+void instruction_pointer_set(struct pt_regs *regs, unsigned long val)
{
regs->ip = val;
}
-static inline unsigned long frame_pointer(struct pt_regs *regs)
+static __always_inline unsigned long frame_pointer(struct pt_regs *regs)
{
return regs->bp;
}
-static inline unsigned long user_stack_pointer(struct pt_regs *regs)
+static __always_inline unsigned long user_stack_pointer(struct pt_regs *regs)
{
return regs->sp;
}
-static inline void user_stack_pointer_set(struct pt_regs *regs,
- unsigned long val)
+static __always_inline
+void user_stack_pointer_set(struct pt_regs *regs, unsigned long val)
{
regs->sp = val;
}
diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
index 91a3fb8ae7ff..367297b188c3 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -528,18 +528,18 @@ static __must_check __always_inline bool user_access_begin(const void __user *pt
#define user_access_save() smap_save()
#define user_access_restore(x) smap_restore(x)
-#define unsafe_put_user(x, ptr, label) \
+#define arch_unsafe_put_user(x, ptr, label) \
__put_user_size((__typeof__(*(ptr)))(x), (ptr), sizeof(*(ptr)), label)
#ifdef CONFIG_CC_HAS_ASM_GOTO_OUTPUT
-#define unsafe_get_user(x, ptr, err_label) \
+#define arch_unsafe_get_user(x, ptr, err_label) \
do { \
__inttype(*(ptr)) __gu_val; \
__get_user_size(__gu_val, (ptr), sizeof(*(ptr)), err_label); \
(x) = (__force __typeof__(*(ptr)))__gu_val; \
} while (0)
#else // !CONFIG_CC_HAS_ASM_GOTO_OUTPUT
-#define unsafe_get_user(x, ptr, err_label) \
+#define arch_unsafe_get_user(x, ptr, err_label) \
do { \
int __gu_err; \
__inttype(*(ptr)) __gu_val; \
@@ -618,11 +618,11 @@ do { \
} while (0)
#ifdef CONFIG_CC_HAS_ASM_GOTO_OUTPUT
-#define __get_kernel_nofault(dst, src, type, err_label) \
+#define arch_get_kernel_nofault(dst, src, type, err_label) \
__get_user_size(*((type *)(dst)), (__force type __user *)(src), \
sizeof(type), err_label)
#else // !CONFIG_CC_HAS_ASM_GOTO_OUTPUT
-#define __get_kernel_nofault(dst, src, type, err_label) \
+#define arch_get_kernel_nofault(dst, src, type, err_label) \
do { \
int __kr_err; \
\
@@ -633,7 +633,7 @@ do { \
} while (0)
#endif // CONFIG_CC_HAS_ASM_GOTO_OUTPUT
-#define __put_kernel_nofault(dst, src, type, err_label) \
+#define arch_put_kernel_nofault(dst, src, type, err_label) \
__put_user_size(*((type *)(src)), (__force type __user *)(dst), \
sizeof(type), err_label)
diff --git a/drivers/hv/mshv_root_main.c b/drivers/hv/mshv_root_main.c
index e3b2bd417c46..a21a0eb0f5be 100644
--- a/drivers/hv/mshv_root_main.c
+++ b/drivers/hv/mshv_root_main.c
@@ -29,6 +29,7 @@
#include <linux/crash_dump.h>
#include <linux/panic_notifier.h>
#include <linux/vmalloc.h>
+#include <linux/rseq.h>
#include "mshv_eventfd.h"
#include "mshv.h"
@@ -560,6 +561,8 @@ static long mshv_run_vp_with_root_scheduler(struct mshv_vp *vp)
}
} while (!vp->run.flags.intercept_suspend);
+ rseq_virt_userspace_exit();
+
return ret;
}
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index e4653bb99946..3eb734c192e9 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -46,7 +46,7 @@
#include <linux/cred.h>
#include <linux/dax.h>
#include <linux/uaccess.h>
-#include <linux/rseq.h>
+#include <uapi/linux/rseq.h>
#include <asm/param.h>
#include <asm/page.h>
diff --git a/fs/exec.c b/fs/exec.c
index 4298e7e08d5d..90e47eb156ab 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1775,7 +1775,7 @@ static int bprm_execve(struct linux_binprm *bprm)
force_fatal_sig(SIGSEGV);
sched_mm_cid_after_execve(current);
- rseq_set_notify_resume(current);
+ rseq_force_update();
current->in_execve = 0;
return retval;
diff --git a/fs/select.c b/fs/select.c
index 082cf60c7e23..65019b8ba3f7 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -776,17 +776,13 @@ static inline int get_sigset_argpack(struct sigset_argpack *to,
{
// the path is hot enough for overhead of copy_from_user() to matter
if (from) {
- if (can_do_masked_user_access())
- from = masked_user_access_begin(from);
- else if (!user_read_access_begin(from, sizeof(*from)))
- return -EFAULT;
- unsafe_get_user(to->p, &from->p, Efault);
- unsafe_get_user(to->size, &from->size, Efault);
- user_read_access_end();
+ scoped_user_read_access(from, Efault) {
+ unsafe_get_user(to->p, &from->p, Efault);
+ unsafe_get_user(to->size, &from->size, Efault);
+ }
}
return 0;
Efault:
- user_read_access_end();
return -EFAULT;
}
diff --git a/include/asm-generic/thread_info_tif.h b/include/asm-generic/thread_info_tif.h
index ee3793e9b1a4..da1610a78f92 100644
--- a/include/asm-generic/thread_info_tif.h
+++ b/include/asm-generic/thread_info_tif.h
@@ -45,4 +45,7 @@
# define _TIF_RESTORE_SIGMASK BIT(TIF_RESTORE_SIGMASK)
#endif
+#define TIF_RSEQ 11 // Run RSEQ fast path
+#define _TIF_RSEQ BIT(TIF_RSEQ)
+
#endif /* _ASM_GENERIC_THREAD_INFO_TIF_H_ */
diff --git a/include/linux/bitmap.h b/include/linux/bitmap.h
index 595217b7a6e7..b0395e4ccf90 100644
--- a/include/linux/bitmap.h
+++ b/include/linux/bitmap.h
@@ -45,6 +45,7 @@ struct device;
* bitmap_copy(dst, src, nbits) *dst = *src
* bitmap_and(dst, src1, src2, nbits) *dst = *src1 & *src2
* bitmap_or(dst, src1, src2, nbits) *dst = *src1 | *src2
+ * bitmap_weighted_or(dst, src1, src2, nbits) *dst = *src1 | *src2. Returns Hamming Weight of dst
* bitmap_xor(dst, src1, src2, nbits) *dst = *src1 ^ *src2
* bitmap_andnot(dst, src1, src2, nbits) *dst = *src1 & ~(*src2)
* bitmap_complement(dst, src, nbits) *dst = ~(*src)
@@ -165,6 +166,8 @@ bool __bitmap_and(unsigned long *dst, const unsigned long *bitmap1,
const unsigned long *bitmap2, unsigned int nbits);
void __bitmap_or(unsigned long *dst, const unsigned long *bitmap1,
const unsigned long *bitmap2, unsigned int nbits);
+unsigned int __bitmap_weighted_or(unsigned long *dst, const unsigned long *bitmap1,
+ const unsigned long *bitmap2, unsigned int nbits);
void __bitmap_xor(unsigned long *dst, const unsigned long *bitmap1,
const unsigned long *bitmap2, unsigned int nbits);
bool __bitmap_andnot(unsigned long *dst, const unsigned long *bitmap1,
@@ -337,6 +340,18 @@ void bitmap_or(unsigned long *dst, const unsigned long *src1,
__bitmap_or(dst, src1, src2, nbits);
}
+static __always_inline
+unsigned int bitmap_weighted_or(unsigned long *dst, const unsigned long *src1,
+ const unsigned long *src2, unsigned int nbits)
+{
+ if (small_const_nbits(nbits)) {
+ *dst = *src1 | *src2;
+ return hweight_long(*dst & BITMAP_LAST_WORD_MASK(nbits));
+ } else {
+ return __bitmap_weighted_or(dst, src1, src2, nbits);
+ }
+}
+
static __always_inline
void bitmap_xor(unsigned long *dst, const unsigned long *src1,
const unsigned long *src2, unsigned int nbits)
diff --git a/include/linux/cleanup.h b/include/linux/cleanup.h
index 2573585b7f06..d1806ac5342c 100644
--- a/include/linux/cleanup.h
+++ b/include/linux/cleanup.h
@@ -208,7 +208,7 @@
*/
#define DEFINE_FREE(_name, _type, _free) \
- static inline void __free_##_name(void *p) { _type _T = *(_type *)p; _free; }
+ static __always_inline void __free_##_name(void *p) { _type _T = *(_type *)p; _free; }
#define __free(_name) __cleanup(__free_##_name)
@@ -220,7 +220,7 @@
__val; \
})
-static inline __must_check
+static __always_inline __must_check
const volatile void * __must_check_fn(const volatile void *val)
{ return val; }
@@ -274,16 +274,16 @@ const volatile void * __must_check_fn(const volatile void *val)
#define DEFINE_CLASS(_name, _type, _exit, _init, _init_args...) \
typedef _type class_##_name##_t; \
-static inline void class_##_name##_destructor(_type *p) \
+static __always_inline void class_##_name##_destructor(_type *p) \
{ _type _T = *p; _exit; } \
-static inline _type class_##_name##_constructor(_init_args) \
+static __always_inline _type class_##_name##_constructor(_init_args) \
{ _type t = _init; return t; }
#define EXTEND_CLASS(_name, ext, _init, _init_args...) \
typedef class_##_name##_t class_##_name##ext##_t; \
-static inline void class_##_name##ext##_destructor(class_##_name##_t *p)\
+static __always_inline void class_##_name##ext##_destructor(class_##_name##_t *p) \
{ class_##_name##_destructor(p); } \
-static inline class_##_name##_t class_##_name##ext##_constructor(_init_args) \
+static __always_inline class_##_name##_t class_##_name##ext##_constructor(_init_args) \
{ class_##_name##_t t = _init; return t; }
#define CLASS(_name, var) \
@@ -347,7 +347,7 @@ static __maybe_unused const bool class_##_name##_is_conditional = _is_cond
})
#define __DEFINE_GUARD_LOCK_PTR(_name, _exp) \
- static inline void *class_##_name##_lock_ptr(class_##_name##_t *_T) \
+ static __always_inline void *class_##_name##_lock_ptr(class_##_name##_t *_T) \
{ \
void *_ptr = (void *)(__force unsigned long)*(_exp); \
if (IS_ERR(_ptr)) { \
@@ -355,7 +355,7 @@ static __maybe_unused const bool class_##_name##_is_conditional = _is_cond
} \
return _ptr; \
} \
- static inline int class_##_name##_lock_err(class_##_name##_t *_T) \
+ static __always_inline int class_##_name##_lock_err(class_##_name##_t *_T) \
{ \
long _rc = (__force unsigned long)*(_exp); \
if (!_rc) { \
@@ -384,9 +384,9 @@ static __maybe_unused const bool class_##_name##_is_conditional = _is_cond
EXTEND_CLASS(_name, _ext, \
({ void *_t = _T; int _RET = (_lock); if (_T && !(_cond)) _t = ERR_PTR(_RET); _t; }), \
class_##_name##_t _T) \
- static inline void * class_##_name##_ext##_lock_ptr(class_##_name##_t *_T) \
+ static __always_inline void * class_##_name##_ext##_lock_ptr(class_##_name##_t *_T) \
{ return class_##_name##_lock_ptr(_T); } \
- static inline int class_##_name##_ext##_lock_err(class_##_name##_t *_T) \
+ static __always_inline int class_##_name##_ext##_lock_err(class_##_name##_t *_T) \
{ return class_##_name##_lock_err(_T); }
/*
@@ -466,7 +466,7 @@ typedef struct { \
__VA_ARGS__; \
} class_##_name##_t; \
\
-static inline void class_##_name##_destructor(class_##_name##_t *_T) \
+static __always_inline void class_##_name##_destructor(class_##_name##_t *_T) \
{ \
if (!__GUARD_IS_ERR(_T->lock)) { _unlock; } \
} \
@@ -474,7 +474,7 @@ static inline void class_##_name##_destructor(class_##_name##_t *_T) \
__DEFINE_GUARD_LOCK_PTR(_name, &_T->lock)
#define __DEFINE_LOCK_GUARD_1(_name, _type, _lock) \
-static inline class_##_name##_t class_##_name##_constructor(_type *l) \
+static __always_inline class_##_name##_t class_##_name##_constructor(_type *l) \
{ \
class_##_name##_t _t = { .lock = l }, *_T = &_t; \
_lock; \
@@ -482,7 +482,7 @@ static inline class_##_name##_t class_##_name##_constructor(_type *l) \
}
#define __DEFINE_LOCK_GUARD_0(_name, _lock) \
-static inline class_##_name##_t class_##_name##_constructor(void) \
+static __always_inline class_##_name##_t class_##_name##_constructor(void) \
{ \
class_##_name##_t _t = { .lock = (void*)1 }, \
*_T __maybe_unused = &_t; \
@@ -508,9 +508,9 @@ __DEFINE_LOCK_GUARD_0(_name, _lock)
if (_T->lock && !(_cond)) _T->lock = ERR_PTR(_RET);\
_t; }), \
typeof_member(class_##_name##_t, lock) l) \
- static inline void * class_##_name##_ext##_lock_ptr(class_##_name##_t *_T) \
+ static __always_inline void * class_##_name##_ext##_lock_ptr(class_##_name##_t *_T) \
{ return class_##_name##_lock_ptr(_T); } \
- static inline int class_##_name##_ext##_lock_err(class_##_name##_t *_T) \
+ static __always_inline int class_##_name##_ext##_lock_err(class_##_name##_t *_T) \
{ return class_##_name##_lock_err(_T); }
#define DEFINE_LOCK_GUARD_1_COND_3(_name, _ext, _lock) \
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index ff8f41ab7ce6..66694ee8d86e 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -126,6 +126,7 @@ extern struct cpumask __cpu_dying_mask;
#define cpu_dying_mask ((const struct cpumask *)&__cpu_dying_mask)
extern atomic_t __num_online_cpus;
+extern unsigned int __num_possible_cpus;
extern cpumask_t cpus_booted_once_mask;
@@ -728,6 +729,22 @@ void cpumask_or(struct cpumask *dstp, const struct cpumask *src1p,
cpumask_bits(src2p), small_cpumask_bits);
}
+/**
+ * cpumask_weighted_or - *dstp = *src1p | *src2p and return the weight of the result
+ * @dstp: the cpumask result
+ * @src1p: the first input
+ * @src2p: the second input
+ *
+ * Return: The number of bits set in the resulting cpumask @dstp
+ */
+static __always_inline
+unsigned int cpumask_weighted_or(struct cpumask *dstp, const struct cpumask *src1p,
+ const struct cpumask *src2p)
+{
+ return bitmap_weighted_or(cpumask_bits(dstp), cpumask_bits(src1p),
+ cpumask_bits(src2p), small_cpumask_bits);
+}
+
/**
* cpumask_xor - *dstp = *src1p ^ *src2p
* @dstp: the cpumask result
@@ -1136,13 +1153,13 @@ void init_cpu_possible(const struct cpumask *src);
#define __assign_cpu(cpu, mask, val) \
__assign_bit(cpumask_check(cpu), cpumask_bits(mask), (val))
-#define set_cpu_possible(cpu, possible) assign_cpu((cpu), &__cpu_possible_mask, (possible))
#define set_cpu_enabled(cpu, enabled) assign_cpu((cpu), &__cpu_enabled_mask, (enabled))
#define set_cpu_present(cpu, present) assign_cpu((cpu), &__cpu_present_mask, (present))
#define set_cpu_active(cpu, active) assign_cpu((cpu), &__cpu_active_mask, (active))
#define set_cpu_dying(cpu, dying) assign_cpu((cpu), &__cpu_dying_mask, (dying))
void set_cpu_online(unsigned int cpu, bool online);
+void set_cpu_possible(unsigned int cpu, bool possible);
/**
* to_cpumask - convert a NR_CPUS bitmap to a struct cpumask *
@@ -1195,7 +1212,12 @@ static __always_inline unsigned int num_online_cpus(void)
{
return raw_atomic_read(&__num_online_cpus);
}
-#define num_possible_cpus() cpumask_weight(cpu_possible_mask)
+
+static __always_inline unsigned int num_possible_cpus(void)
+{
+ return __num_possible_cpus;
+}
+
#define num_enabled_cpus() cpumask_weight(cpu_enabled_mask)
#define num_present_cpus() cpumask_weight(cpu_present_mask)
#define num_active_cpus() cpumask_weight(cpu_active_mask)
diff --git a/include/linux/entry-common.h b/include/linux/entry-common.h
index 7177436f0f9e..87efb38b7081 100644
--- a/include/linux/entry-common.h
+++ b/include/linux/entry-common.h
@@ -3,11 +3,11 @@
#define __LINUX_ENTRYCOMMON_H
#include <linux/irq-entry-common.h>
+#include <linux/livepatch.h>
#include <linux/ptrace.h>
+#include <linux/resume_user_mode.h>
#include <linux/seccomp.h>
#include <linux/sched.h>
-#include <linux/livepatch.h>
-#include <linux/resume_user_mode.h>
#include <asm/entry-common.h>
#include <asm/syscall.h>
@@ -37,6 +37,7 @@
SYSCALL_WORK_SYSCALL_AUDIT | \
SYSCALL_WORK_SYSCALL_USER_DISPATCH | \
ARCH_SYSCALL_WORK_ENTER)
+
#define SYSCALL_WORK_EXIT (SYSCALL_WORK_SYSCALL_TRACEPOINT | \
SYSCALL_WORK_SYSCALL_TRACE | \
SYSCALL_WORK_SYSCALL_AUDIT | \
@@ -44,25 +45,7 @@
SYSCALL_WORK_SYSCALL_EXIT_TRAP | \
ARCH_SYSCALL_WORK_EXIT)
-/**
- * syscall_enter_from_user_mode_prepare - Establish state and enable interrupts
- * @regs: Pointer to currents pt_regs
- *
- * Invoked from architecture specific syscall entry code with interrupts
- * disabled. The calling code has to be non-instrumentable. When the
- * function returns all state is correct, interrupts are enabled and the
- * subsequent functions can be instrumented.
- *
- * This handles lockdep, RCU (context tracking) and tracing state, i.e.
- * the functionality provided by enter_from_user_mode().
- *
- * This is invoked when there is extra architecture specific functionality
- * to be done between establishing state and handling user mode entry work.
- */
-void syscall_enter_from_user_mode_prepare(struct pt_regs *regs);
-
-long syscall_trace_enter(struct pt_regs *regs, long syscall,
- unsigned long work);
+long syscall_trace_enter(struct pt_regs *regs, long syscall, unsigned long work);
/**
* syscall_enter_from_user_mode_work - Check and handle work before invoking
@@ -71,8 +54,8 @@ long syscall_trace_enter(struct pt_regs *regs, long syscall,
* @syscall: The syscall number
*
* Invoked from architecture specific syscall entry code with interrupts
- * enabled after invoking syscall_enter_from_user_mode_prepare() and extra
- * architecture specific work.
+ * enabled after invoking enter_from_user_mode(), enabling interrupts and
+ * extra architecture specific work.
*
* Returns: The original or a modified syscall number
*
@@ -108,8 +91,9 @@ static __always_inline long syscall_enter_from_user_mode_work(struct pt_regs *re
* function returns all state is correct, interrupts are enabled and the
* subsequent functions can be instrumented.
*
- * This is combination of syscall_enter_from_user_mode_prepare() and
- * syscall_enter_from_user_mode_work().
+ * This is the combination of enter_from_user_mode() and
+ * syscall_enter_from_user_mode_work() to be used when there is no
+ * architecture specific work to be done between the two.
*
* Returns: The original or a modified syscall number. See
* syscall_enter_from_user_mode_work() for further explanation.
@@ -162,7 +146,7 @@ static __always_inline void syscall_exit_to_user_mode_work(struct pt_regs *regs)
local_irq_enable();
}
- rseq_syscall(regs);
+ rseq_debug_syscall_return(regs);
/*
* Do one-time syscall specific work. If these work items are
@@ -172,7 +156,7 @@ static __always_inline void syscall_exit_to_user_mode_work(struct pt_regs *regs)
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_prepare(regs);
}
/**
diff --git a/include/linux/irq-entry-common.h b/include/linux/irq-entry-common.h
index d643c7c87822..72e3f7a59469 100644
--- a/include/linux/irq-entry-common.h
+++ b/include/linux/irq-entry-common.h
@@ -2,11 +2,12 @@
#ifndef __LINUX_IRQENTRYCOMMON_H
#define __LINUX_IRQENTRYCOMMON_H
+#include <linux/context_tracking.h>
+#include <linux/kmsan.h>
+#include <linux/rseq_entry.h>
#include <linux/static_call_types.h>
#include <linux/syscalls.h>
-#include <linux/context_tracking.h>
#include <linux/tick.h>
-#include <linux/kmsan.h>
#include <linux/unwind_deferred.h>
#include <asm/entry-common.h>
@@ -29,7 +30,7 @@
#define EXIT_TO_USER_MODE_WORK \
(_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_UPROBE | \
_TIF_NEED_RESCHED | _TIF_NEED_RESCHED_LAZY | \
- _TIF_PATCH_PENDING | _TIF_NOTIFY_SIGNAL | \
+ _TIF_PATCH_PENDING | _TIF_NOTIFY_SIGNAL | _TIF_RSEQ | \
ARCH_EXIT_TO_USER_MODE_WORK)
/**
@@ -67,6 +68,7 @@ static __always_inline bool arch_in_rcu_eqs(void) { return false; }
/**
* enter_from_user_mode - Establish state when coming from user mode
+ * @regs: Pointer to currents pt_regs
*
* Syscall/interrupt entry disables interrupts, but user mode is traced as
* interrupts enabled. Also with NO_HZ_FULL RCU might be idle.
@@ -195,14 +197,11 @@ static __always_inline void arch_exit_to_user_mode(void) { }
*/
void arch_do_signal_or_restart(struct pt_regs *regs);
-/**
- * exit_to_user_mode_loop - do any pending work before leaving to user space
- */
-unsigned long exit_to_user_mode_loop(struct pt_regs *regs,
- unsigned long ti_work);
+/* Handle pending TIF work */
+unsigned long exit_to_user_mode_loop(struct pt_regs *regs, unsigned long ti_work);
/**
- * exit_to_user_mode_prepare - call exit_to_user_mode_loop() if required
+ * __exit_to_user_mode_prepare - call exit_to_user_mode_loop() if required
* @regs: Pointer to pt_regs on entry stack
*
* 1) check that interrupts are disabled
@@ -210,8 +209,10 @@ unsigned long exit_to_user_mode_loop(struct pt_regs *regs,
* 3) call exit_to_user_mode_loop() if any flags from
* EXIT_TO_USER_MODE_WORK are set
* 4) check that interrupts are still disabled
+ *
+ * Don't invoke directly, use the syscall/irqentry_ prefixed variants below
*/
-static __always_inline void exit_to_user_mode_prepare(struct pt_regs *regs)
+static __always_inline void __exit_to_user_mode_prepare(struct pt_regs *regs)
{
unsigned long ti_work;
@@ -225,13 +226,52 @@ static __always_inline void exit_to_user_mode_prepare(struct pt_regs *regs)
ti_work = exit_to_user_mode_loop(regs, ti_work);
arch_exit_to_user_mode_prepare(regs, ti_work);
+}
+static __always_inline void __exit_to_user_mode_validate(void)
+{
/* Ensure that kernel state is sane for a return to userspace */
kmap_assert_nomap();
lockdep_assert_irqs_disabled();
lockdep_sys_exit();
}
+/* Temporary workaround to keep ARM64 alive */
+static __always_inline void exit_to_user_mode_prepare_legacy(struct pt_regs *regs)
+{
+ __exit_to_user_mode_prepare(regs);
+ rseq_exit_to_user_mode_legacy();
+ __exit_to_user_mode_validate();
+}
+
+/**
+ * syscall_exit_to_user_mode_prepare - call exit_to_user_mode_loop() if required
+ * @regs: Pointer to pt_regs on entry stack
+ *
+ * Wrapper around __exit_to_user_mode_prepare() to separate the exit work for
+ * syscalls and interrupts.
+ */
+static __always_inline void syscall_exit_to_user_mode_prepare(struct pt_regs *regs)
+{
+ __exit_to_user_mode_prepare(regs);
+ rseq_syscall_exit_to_user_mode();
+ __exit_to_user_mode_validate();
+}
+
+/**
+ * irqentry_exit_to_user_mode_prepare - call exit_to_user_mode_loop() if required
+ * @regs: Pointer to pt_regs on entry stack
+ *
+ * Wrapper around __exit_to_user_mode_prepare() to separate the exit work for
+ * syscalls and interrupts.
+ */
+static __always_inline void irqentry_exit_to_user_mode_prepare(struct pt_regs *regs)
+{
+ __exit_to_user_mode_prepare(regs);
+ rseq_irqentry_exit_to_user_mode();
+ __exit_to_user_mode_validate();
+}
+
/**
* exit_to_user_mode - Fixup state when exiting to user mode
*
@@ -274,7 +314,11 @@ static __always_inline void exit_to_user_mode(void)
*
* The function establishes state (lockdep, RCU (context tracking), tracing)
*/
-void irqentry_enter_from_user_mode(struct pt_regs *regs);
+static __always_inline void irqentry_enter_from_user_mode(struct pt_regs *regs)
+{
+ enter_from_user_mode(regs);
+ rseq_note_user_irq_entry();
+}
/**
* irqentry_exit_to_user_mode - Interrupt exit work
@@ -289,7 +333,13 @@ void irqentry_enter_from_user_mode(struct pt_regs *regs);
* Interrupt exit is not invoking #1 which is the syscall specific one time
* work.
*/
-void irqentry_exit_to_user_mode(struct pt_regs *regs);
+static __always_inline void irqentry_exit_to_user_mode(struct pt_regs *regs)
+{
+ instrumentation_begin();
+ irqentry_exit_to_user_mode_prepare(regs);
+ instrumentation_end();
+ exit_to_user_mode();
+}
#ifndef irqentry_state
/**
@@ -354,6 +404,7 @@ irqentry_state_t noinstr irqentry_enter(struct pt_regs *regs);
* Conditional reschedule with additional sanity checks.
*/
void raw_irqentry_exit_cond_resched(void);
+
#ifdef CONFIG_PREEMPT_DYNAMIC
#if defined(CONFIG_HAVE_PREEMPT_DYNAMIC_CALL)
#define irqentry_exit_cond_resched_dynamic_enabled raw_irqentry_exit_cond_resched
diff --git a/include/linux/irq_work.h b/include/linux/irq_work.h
index 136f2980cba3..c5afd053ae32 100644
--- a/include/linux/irq_work.h
+++ b/include/linux/irq_work.h
@@ -2,8 +2,9 @@
#ifndef _LINUX_IRQ_WORK_H
#define _LINUX_IRQ_WORK_H
-#include <linux/smp_types.h>
+#include <linux/irq_work_types.h>
#include <linux/rcuwait.h>
+#include <linux/smp_types.h>
/*
* An entry can be in one of four states:
@@ -14,12 +15,6 @@
* busy NULL, 2 -> {free, claimed} : callback in progress, can be claimed
*/
-struct irq_work {
- struct __call_single_node node;
- void (*func)(struct irq_work *);
- struct rcuwait irqwait;
-};
-
#define __IRQ_WORK_INIT(_func, _flags) (struct irq_work){ \
.node = { .u_flags = (_flags), }, \
.func = (_func), \
diff --git a/include/linux/irq_work_types.h b/include/linux/irq_work_types.h
new file mode 100644
index 000000000000..73abec5bb06e
--- /dev/null
+++ b/include/linux/irq_work_types.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_IRQ_WORK_TYPES_H
+#define _LINUX_IRQ_WORK_TYPES_H
+
+#include <linux/smp_types.h>
+#include <linux/types.h>
+
+struct irq_work {
+ struct __call_single_node node;
+ void (*func)(struct irq_work *);
+ struct rcuwait irqwait;
+};
+
+#endif
diff --git a/include/linux/mm.h b/include/linux/mm.h
index d16b33bacc32..17cfbba9914c 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -2401,31 +2401,6 @@ struct zap_details {
/* Set in unmap_vmas() to indicate a final unmap call. Only used by hugetlb */
#define ZAP_FLAG_UNMAP ((__force zap_flags_t) BIT(1))
-#ifdef CONFIG_SCHED_MM_CID
-void sched_mm_cid_before_execve(struct task_struct *t);
-void sched_mm_cid_after_execve(struct task_struct *t);
-void sched_mm_cid_fork(struct task_struct *t);
-void sched_mm_cid_exit_signals(struct task_struct *t);
-static inline int task_mm_cid(struct task_struct *t)
-{
- return t->mm_cid;
-}
-#else
-static inline void sched_mm_cid_before_execve(struct task_struct *t) { }
-static inline void sched_mm_cid_after_execve(struct task_struct *t) { }
-static inline void sched_mm_cid_fork(struct task_struct *t) { }
-static inline void sched_mm_cid_exit_signals(struct task_struct *t) { }
-static inline int task_mm_cid(struct task_struct *t)
-{
- /*
- * Use the processor id as a fall-back when the mm cid feature is
- * disabled. This provides functional per-cpu data structure accesses
- * in user-space, althrough it won't provide the memory usage benefits.
- */
- return raw_smp_processor_id();
-}
-#endif
-
#ifdef CONFIG_MMU
extern bool can_do_mlock(void);
#else
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 90e5790c318f..3b7d05e7169c 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -20,6 +20,7 @@
#include <linux/seqlock.h>
#include <linux/percpu_counter.h>
#include <linux/types.h>
+#include <linux/rseq_types.h>
#include <linux/bitmap.h>
#include <asm/mmu.h>
@@ -922,14 +923,6 @@ struct vm_area_struct {
#define vma_policy(vma) NULL
#endif
-#ifdef CONFIG_SCHED_MM_CID
-struct mm_cid {
- u64 time;
- int cid;
- int recent_cid;
-};
-#endif
-
/*
* Opaque type representing current mm_struct flag state. Must be accessed via
* mm_flags_xxx() helper functions.
@@ -991,44 +984,9 @@ struct mm_struct {
*/
atomic_t mm_users;
-#ifdef CONFIG_SCHED_MM_CID
- /**
- * @pcpu_cid: Per-cpu current cid.
- *
- * Keep track of the currently allocated mm_cid for each cpu.
- * The per-cpu mm_cid values are serialized by their respective
- * runqueue locks.
- */
- struct mm_cid __percpu *pcpu_cid;
- /*
- * @mm_cid_next_scan: Next mm_cid scan (in jiffies).
- *
- * When the next mm_cid scan is due (in jiffies).
- */
- unsigned long mm_cid_next_scan;
- /**
- * @nr_cpus_allowed: Number of CPUs allowed for mm.
- *
- * Number of CPUs allowed in the union of all mm's
- * threads allowed CPUs.
- */
- unsigned int nr_cpus_allowed;
- /**
- * @max_nr_cid: Maximum number of allowed concurrency
- * IDs allocated.
- *
- * Track the highest number of allowed concurrency IDs
- * allocated for the mm.
- */
- atomic_t max_nr_cid;
- /**
- * @cpus_allowed_lock: Lock protecting mm cpus_allowed.
- *
- * Provide mutual exclusion for mm cpus_allowed and
- * mm nr_cpus_allowed updates.
- */
- raw_spinlock_t cpus_allowed_lock;
-#endif
+ /* MM CID related storage */
+ struct mm_mm_cid mm_cid;
+
#ifdef CONFIG_MMU
atomic_long_t pgtables_bytes; /* size of all page tables */
#endif
@@ -1370,37 +1328,6 @@ static inline void vma_iter_init(struct vma_iterator *vmi,
}
#ifdef CONFIG_SCHED_MM_CID
-
-enum mm_cid_state {
- MM_CID_UNSET = -1U, /* Unset state has lazy_put flag set. */
- MM_CID_LAZY_PUT = (1U << 31),
-};
-
-static inline bool mm_cid_is_unset(int cid)
-{
- return cid == MM_CID_UNSET;
-}
-
-static inline bool mm_cid_is_lazy_put(int cid)
-{
- return !mm_cid_is_unset(cid) && (cid & MM_CID_LAZY_PUT);
-}
-
-static inline bool mm_cid_is_valid(int cid)
-{
- return !(cid & MM_CID_LAZY_PUT);
-}
-
-static inline int mm_cid_set_lazy_put(int cid)
-{
- return cid | MM_CID_LAZY_PUT;
-}
-
-static inline int mm_cid_clear_lazy_put(int cid)
-{
- return cid & ~MM_CID_LAZY_PUT;
-}
-
/*
* mm_cpus_allowed: Union of all mm's threads allowed CPUs.
*/
@@ -1415,37 +1342,21 @@ static inline cpumask_t *mm_cpus_allowed(struct mm_struct *mm)
}
/* Accessor for struct mm_struct's cidmask. */
-static inline cpumask_t *mm_cidmask(struct mm_struct *mm)
+static inline unsigned long *mm_cidmask(struct mm_struct *mm)
{
unsigned long cid_bitmap = (unsigned long)mm_cpus_allowed(mm);
/* Skip mm_cpus_allowed */
cid_bitmap += cpumask_size();
- return (struct cpumask *)cid_bitmap;
+ return (unsigned long *)cid_bitmap;
}
-static inline void mm_init_cid(struct mm_struct *mm, struct task_struct *p)
-{
- int i;
-
- for_each_possible_cpu(i) {
- struct mm_cid *pcpu_cid = per_cpu_ptr(mm->pcpu_cid, i);
-
- pcpu_cid->cid = MM_CID_UNSET;
- pcpu_cid->recent_cid = MM_CID_UNSET;
- pcpu_cid->time = 0;
- }
- mm->nr_cpus_allowed = p->nr_cpus_allowed;
- atomic_set(&mm->max_nr_cid, 0);
- raw_spin_lock_init(&mm->cpus_allowed_lock);
- cpumask_copy(mm_cpus_allowed(mm), &p->cpus_mask);
- cpumask_clear(mm_cidmask(mm));
-}
+void mm_init_cid(struct mm_struct *mm, struct task_struct *p);
static inline int mm_alloc_cid_noprof(struct mm_struct *mm, struct task_struct *p)
{
- mm->pcpu_cid = alloc_percpu_noprof(struct mm_cid);
- if (!mm->pcpu_cid)
+ mm->mm_cid.pcpu = alloc_percpu_noprof(struct mm_cid_pcpu);
+ if (!mm->mm_cid.pcpu)
return -ENOMEM;
mm_init_cid(mm, p);
return 0;
@@ -1454,37 +1365,24 @@ static inline int mm_alloc_cid_noprof(struct mm_struct *mm, struct task_struct *
static inline void mm_destroy_cid(struct mm_struct *mm)
{
- free_percpu(mm->pcpu_cid);
- mm->pcpu_cid = NULL;
+ free_percpu(mm->mm_cid.pcpu);
+ mm->mm_cid.pcpu = NULL;
}
static inline unsigned int mm_cid_size(void)
{
- return 2 * cpumask_size(); /* mm_cpus_allowed(), mm_cidmask(). */
+ /* mm_cpus_allowed(), mm_cidmask(). */
+ return cpumask_size() + bitmap_size(num_possible_cpus());
}
-static inline void mm_set_cpus_allowed(struct mm_struct *mm, const struct cpumask *cpumask)
-{
- struct cpumask *mm_allowed = mm_cpus_allowed(mm);
-
- if (!mm)
- return;
- /* The mm_cpus_allowed is the union of each thread allowed CPUs masks. */
- raw_spin_lock(&mm->cpus_allowed_lock);
- cpumask_or(mm_allowed, mm_allowed, cpumask);
- WRITE_ONCE(mm->nr_cpus_allowed, cpumask_weight(mm_allowed));
- raw_spin_unlock(&mm->cpus_allowed_lock);
-}
#else /* CONFIG_SCHED_MM_CID */
static inline void mm_init_cid(struct mm_struct *mm, struct task_struct *p) { }
static inline int mm_alloc_cid(struct mm_struct *mm, struct task_struct *p) { return 0; }
static inline void mm_destroy_cid(struct mm_struct *mm) { }
-
static inline unsigned int mm_cid_size(void)
{
return 0;
}
-static inline void mm_set_cpus_allowed(struct mm_struct *mm, const struct cpumask *cpumask) { }
#endif /* CONFIG_SCHED_MM_CID */
struct mmu_gather;
diff --git a/include/linux/resume_user_mode.h b/include/linux/resume_user_mode.h
index e0135e0adae0..bf92227c78d0 100644
--- a/include/linux/resume_user_mode.h
+++ b/include/linux/resume_user_mode.h
@@ -59,7 +59,7 @@ static inline void resume_user_mode_work(struct pt_regs *regs)
mem_cgroup_handle_over_high(GFP_KERNEL);
blkcg_maybe_throttle_current();
- rseq_handle_notify_resume(NULL, regs);
+ rseq_handle_slowpath(regs);
}
#endif /* LINUX_RESUME_USER_MODE_H */
diff --git a/include/linux/rseq.h b/include/linux/rseq.h
index 69553e7c14c1..2266f4dc77b6 100644
--- a/include/linux/rseq.h
+++ b/include/linux/rseq.h
@@ -3,134 +3,164 @@
#define _LINUX_RSEQ_H
#ifdef CONFIG_RSEQ
-
-#include <linux/preempt.h>
#include <linux/sched.h>
-#ifdef CONFIG_MEMBARRIER
-# define RSEQ_EVENT_GUARD irq
-#else
-# define RSEQ_EVENT_GUARD preempt
-#endif
-
-/*
- * Map the event mask on the user-space ABI enum rseq_cs_flags
- * for direct mask checks.
- */
-enum rseq_event_mask_bits {
- RSEQ_EVENT_PREEMPT_BIT = RSEQ_CS_FLAG_NO_RESTART_ON_PREEMPT_BIT,
- RSEQ_EVENT_SIGNAL_BIT = RSEQ_CS_FLAG_NO_RESTART_ON_SIGNAL_BIT,
- RSEQ_EVENT_MIGRATE_BIT = RSEQ_CS_FLAG_NO_RESTART_ON_MIGRATE_BIT,
-};
-
-enum rseq_event_mask {
- RSEQ_EVENT_PREEMPT = (1U << RSEQ_EVENT_PREEMPT_BIT),
- RSEQ_EVENT_SIGNAL = (1U << RSEQ_EVENT_SIGNAL_BIT),
- RSEQ_EVENT_MIGRATE = (1U << RSEQ_EVENT_MIGRATE_BIT),
-};
-
-static inline void rseq_set_notify_resume(struct task_struct *t)
-{
- if (t->rseq)
- set_tsk_thread_flag(t, TIF_NOTIFY_RESUME);
-}
+#include <uapi/linux/rseq.h>
-void __rseq_handle_notify_resume(struct ksignal *sig, struct pt_regs *regs);
+void __rseq_handle_slowpath(struct pt_regs *regs);
-static inline void rseq_handle_notify_resume(struct ksignal *ksig,
- struct pt_regs *regs)
+/* Invoked from resume_user_mode_work() */
+static inline void rseq_handle_slowpath(struct pt_regs *regs)
{
- if (current->rseq)
- __rseq_handle_notify_resume(ksig, regs);
+ if (IS_ENABLED(CONFIG_GENERIC_ENTRY)) {
+ if (current->rseq.event.slowpath)
+ __rseq_handle_slowpath(regs);
+ } else {
+ /* '&' is intentional to spare one conditional branch */
+ if (current->rseq.event.sched_switch & current->rseq.event.has_rseq)
+ __rseq_handle_slowpath(regs);
+ }
}
-static inline void rseq_signal_deliver(struct ksignal *ksig,
- struct pt_regs *regs)
-{
- scoped_guard(RSEQ_EVENT_GUARD)
- __set_bit(RSEQ_EVENT_SIGNAL_BIT, ¤t->rseq_event_mask);
- rseq_handle_notify_resume(ksig, regs);
-}
+void __rseq_signal_deliver(int sig, struct pt_regs *regs);
-/* rseq_preempt() requires preemption to be disabled. */
-static inline void rseq_preempt(struct task_struct *t)
+/*
+ * Invoked from signal delivery to fixup based on the register context before
+ * switching to the signal delivery context.
+ */
+static inline void rseq_signal_deliver(struct ksignal *ksig, struct pt_regs *regs)
{
- __set_bit(RSEQ_EVENT_PREEMPT_BIT, &t->rseq_event_mask);
- rseq_set_notify_resume(t);
+ if (IS_ENABLED(CONFIG_GENERIC_IRQ_ENTRY)) {
+ /* '&' is intentional to spare one conditional branch */
+ if (current->rseq.event.has_rseq & current->rseq.event.user_irq)
+ __rseq_signal_deliver(ksig->sig, regs);
+ } else {
+ if (current->rseq.event.has_rseq)
+ __rseq_signal_deliver(ksig->sig, regs);
+ }
}
-/* rseq_migrate() requires preemption to be disabled. */
-static inline void rseq_migrate(struct task_struct *t)
+static inline void rseq_raise_notify_resume(struct task_struct *t)
{
- __set_bit(RSEQ_EVENT_MIGRATE_BIT, &t->rseq_event_mask);
- rseq_set_notify_resume(t);
+ set_tsk_thread_flag(t, TIF_RSEQ);
}
-/*
- * If parent process has a registered restartable sequences area, the
- * child inherits. Unregister rseq for a clone with CLONE_VM set.
- */
-static inline void rseq_fork(struct task_struct *t, u64 clone_flags)
+/* Invoked from context switch to force evaluation on exit to user */
+static __always_inline void rseq_sched_switch_event(struct task_struct *t)
{
- if (clone_flags & CLONE_VM) {
- t->rseq = NULL;
- t->rseq_len = 0;
- t->rseq_sig = 0;
- t->rseq_event_mask = 0;
+ struct rseq_event *ev = &t->rseq.event;
+
+ if (IS_ENABLED(CONFIG_GENERIC_IRQ_ENTRY)) {
+ /*
+ * Avoid a boat load of conditionals by using simple logic
+ * to determine whether NOTIFY_RESUME needs to be raised.
+ *
+ * It's required when the CPU or MM CID has changed or
+ * the entry was from user space.
+ */
+ bool raise = (ev->user_irq | ev->ids_changed) & ev->has_rseq;
+
+ if (raise) {
+ ev->sched_switch = true;
+ rseq_raise_notify_resume(t);
+ }
} else {
- t->rseq = current->rseq;
- t->rseq_len = current->rseq_len;
- t->rseq_sig = current->rseq_sig;
- t->rseq_event_mask = current->rseq_event_mask;
+ if (ev->has_rseq) {
+ t->rseq.event.sched_switch = true;
+ rseq_raise_notify_resume(t);
+ }
}
}
-static inline void rseq_execve(struct task_struct *t)
+/*
+ * Invoked from __set_task_cpu() when a task migrates or from
+ * mm_cid_schedin() when the CID changes to enforce an IDs update.
+ *
+ * This does not raise TIF_NOTIFY_RESUME as that happens in
+ * rseq_sched_switch_event().
+ */
+static __always_inline void rseq_sched_set_ids_changed(struct task_struct *t)
{
- t->rseq = NULL;
- t->rseq_len = 0;
- t->rseq_sig = 0;
- t->rseq_event_mask = 0;
+ t->rseq.event.ids_changed = true;
}
-#else
-
-static inline void rseq_set_notify_resume(struct task_struct *t)
-{
-}
-static inline void rseq_handle_notify_resume(struct ksignal *ksig,
- struct pt_regs *regs)
+/* Enforce a full update after RSEQ registration and when execve() failed */
+static inline void rseq_force_update(void)
{
+ if (current->rseq.event.has_rseq) {
+ current->rseq.event.ids_changed = true;
+ current->rseq.event.sched_switch = true;
+ rseq_raise_notify_resume(current);
+ }
}
-static inline void rseq_signal_deliver(struct ksignal *ksig,
- struct pt_regs *regs)
+
+/*
+ * KVM/HYPERV invoke resume_user_mode_work() before entering guest mode,
+ * which clears TIF_NOTIFY_RESUME on architectures that don't use the
+ * generic TIF bits and therefore can't provide a separate TIF_RSEQ flag.
+ *
+ * To avoid updating user space RSEQ in that case just to do it eventually
+ * again before returning to user space, because __rseq_handle_slowpath()
+ * does nothing when invoked with NULL register state.
+ *
+ * After returning from guest mode, before exiting to userspace, hypervisors
+ * must invoke this function to re-raise TIF_NOTIFY_RESUME if necessary.
+ */
+static inline void rseq_virt_userspace_exit(void)
{
+ /*
+ * The generic optimization for deferring RSEQ updates until the next
+ * exit relies on having a dedicated TIF_RSEQ.
+ */
+ if (!IS_ENABLED(CONFIG_HAVE_GENERIC_TIF_BITS) &&
+ current->rseq.event.sched_switch)
+ rseq_raise_notify_resume(current);
}
-static inline void rseq_preempt(struct task_struct *t)
+
+static inline void rseq_reset(struct task_struct *t)
{
+ memset(&t->rseq, 0, sizeof(t->rseq));
+ t->rseq.ids.cpu_id = RSEQ_CPU_ID_UNINITIALIZED;
}
-static inline void rseq_migrate(struct task_struct *t)
+
+static inline void rseq_execve(struct task_struct *t)
{
+ rseq_reset(t);
}
+
+/*
+ * If parent process has a registered restartable sequences area, the
+ * child inherits. Unregister rseq for a clone with CLONE_VM set.
+ *
+ * On fork, keep the IDs (CPU, MMCID) of the parent, which avoids a fault
+ * on the COW page on exit to user space, when the child stays on the same
+ * CPU as the parent. That's obviously not guaranteed, but in overcommit
+ * scenarios it is more likely and optimizes for the fork/exec case without
+ * taking the fault.
+ */
static inline void rseq_fork(struct task_struct *t, u64 clone_flags)
{
-}
-static inline void rseq_execve(struct task_struct *t)
-{
+ if (clone_flags & CLONE_VM)
+ rseq_reset(t);
+ else
+ t->rseq = current->rseq;
}
-#endif
+#else /* CONFIG_RSEQ */
+static inline void rseq_handle_slowpath(struct pt_regs *regs) { }
+static inline void rseq_signal_deliver(struct ksignal *ksig, struct pt_regs *regs) { }
+static inline void rseq_sched_switch_event(struct task_struct *t) { }
+static inline void rseq_sched_set_ids_changed(struct task_struct *t) { }
+static inline void rseq_force_update(void) { }
+static inline void rseq_virt_userspace_exit(void) { }
+static inline void rseq_fork(struct task_struct *t, u64 clone_flags) { }
+static inline void rseq_execve(struct task_struct *t) { }
+#endif /* !CONFIG_RSEQ */
#ifdef CONFIG_DEBUG_RSEQ
-
void rseq_syscall(struct pt_regs *regs);
-
-#else
-
-static inline void rseq_syscall(struct pt_regs *regs)
-{
-}
-
-#endif
+#else /* CONFIG_DEBUG_RSEQ */
+static inline void rseq_syscall(struct pt_regs *regs) { }
+#endif /* !CONFIG_DEBUG_RSEQ */
#endif /* _LINUX_RSEQ_H */
diff --git a/include/linux/rseq_entry.h b/include/linux/rseq_entry.h
new file mode 100644
index 000000000000..c92167ff8a7f
--- /dev/null
+++ b/include/linux/rseq_entry.h
@@ -0,0 +1,616 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_RSEQ_ENTRY_H
+#define _LINUX_RSEQ_ENTRY_H
+
+/* Must be outside the CONFIG_RSEQ guard to resolve the stubs */
+#ifdef CONFIG_RSEQ_STATS
+#include <linux/percpu.h>
+
+struct rseq_stats {
+ unsigned long exit;
+ unsigned long signal;
+ unsigned long slowpath;
+ unsigned long fastpath;
+ unsigned long ids;
+ unsigned long cs;
+ unsigned long clear;
+ unsigned long fixup;
+};
+
+DECLARE_PER_CPU(struct rseq_stats, rseq_stats);
+
+/*
+ * Slow path has interrupts and preemption enabled, but the fast path
+ * runs with interrupts disabled so there is no point in having the
+ * preemption checks implied in __this_cpu_inc() for every operation.
+ */
+#ifdef RSEQ_BUILD_SLOW_PATH
+#define rseq_stat_inc(which) this_cpu_inc((which))
+#else
+#define rseq_stat_inc(which) raw_cpu_inc((which))
+#endif
+
+#else /* CONFIG_RSEQ_STATS */
+#define rseq_stat_inc(x) do { } while (0)
+#endif /* !CONFIG_RSEQ_STATS */
+
+#ifdef CONFIG_RSEQ
+#include <linux/jump_label.h>
+#include <linux/rseq.h>
+#include <linux/uaccess.h>
+
+#include <linux/tracepoint-defs.h>
+
+#ifdef CONFIG_TRACEPOINTS
+DECLARE_TRACEPOINT(rseq_update);
+DECLARE_TRACEPOINT(rseq_ip_fixup);
+void __rseq_trace_update(struct task_struct *t);
+void __rseq_trace_ip_fixup(unsigned long ip, unsigned long start_ip,
+ unsigned long offset, unsigned long abort_ip);
+
+static inline void rseq_trace_update(struct task_struct *t, struct rseq_ids *ids)
+{
+ if (tracepoint_enabled(rseq_update) && ids)
+ __rseq_trace_update(t);
+}
+
+static inline void rseq_trace_ip_fixup(unsigned long ip, unsigned long start_ip,
+ unsigned long offset, unsigned long abort_ip)
+{
+ if (tracepoint_enabled(rseq_ip_fixup))
+ __rseq_trace_ip_fixup(ip, start_ip, offset, abort_ip);
+}
+
+#else /* CONFIG_TRACEPOINT */
+static inline void rseq_trace_update(struct task_struct *t, struct rseq_ids *ids) { }
+static inline void rseq_trace_ip_fixup(unsigned long ip, unsigned long start_ip,
+ unsigned long offset, unsigned long abort_ip) { }
+#endif /* !CONFIG_TRACEPOINT */
+
+DECLARE_STATIC_KEY_MAYBE(CONFIG_RSEQ_DEBUG_DEFAULT_ENABLE, rseq_debug_enabled);
+
+#ifdef RSEQ_BUILD_SLOW_PATH
+#define rseq_inline
+#else
+#define rseq_inline __always_inline
+#endif
+
+bool rseq_debug_update_user_cs(struct task_struct *t, struct pt_regs *regs, unsigned long csaddr);
+bool rseq_debug_validate_ids(struct task_struct *t);
+
+static __always_inline void rseq_note_user_irq_entry(void)
+{
+ if (IS_ENABLED(CONFIG_GENERIC_IRQ_ENTRY))
+ current->rseq.event.user_irq = true;
+}
+
+/*
+ * Check whether there is a valid critical section and whether the
+ * instruction pointer in @regs is inside the critical section.
+ *
+ * - If the critical section is invalid, terminate the task.
+ *
+ * - If valid and the instruction pointer is inside, set it to the abort IP.
+ *
+ * - If valid and the instruction pointer is outside, clear the critical
+ * section address.
+ *
+ * Returns true, if the section was valid and either fixup or clear was
+ * done, false otherwise.
+ *
+ * In the failure case task::rseq_event::fatal is set when a invalid
+ * section was found. It's clear when the failure was an unresolved page
+ * fault.
+ *
+ * If inlined into the exit to user path with interrupts disabled, the
+ * caller has to protect against page faults with pagefault_disable().
+ *
+ * In preemptible task context this would be counterproductive as the page
+ * faults could not be fully resolved. As a consequence unresolved page
+ * faults in task context are fatal too.
+ */
+
+#ifdef RSEQ_BUILD_SLOW_PATH
+/*
+ * The debug version is put out of line, but kept here so the code stays
+ * together.
+ *
+ * @csaddr has already been checked by the caller to be in user space
+ */
+bool rseq_debug_update_user_cs(struct task_struct *t, struct pt_regs *regs,
+ unsigned long csaddr)
+{
+ struct rseq_cs __user *ucs = (struct rseq_cs __user *)(unsigned long)csaddr;
+ u64 start_ip, abort_ip, offset, cs_end, head, tasksize = TASK_SIZE;
+ unsigned long ip = instruction_pointer(regs);
+ u64 __user *uc_head = (u64 __user *) ucs;
+ u32 usig, __user *uc_sig;
+
+ scoped_user_rw_access(ucs, efault) {
+ /*
+ * Evaluate the user pile and exit if one of the conditions
+ * is not fulfilled.
+ */
+ unsafe_get_user(start_ip, &ucs->start_ip, efault);
+ if (unlikely(start_ip >= tasksize))
+ goto die;
+ /* If outside, just clear the critical section. */
+ if (ip < start_ip)
+ goto clear;
+
+ unsafe_get_user(offset, &ucs->post_commit_offset, efault);
+ cs_end = start_ip + offset;
+ /* Check for overflow and wraparound */
+ if (unlikely(cs_end >= tasksize || cs_end < start_ip))
+ goto die;
+
+ /* If not inside, clear it. */
+ if (ip >= cs_end)
+ goto clear;
+
+ unsafe_get_user(abort_ip, &ucs->abort_ip, efault);
+ /* Ensure it's "valid" */
+ if (unlikely(abort_ip >= tasksize || abort_ip < sizeof(*uc_sig)))
+ goto die;
+ /* Validate that the abort IP is not in the critical section */
+ if (unlikely(abort_ip - start_ip < offset))
+ goto die;
+
+ /*
+ * Check version and flags for 0. No point in emitting
+ * deprecated warnings before dying. That could be done in
+ * the slow path eventually, but *shrug*.
+ */
+ unsafe_get_user(head, uc_head, efault);
+ if (unlikely(head))
+ goto die;
+
+ /* abort_ip - 4 is >= 0. See abort_ip check above */
+ uc_sig = (u32 __user *)(unsigned long)(abort_ip - sizeof(*uc_sig));
+ unsafe_get_user(usig, uc_sig, efault);
+ if (unlikely(usig != t->rseq.sig))
+ goto die;
+
+ /* rseq_event.user_irq is only valid if CONFIG_GENERIC_IRQ_ENTRY=y */
+ if (IS_ENABLED(CONFIG_GENERIC_IRQ_ENTRY)) {
+ /* If not in interrupt from user context, let it die */
+ if (unlikely(!t->rseq.event.user_irq))
+ goto die;
+ }
+ unsafe_put_user(0ULL, &t->rseq.usrptr->rseq_cs, efault);
+ instruction_pointer_set(regs, (unsigned long)abort_ip);
+ rseq_stat_inc(rseq_stats.fixup);
+ break;
+ clear:
+ unsafe_put_user(0ULL, &t->rseq.usrptr->rseq_cs, efault);
+ rseq_stat_inc(rseq_stats.clear);
+ abort_ip = 0ULL;
+ }
+
+ if (unlikely(abort_ip))
+ rseq_trace_ip_fixup(ip, start_ip, offset, abort_ip);
+ return true;
+die:
+ t->rseq.event.fatal = true;
+efault:
+ return false;
+}
+
+/*
+ * On debug kernels validate that user space did not mess with it if the
+ * debug branch is enabled.
+ */
+bool rseq_debug_validate_ids(struct task_struct *t)
+{
+ struct rseq __user *rseq = t->rseq.usrptr;
+ u32 cpu_id, uval, node_id;
+
+ /*
+ * On the first exit after registering the rseq region CPU ID is
+ * RSEQ_CPU_ID_UNINITIALIZED and node_id in user space is 0!
+ */
+ node_id = t->rseq.ids.cpu_id != RSEQ_CPU_ID_UNINITIALIZED ?
+ cpu_to_node(t->rseq.ids.cpu_id) : 0;
+
+ scoped_user_read_access(rseq, efault) {
+ unsafe_get_user(cpu_id, &rseq->cpu_id_start, efault);
+ if (cpu_id != t->rseq.ids.cpu_id)
+ goto die;
+ unsafe_get_user(uval, &rseq->cpu_id, efault);
+ if (uval != cpu_id)
+ goto die;
+ unsafe_get_user(uval, &rseq->node_id, efault);
+ if (uval != node_id)
+ goto die;
+ unsafe_get_user(uval, &rseq->mm_cid, efault);
+ if (uval != t->rseq.ids.mm_cid)
+ goto die;
+ }
+ return true;
+die:
+ t->rseq.event.fatal = true;
+efault:
+ return false;
+}
+
+#endif /* RSEQ_BUILD_SLOW_PATH */
+
+/*
+ * This only ensures that abort_ip is in the user address space and
+ * validates that it is preceded by the signature.
+ *
+ * No other sanity checks are done here, that's what the debug code is for.
+ */
+static rseq_inline bool
+rseq_update_user_cs(struct task_struct *t, struct pt_regs *regs, unsigned long csaddr)
+{
+ struct rseq_cs __user *ucs = (struct rseq_cs __user *)(unsigned long)csaddr;
+ unsigned long ip = instruction_pointer(regs);
+ unsigned long tasksize = TASK_SIZE;
+ u64 start_ip, abort_ip, offset;
+ u32 usig, __user *uc_sig;
+
+ rseq_stat_inc(rseq_stats.cs);
+
+ if (unlikely(csaddr >= tasksize)) {
+ t->rseq.event.fatal = true;
+ return false;
+ }
+
+ if (static_branch_unlikely(&rseq_debug_enabled))
+ return rseq_debug_update_user_cs(t, regs, csaddr);
+
+ scoped_user_rw_access(ucs, efault) {
+ unsafe_get_user(start_ip, &ucs->start_ip, efault);
+ unsafe_get_user(offset, &ucs->post_commit_offset, efault);
+ unsafe_get_user(abort_ip, &ucs->abort_ip, efault);
+
+ /*
+ * No sanity checks. If user space screwed it up, it can
+ * keep the pieces. That's what debug code is for.
+ *
+ * If outside, just clear the critical section.
+ */
+ if (ip - start_ip >= offset)
+ goto clear;
+
+ /*
+ * Two requirements for @abort_ip:
+ * - Must be in user space as x86 IRET would happily return to
+ * the kernel.
+ * - The four bytes preceding the instruction at @abort_ip must
+ * contain the signature.
+ *
+ * The latter protects against the following attack vector:
+ *
+ * An attacker with limited abilities to write, creates a critical
+ * section descriptor, sets the abort IP to a library function or
+ * some other ROP gadget and stores the address of the descriptor
+ * in TLS::rseq::rseq_cs. An RSEQ abort would then evade ROP
+ * protection.
+ */
+ if (unlikely(abort_ip >= tasksize || abort_ip < sizeof(*uc_sig)))
+ goto die;
+
+ /* The address is guaranteed to be >= 0 and < TASK_SIZE */
+ uc_sig = (u32 __user *)(unsigned long)(abort_ip - sizeof(*uc_sig));
+ unsafe_get_user(usig, uc_sig, efault);
+ if (unlikely(usig != t->rseq.sig))
+ goto die;
+
+ /* Invalidate the critical section */
+ unsafe_put_user(0ULL, &t->rseq.usrptr->rseq_cs, efault);
+ /* Update the instruction pointer */
+ instruction_pointer_set(regs, (unsigned long)abort_ip);
+ rseq_stat_inc(rseq_stats.fixup);
+ break;
+ clear:
+ unsafe_put_user(0ULL, &t->rseq.usrptr->rseq_cs, efault);
+ rseq_stat_inc(rseq_stats.clear);
+ abort_ip = 0ULL;
+ }
+
+ if (unlikely(abort_ip))
+ rseq_trace_ip_fixup(ip, start_ip, offset, abort_ip);
+ return true;
+die:
+ t->rseq.event.fatal = true;
+efault:
+ return false;
+}
+
+/*
+ * Updates CPU ID, Node ID and MM CID and reads the critical section
+ * address, when @csaddr != NULL. This allows to put the ID update and the
+ * read under the same uaccess region to spare a separate begin/end.
+ *
+ * As this is either invoked from a C wrapper with @csaddr = NULL or from
+ * the fast path code with a valid pointer, a clever compiler should be
+ * able to optimize the read out. Spares a duplicate implementation.
+ *
+ * Returns true, if the operation was successful, false otherwise.
+ *
+ * In the failure case task::rseq_event::fatal is set when invalid data
+ * was found on debug kernels. It's clear when the failure was an unresolved page
+ * fault.
+ *
+ * If inlined into the exit to user path with interrupts disabled, the
+ * caller has to protect against page faults with pagefault_disable().
+ *
+ * In preemptible task context this would be counterproductive as the page
+ * faults could not be fully resolved. As a consequence unresolved page
+ * faults in task context are fatal too.
+ */
+static rseq_inline
+bool rseq_set_ids_get_csaddr(struct task_struct *t, struct rseq_ids *ids,
+ u32 node_id, u64 *csaddr)
+{
+ struct rseq __user *rseq = t->rseq.usrptr;
+
+ if (static_branch_unlikely(&rseq_debug_enabled)) {
+ if (!rseq_debug_validate_ids(t))
+ return false;
+ }
+
+ scoped_user_rw_access(rseq, efault) {
+ unsafe_put_user(ids->cpu_id, &rseq->cpu_id_start, efault);
+ unsafe_put_user(ids->cpu_id, &rseq->cpu_id, efault);
+ unsafe_put_user(node_id, &rseq->node_id, efault);
+ unsafe_put_user(ids->mm_cid, &rseq->mm_cid, efault);
+ if (csaddr)
+ unsafe_get_user(*csaddr, &rseq->rseq_cs, efault);
+ }
+
+ /* Cache the new values */
+ t->rseq.ids.cpu_cid = ids->cpu_cid;
+ rseq_stat_inc(rseq_stats.ids);
+ rseq_trace_update(t, ids);
+ return true;
+efault:
+ return false;
+}
+
+/*
+ * Update user space with new IDs and conditionally check whether the task
+ * is in a critical section.
+ */
+static rseq_inline bool rseq_update_usr(struct task_struct *t, struct pt_regs *regs,
+ struct rseq_ids *ids, u32 node_id)
+{
+ u64 csaddr;
+
+ if (!rseq_set_ids_get_csaddr(t, ids, node_id, &csaddr))
+ return false;
+
+ /*
+ * On architectures which utilize the generic entry code this
+ * allows to skip the critical section when the entry was not from
+ * a user space interrupt, unless debug mode is enabled.
+ */
+ if (IS_ENABLED(CONFIG_GENERIC_IRQ_ENTRY)) {
+ if (!static_branch_unlikely(&rseq_debug_enabled)) {
+ if (likely(!t->rseq.event.user_irq))
+ return true;
+ }
+ }
+ if (likely(!csaddr))
+ return true;
+ /* Sigh, this really needs to do work */
+ return rseq_update_user_cs(t, regs, csaddr);
+}
+
+/*
+ * If you want to use this then convert your architecture to the generic
+ * entry code. I'm tired of building workarounds for people who can't be
+ * bothered to make the maintenance of generic infrastructure less
+ * burdensome. Just sucking everything into the architecture code and
+ * thereby making others chase the horrible hacks and keep them working is
+ * neither acceptable nor sustainable.
+ */
+#ifdef CONFIG_GENERIC_ENTRY
+
+/*
+ * This is inlined into the exit path because:
+ *
+ * 1) It's a one time comparison in the fast path when there is no event to
+ * handle
+ *
+ * 2) The access to the user space rseq memory (TLS) is unlikely to fault
+ * so the straight inline operation is:
+ *
+ * - Four 32-bit stores only if CPU ID/ MM CID need to be updated
+ * - One 64-bit load to retrieve the critical section address
+ *
+ * 3) In the unlikely case that the critical section address is != NULL:
+ *
+ * - One 64-bit load to retrieve the start IP
+ * - One 64-bit load to retrieve the offset for calculating the end
+ * - One 64-bit load to retrieve the abort IP
+ * - One 64-bit load to retrieve the signature
+ * - One store to clear the critical section address
+ *
+ * The non-debug case implements only the minimal required checking. It
+ * provides protection against a rogue abort IP in kernel space, which
+ * would be exploitable at least on x86, and also against a rogue CS
+ * descriptor by checking the signature at the abort IP. Any fallout from
+ * invalid critical section descriptors is a user space problem. The debug
+ * case provides the full set of checks and terminates the task if a
+ * condition is not met.
+ *
+ * In case of a fault or an invalid value, this sets TIF_NOTIFY_RESUME and
+ * tells the caller to loop back into exit_to_user_mode_loop(). The rseq
+ * slow path there will handle the failure.
+ */
+static __always_inline bool rseq_exit_user_update(struct pt_regs *regs, struct task_struct *t)
+{
+ /*
+ * Page faults need to be disabled as this is called with
+ * interrupts disabled
+ */
+ guard(pagefault)();
+ if (likely(!t->rseq.event.ids_changed)) {
+ struct rseq __user *rseq = t->rseq.usrptr;
+ /*
+ * If IDs have not changed rseq_event::user_irq must be true
+ * See rseq_sched_switch_event().
+ */
+ u64 csaddr;
+
+ if (unlikely(get_user_inline(csaddr, &rseq->rseq_cs)))
+ return false;
+
+ if (static_branch_unlikely(&rseq_debug_enabled) || unlikely(csaddr)) {
+ if (unlikely(!rseq_update_user_cs(t, regs, csaddr)))
+ return false;
+ }
+ return true;
+ }
+
+ struct rseq_ids ids = {
+ .cpu_id = task_cpu(t),
+ .mm_cid = task_mm_cid(t),
+ };
+ u32 node_id = cpu_to_node(ids.cpu_id);
+
+ return rseq_update_usr(t, regs, &ids, node_id);
+}
+
+static __always_inline bool __rseq_exit_to_user_mode_restart(struct pt_regs *regs)
+{
+ struct task_struct *t = current;
+
+ /*
+ * If the task did not go through schedule or got the flag enforced
+ * by the rseq syscall or execve, then nothing to do here.
+ *
+ * CPU ID and MM CID can only change when going through a context
+ * switch.
+ *
+ * rseq_sched_switch_event() sets the rseq_event::sched_switch bit
+ * only when rseq_event::has_rseq is true. That conditional is
+ * required to avoid setting the TIF bit if RSEQ is not registered
+ * for a task. rseq_event::sched_switch is cleared when RSEQ is
+ * unregistered by a task so it's sufficient to check for the
+ * sched_switch bit alone.
+ *
+ * A sane compiler requires three instructions for the nothing to do
+ * case including clearing the events, but your mileage might vary.
+ */
+ if (unlikely((t->rseq.event.sched_switch))) {
+ rseq_stat_inc(rseq_stats.fastpath);
+
+ if (unlikely(!rseq_exit_user_update(regs, t)))
+ return true;
+ }
+ /* Clear state so next entry starts from a clean slate */
+ t->rseq.event.events = 0;
+ return false;
+}
+
+/* Required to allow conversion to GENERIC_ENTRY w/o GENERIC_TIF_BITS */
+#ifdef CONFIG_HAVE_GENERIC_TIF_BITS
+static __always_inline bool test_tif_rseq(unsigned long ti_work)
+{
+ return ti_work & _TIF_RSEQ;
+}
+
+static __always_inline void clear_tif_rseq(void)
+{
+ static_assert(TIF_RSEQ != TIF_NOTIFY_RESUME);
+ clear_thread_flag(TIF_RSEQ);
+}
+#else
+static __always_inline bool test_tif_rseq(unsigned long ti_work) { return true; }
+static __always_inline void clear_tif_rseq(void) { }
+#endif
+
+static __always_inline bool
+rseq_exit_to_user_mode_restart(struct pt_regs *regs, unsigned long ti_work)
+{
+ if (likely(!test_tif_rseq(ti_work)))
+ return false;
+
+ if (unlikely(__rseq_exit_to_user_mode_restart(regs))) {
+ current->rseq.event.slowpath = true;
+ set_tsk_thread_flag(current, TIF_NOTIFY_RESUME);
+ return true;
+ }
+
+ clear_tif_rseq();
+ return false;
+}
+
+#else /* CONFIG_GENERIC_ENTRY */
+static inline bool rseq_exit_to_user_mode_restart(struct pt_regs *regs, unsigned long ti_work)
+{
+ return false;
+}
+#endif /* !CONFIG_GENERIC_ENTRY */
+
+static __always_inline void rseq_syscall_exit_to_user_mode(void)
+{
+ struct rseq_event *ev = ¤t->rseq.event;
+
+ rseq_stat_inc(rseq_stats.exit);
+
+ /* Needed to remove the store for the !lockdep case */
+ if (IS_ENABLED(CONFIG_LOCKDEP)) {
+ WARN_ON_ONCE(ev->sched_switch);
+ ev->events = 0;
+ }
+}
+
+static __always_inline void rseq_irqentry_exit_to_user_mode(void)
+{
+ struct rseq_event *ev = ¤t->rseq.event;
+
+ rseq_stat_inc(rseq_stats.exit);
+
+ lockdep_assert_once(!ev->sched_switch);
+
+ /*
+ * Ensure that event (especially user_irq) is cleared when the
+ * interrupt did not result in a schedule and therefore the
+ * rseq processing could not clear it.
+ */
+ ev->events = 0;
+}
+
+/* Required to keep ARM64 working */
+static __always_inline void rseq_exit_to_user_mode_legacy(void)
+{
+ struct rseq_event *ev = ¤t->rseq.event;
+
+ rseq_stat_inc(rseq_stats.exit);
+
+ if (static_branch_unlikely(&rseq_debug_enabled))
+ WARN_ON_ONCE(ev->sched_switch);
+
+ /*
+ * Ensure that event (especially user_irq) is cleared when the
+ * interrupt did not result in a schedule and therefore the
+ * rseq processing did not clear it.
+ */
+ ev->events = 0;
+}
+
+void __rseq_debug_syscall_return(struct pt_regs *regs);
+
+static inline void rseq_debug_syscall_return(struct pt_regs *regs)
+{
+ if (static_branch_unlikely(&rseq_debug_enabled))
+ __rseq_debug_syscall_return(regs);
+}
+#else /* CONFIG_RSEQ */
+static inline void rseq_note_user_irq_entry(void) { }
+static inline bool rseq_exit_to_user_mode_restart(struct pt_regs *regs, unsigned long ti_work)
+{
+ return false;
+}
+static inline void rseq_syscall_exit_to_user_mode(void) { }
+static inline void rseq_irqentry_exit_to_user_mode(void) { }
+static inline void rseq_exit_to_user_mode_legacy(void) { }
+static inline void rseq_debug_syscall_return(struct pt_regs *regs) { }
+#endif /* !CONFIG_RSEQ */
+
+#endif /* _LINUX_RSEQ_ENTRY_H */
diff --git a/include/linux/rseq_types.h b/include/linux/rseq_types.h
new file mode 100644
index 000000000000..332dc14b81c9
--- /dev/null
+++ b/include/linux/rseq_types.h
@@ -0,0 +1,164 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_RSEQ_TYPES_H
+#define _LINUX_RSEQ_TYPES_H
+
+#include <linux/irq_work_types.h>
+#include <linux/types.h>
+#include <linux/workqueue_types.h>
+
+#ifdef CONFIG_RSEQ
+struct rseq;
+
+/**
+ * struct rseq_event - Storage for rseq related event management
+ * @all: Compound to initialize and clear the data efficiently
+ * @events: Compound to access events with a single load/store
+ * @sched_switch: True if the task was scheduled and needs update on
+ * exit to user
+ * @ids_changed: Indicator that IDs need to be updated
+ * @user_irq: True on interrupt entry from user mode
+ * @has_rseq: True if the task has a rseq pointer installed
+ * @error: Compound error code for the slow path to analyze
+ * @fatal: User space data corrupted or invalid
+ * @slowpath: Indicator that slow path processing via TIF_NOTIFY_RESUME
+ * is required
+ *
+ * @sched_switch and @ids_changed must be adjacent and the combo must be
+ * 16bit aligned to allow a single store, when both are set at the same
+ * time in the scheduler.
+ */
+struct rseq_event {
+ union {
+ u64 all;
+ struct {
+ union {
+ u32 events;
+ struct {
+ u8 sched_switch;
+ u8 ids_changed;
+ u8 user_irq;
+ };
+ };
+
+ u8 has_rseq;
+ u8 __pad;
+ union {
+ u16 error;
+ struct {
+ u8 fatal;
+ u8 slowpath;
+ };
+ };
+ };
+ };
+};
+
+/**
+ * struct rseq_ids - Cache for ids, which need to be updated
+ * @cpu_cid: Compound of @cpu_id and @mm_cid to make the
+ * compiler emit a single compare on 64-bit
+ * @cpu_id: The CPU ID which was written last to user space
+ * @mm_cid: The MM CID which was written last to user space
+ *
+ * @cpu_id and @mm_cid are updated when the data is written to user space.
+ */
+struct rseq_ids {
+ union {
+ u64 cpu_cid;
+ struct {
+ u32 cpu_id;
+ u32 mm_cid;
+ };
+ };
+};
+
+/**
+ * struct rseq_data - Storage for all rseq related data
+ * @usrptr: Pointer to the registered user space RSEQ memory
+ * @len: Length of the RSEQ region
+ * @sig: Signature of critial section abort IPs
+ * @event: Storage for event management
+ * @ids: Storage for cached CPU ID and MM CID
+ */
+struct rseq_data {
+ struct rseq __user *usrptr;
+ u32 len;
+ u32 sig;
+ struct rseq_event event;
+ struct rseq_ids ids;
+};
+
+#else /* CONFIG_RSEQ */
+struct rseq_data { };
+#endif /* !CONFIG_RSEQ */
+
+#ifdef CONFIG_SCHED_MM_CID
+
+#define MM_CID_UNSET BIT(31)
+#define MM_CID_ONCPU BIT(30)
+#define MM_CID_TRANSIT BIT(29)
+
+/**
+ * struct sched_mm_cid - Storage for per task MM CID data
+ * @active: MM CID is active for the task
+ * @cid: The CID associated to the task either permanently or
+ * borrowed from the CPU
+ */
+struct sched_mm_cid {
+ unsigned int active;
+ unsigned int cid;
+};
+
+/**
+ * struct mm_cid_pcpu - Storage for per CPU MM_CID data
+ * @cid: The CID associated to the CPU either permanently or
+ * while a task with a CID is running
+ */
+struct mm_cid_pcpu {
+ unsigned int cid;
+}____cacheline_aligned_in_smp;
+
+/**
+ * struct mm_mm_cid - Storage for per MM CID data
+ * @pcpu: Per CPU storage for CIDs associated to a CPU
+ * @percpu: Set, when CIDs are in per CPU mode
+ * @transit: Set to MM_CID_TRANSIT during a mode change transition phase
+ * @max_cids: The exclusive maximum CID value for allocation and convergence
+ * @irq_work: irq_work to handle the affinity mode change case
+ * @work: Regular work to handle the affinity mode change case
+ * @lock: Spinlock to protect against affinity setting which can't take @mutex
+ * @mutex: Mutex to serialize forks and exits related to this mm
+ * @nr_cpus_allowed: The number of CPUs in the per MM allowed CPUs map. The map
+ * is growth only.
+ * @users: The number of tasks sharing this MM. Separate from mm::mm_users
+ * as that is modified by mmget()/mm_put() by other entities which
+ * do not actually share the MM.
+ * @pcpu_thrs: Threshold for switching back from per CPU mode
+ * @update_deferred: A deferred switch back to per task mode is pending.
+ */
+struct mm_mm_cid {
+ /* Hotpath read mostly members */
+ struct mm_cid_pcpu __percpu *pcpu;
+ unsigned int percpu;
+ unsigned int transit;
+ unsigned int max_cids;
+
+ /* Rarely used. Moves @lock and @mutex into the second cacheline */
+ struct irq_work irq_work;
+ struct work_struct work;
+
+ raw_spinlock_t lock;
+ struct mutex mutex;
+
+ /* Low frequency modified */
+ unsigned int nr_cpus_allowed;
+ unsigned int users;
+ unsigned int pcpu_thrs;
+ unsigned int update_deferred;
+}____cacheline_aligned_in_smp;
+#else /* CONFIG_SCHED_MM_CID */
+struct mm_mm_cid { };
+struct sched_mm_cid { };
+#endif /* !CONFIG_SCHED_MM_CID */
+
+#endif
diff --git a/include/linux/sched.h b/include/linux/sched.h
index b469878de25c..9eec409745f8 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -41,7 +41,7 @@
#include <linux/task_io_accounting.h>
#include <linux/posix-timers_types.h>
#include <linux/restart_block.h>
-#include <uapi/linux/rseq.h>
+#include <linux/rseq_types.h>
#include <linux/seqlock_types.h>
#include <linux/kcsan.h>
#include <linux/rv.h>
@@ -1406,33 +1406,8 @@ struct task_struct {
unsigned long numa_pages_migrated;
#endif /* CONFIG_NUMA_BALANCING */
-#ifdef CONFIG_RSEQ
- struct rseq __user *rseq;
- u32 rseq_len;
- u32 rseq_sig;
- /*
- * RmW on rseq_event_mask must be performed atomically
- * with respect to preemption.
- */
- unsigned long rseq_event_mask;
-# ifdef CONFIG_DEBUG_RSEQ
- /*
- * This is a place holder to save a copy of the rseq fields for
- * validation of read-only fields. The struct rseq has a
- * variable-length array at the end, so it cannot be used
- * directly. Reserve a size large enough for the known fields.
- */
- char rseq_fields[sizeof(struct rseq)];
-# endif
-#endif
-
-#ifdef CONFIG_SCHED_MM_CID
- int mm_cid; /* Current cid in mm */
- int last_mm_cid; /* Most recent cid in mm */
- int migrate_from_cpu;
- int mm_cid_active; /* Whether cid bitmap is active */
- struct callback_head cid_work;
-#endif
+ struct rseq_data rseq;
+ struct sched_mm_cid mm_cid;
struct tlbflush_unmap_batch tlb_ubc;
@@ -2318,6 +2293,32 @@ static __always_inline void alloc_tag_restore(struct alloc_tag *tag, struct allo
#define alloc_tag_restore(_tag, _old) do {} while (0)
#endif
+/* Avoids recursive inclusion hell */
+#ifdef CONFIG_SCHED_MM_CID
+void sched_mm_cid_before_execve(struct task_struct *t);
+void sched_mm_cid_after_execve(struct task_struct *t);
+void sched_mm_cid_fork(struct task_struct *t);
+void sched_mm_cid_exit(struct task_struct *t);
+static __always_inline int task_mm_cid(struct task_struct *t)
+{
+ return t->mm_cid.cid & ~(MM_CID_ONCPU | MM_CID_TRANSIT);
+}
+#else
+static inline void sched_mm_cid_before_execve(struct task_struct *t) { }
+static inline void sched_mm_cid_after_execve(struct task_struct *t) { }
+static inline void sched_mm_cid_fork(struct task_struct *t) { }
+static inline void sched_mm_cid_exit(struct task_struct *t) { }
+static __always_inline int task_mm_cid(struct task_struct *t)
+{
+ /*
+ * Use the processor id as a fall-back when the mm cid feature is
+ * disabled. This provides functional per-cpu data structure accesses
+ * in user-space, althrough it won't provide the memory usage benefits.
+ */
+ return task_cpu(t);
+}
+#endif
+
#ifndef MODULE
#ifndef COMPILE_OFFSETS
diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h
index dd925d84fa46..b40de9bab4b7 100644
--- a/include/linux/thread_info.h
+++ b/include/linux/thread_info.h
@@ -67,6 +67,11 @@ enum syscall_work_bit {
#define _TIF_NEED_RESCHED_LAZY _TIF_NEED_RESCHED
#endif
+#ifndef TIF_RSEQ
+# define TIF_RSEQ TIF_NOTIFY_RESUME
+# define _TIF_RSEQ _TIF_NOTIFY_RESUME
+#endif
+
#ifdef __KERNEL__
#ifndef arch_set_restart_data
diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index 1beb5b395d81..be395f5f7ee3 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -2,6 +2,7 @@
#ifndef __LINUX_UACCESS_H__
#define __LINUX_UACCESS_H__
+#include <linux/cleanup.h>
#include <linux/fault-inject-usercopy.h>
#include <linux/instrumented.h>
#include <linux/minmax.h>
@@ -35,9 +36,17 @@
#ifdef masked_user_access_begin
#define can_do_masked_user_access() 1
+# ifndef masked_user_write_access_begin
+# define masked_user_write_access_begin masked_user_access_begin
+# endif
+# ifndef masked_user_read_access_begin
+# define masked_user_read_access_begin masked_user_access_begin
+#endif
#else
#define can_do_masked_user_access() 0
#define masked_user_access_begin(src) NULL
+ #define masked_user_read_access_begin(src) NULL
+ #define masked_user_write_access_begin(src) NULL
#define mask_user_address(src) (src)
#endif
@@ -518,7 +527,34 @@ long strncpy_from_user_nofault(char *dst, const void __user *unsafe_addr,
long count);
long strnlen_user_nofault(const void __user *unsafe_addr, long count);
-#ifndef __get_kernel_nofault
+#ifdef arch_get_kernel_nofault
+/*
+ * Wrap the architecture implementation so that @label can be outside of a
+ * cleanup() scope. A regular C goto works correctly, but ASM goto does
+ * not. Clang rejects such an attempt, but GCC silently emits buggy code.
+ */
+#define __get_kernel_nofault(dst, src, type, label) \
+do { \
+ __label__ local_label; \
+ arch_get_kernel_nofault(dst, src, type, local_label); \
+ if (0) { \
+ local_label: \
+ goto label; \
+ } \
+} while (0)
+
+#define __put_kernel_nofault(dst, src, type, label) \
+do { \
+ __label__ local_label; \
+ arch_put_kernel_nofault(dst, src, type, local_label); \
+ if (0) { \
+ local_label: \
+ goto label; \
+ } \
+} while (0)
+
+#elif !defined(__get_kernel_nofault) /* arch_get_kernel_nofault */
+
#define __get_kernel_nofault(dst, src, type, label) \
do { \
type __user *p = (type __force __user *)(src); \
@@ -535,7 +571,8 @@ do { \
if (__put_user(data, p)) \
goto label; \
} while (0)
-#endif
+
+#endif /* !__get_kernel_nofault */
/**
* get_kernel_nofault(): safely attempt to read from a location
@@ -549,7 +586,42 @@ do { \
copy_from_kernel_nofault(&(val), __gk_ptr, sizeof(val));\
})
-#ifndef user_access_begin
+#ifdef user_access_begin
+
+#ifdef arch_unsafe_get_user
+/*
+ * Wrap the architecture implementation so that @label can be outside of a
+ * cleanup() scope. A regular C goto works correctly, but ASM goto does
+ * not. Clang rejects such an attempt, but GCC silently emits buggy code.
+ *
+ * Some architectures use internal local labels already, but this extra
+ * indirection here is harmless because the compiler optimizes it out
+ * completely in any case. This construct just ensures that the ASM GOTO
+ * target is always in the local scope. The C goto 'label' works correctly
+ * when leaving a cleanup() scope.
+ */
+#define unsafe_get_user(x, ptr, label) \
+do { \
+ __label__ local_label; \
+ arch_unsafe_get_user(x, ptr, local_label); \
+ if (0) { \
+ local_label: \
+ goto label; \
+ } \
+} while (0)
+
+#define unsafe_put_user(x, ptr, label) \
+do { \
+ __label__ local_label; \
+ arch_unsafe_put_user(x, ptr, local_label); \
+ if (0) { \
+ local_label: \
+ goto label; \
+ } \
+} while (0)
+#endif /* arch_unsafe_get_user */
+
+#else /* user_access_begin */
#define user_access_begin(ptr,len) access_ok(ptr, len)
#define user_access_end() do { } while (0)
#define unsafe_op_wrap(op, err) do { if (unlikely(op)) goto err; } while (0)
@@ -559,7 +631,8 @@ do { \
#define unsafe_copy_from_user(d,s,l,e) unsafe_op_wrap(__copy_from_user(d,s,l),e)
static inline unsigned long user_access_save(void) { return 0UL; }
static inline void user_access_restore(unsigned long flags) { }
-#endif
+#endif /* !user_access_begin */
+
#ifndef user_write_access_begin
#define user_write_access_begin user_access_begin
#define user_write_access_end user_access_end
@@ -569,6 +642,239 @@ static inline void user_access_restore(unsigned long flags) { }
#define user_read_access_end user_access_end
#endif
+/* Define RW variant so the below _mode macro expansion works */
+#define masked_user_rw_access_begin(u) masked_user_access_begin(u)
+#define user_rw_access_begin(u, s) user_access_begin(u, s)
+#define user_rw_access_end() user_access_end()
+
+/* Scoped user access */
+#define USER_ACCESS_GUARD(_mode) \
+static __always_inline void __user * \
+class_user_##_mode##_begin(void __user *ptr) \
+{ \
+ return ptr; \
+} \
+ \
+static __always_inline void \
+class_user_##_mode##_end(void __user *ptr) \
+{ \
+ user_##_mode##_access_end(); \
+} \
+ \
+DEFINE_CLASS(user_ ##_mode## _access, void __user *, \
+ class_user_##_mode##_end(_T), \
+ class_user_##_mode##_begin(ptr), void __user *ptr) \
+ \
+static __always_inline class_user_##_mode##_access_t \
+class_user_##_mode##_access_ptr(void __user *scope) \
+{ \
+ return scope; \
+}
+
+USER_ACCESS_GUARD(read)
+USER_ACCESS_GUARD(write)
+USER_ACCESS_GUARD(rw)
+#undef USER_ACCESS_GUARD
+
+/**
+ * __scoped_user_access_begin - Start a scoped user access
+ * @mode: The mode of the access class (read, write, rw)
+ * @uptr: The pointer to access user space memory
+ * @size: Size of the access
+ * @elbl: Error label to goto when the access region is rejected
+ *
+ * Internal helper for __scoped_user_access(). Don't use directly.
+ */
+#define __scoped_user_access_begin(mode, uptr, size, elbl) \
+({ \
+ typeof(uptr) __retptr; \
+ \
+ if (can_do_masked_user_access()) { \
+ __retptr = masked_user_##mode##_access_begin(uptr); \
+ } else { \
+ __retptr = uptr; \
+ if (!user_##mode##_access_begin(uptr, size)) \
+ goto elbl; \
+ } \
+ __retptr; \
+})
+
+/**
+ * __scoped_user_access - Open a scope for user access
+ * @mode: The mode of the access class (read, write, rw)
+ * @uptr: The pointer to access user space memory
+ * @size: Size of the access
+ * @elbl: Error label to goto when the access region is rejected. It
+ * must be placed outside the scope
+ *
+ * If the user access function inside the scope requires a fault label, it
+ * can use @elbl or a different label outside the scope, which requires
+ * that user access which is implemented with ASM GOTO has been properly
+ * wrapped. See unsafe_get_user() for reference.
+ *
+ * scoped_user_rw_access(ptr, efault) {
+ * unsafe_get_user(rval, &ptr->rval, efault);
+ * unsafe_put_user(wval, &ptr->wval, efault);
+ * }
+ * return 0;
+ * efault:
+ * return -EFAULT;
+ *
+ * The scope is internally implemented as a autoterminating nested for()
+ * loop, which can be left with 'return', 'break' and 'goto' at any
+ * point.
+ *
+ * When the scope is left user_##@_mode##_access_end() is automatically
+ * invoked.
+ *
+ * When the architecture supports masked user access and the access region
+ * which is determined by @uptr and @size is not a valid user space
+ * address, i.e. < TASK_SIZE, the scope sets the pointer to a faulting user
+ * space address and does not terminate early. This optimizes for the good
+ * case and lets the performance uncritical bad case go through the fault.
+ *
+ * The eventual modification of the pointer is limited to the scope.
+ * Outside of the scope the original pointer value is unmodified, so that
+ * the original pointer value is available for diagnostic purposes in an
+ * out of scope fault path.
+ *
+ * Nesting scoped user access into a user access scope is invalid and fails
+ * the build. Nesting into other guards, e.g. pagefault is safe.
+ *
+ * The masked variant does not check the size of the access and relies on a
+ * mapping hole (e.g. guard page) to catch an out of range pointer, the
+ * first access to user memory inside the scope has to be within
+ * @uptr ... @uptr + PAGE_SIZE - 1
+ *
+ * Don't use directly. Use scoped_masked_user_$MODE_access() instead.
+ */
+#define __scoped_user_access(mode, uptr, size, elbl) \
+for (bool done = false; !done; done = true) \
+ for (void __user *_tmpptr = __scoped_user_access_begin(mode, uptr, size, elbl); \
+ !done; done = true) \
+ for (CLASS(user_##mode##_access, scope)(_tmpptr); !done; done = true) \
+ /* Force modified pointer usage within the scope */ \
+ for (const typeof(uptr) uptr = _tmpptr; !done; done = true)
+
+/**
+ * scoped_user_read_access_size - Start a scoped user read access with given size
+ * @usrc: Pointer to the user space address to read from
+ * @size: Size of the access starting from @usrc
+ * @elbl: Error label to goto when the access region is rejected
+ *
+ * For further information see __scoped_user_access() above.
+ */
+#define scoped_user_read_access_size(usrc, size, elbl) \
+ __scoped_user_access(read, usrc, size, elbl)
+
+/**
+ * scoped_user_read_access - Start a scoped user read access
+ * @usrc: Pointer to the user space address to read from
+ * @elbl: Error label to goto when the access region is rejected
+ *
+ * The size of the access starting from @usrc is determined via sizeof(*@usrc)).
+ *
+ * For further information see __scoped_user_access() above.
+ */
+#define scoped_user_read_access(usrc, elbl) \
+ scoped_user_read_access_size(usrc, sizeof(*(usrc)), elbl)
+
+/**
+ * scoped_user_write_access_size - Start a scoped user write access with given size
+ * @udst: Pointer to the user space address to write to
+ * @size: Size of the access starting from @udst
+ * @elbl: Error label to goto when the access region is rejected
+ *
+ * For further information see __scoped_user_access() above.
+ */
+#define scoped_user_write_access_size(udst, size, elbl) \
+ __scoped_user_access(write, udst, size, elbl)
+
+/**
+ * scoped_user_write_access - Start a scoped user write access
+ * @udst: Pointer to the user space address to write to
+ * @elbl: Error label to goto when the access region is rejected
+ *
+ * The size of the access starting from @udst is determined via sizeof(*@udst)).
+ *
+ * For further information see __scoped_user_access() above.
+ */
+#define scoped_user_write_access(udst, elbl) \
+ scoped_user_write_access_size(udst, sizeof(*(udst)), elbl)
+
+/**
+ * scoped_user_rw_access_size - Start a scoped user read/write access with given size
+ * @uptr Pointer to the user space address to read from and write to
+ * @size: Size of the access starting from @uptr
+ * @elbl: Error label to goto when the access region is rejected
+ *
+ * For further information see __scoped_user_access() above.
+ */
+#define scoped_user_rw_access_size(uptr, size, elbl) \
+ __scoped_user_access(rw, uptr, size, elbl)
+
+/**
+ * scoped_user_rw_access - Start a scoped user read/write access
+ * @uptr Pointer to the user space address to read from and write to
+ * @elbl: Error label to goto when the access region is rejected
+ *
+ * The size of the access starting from @uptr is determined via sizeof(*@uptr)).
+ *
+ * For further information see __scoped_user_access() above.
+ */
+#define scoped_user_rw_access(uptr, elbl) \
+ scoped_user_rw_access_size(uptr, sizeof(*(uptr)), elbl)
+
+/**
+ * get_user_inline - Read user data inlined
+ * @val: The variable to store the value read from user memory
+ * @usrc: Pointer to the user space memory to read from
+ *
+ * Return: 0 if successful, -EFAULT when faulted
+ *
+ * Inlined variant of get_user(). Only use when there is a demonstrable
+ * performance reason.
+ */
+#define get_user_inline(val, usrc) \
+({ \
+ __label__ efault; \
+ typeof(usrc) _tmpsrc = usrc; \
+ int _ret = 0; \
+ \
+ scoped_user_read_access(_tmpsrc, efault) \
+ unsafe_get_user(val, _tmpsrc, efault); \
+ if (0) { \
+ efault: \
+ _ret = -EFAULT; \
+ } \
+ _ret; \
+})
+
+/**
+ * put_user_inline - Write to user memory inlined
+ * @val: The value to write
+ * @udst: Pointer to the user space memory to write to
+ *
+ * Return: 0 if successful, -EFAULT when faulted
+ *
+ * Inlined variant of put_user(). Only use when there is a demonstrable
+ * performance reason.
+ */
+#define put_user_inline(val, udst) \
+({ \
+ __label__ efault; \
+ typeof(udst) _tmpdst = udst; \
+ int _ret = 0; \
+ \
+ scoped_user_write_access(_tmpdst, efault) \
+ unsafe_put_user(val, _tmpdst, efault); \
+ if (0) { \
+ efault: \
+ _ret = -EFAULT; \
+ } \
+ _ret; \
+})
+
#ifdef CONFIG_HARDENED_USERCOPY
void __noreturn usercopy_abort(const char *name, const char *detail,
bool to_user, unsigned long offset,
diff --git a/include/trace/events/rseq.h b/include/trace/events/rseq.h
index 823b47d1ba1e..ce85d650bf4b 100644
--- a/include/trace/events/rseq.h
+++ b/include/trace/events/rseq.h
@@ -21,9 +21,9 @@ TRACE_EVENT(rseq_update,
),
TP_fast_assign(
- __entry->cpu_id = raw_smp_processor_id();
+ __entry->cpu_id = t->rseq.ids.cpu_id;
__entry->node_id = cpu_to_node(__entry->cpu_id);
- __entry->mm_cid = task_mm_cid(t);
+ __entry->mm_cid = t->rseq.ids.mm_cid;
),
TP_printk("cpu_id=%d node_id=%d mm_cid=%d", __entry->cpu_id,
diff --git a/include/uapi/linux/rseq.h b/include/uapi/linux/rseq.h
index c233aae5eac9..1b76d508400c 100644
--- a/include/uapi/linux/rseq.h
+++ b/include/uapi/linux/rseq.h
@@ -114,20 +114,13 @@ struct rseq {
/*
* Restartable sequences flags field.
*
- * This field should only be updated by the thread which
- * registered this data structure. Read by the kernel.
- * Mainly used for single-stepping through rseq critical sections
- * with debuggers.
- *
- * - RSEQ_CS_FLAG_NO_RESTART_ON_PREEMPT
- * Inhibit instruction sequence block restart on preemption
- * for this thread.
- * - RSEQ_CS_FLAG_NO_RESTART_ON_SIGNAL
- * Inhibit instruction sequence block restart on signal
- * delivery for this thread.
- * - RSEQ_CS_FLAG_NO_RESTART_ON_MIGRATE
- * Inhibit instruction sequence block restart on migration for
- * this thread.
+ * This field was initially intended to allow event masking for
+ * single-stepping through rseq critical sections with debuggers.
+ * The kernel does not support this anymore and the relevant bits
+ * are checked for being always false:
+ * - RSEQ_CS_FLAG_NO_RESTART_ON_PREEMPT
+ * - RSEQ_CS_FLAG_NO_RESTART_ON_SIGNAL
+ * - RSEQ_CS_FLAG_NO_RESTART_ON_MIGRATE
*/
__u32 flags;
diff --git a/init/Kconfig b/init/Kconfig
index cab3ad28ca49..d1c606ec632e 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -1913,10 +1913,36 @@ config RSEQ
If unsure, say Y.
+config RSEQ_STATS
+ default n
+ bool "Enable lightweight statistics of restartable sequences" if EXPERT
+ depends on RSEQ && DEBUG_FS
+ help
+ Enable lightweight counters which expose information about the
+ frequency of RSEQ operations via debugfs. Mostly interesting for
+ kernel debugging or performance analysis. While lightweight it's
+ still adding code into the user/kernel mode transitions.
+
+ If unsure, say N.
+
+config RSEQ_DEBUG_DEFAULT_ENABLE
+ default n
+ bool "Enable restartable sequences debug mode by default" if EXPERT
+ depends on RSEQ
+ help
+ This enables the static branch for debug mode of restartable
+ sequences.
+
+ This also can be controlled on the kernel command line via the
+ command line parameter "rseq_debug=0/1" and through debugfs.
+
+ If unsure, say N.
+
config DEBUG_RSEQ
default n
bool "Enable debugging of rseq() system call" if EXPERT
- depends on RSEQ && DEBUG_KERNEL
+ depends on RSEQ && DEBUG_KERNEL && !GENERIC_ENTRY
+ select RSEQ_DEBUG_DEFAULT_ENABLE
help
Enable extra debugging checks for the rseq system call.
diff --git a/init/init_task.c b/init/init_task.c
index a55e2189206f..5d122699b664 100644
--- a/init/init_task.c
+++ b/init/init_task.c
@@ -223,6 +223,9 @@ struct task_struct init_task __aligned(L1_CACHE_BYTES) = {
#ifdef CONFIG_SECCOMP_FILTER
.seccomp = { .filter_count = ATOMIC_INIT(0) },
#endif
+#ifdef CONFIG_SCHED_MM_CID
+ .mm_cid = { .cid = MM_CID_UNSET, },
+#endif
};
EXPORT_SYMBOL(init_task);
diff --git a/kernel/cpu.c b/kernel/cpu.c
index db9f6c539b28..b674fdf96208 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -3085,10 +3085,13 @@ EXPORT_SYMBOL(cpu_all_bits);
#ifdef CONFIG_INIT_ALL_POSSIBLE
struct cpumask __cpu_possible_mask __ro_after_init
= {CPU_BITS_ALL};
+unsigned int __num_possible_cpus __ro_after_init = NR_CPUS;
#else
struct cpumask __cpu_possible_mask __ro_after_init;
+unsigned int __num_possible_cpus __ro_after_init;
#endif
EXPORT_SYMBOL(__cpu_possible_mask);
+EXPORT_SYMBOL(__num_possible_cpus);
struct cpumask __cpu_online_mask __read_mostly;
EXPORT_SYMBOL(__cpu_online_mask);
@@ -3116,6 +3119,7 @@ void init_cpu_present(const struct cpumask *src)
void init_cpu_possible(const struct cpumask *src)
{
cpumask_copy(&__cpu_possible_mask, src);
+ __num_possible_cpus = cpumask_weight(&__cpu_possible_mask);
}
void set_cpu_online(unsigned int cpu, bool online)
@@ -3139,6 +3143,21 @@ void set_cpu_online(unsigned int cpu, bool online)
}
}
+/*
+ * This should be marked __init, but there is a boatload of call sites
+ * which need to be fixed up to do so. Sigh...
+ */
+void set_cpu_possible(unsigned int cpu, bool possible)
+{
+ if (possible) {
+ if (!cpumask_test_and_set_cpu(cpu, &__cpu_possible_mask))
+ __num_possible_cpus++;
+ } else {
+ if (cpumask_test_and_clear_cpu(cpu, &__cpu_possible_mask))
+ __num_possible_cpus--;
+ }
+}
+
/*
* Activate the first processor.
*/
diff --git a/kernel/entry/common.c b/kernel/entry/common.c
index f62e1d1b2063..5c792b30c58a 100644
--- a/kernel/entry/common.c
+++ b/kernel/entry/common.c
@@ -11,19 +11,20 @@
/* Workaround to allow gradual conversion of architecture code */
void __weak arch_do_signal_or_restart(struct pt_regs *regs) { }
-/**
- * exit_to_user_mode_loop - do any pending work before leaving to user space
- * @regs: Pointer to pt_regs on entry stack
- * @ti_work: TIF work flags as read by the caller
- */
-__always_inline unsigned long exit_to_user_mode_loop(struct pt_regs *regs,
- unsigned long ti_work)
+#ifdef CONFIG_HAVE_GENERIC_TIF_BITS
+#define EXIT_TO_USER_MODE_WORK_LOOP (EXIT_TO_USER_MODE_WORK & ~_TIF_RSEQ)
+#else
+#define EXIT_TO_USER_MODE_WORK_LOOP (EXIT_TO_USER_MODE_WORK)
+#endif
+
+static __always_inline unsigned long __exit_to_user_mode_loop(struct pt_regs *regs,
+ unsigned long ti_work)
{
/*
* Before returning to user space ensure that all pending work
* items have been completed.
*/
- while (ti_work & EXIT_TO_USER_MODE_WORK) {
+ while (ti_work & EXIT_TO_USER_MODE_WORK_LOOP) {
local_irq_enable_exit_to_user(ti_work);
@@ -62,17 +63,21 @@ __always_inline unsigned long exit_to_user_mode_loop(struct pt_regs *regs,
return ti_work;
}
-noinstr void irqentry_enter_from_user_mode(struct pt_regs *regs)
+/**
+ * exit_to_user_mode_loop - do any pending work before leaving to user space
+ * @regs: Pointer to pt_regs on entry stack
+ * @ti_work: TIF work flags as read by the caller
+ */
+__always_inline unsigned long exit_to_user_mode_loop(struct pt_regs *regs,
+ unsigned long ti_work)
{
- enter_from_user_mode(regs);
-}
+ for (;;) {
+ ti_work = __exit_to_user_mode_loop(regs, ti_work);
-noinstr void irqentry_exit_to_user_mode(struct pt_regs *regs)
-{
- instrumentation_begin();
- exit_to_user_mode_prepare(regs);
- instrumentation_end();
- exit_to_user_mode();
+ if (likely(!rseq_exit_to_user_mode_restart(regs, ti_work)))
+ return ti_work;
+ ti_work = read_thread_flags();
+ }
}
noinstr irqentry_state_t irqentry_enter(struct pt_regs *regs)
diff --git a/kernel/entry/syscall-common.c b/kernel/entry/syscall-common.c
index 66e6ba7fa80c..940a597ded40 100644
--- a/kernel/entry/syscall-common.c
+++ b/kernel/entry/syscall-common.c
@@ -63,14 +63,6 @@ long syscall_trace_enter(struct pt_regs *regs, long syscall,
return ret ? : syscall;
}
-noinstr void syscall_enter_from_user_mode_prepare(struct pt_regs *regs)
-{
- enter_from_user_mode(regs);
- instrumentation_begin();
- local_irq_enable();
- instrumentation_end();
-}
-
/*
* If SYSCALL_EMU is set, then the only reason to report is when
* SINGLESTEP is set (i.e. PTRACE_SYSEMU_SINGLESTEP). This syscall
diff --git a/kernel/exit.c b/kernel/exit.c
index 9f74e8f1c431..324616f690b7 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -910,6 +910,7 @@ void __noreturn do_exit(long code)
user_events_exit(tsk);
io_uring_files_cancel();
+ sched_mm_cid_exit(tsk);
exit_signals(tsk); /* sets PF_EXITING */
seccomp_filter_release(tsk);
diff --git a/kernel/fork.c b/kernel/fork.c
index 3da0f08615a9..8475958e029b 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -955,10 +955,8 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node)
#endif
#ifdef CONFIG_SCHED_MM_CID
- tsk->mm_cid = -1;
- tsk->last_mm_cid = -1;
- tsk->mm_cid_active = 0;
- tsk->migrate_from_cpu = -1;
+ tsk->mm_cid.cid = MM_CID_UNSET;
+ tsk->mm_cid.active = 0;
#endif
return tsk;
@@ -2456,6 +2454,7 @@ __latent_entropy struct task_struct *copy_process(
exit_task_namespaces(p);
bad_fork_cleanup_mm:
if (p->mm) {
+ sched_mm_cid_exit(p);
mm_clear_owner(p->mm, p);
mmput(p->mm);
}
diff --git a/kernel/futex/core.c b/kernel/futex/core.c
index 125804fbb5cb..ebcccb16ae0b 100644
--- a/kernel/futex/core.c
+++ b/kernel/futex/core.c
@@ -581,7 +581,7 @@ int get_futex_key(u32 __user *uaddr, unsigned int flags, union futex_key *key,
if (flags & FLAGS_NUMA) {
u32 __user *naddr = (void *)uaddr + size / 2;
- if (futex_get_value(&node, naddr))
+ if (get_user_inline(node, naddr))
return -EFAULT;
if ((node != FUTEX_NO_NODE) &&
@@ -601,7 +601,7 @@ int get_futex_key(u32 __user *uaddr, unsigned int flags, union futex_key *key,
node = numa_node_id();
node_updated = true;
}
- if (node_updated && futex_put_value(node, naddr))
+ if (node_updated && put_user_inline(node, naddr))
return -EFAULT;
}
diff --git a/kernel/futex/futex.h b/kernel/futex/futex.h
index 2cd57096c38e..30c2afa03889 100644
--- a/kernel/futex/futex.h
+++ b/kernel/futex/futex.h
@@ -281,63 +281,11 @@ static inline int futex_cmpxchg_value_locked(u32 *curval, u32 __user *uaddr, u32
return ret;
}
-/*
- * This does a plain atomic user space read, and the user pointer has
- * already been verified earlier by get_futex_key() to be both aligned
- * and actually in user space, just like futex_atomic_cmpxchg_inatomic().
- *
- * We still want to avoid any speculation, and while __get_user() is
- * the traditional model for this, it's actually slower than doing
- * this manually these days.
- *
- * We could just have a per-architecture special function for it,
- * the same way we do futex_atomic_cmpxchg_inatomic(), but rather
- * than force everybody to do that, write it out long-hand using
- * the low-level user-access infrastructure.
- *
- * This looks a bit overkill, but generally just results in a couple
- * of instructions.
- */
-static __always_inline int futex_get_value(u32 *dest, u32 __user *from)
-{
- u32 val;
-
- if (can_do_masked_user_access())
- from = masked_user_access_begin(from);
- else if (!user_read_access_begin(from, sizeof(*from)))
- return -EFAULT;
- unsafe_get_user(val, from, Efault);
- user_read_access_end();
- *dest = val;
- return 0;
-Efault:
- user_read_access_end();
- return -EFAULT;
-}
-
-static __always_inline int futex_put_value(u32 val, u32 __user *to)
-{
- if (can_do_masked_user_access())
- to = masked_user_access_begin(to);
- else if (!user_write_access_begin(to, sizeof(*to)))
- return -EFAULT;
- unsafe_put_user(val, to, Efault);
- user_write_access_end();
- return 0;
-Efault:
- user_write_access_end();
- return -EFAULT;
-}
-
+/* Read from user memory with pagefaults disabled */
static inline int futex_get_value_locked(u32 *dest, u32 __user *from)
{
- int ret;
-
- pagefault_disable();
- ret = futex_get_value(dest, from);
- pagefault_enable();
-
- return ret;
+ guard(pagefault)();
+ return get_user_inline(*dest, from);
}
extern void __futex_unqueue(struct futex_q *q);
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index 75a84efad40f..392ec2f75f01 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -793,9 +793,9 @@ static long ptrace_get_rseq_configuration(struct task_struct *task,
unsigned long size, void __user *data)
{
struct ptrace_rseq_configuration conf = {
- .rseq_abi_pointer = (u64)(uintptr_t)task->rseq,
- .rseq_abi_size = task->rseq_len,
- .signature = task->rseq_sig,
+ .rseq_abi_pointer = (u64)(uintptr_t)task->rseq.usrptr,
+ .rseq_abi_size = task->rseq.len,
+ .signature = task->rseq.sig,
.flags = 0,
};
diff --git a/kernel/rseq.c b/kernel/rseq.c
index 2452b7366b00..395d8b002350 100644
--- a/kernel/rseq.c
+++ b/kernel/rseq.c
@@ -8,98 +8,7 @@
* Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
*/
-#include <linux/sched.h>
-#include <linux/uaccess.h>
-#include <linux/syscalls.h>
-#include <linux/rseq.h>
-#include <linux/types.h>
-#include <linux/ratelimit.h>
-#include <asm/ptrace.h>
-
-#define CREATE_TRACE_POINTS
-#include <trace/events/rseq.h>
-
-/* The original rseq structure size (including padding) is 32 bytes. */
-#define ORIG_RSEQ_SIZE 32
-
-#define RSEQ_CS_NO_RESTART_FLAGS (RSEQ_CS_FLAG_NO_RESTART_ON_PREEMPT | \
- RSEQ_CS_FLAG_NO_RESTART_ON_SIGNAL | \
- RSEQ_CS_FLAG_NO_RESTART_ON_MIGRATE)
-
-#ifdef CONFIG_DEBUG_RSEQ
-static struct rseq *rseq_kernel_fields(struct task_struct *t)
-{
- return (struct rseq *) t->rseq_fields;
-}
-
-static int rseq_validate_ro_fields(struct task_struct *t)
-{
- static DEFINE_RATELIMIT_STATE(_rs,
- DEFAULT_RATELIMIT_INTERVAL,
- DEFAULT_RATELIMIT_BURST);
- u32 cpu_id_start, cpu_id, node_id, mm_cid;
- struct rseq __user *rseq = t->rseq;
-
- /*
- * Validate fields which are required to be read-only by
- * user-space.
- */
- if (!user_read_access_begin(rseq, t->rseq_len))
- goto efault;
- unsafe_get_user(cpu_id_start, &rseq->cpu_id_start, efault_end);
- unsafe_get_user(cpu_id, &rseq->cpu_id, efault_end);
- unsafe_get_user(node_id, &rseq->node_id, efault_end);
- unsafe_get_user(mm_cid, &rseq->mm_cid, efault_end);
- user_read_access_end();
-
- if ((cpu_id_start != rseq_kernel_fields(t)->cpu_id_start ||
- cpu_id != rseq_kernel_fields(t)->cpu_id ||
- node_id != rseq_kernel_fields(t)->node_id ||
- mm_cid != rseq_kernel_fields(t)->mm_cid) && __ratelimit(&_rs)) {
-
- pr_warn("Detected rseq corruption for pid: %d, name: %s\n"
- "\tcpu_id_start: %u ?= %u\n"
- "\tcpu_id: %u ?= %u\n"
- "\tnode_id: %u ?= %u\n"
- "\tmm_cid: %u ?= %u\n",
- t->pid, t->comm,
- cpu_id_start, rseq_kernel_fields(t)->cpu_id_start,
- cpu_id, rseq_kernel_fields(t)->cpu_id,
- node_id, rseq_kernel_fields(t)->node_id,
- mm_cid, rseq_kernel_fields(t)->mm_cid);
- }
-
- /* For now, only print a console warning on mismatch. */
- return 0;
-
-efault_end:
- user_read_access_end();
-efault:
- return -EFAULT;
-}
-
-/*
- * Update an rseq field and its in-kernel copy in lock-step to keep a coherent
- * state.
- */
-#define rseq_unsafe_put_user(t, value, field, error_label) \
- do { \
- unsafe_put_user(value, &t->rseq->field, error_label); \
- rseq_kernel_fields(t)->field = value; \
- } while (0)
-
-#else
-static int rseq_validate_ro_fields(struct task_struct *t)
-{
- return 0;
-}
-
-#define rseq_unsafe_put_user(t, value, field, error_label) \
- unsafe_put_user(value, &t->rseq->field, error_label)
-#endif
-
/*
- *
* Restartable sequences are a lightweight interface that allows
* user-level code to be executed atomically relative to scheduler
* preemption and signal delivery. Typically used for implementing
@@ -158,356 +67,356 @@ static int rseq_validate_ro_fields(struct task_struct *t)
* F1. <failure>
*/
-static int rseq_update_cpu_node_id(struct task_struct *t)
-{
- struct rseq __user *rseq = t->rseq;
- u32 cpu_id = raw_smp_processor_id();
- u32 node_id = cpu_to_node(cpu_id);
- u32 mm_cid = task_mm_cid(t);
+/* Required to select the proper per_cpu ops for rseq_stats_inc() */
+#define RSEQ_BUILD_SLOW_PATH
- /*
- * Validate read-only rseq fields.
- */
- if (rseq_validate_ro_fields(t))
- goto efault;
- WARN_ON_ONCE((int) mm_cid < 0);
- if (!user_write_access_begin(rseq, t->rseq_len))
- goto efault;
+#include <linux/debugfs.h>
+#include <linux/ratelimit.h>
+#include <linux/rseq_entry.h>
+#include <linux/sched.h>
+#include <linux/syscalls.h>
+#include <linux/uaccess.h>
+#include <linux/types.h>
+#include <asm/ptrace.h>
- rseq_unsafe_put_user(t, cpu_id, cpu_id_start, efault_end);
- rseq_unsafe_put_user(t, cpu_id, cpu_id, efault_end);
- rseq_unsafe_put_user(t, node_id, node_id, efault_end);
- rseq_unsafe_put_user(t, mm_cid, mm_cid, efault_end);
+#define CREATE_TRACE_POINTS
+#include <trace/events/rseq.h>
- /*
- * Additional feature fields added after ORIG_RSEQ_SIZE
- * need to be conditionally updated only if
- * t->rseq_len != ORIG_RSEQ_SIZE.
- */
- user_write_access_end();
- trace_rseq_update(t);
- return 0;
+DEFINE_STATIC_KEY_MAYBE(CONFIG_RSEQ_DEBUG_DEFAULT_ENABLE, rseq_debug_enabled);
-efault_end:
- user_write_access_end();
-efault:
- return -EFAULT;
+static inline void rseq_control_debug(bool on)
+{
+ if (on)
+ static_branch_enable(&rseq_debug_enabled);
+ else
+ static_branch_disable(&rseq_debug_enabled);
}
-static int rseq_reset_rseq_cpu_node_id(struct task_struct *t)
+static int __init rseq_setup_debug(char *str)
{
- struct rseq __user *rseq = t->rseq;
- u32 cpu_id_start = 0, cpu_id = RSEQ_CPU_ID_UNINITIALIZED, node_id = 0,
- mm_cid = 0;
-
- /*
- * Validate read-only rseq fields.
- */
- if (rseq_validate_ro_fields(t))
- goto efault;
+ bool on;
- if (!user_write_access_begin(rseq, t->rseq_len))
- goto efault;
-
- /*
- * Reset all fields to their initial state.
- *
- * All fields have an initial state of 0 except cpu_id which is set to
- * RSEQ_CPU_ID_UNINITIALIZED, so that any user coming in after
- * unregistration can figure out that rseq needs to be registered
- * again.
- */
- rseq_unsafe_put_user(t, cpu_id_start, cpu_id_start, efault_end);
- rseq_unsafe_put_user(t, cpu_id, cpu_id, efault_end);
- rseq_unsafe_put_user(t, node_id, node_id, efault_end);
- rseq_unsafe_put_user(t, mm_cid, mm_cid, efault_end);
-
- /*
- * Additional feature fields added after ORIG_RSEQ_SIZE
- * need to be conditionally reset only if
- * t->rseq_len != ORIG_RSEQ_SIZE.
- */
- user_write_access_end();
- return 0;
-
-efault_end:
- user_write_access_end();
-efault:
- return -EFAULT;
+ if (kstrtobool(str, &on))
+ return -EINVAL;
+ rseq_control_debug(on);
+ return 1;
}
+__setup("rseq_debug=", rseq_setup_debug);
+#ifdef CONFIG_TRACEPOINTS
/*
- * Get the user-space pointer value stored in the 'rseq_cs' field.
+ * Out of line, so the actual update functions can be in a header to be
+ * inlined into the exit to user code.
*/
-static int rseq_get_rseq_cs_ptr_val(struct rseq __user *rseq, u64 *rseq_cs)
+void __rseq_trace_update(struct task_struct *t)
{
- if (!rseq_cs)
- return -EFAULT;
-
-#ifdef CONFIG_64BIT
- if (get_user(*rseq_cs, &rseq->rseq_cs))
- return -EFAULT;
-#else
- if (copy_from_user(rseq_cs, &rseq->rseq_cs, sizeof(*rseq_cs)))
- return -EFAULT;
-#endif
+ trace_rseq_update(t);
+}
- return 0;
+void __rseq_trace_ip_fixup(unsigned long ip, unsigned long start_ip,
+ unsigned long offset, unsigned long abort_ip)
+{
+ trace_rseq_ip_fixup(ip, start_ip, offset, abort_ip);
}
+#endif /* CONFIG_TRACEPOINTS */
-/*
- * If the rseq_cs field of 'struct rseq' contains a valid pointer to
- * user-space, copy 'struct rseq_cs' from user-space and validate its fields.
- */
-static int rseq_get_rseq_cs(struct task_struct *t, struct rseq_cs *rseq_cs)
+#ifdef CONFIG_DEBUG_FS
+#ifdef CONFIG_RSEQ_STATS
+DEFINE_PER_CPU(struct rseq_stats, rseq_stats);
+
+static int rseq_stats_show(struct seq_file *m, void *p)
{
- struct rseq_cs __user *urseq_cs;
- u64 ptr;
- u32 __user *usig;
- u32 sig;
- int ret;
-
- ret = rseq_get_rseq_cs_ptr_val(t->rseq, &ptr);
- if (ret)
- return ret;
-
- /* If the rseq_cs pointer is NULL, return a cleared struct rseq_cs. */
- if (!ptr) {
- memset(rseq_cs, 0, sizeof(*rseq_cs));
- return 0;
+ struct rseq_stats stats = { };
+ unsigned int cpu;
+
+ for_each_possible_cpu(cpu) {
+ stats.exit += data_race(per_cpu(rseq_stats.exit, cpu));
+ stats.signal += data_race(per_cpu(rseq_stats.signal, cpu));
+ stats.slowpath += data_race(per_cpu(rseq_stats.slowpath, cpu));
+ stats.fastpath += data_race(per_cpu(rseq_stats.fastpath, cpu));
+ stats.ids += data_race(per_cpu(rseq_stats.ids, cpu));
+ stats.cs += data_race(per_cpu(rseq_stats.cs, cpu));
+ stats.clear += data_race(per_cpu(rseq_stats.clear, cpu));
+ stats.fixup += data_race(per_cpu(rseq_stats.fixup, cpu));
}
- /* Check that the pointer value fits in the user-space process space. */
- if (ptr >= TASK_SIZE)
- return -EINVAL;
- urseq_cs = (struct rseq_cs __user *)(unsigned long)ptr;
- if (copy_from_user(rseq_cs, urseq_cs, sizeof(*rseq_cs)))
- return -EFAULT;
- if (rseq_cs->start_ip >= TASK_SIZE ||
- rseq_cs->start_ip + rseq_cs->post_commit_offset >= TASK_SIZE ||
- rseq_cs->abort_ip >= TASK_SIZE ||
- rseq_cs->version > 0)
- return -EINVAL;
- /* Check for overflow. */
- if (rseq_cs->start_ip + rseq_cs->post_commit_offset < rseq_cs->start_ip)
- return -EINVAL;
- /* Ensure that abort_ip is not in the critical section. */
- if (rseq_cs->abort_ip - rseq_cs->start_ip < rseq_cs->post_commit_offset)
- return -EINVAL;
+ seq_printf(m, "exit: %16lu\n", stats.exit);
+ seq_printf(m, "signal: %16lu\n", stats.signal);
+ seq_printf(m, "slowp: %16lu\n", stats.slowpath);
+ seq_printf(m, "fastp: %16lu\n", stats.fastpath);
+ seq_printf(m, "ids: %16lu\n", stats.ids);
+ seq_printf(m, "cs: %16lu\n", stats.cs);
+ seq_printf(m, "clear: %16lu\n", stats.clear);
+ seq_printf(m, "fixup: %16lu\n", stats.fixup);
+ return 0;
+}
- usig = (u32 __user *)(unsigned long)(rseq_cs->abort_ip - sizeof(u32));
- ret = get_user(sig, usig);
- if (ret)
- return ret;
+static int rseq_stats_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, rseq_stats_show, inode->i_private);
+}
- if (current->rseq_sig != sig) {
- printk_ratelimited(KERN_WARNING
- "Possible attack attempt. Unexpected rseq signature 0x%x, expecting 0x%x (pid=%d, addr=%p).\n",
- sig, current->rseq_sig, current->pid, usig);
- return -EINVAL;
- }
+static const struct file_operations stat_ops = {
+ .open = rseq_stats_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
+static int __init rseq_stats_init(struct dentry *root_dir)
+{
+ debugfs_create_file("stats", 0444, root_dir, NULL, &stat_ops);
return 0;
}
+#else
+static inline void rseq_stats_init(struct dentry *root_dir) { }
+#endif /* CONFIG_RSEQ_STATS */
-static bool rseq_warn_flags(const char *str, u32 flags)
+static int rseq_debug_show(struct seq_file *m, void *p)
{
- u32 test_flags;
-
- if (!flags)
- return false;
- test_flags = flags & RSEQ_CS_NO_RESTART_FLAGS;
- if (test_flags)
- pr_warn_once("Deprecated flags (%u) in %s ABI structure", test_flags, str);
- test_flags = flags & ~RSEQ_CS_NO_RESTART_FLAGS;
- if (test_flags)
- pr_warn_once("Unknown flags (%u) in %s ABI structure", test_flags, str);
- return true;
+ bool on = static_branch_unlikely(&rseq_debug_enabled);
+
+ seq_printf(m, "%d\n", on);
+ return 0;
}
-static int rseq_need_restart(struct task_struct *t, u32 cs_flags)
+static ssize_t rseq_debug_write(struct file *file, const char __user *ubuf,
+ size_t count, loff_t *ppos)
{
- u32 flags, event_mask;
- int ret;
+ bool on;
- if (rseq_warn_flags("rseq_cs", cs_flags))
+ if (kstrtobool_from_user(ubuf, count, &on))
return -EINVAL;
- /* Get thread flags. */
- ret = get_user(flags, &t->rseq->flags);
- if (ret)
- return ret;
+ rseq_control_debug(on);
+ return count;
+}
- if (rseq_warn_flags("rseq", flags))
- return -EINVAL;
+static int rseq_debug_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, rseq_debug_show, inode->i_private);
+}
- /*
- * Load and clear event mask atomically with respect to
- * scheduler preemption and membarrier IPIs.
- */
- scoped_guard(RSEQ_EVENT_GUARD) {
- event_mask = t->rseq_event_mask;
- t->rseq_event_mask = 0;
- }
+static const struct file_operations debug_ops = {
+ .open = rseq_debug_open,
+ .read = seq_read,
+ .write = rseq_debug_write,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
+static int __init rseq_debugfs_init(void)
+{
+ struct dentry *root_dir = debugfs_create_dir("rseq", NULL);
- return !!event_mask;
+ debugfs_create_file("debug", 0644, root_dir, NULL, &debug_ops);
+ rseq_stats_init(root_dir);
+ return 0;
}
+__initcall(rseq_debugfs_init);
+#endif /* CONFIG_DEBUG_FS */
-static int clear_rseq_cs(struct rseq __user *rseq)
+static bool rseq_set_ids(struct task_struct *t, struct rseq_ids *ids, u32 node_id)
{
- /*
- * The rseq_cs field is set to NULL on preemption or signal
- * delivery on top of rseq assembly block, as well as on top
- * of code outside of the rseq assembly block. This performs
- * a lazy clear of the rseq_cs field.
- *
- * Set rseq_cs to NULL.
- */
-#ifdef CONFIG_64BIT
- return put_user(0UL, &rseq->rseq_cs);
-#else
- if (clear_user(&rseq->rseq_cs, sizeof(rseq->rseq_cs)))
- return -EFAULT;
- return 0;
-#endif
+ return rseq_set_ids_get_csaddr(t, ids, node_id, NULL);
}
-/*
- * Unsigned comparison will be true when ip >= start_ip, and when
- * ip < start_ip + post_commit_offset.
- */
-static bool in_rseq_cs(unsigned long ip, struct rseq_cs *rseq_cs)
+static bool rseq_handle_cs(struct task_struct *t, struct pt_regs *regs)
{
- return ip - rseq_cs->start_ip < rseq_cs->post_commit_offset;
+ struct rseq __user *urseq = t->rseq.usrptr;
+ u64 csaddr;
+
+ scoped_user_read_access(urseq, efault)
+ unsafe_get_user(csaddr, &urseq->rseq_cs, efault);
+ if (likely(!csaddr))
+ return true;
+ return rseq_update_user_cs(t, regs, csaddr);
+efault:
+ return false;
}
-static int rseq_ip_fixup(struct pt_regs *regs)
+static void rseq_slowpath_update_usr(struct pt_regs *regs)
{
- unsigned long ip = instruction_pointer(regs);
+ /*
+ * Preserve rseq state and user_irq state. The generic entry code
+ * clears user_irq on the way out, the non-generic entry
+ * architectures are not having user_irq.
+ */
+ const struct rseq_event evt_mask = { .has_rseq = true, .user_irq = true, };
struct task_struct *t = current;
- struct rseq_cs rseq_cs;
- int ret;
+ struct rseq_ids ids;
+ u32 node_id;
+ bool event;
+
+ if (unlikely(t->flags & PF_EXITING))
+ return;
- ret = rseq_get_rseq_cs(t, &rseq_cs);
- if (ret)
- return ret;
+ rseq_stat_inc(rseq_stats.slowpath);
/*
- * Handle potentially not being within a critical section.
- * If not nested over a rseq critical section, restart is useless.
- * Clear the rseq_cs pointer and return.
+ * Read and clear the event pending bit first. If the task
+ * was not preempted or migrated or a signal is on the way,
+ * there is no point in doing any of the heavy lifting here
+ * on production kernels. In that case TIF_NOTIFY_RESUME
+ * was raised by some other functionality.
+ *
+ * This is correct because the read/clear operation is
+ * guarded against scheduler preemption, which makes it CPU
+ * local atomic. If the task is preempted right after
+ * re-enabling preemption then TIF_NOTIFY_RESUME is set
+ * again and this function is invoked another time _before_
+ * the task is able to return to user mode.
+ *
+ * On a debug kernel, invoke the fixup code unconditionally
+ * with the result handed in to allow the detection of
+ * inconsistencies.
*/
- if (!in_rseq_cs(ip, &rseq_cs))
- return clear_rseq_cs(t->rseq);
- ret = rseq_need_restart(t, rseq_cs.flags);
- if (ret <= 0)
- return ret;
- ret = clear_rseq_cs(t->rseq);
- if (ret)
- return ret;
- trace_rseq_ip_fixup(ip, rseq_cs.start_ip, rseq_cs.post_commit_offset,
- rseq_cs.abort_ip);
- instruction_pointer_set(regs, (unsigned long)rseq_cs.abort_ip);
- return 0;
+ scoped_guard(irq) {
+ event = t->rseq.event.sched_switch;
+ t->rseq.event.all &= evt_mask.all;
+ ids.cpu_id = task_cpu(t);
+ ids.mm_cid = task_mm_cid(t);
+ }
+
+ if (!event)
+ return;
+
+ node_id = cpu_to_node(ids.cpu_id);
+
+ if (unlikely(!rseq_update_usr(t, regs, &ids, node_id))) {
+ /*
+ * Clear the errors just in case this might survive magically, but
+ * leave the rest intact.
+ */
+ t->rseq.event.error = 0;
+ force_sig(SIGSEGV);
+ }
}
-/*
- * This resume handler must always be executed between any of:
- * - preemption,
- * - signal delivery,
- * and return to user-space.
- *
- * This is how we can ensure that the entire rseq critical section
- * will issue the commit instruction only if executed atomically with
- * respect to other threads scheduled on the same CPU, and with respect
- * to signal handlers.
- */
-void __rseq_handle_notify_resume(struct ksignal *ksig, struct pt_regs *regs)
+void __rseq_handle_slowpath(struct pt_regs *regs)
{
- struct task_struct *t = current;
- int ret, sig;
-
- if (unlikely(t->flags & PF_EXITING))
+ /*
+ * If invoked from hypervisors before entering the guest via
+ * resume_user_mode_work(), then @regs is a NULL pointer.
+ *
+ * resume_user_mode_work() clears TIF_NOTIFY_RESUME and re-raises
+ * it before returning from the ioctl() to user space when
+ * rseq_event.sched_switch is set.
+ *
+ * So it's safe to ignore here instead of pointlessly updating it
+ * in the vcpu_run() loop.
+ */
+ if (!regs)
return;
+ rseq_slowpath_update_usr(regs);
+}
+
+void __rseq_signal_deliver(int sig, struct pt_regs *regs)
+{
+ rseq_stat_inc(rseq_stats.signal);
/*
- * regs is NULL if and only if the caller is in a syscall path. Skip
- * fixup and leave rseq_cs as is so that rseq_sycall() will detect and
- * kill a misbehaving userspace on debug kernels.
+ * Don't update IDs, they are handled on exit to user if
+ * necessary. The important thing is to abort a critical section of
+ * the interrupted context as after this point the instruction
+ * pointer in @regs points to the signal handler.
*/
- if (regs) {
- ret = rseq_ip_fixup(regs);
- if (unlikely(ret < 0))
- goto error;
+ if (unlikely(!rseq_handle_cs(current, regs))) {
+ /*
+ * Clear the errors just in case this might survive
+ * magically, but leave the rest intact.
+ */
+ current->rseq.event.error = 0;
+ force_sigsegv(sig);
}
- if (unlikely(rseq_update_cpu_node_id(t)))
- goto error;
- return;
-
-error:
- sig = ksig ? ksig->sig : 0;
- force_sigsegv(sig);
}
-#ifdef CONFIG_DEBUG_RSEQ
-
/*
* Terminate the process if a syscall is issued within a restartable
* sequence.
*/
-void rseq_syscall(struct pt_regs *regs)
+void __rseq_debug_syscall_return(struct pt_regs *regs)
{
- unsigned long ip = instruction_pointer(regs);
struct task_struct *t = current;
- struct rseq_cs rseq_cs;
+ u64 csaddr;
- if (!t->rseq)
+ if (!t->rseq.event.has_rseq)
return;
- if (rseq_get_rseq_cs(t, &rseq_cs) || in_rseq_cs(ip, &rseq_cs))
- force_sig(SIGSEGV);
+ if (get_user(csaddr, &t->rseq.usrptr->rseq_cs))
+ goto fail;
+ if (likely(!csaddr))
+ return;
+ if (unlikely(csaddr >= TASK_SIZE))
+ goto fail;
+ if (rseq_debug_update_user_cs(t, regs, csaddr))
+ return;
+fail:
+ force_sig(SIGSEGV);
}
+#ifdef CONFIG_DEBUG_RSEQ
+/* Kept around to keep GENERIC_ENTRY=n architectures supported. */
+void rseq_syscall(struct pt_regs *regs)
+{
+ __rseq_debug_syscall_return(regs);
+}
#endif
+static bool rseq_reset_ids(void)
+{
+ struct rseq_ids ids = {
+ .cpu_id = RSEQ_CPU_ID_UNINITIALIZED,
+ .mm_cid = 0,
+ };
+
+ /*
+ * If this fails, terminate it because this leaves the kernel in
+ * stupid state as exit to user space will try to fixup the ids
+ * again.
+ */
+ if (rseq_set_ids(current, &ids, 0))
+ return true;
+
+ force_sig(SIGSEGV);
+ return false;
+}
+
+/* The original rseq structure size (including padding) is 32 bytes. */
+#define ORIG_RSEQ_SIZE 32
+
/*
* sys_rseq - setup restartable sequences for caller thread.
*/
-SYSCALL_DEFINE4(rseq, struct rseq __user *, rseq, u32, rseq_len,
- int, flags, u32, sig)
+SYSCALL_DEFINE4(rseq, struct rseq __user *, rseq, u32, rseq_len, int, flags, u32, sig)
{
- int ret;
- u64 rseq_cs;
-
if (flags & RSEQ_FLAG_UNREGISTER) {
if (flags & ~RSEQ_FLAG_UNREGISTER)
return -EINVAL;
/* Unregister rseq for current thread. */
- if (current->rseq != rseq || !current->rseq)
+ if (current->rseq.usrptr != rseq || !current->rseq.usrptr)
return -EINVAL;
- if (rseq_len != current->rseq_len)
+ if (rseq_len != current->rseq.len)
return -EINVAL;
- if (current->rseq_sig != sig)
+ if (current->rseq.sig != sig)
return -EPERM;
- ret = rseq_reset_rseq_cpu_node_id(current);
- if (ret)
- return ret;
- current->rseq = NULL;
- current->rseq_sig = 0;
- current->rseq_len = 0;
+ if (!rseq_reset_ids())
+ return -EFAULT;
+ rseq_reset(current);
return 0;
}
if (unlikely(flags))
return -EINVAL;
- if (current->rseq) {
+ if (current->rseq.usrptr) {
/*
* If rseq is already registered, check whether
* the provided address differs from the prior
* one.
*/
- if (current->rseq != rseq || rseq_len != current->rseq_len)
+ if (current->rseq.usrptr != rseq || rseq_len != current->rseq.len)
return -EINVAL;
- if (current->rseq_sig != sig)
+ if (current->rseq.sig != sig)
return -EPERM;
/* Already registered. */
return -EBUSY;
@@ -531,43 +440,39 @@ SYSCALL_DEFINE4(rseq, struct rseq __user *, rseq, u32, rseq_len,
if (!access_ok(rseq, rseq_len))
return -EFAULT;
- /*
- * If the rseq_cs pointer is non-NULL on registration, clear it to
- * avoid a potential segfault on return to user-space. The proper thing
- * to do would have been to fail the registration but this would break
- * older libcs that reuse the rseq area for new threads without
- * clearing the fields.
- */
- if (rseq_get_rseq_cs_ptr_val(rseq, &rseq_cs))
- return -EFAULT;
- if (rseq_cs && clear_rseq_cs(rseq))
- return -EFAULT;
+ scoped_user_write_access(rseq, efault) {
+ /*
+ * If the rseq_cs pointer is non-NULL on registration, clear it to
+ * avoid a potential segfault on return to user-space. The proper thing
+ * to do would have been to fail the registration but this would break
+ * older libcs that reuse the rseq area for new threads without
+ * clearing the fields. Don't bother reading it, just reset it.
+ */
+ unsafe_put_user(0UL, &rseq->rseq_cs, efault);
+ /* Initialize IDs in user space */
+ unsafe_put_user(RSEQ_CPU_ID_UNINITIALIZED, &rseq->cpu_id_start, efault);
+ unsafe_put_user(RSEQ_CPU_ID_UNINITIALIZED, &rseq->cpu_id, efault);
+ unsafe_put_user(0U, &rseq->node_id, efault);
+ unsafe_put_user(0U, &rseq->mm_cid, efault);
+ }
-#ifdef CONFIG_DEBUG_RSEQ
- /*
- * Initialize the in-kernel rseq fields copy for validation of
- * read-only fields.
- */
- if (get_user(rseq_kernel_fields(current)->cpu_id_start, &rseq->cpu_id_start) ||
- get_user(rseq_kernel_fields(current)->cpu_id, &rseq->cpu_id) ||
- get_user(rseq_kernel_fields(current)->node_id, &rseq->node_id) ||
- get_user(rseq_kernel_fields(current)->mm_cid, &rseq->mm_cid))
- return -EFAULT;
-#endif
/*
* Activate the registration by setting the rseq area address, length
* and signature in the task struct.
*/
- current->rseq = rseq;
- current->rseq_len = rseq_len;
- current->rseq_sig = sig;
+ current->rseq.usrptr = rseq;
+ current->rseq.len = rseq_len;
+ current->rseq.sig = sig;
/*
* If rseq was previously inactive, and has just been
* registered, ensure the cpu_id_start and cpu_id fields
* are updated before returning to user-space.
*/
- rseq_set_notify_resume(current);
-
+ current->rseq.event.has_rseq = true;
+ rseq_force_update();
return 0;
+
+efault:
+ return -EFAULT;
}
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index f1ebf67b48e2..62235f1dc04e 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2128,8 +2128,6 @@ void activate_task(struct rq *rq, struct task_struct *p, int flags)
{
if (task_on_rq_migrating(p))
flags |= ENQUEUE_MIGRATED;
- if (flags & ENQUEUE_MIGRATED)
- sched_mm_cid_migrate_to(rq, p);
enqueue_task(rq, p, flags);
@@ -2671,6 +2669,8 @@ int push_cpu_stop(void *arg)
return 0;
}
+static inline void mm_update_cpus_allowed(struct mm_struct *mm, const cpumask_t *affmask);
+
/*
* sched_class::set_cpus_allowed must do the below, but is not required to
* actually call this function.
@@ -2684,6 +2684,7 @@ void set_cpus_allowed_common(struct task_struct *p, struct affinity_context *ctx
cpumask_copy(&p->cpus_mask, ctx->new_mask);
p->nr_cpus_allowed = cpumask_weight(ctx->new_mask);
+ mm_update_cpus_allowed(p->mm, ctx->new_mask);
/*
* Swap in a new user_cpus_ptr if SCA_USER flag set
@@ -2730,7 +2731,6 @@ __do_set_cpus_allowed(struct task_struct *p, struct affinity_context *ctx)
put_prev_task(rq, p);
p->sched_class->set_cpus_allowed(p, ctx);
- mm_set_cpus_allowed(p->mm, ctx->new_mask);
if (queued)
enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK);
@@ -3329,8 +3329,6 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
if (p->sched_class->migrate_task_rq)
p->sched_class->migrate_task_rq(p, new_cpu);
p->se.nr_migrations++;
- rseq_migrate(p);
- sched_mm_cid_migrate_from(p);
perf_event_task_migrate(p);
}
@@ -4487,7 +4485,6 @@ static void __sched_fork(u64 clone_flags, struct task_struct *p)
init_numa_balancing(clone_flags, p);
p->wake_entry.u_flags = CSD_TYPE_TTWU;
p->migration_pending = NULL;
- init_sched_mm_cid(p);
}
DEFINE_STATIC_KEY_FALSE(sched_numa_balancing);
@@ -4763,7 +4760,6 @@ int sched_cgroup_fork(struct task_struct *p, struct kernel_clone_args *kargs)
p->sched_task_group = tg;
}
#endif
- rseq_migrate(p);
/*
* We're setting the CPU for the first time, we don't migrate,
* so use __set_task_cpu().
@@ -4827,7 +4823,6 @@ void wake_up_new_task(struct task_struct *p)
* as we're not fully set-up yet.
*/
p->recent_used_cpu = task_cpu(p);
- rseq_migrate(p);
__set_task_cpu(p, select_task_rq(p, task_cpu(p), &wake_flags));
rq = __task_rq_lock(p, &rf);
update_rq_clock(rq);
@@ -5121,7 +5116,6 @@ prepare_task_switch(struct rq *rq, struct task_struct *prev,
kcov_prepare_switch(prev);
sched_info_switch(rq, prev, next);
perf_event_task_sched_out(prev, next);
- rseq_preempt(prev);
fire_sched_out_preempt_notifiers(prev, next);
kmap_local_sched_out();
prepare_task(next);
@@ -5284,19 +5278,16 @@ context_switch(struct rq *rq, struct task_struct *prev,
*
* kernel -> user switch + mmdrop_lazy_tlb() active
* user -> user switch
- *
- * switch_mm_cid() needs to be updated if the barriers provided
- * by context_switch() are modified.
*/
- if (!next->mm) { // to kernel
+ if (!next->mm) { // to kernel
enter_lazy_tlb(prev->active_mm, next);
next->active_mm = prev->active_mm;
- if (prev->mm) // from user
+ if (prev->mm) // from user
mmgrab_lazy_tlb(prev->active_mm);
else
prev->active_mm = NULL;
- } else { // to user
+ } else { // to user
membarrier_switch_mm(rq, prev->active_mm, next->mm);
/*
* sys_membarrier() requires an smp_mb() between setting
@@ -5309,15 +5300,20 @@ context_switch(struct rq *rq, struct task_struct *prev,
switch_mm_irqs_off(prev->active_mm, next->mm, next);
lru_gen_use_mm(next->mm);
- if (!prev->mm) { // from kernel
+ if (!prev->mm) { // from kernel
/* will mmdrop_lazy_tlb() in finish_task_switch(). */
rq->prev_mm = prev->active_mm;
prev->active_mm = NULL;
}
}
- /* switch_mm_cid() requires the memory barriers above. */
- switch_mm_cid(rq, prev, next);
+ mm_cid_switch_to(prev, next);
+
+ /*
+ * Tell rseq that the task was scheduled in. Must be after
+ * switch_mm_cid() to get the TIF flag set.
+ */
+ rseq_sched_switch_event(next);
prepare_lock_switch(rq, next, rf);
@@ -5602,7 +5598,6 @@ void sched_tick(void)
resched_latency = cpu_resched_latency(rq);
calc_global_load_tick(rq);
sched_core_tick(rq);
- task_tick_mm_cid(rq, donor);
scx_tick(rq);
rq_unlock(rq, &rf);
@@ -10374,525 +10369,501 @@ void call_trace_sched_update_nr_running(struct rq *rq, int count)
}
#ifdef CONFIG_SCHED_MM_CID
-
/*
- * @cid_lock: Guarantee forward-progress of cid allocation.
+ * Concurrency IDentifier management
*
- * Concurrency ID allocation within a bitmap is mostly lock-free. The cid_lock
- * is only used when contention is detected by the lock-free allocation so
- * forward progress can be guaranteed.
- */
-DEFINE_RAW_SPINLOCK(cid_lock);
-
-/*
- * @use_cid_lock: Select cid allocation behavior: lock-free vs spinlock.
- *
- * When @use_cid_lock is 0, the cid allocation is lock-free. When contention is
- * detected, it is set to 1 to ensure that all newly coming allocations are
- * serialized by @cid_lock until the allocation which detected contention
- * completes and sets @use_cid_lock back to 0. This guarantees forward progress
- * of a cid allocation.
- */
-int use_cid_lock;
-
-/*
- * mm_cid remote-clear implements a lock-free algorithm to clear per-mm/cpu cid
- * concurrently with respect to the execution of the source runqueue context
- * switch.
- *
- * There is one basic properties we want to guarantee here:
- *
- * (1) Remote-clear should _never_ mark a per-cpu cid UNSET when it is actively
- * used by a task. That would lead to concurrent allocation of the cid and
- * userspace corruption.
- *
- * Provide this guarantee by introducing a Dekker memory ordering to guarantee
- * that a pair of loads observe at least one of a pair of stores, which can be
- * shown as:
+ * Serialization rules:
*
- * X = Y = 0
+ * mm::mm_cid::mutex: Serializes fork() and exit() and therefore
+ * protects mm::mm_cid::users.
*
- * w[X]=1 w[Y]=1
- * MB MB
- * r[Y]=y r[X]=x
+ * mm::mm_cid::lock: Serializes mm_update_max_cids() and
+ * mm_update_cpus_allowed(). Nests in mm_cid::mutex
+ * and runqueue lock.
*
- * Which guarantees that x==0 && y==0 is impossible. But rather than using
- * values 0 and 1, this algorithm cares about specific state transitions of the
- * runqueue current task (as updated by the scheduler context switch), and the
- * per-mm/cpu cid value.
+ * The mm_cidmask bitmap is not protected by any of the mm::mm_cid locks
+ * and can only be modified with atomic operations.
*
- * Let's introduce task (Y) which has task->mm == mm and task (N) which has
- * task->mm != mm for the rest of the discussion. There are two scheduler state
- * transitions on context switch we care about:
+ * The mm::mm_cid:pcpu per CPU storage is protected by the CPUs runqueue
+ * lock.
*
- * (TSA) Store to rq->curr with transition from (N) to (Y)
+ * CID ownership:
*
- * (TSB) Store to rq->curr with transition from (Y) to (N)
+ * A CID is either owned by a task (stored in task_struct::mm_cid.cid) or
+ * by a CPU (stored in mm::mm_cid.pcpu::cid). CIDs owned by CPUs have the
+ * MM_CID_ONCPU bit set. During transition from CPU to task ownership mode,
+ * MM_CID_TRANSIT is set on the per task CIDs. When this bit is set the
+ * task needs to drop the CID into the pool when scheduling out. Both bits
+ * (ONCPU and TRANSIT) are filtered out by task_cid() when the CID is
+ * actually handed over to user space in the RSEQ memory.
*
- * On the remote-clear side, there is one transition we care about:
+ * Mode switching:
*
- * (TMA) cmpxchg to *pcpu_cid to set the LAZY flag
+ * Switching to per CPU mode happens when the user count becomes greater
+ * than the maximum number of CIDs, which is calculated by:
*
- * There is also a transition to UNSET state which can be performed from all
- * sides (scheduler, remote-clear). It is always performed with a cmpxchg which
- * guarantees that only a single thread will succeed:
+ * opt_cids = min(mm_cid::nr_cpus_allowed, mm_cid::users);
+ * max_cids = min(1.25 * opt_cids, num_possible_cpus());
*
- * (TMB) cmpxchg to *pcpu_cid to mark UNSET
+ * The +25% allowance is useful for tight CPU masks in scenarios where only
+ * a few threads are created and destroyed to avoid frequent mode
+ * switches. Though this allowance shrinks, the closer opt_cids becomes to
+ * num_possible_cpus(), which is the (unfortunate) hard ABI limit.
*
- * Just to be clear, what we do _not_ want to happen is a transition to UNSET
- * when a thread is actively using the cid (property (1)).
+ * At the point of switching to per CPU mode the new user is not yet
+ * visible in the system, so the task which initiated the fork() runs the
+ * fixup function: mm_cid_fixup_tasks_to_cpu() walks the thread list and
+ * either transfers each tasks owned CID to the CPU the task runs on or
+ * drops it into the CID pool if a task is not on a CPU at that point in
+ * time. Tasks which schedule in before the task walk reaches them do the
+ * handover in mm_cid_schedin(). When mm_cid_fixup_tasks_to_cpus() completes
+ * it's guaranteed that no task related to that MM owns a CID anymore.
*
- * Let's looks at the relevant combinations of TSA/TSB, and TMA transitions.
+ * Switching back to task mode happens when the user count goes below the
+ * threshold which was recorded on the per CPU mode switch:
*
- * Scenario A) (TSA)+(TMA) (from next task perspective)
+ * pcpu_thrs = min(opt_cids - (opt_cids / 4), num_possible_cpus() / 2);
*
- * CPU0 CPU1
+ * This threshold is updated when a affinity change increases the number of
+ * allowed CPUs for the MM, which might cause a switch back to per task
+ * mode.
*
- * Context switch CS-1 Remote-clear
- * - store to rq->curr: (N)->(Y) (TSA) - cmpxchg to *pcpu_id to LAZY (TMA)
- * (implied barrier after cmpxchg)
- * - switch_mm_cid()
- * - memory barrier (see switch_mm_cid()
- * comment explaining how this barrier
- * is combined with other scheduler
- * barriers)
- * - mm_cid_get (next)
- * - READ_ONCE(*pcpu_cid) - rcu_dereference(src_rq->curr)
+ * If the switch back was initiated by a exiting task, then that task runs
+ * the fixup function. If it was initiated by a affinity change, then it's
+ * run either in the deferred update function in context of a workqueue or
+ * by a task which forks a new one or by a task which exits. Whatever
+ * happens first. mm_cid_fixup_cpus_to_task() walks through the possible
+ * CPUs and either transfers the CPU owned CIDs to a related task which
+ * runs on the CPU or drops it into the pool. Tasks which schedule in on a
+ * CPU which the walk did not cover yet do the handover themself.
*
- * This Dekker ensures that either task (Y) is observed by the
- * rcu_dereference() or the LAZY flag is observed by READ_ONCE(), or both are
- * observed.
+ * This transition from CPU to per task ownership happens in two phases:
*
- * If task (Y) store is observed by rcu_dereference(), it means that there is
- * still an active task on the cpu. Remote-clear will therefore not transition
- * to UNSET, which fulfills property (1).
+ * 1) mm:mm_cid.transit contains MM_CID_TRANSIT This is OR'ed on the task
+ * CID and denotes that the CID is only temporarily owned by the
+ * task. When it schedules out the task drops the CID back into the
+ * pool if this bit is set.
*
- * If task (Y) is not observed, but the lazy flag is observed by READ_ONCE(),
- * it will move its state to UNSET, which clears the percpu cid perhaps
- * uselessly (which is not an issue for correctness). Because task (Y) is not
- * observed, CPU1 can move ahead to set the state to UNSET. Because moving
- * state to UNSET is done with a cmpxchg expecting that the old state has the
- * LAZY flag set, only one thread will successfully UNSET.
+ * 2) The initiating context walks the per CPU space and after completion
+ * clears mm:mm_cid.transit. So after that point the CIDs are strictly
+ * task owned again.
*
- * If both states (LAZY flag and task (Y)) are observed, the thread on CPU0
- * will observe the LAZY flag and transition to UNSET (perhaps uselessly), and
- * CPU1 will observe task (Y) and do nothing more, which is fine.
+ * This two phase transition is required to prevent CID space exhaustion
+ * during the transition as a direct transfer of ownership would fail if
+ * two tasks are scheduled in on the same CPU before the fixup freed per
+ * CPU CIDs.
*
- * What we are effectively preventing with this Dekker is a scenario where
- * neither LAZY flag nor store (Y) are observed, which would fail property (1)
- * because this would UNSET a cid which is actively used.
+ * When mm_cid_fixup_cpus_to_tasks() completes it's guaranteed that no CID
+ * related to that MM is owned by a CPU anymore.
*/
-void sched_mm_cid_migrate_from(struct task_struct *t)
-{
- t->migrate_from_cpu = task_cpu(t);
-}
-
-static
-int __sched_mm_cid_migrate_from_fetch_cid(struct rq *src_rq,
- struct task_struct *t,
- struct mm_cid *src_pcpu_cid)
+/*
+ * Update the CID range properties when the constraints change. Invoked via
+ * fork(), exit() and affinity changes
+ */
+static void __mm_update_max_cids(struct mm_mm_cid *mc)
{
- struct mm_struct *mm = t->mm;
- struct task_struct *src_task;
- int src_cid, last_mm_cid;
+ unsigned int opt_cids, max_cids;
- if (!mm)
- return -1;
+ /* Calculate the new optimal constraint */
+ opt_cids = min(mc->nr_cpus_allowed, mc->users);
- last_mm_cid = t->last_mm_cid;
- /*
- * If the migrated task has no last cid, or if the current
- * task on src rq uses the cid, it means the source cid does not need
- * to be moved to the destination cpu.
- */
- if (last_mm_cid == -1)
- return -1;
- src_cid = READ_ONCE(src_pcpu_cid->cid);
- if (!mm_cid_is_valid(src_cid) || last_mm_cid != src_cid)
- return -1;
+ /* Adjust the maximum CIDs to +25% limited by the number of possible CPUs */
+ max_cids = min(opt_cids + (opt_cids / 4), num_possible_cpus());
+ WRITE_ONCE(mc->max_cids, max_cids);
+}
- /*
- * If we observe an active task using the mm on this rq, it means we
- * are not the last task to be migrated from this cpu for this mm, so
- * there is no need to move src_cid to the destination cpu.
- */
- guard(rcu)();
- src_task = rcu_dereference(src_rq->curr);
- if (READ_ONCE(src_task->mm_cid_active) && src_task->mm == mm) {
- t->last_mm_cid = -1;
- return -1;
- }
+static inline unsigned int mm_cid_calc_pcpu_thrs(struct mm_mm_cid *mc)
+{
+ unsigned int opt_cids;
- return src_cid;
+ opt_cids = min(mc->nr_cpus_allowed, mc->users);
+ /* Has to be at least 1 because 0 indicates PCPU mode off */
+ return max(min(opt_cids - opt_cids / 4, num_possible_cpus() / 2), 1);
}
-static
-int __sched_mm_cid_migrate_from_try_steal_cid(struct rq *src_rq,
- struct task_struct *t,
- struct mm_cid *src_pcpu_cid,
- int src_cid)
+static bool mm_update_max_cids(struct mm_struct *mm)
{
- struct task_struct *src_task;
- struct mm_struct *mm = t->mm;
- int lazy_cid;
+ struct mm_mm_cid *mc = &mm->mm_cid;
- if (src_cid == -1)
- return -1;
-
- /*
- * Attempt to clear the source cpu cid to move it to the destination
- * cpu.
- */
- lazy_cid = mm_cid_set_lazy_put(src_cid);
- if (!try_cmpxchg(&src_pcpu_cid->cid, &src_cid, lazy_cid))
- return -1;
+ lockdep_assert_held(&mm->mm_cid.lock);
- /*
- * The implicit barrier after cmpxchg per-mm/cpu cid before loading
- * rq->curr->mm matches the scheduler barrier in context_switch()
- * between store to rq->curr and load of prev and next task's
- * per-mm/cpu cid.
- *
- * The implicit barrier after cmpxchg per-mm/cpu cid before loading
- * rq->curr->mm_cid_active matches the barrier in
- * sched_mm_cid_exit_signals(), sched_mm_cid_before_execve(), and
- * sched_mm_cid_after_execve() between store to t->mm_cid_active and
- * load of per-mm/cpu cid.
- */
+ /* Clear deferred mode switch flag. A change is handled by the caller */
+ mc->update_deferred = false;
+ __mm_update_max_cids(mc);
- /*
- * If we observe an active task using the mm on this rq after setting
- * the lazy-put flag, this task will be responsible for transitioning
- * from lazy-put flag set to MM_CID_UNSET.
- */
- scoped_guard (rcu) {
- src_task = rcu_dereference(src_rq->curr);
- if (READ_ONCE(src_task->mm_cid_active) && src_task->mm == mm) {
- /*
- * We observed an active task for this mm, there is therefore
- * no point in moving this cid to the destination cpu.
- */
- t->last_mm_cid = -1;
- return -1;
- }
+ /* Check whether owner mode must be changed */
+ if (!mc->percpu) {
+ /* Enable per CPU mode when the number of users is above max_cids */
+ if (mc->users > mc->max_cids)
+ mc->pcpu_thrs = mm_cid_calc_pcpu_thrs(mc);
+ } else {
+ /* Switch back to per task if user count under threshold */
+ if (mc->users < mc->pcpu_thrs)
+ mc->pcpu_thrs = 0;
}
- /*
- * The src_cid is unused, so it can be unset.
- */
- if (!try_cmpxchg(&src_pcpu_cid->cid, &lazy_cid, MM_CID_UNSET))
- return -1;
- WRITE_ONCE(src_pcpu_cid->recent_cid, MM_CID_UNSET);
- return src_cid;
+ /* Mode change required? */
+ if (!!mc->percpu == !!mc->pcpu_thrs)
+ return false;
+ /* When switching back to per TASK mode, set the transition flag */
+ if (!mc->pcpu_thrs)
+ WRITE_ONCE(mc->transit, MM_CID_TRANSIT);
+ WRITE_ONCE(mc->percpu, !!mc->pcpu_thrs);
+ return true;
}
-/*
- * Migration to dst cpu. Called with dst_rq lock held.
- * Interrupts are disabled, which keeps the window of cid ownership without the
- * source rq lock held small.
- */
-void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t)
+static inline void mm_update_cpus_allowed(struct mm_struct *mm, const struct cpumask *affmsk)
{
- struct mm_cid *src_pcpu_cid, *dst_pcpu_cid;
- struct mm_struct *mm = t->mm;
- int src_cid, src_cpu;
- bool dst_cid_is_set;
- struct rq *src_rq;
-
- lockdep_assert_rq_held(dst_rq);
+ struct cpumask *mm_allowed;
+ struct mm_mm_cid *mc;
+ unsigned int weight;
- if (!mm)
+ if (!mm || !READ_ONCE(mm->mm_cid.users))
return;
- src_cpu = t->migrate_from_cpu;
- if (src_cpu == -1) {
- t->last_mm_cid = -1;
- return;
- }
/*
- * Move the src cid if the dst cid is unset. This keeps id
- * allocation closest to 0 in cases where few threads migrate around
- * many CPUs.
- *
- * If destination cid or recent cid is already set, we may have
- * to just clear the src cid to ensure compactness in frequent
- * migrations scenarios.
- *
- * It is not useful to clear the src cid when the number of threads is
- * greater or equal to the number of allowed CPUs, because user-space
- * can expect that the number of allowed cids can reach the number of
- * allowed CPUs.
- */
- dst_pcpu_cid = per_cpu_ptr(mm->pcpu_cid, cpu_of(dst_rq));
- dst_cid_is_set = !mm_cid_is_unset(READ_ONCE(dst_pcpu_cid->cid)) ||
- !mm_cid_is_unset(READ_ONCE(dst_pcpu_cid->recent_cid));
- if (dst_cid_is_set && atomic_read(&mm->mm_users) >= READ_ONCE(mm->nr_cpus_allowed))
+ * mm::mm_cid::mm_cpus_allowed is the superset of each threads
+ * allowed CPUs mask which means it can only grow.
+ */
+ mc = &mm->mm_cid;
+ guard(raw_spinlock)(&mc->lock);
+ mm_allowed = mm_cpus_allowed(mm);
+ weight = cpumask_weighted_or(mm_allowed, mm_allowed, affmsk);
+ if (weight == mc->nr_cpus_allowed)
return;
- src_pcpu_cid = per_cpu_ptr(mm->pcpu_cid, src_cpu);
- src_rq = cpu_rq(src_cpu);
- src_cid = __sched_mm_cid_migrate_from_fetch_cid(src_rq, t, src_pcpu_cid);
- if (src_cid == -1)
+
+ WRITE_ONCE(mc->nr_cpus_allowed, weight);
+ __mm_update_max_cids(mc);
+ if (!mc->percpu)
return;
- src_cid = __sched_mm_cid_migrate_from_try_steal_cid(src_rq, t, src_pcpu_cid,
- src_cid);
- if (src_cid == -1)
+
+ /* Adjust the threshold to the wider set */
+ mc->pcpu_thrs = mm_cid_calc_pcpu_thrs(mc);
+ /* Switch back to per task mode? */
+ if (mc->users >= mc->pcpu_thrs)
return;
- if (dst_cid_is_set) {
- __mm_cid_put(mm, src_cid);
+
+ /* Don't queue twice */
+ if (mc->update_deferred)
return;
- }
- /* Move src_cid to dst cpu. */
- mm_cid_snapshot_time(dst_rq, mm);
- WRITE_ONCE(dst_pcpu_cid->cid, src_cid);
- WRITE_ONCE(dst_pcpu_cid->recent_cid, src_cid);
+
+ /* Queue the irq work, which schedules the real work */
+ mc->update_deferred = true;
+ irq_work_queue(&mc->irq_work);
}
-static void sched_mm_cid_remote_clear(struct mm_struct *mm, struct mm_cid *pcpu_cid,
- int cpu)
+static inline void mm_cid_transit_to_task(struct task_struct *t, struct mm_cid_pcpu *pcp)
{
- struct rq *rq = cpu_rq(cpu);
- struct task_struct *t;
- int cid, lazy_cid;
+ if (cid_on_cpu(t->mm_cid.cid)) {
+ unsigned int cid = cpu_cid_to_cid(t->mm_cid.cid);
- cid = READ_ONCE(pcpu_cid->cid);
- if (!mm_cid_is_valid(cid))
- return;
+ t->mm_cid.cid = cid_to_transit_cid(cid);
+ pcp->cid = t->mm_cid.cid;
+ }
+}
- /*
- * Clear the cpu cid if it is set to keep cid allocation compact. If
- * there happens to be other tasks left on the source cpu using this
- * mm, the next task using this mm will reallocate its cid on context
- * switch.
- */
- lazy_cid = mm_cid_set_lazy_put(cid);
- if (!try_cmpxchg(&pcpu_cid->cid, &cid, lazy_cid))
- return;
+static void mm_cid_fixup_cpus_to_tasks(struct mm_struct *mm)
+{
+ unsigned int cpu;
- /*
- * The implicit barrier after cmpxchg per-mm/cpu cid before loading
- * rq->curr->mm matches the scheduler barrier in context_switch()
- * between store to rq->curr and load of prev and next task's
- * per-mm/cpu cid.
- *
- * The implicit barrier after cmpxchg per-mm/cpu cid before loading
- * rq->curr->mm_cid_active matches the barrier in
- * sched_mm_cid_exit_signals(), sched_mm_cid_before_execve(), and
- * sched_mm_cid_after_execve() between store to t->mm_cid_active and
- * load of per-mm/cpu cid.
- */
+ /* Walk the CPUs and fixup all stale CIDs */
+ for_each_possible_cpu(cpu) {
+ struct mm_cid_pcpu *pcp = per_cpu_ptr(mm->mm_cid.pcpu, cpu);
+ struct rq *rq = cpu_rq(cpu);
- /*
- * If we observe an active task using the mm on this rq after setting
- * the lazy-put flag, that task will be responsible for transitioning
- * from lazy-put flag set to MM_CID_UNSET.
- */
- scoped_guard (rcu) {
- t = rcu_dereference(rq->curr);
- if (READ_ONCE(t->mm_cid_active) && t->mm == mm)
- return;
+ /* Remote access to mm::mm_cid::pcpu requires rq_lock */
+ guard(rq_lock_irq)(rq);
+ /* Is the CID still owned by the CPU? */
+ if (cid_on_cpu(pcp->cid)) {
+ /*
+ * If rq->curr has @mm, transfer it with the
+ * transition bit set. Otherwise drop it.
+ */
+ if (rq->curr->mm == mm && rq->curr->mm_cid.active)
+ mm_cid_transit_to_task(rq->curr, pcp);
+ else
+ mm_drop_cid_on_cpu(mm, pcp);
+
+ } else if (rq->curr->mm == mm && rq->curr->mm_cid.active) {
+ unsigned int cid = rq->curr->mm_cid.cid;
+
+ /* Ensure it has the transition bit set */
+ if (!cid_in_transit(cid)) {
+ cid = cid_to_transit_cid(cid);
+ rq->curr->mm_cid.cid = cid;
+ pcp->cid = cid;
+ }
+ }
}
+ /* Clear the transition bit */
+ WRITE_ONCE(mm->mm_cid.transit, 0);
+}
- /*
- * The cid is unused, so it can be unset.
- * Disable interrupts to keep the window of cid ownership without rq
- * lock small.
- */
- scoped_guard (irqsave) {
- if (try_cmpxchg(&pcpu_cid->cid, &lazy_cid, MM_CID_UNSET))
- __mm_cid_put(mm, cid);
+static inline void mm_cid_transfer_to_cpu(struct task_struct *t, struct mm_cid_pcpu *pcp)
+{
+ if (cid_on_task(t->mm_cid.cid)) {
+ t->mm_cid.cid = cid_to_cpu_cid(t->mm_cid.cid);
+ pcp->cid = t->mm_cid.cid;
}
}
-static void sched_mm_cid_remote_clear_old(struct mm_struct *mm, int cpu)
+static bool mm_cid_fixup_task_to_cpu(struct task_struct *t, struct mm_struct *mm)
{
- struct rq *rq = cpu_rq(cpu);
- struct mm_cid *pcpu_cid;
- struct task_struct *curr;
- u64 rq_clock;
+ /* Remote access to mm::mm_cid::pcpu requires rq_lock */
+ guard(task_rq_lock)(t);
+ /* If the task is not active it is not in the users count */
+ if (!t->mm_cid.active)
+ return false;
+ if (cid_on_task(t->mm_cid.cid)) {
+ /* If running on the CPU, transfer the CID, otherwise drop it */
+ if (task_rq(t)->curr == t)
+ mm_cid_transfer_to_cpu(t, per_cpu_ptr(mm->mm_cid.pcpu, task_cpu(t)));
+ else
+ mm_unset_cid_on_task(t);
+ }
+ return true;
+}
- /*
- * rq->clock load is racy on 32-bit but one spurious clear once in a
- * while is irrelevant.
- */
- rq_clock = READ_ONCE(rq->clock);
- pcpu_cid = per_cpu_ptr(mm->pcpu_cid, cpu);
+static void mm_cid_fixup_tasks_to_cpus(void)
+{
+ struct mm_struct *mm = current->mm;
+ struct task_struct *p, *t;
+ unsigned int users;
/*
- * In order to take care of infrequently scheduled tasks, bump the time
- * snapshot associated with this cid if an active task using the mm is
- * observed on this rq.
+ * This can obviously race with a concurrent affinity change, which
+ * increases the number of allowed CPUs for this mm, but that does
+ * not affect the mode and only changes the CID constraints. A
+ * possible switch back to per task mode happens either in the
+ * deferred handler function or in the next fork()/exit().
+ *
+ * The caller has already transferred. The newly incoming task is
+ * already accounted for, but not yet visible.
*/
- scoped_guard (rcu) {
- curr = rcu_dereference(rq->curr);
- if (READ_ONCE(curr->mm_cid_active) && curr->mm == mm) {
- WRITE_ONCE(pcpu_cid->time, rq_clock);
- return;
- }
+ users = mm->mm_cid.users - 2;
+ if (!users)
+ return;
+
+ guard(rcu)();
+ for_other_threads(current, t) {
+ if (mm_cid_fixup_task_to_cpu(t, mm))
+ users--;
}
- if (rq_clock < pcpu_cid->time + SCHED_MM_CID_PERIOD_NS)
+ if (!users)
return;
- sched_mm_cid_remote_clear(mm, pcpu_cid, cpu);
+
+ /* Happens only for VM_CLONE processes. */
+ for_each_process_thread(p, t) {
+ if (t == current || t->mm != mm)
+ continue;
+ if (mm_cid_fixup_task_to_cpu(t, mm)) {
+ if (--users == 0)
+ return;
+ }
+ }
}
-static void sched_mm_cid_remote_clear_weight(struct mm_struct *mm, int cpu,
- int weight)
+static bool sched_mm_cid_add_user(struct task_struct *t, struct mm_struct *mm)
{
- struct mm_cid *pcpu_cid;
- int cid;
-
- pcpu_cid = per_cpu_ptr(mm->pcpu_cid, cpu);
- cid = READ_ONCE(pcpu_cid->cid);
- if (!mm_cid_is_valid(cid) || cid < weight)
- return;
- sched_mm_cid_remote_clear(mm, pcpu_cid, cpu);
+ t->mm_cid.active = 1;
+ mm->mm_cid.users++;
+ return mm_update_max_cids(mm);
}
-static void task_mm_cid_work(struct callback_head *work)
+void sched_mm_cid_fork(struct task_struct *t)
{
- unsigned long now = jiffies, old_scan, next_scan;
- struct task_struct *t = current;
- struct cpumask *cidmask;
- struct mm_struct *mm;
- int weight, cpu;
+ struct mm_struct *mm = t->mm;
+ bool percpu;
- WARN_ON_ONCE(t != container_of(work, struct task_struct, cid_work));
+ WARN_ON_ONCE(!mm || t->mm_cid.cid != MM_CID_UNSET);
- work->next = work; /* Prevent double-add */
- if (t->flags & PF_EXITING)
- return;
- mm = t->mm;
- if (!mm)
- return;
- old_scan = READ_ONCE(mm->mm_cid_next_scan);
- next_scan = now + msecs_to_jiffies(MM_CID_SCAN_DELAY);
- if (!old_scan) {
- unsigned long res;
-
- res = cmpxchg(&mm->mm_cid_next_scan, old_scan, next_scan);
- if (res != old_scan)
- old_scan = res;
+ guard(mutex)(&mm->mm_cid.mutex);
+ scoped_guard(raw_spinlock_irq, &mm->mm_cid.lock) {
+ struct mm_cid_pcpu *pcp = this_cpu_ptr(mm->mm_cid.pcpu);
+
+ /* First user ? */
+ if (!mm->mm_cid.users) {
+ sched_mm_cid_add_user(t, mm);
+ t->mm_cid.cid = mm_get_cid(mm);
+ /* Required for execve() */
+ pcp->cid = t->mm_cid.cid;
+ return;
+ }
+
+ if (!sched_mm_cid_add_user(t, mm)) {
+ if (!mm->mm_cid.percpu)
+ t->mm_cid.cid = mm_get_cid(mm);
+ return;
+ }
+
+ /* Handle the mode change and transfer current's CID */
+ percpu = !!mm->mm_cid.percpu;
+ if (!percpu)
+ mm_cid_transit_to_task(current, pcp);
else
- old_scan = next_scan;
+ mm_cid_transfer_to_cpu(current, pcp);
}
- if (time_before(now, old_scan))
- return;
- if (!try_cmpxchg(&mm->mm_cid_next_scan, &old_scan, next_scan))
- return;
- cidmask = mm_cidmask(mm);
- /* Clear cids that were not recently used. */
- for_each_possible_cpu(cpu)
- sched_mm_cid_remote_clear_old(mm, cpu);
- weight = cpumask_weight(cidmask);
- /*
- * Clear cids that are greater or equal to the cidmask weight to
- * recompact it.
- */
- for_each_possible_cpu(cpu)
- sched_mm_cid_remote_clear_weight(mm, cpu, weight);
-}
-
-void init_sched_mm_cid(struct task_struct *t)
-{
- struct mm_struct *mm = t->mm;
- int mm_users = 0;
- if (mm) {
- mm_users = atomic_read(&mm->mm_users);
- if (mm_users == 1)
- mm->mm_cid_next_scan = jiffies + msecs_to_jiffies(MM_CID_SCAN_DELAY);
+ if (percpu) {
+ mm_cid_fixup_tasks_to_cpus();
+ } else {
+ mm_cid_fixup_cpus_to_tasks(mm);
+ t->mm_cid.cid = mm_get_cid(mm);
}
- t->cid_work.next = &t->cid_work; /* Protect against double add */
- init_task_work(&t->cid_work, task_mm_cid_work);
}
-void task_tick_mm_cid(struct rq *rq, struct task_struct *curr)
+static bool sched_mm_cid_remove_user(struct task_struct *t)
{
- struct callback_head *work = &curr->cid_work;
- unsigned long now = jiffies;
-
- if (!curr->mm || (curr->flags & (PF_EXITING | PF_KTHREAD)) ||
- work->next != work)
- return;
- if (time_before(now, READ_ONCE(curr->mm->mm_cid_next_scan)))
- return;
-
- /* No page allocation under rq lock */
- task_work_add(curr, work, TWA_RESUME);
+ t->mm_cid.active = 0;
+ scoped_guard(preempt) {
+ /* Clear the transition bit */
+ t->mm_cid.cid = cid_from_transit_cid(t->mm_cid.cid);
+ mm_unset_cid_on_task(t);
+ }
+ t->mm->mm_cid.users--;
+ return mm_update_max_cids(t->mm);
}
-void sched_mm_cid_exit_signals(struct task_struct *t)
+static bool __sched_mm_cid_exit(struct task_struct *t)
{
struct mm_struct *mm = t->mm;
- struct rq *rq;
-
- if (!mm)
- return;
- preempt_disable();
- rq = this_rq();
- guard(rq_lock_irqsave)(rq);
- preempt_enable_no_resched(); /* holding spinlock */
- WRITE_ONCE(t->mm_cid_active, 0);
+ if (!sched_mm_cid_remove_user(t))
+ return false;
+ /*
+ * Contrary to fork() this only deals with a switch back to per
+ * task mode either because the above decreased users or an
+ * affinity change increased the number of allowed CPUs and the
+ * deferred fixup did not run yet.
+ */
+ if (WARN_ON_ONCE(mm->mm_cid.percpu))
+ return false;
/*
- * Store t->mm_cid_active before loading per-mm/cpu cid.
- * Matches barrier in sched_mm_cid_remote_clear_old().
+ * A failed fork(2) cleanup never gets here, so @current must have
+ * the same MM as @t. That's true for exit() and the failed
+ * pthread_create() cleanup case.
*/
- smp_mb();
- mm_cid_put(mm);
- t->last_mm_cid = t->mm_cid = -1;
+ if (WARN_ON_ONCE(current->mm != mm))
+ return false;
+ return true;
}
-void sched_mm_cid_before_execve(struct task_struct *t)
+/*
+ * When a task exits, the MM CID held by the task is not longer required as
+ * the task cannot return to user space.
+ */
+void sched_mm_cid_exit(struct task_struct *t)
{
struct mm_struct *mm = t->mm;
- struct rq *rq;
- if (!mm)
+ if (!mm || !t->mm_cid.active)
return;
+ /*
+ * Ensure that only one instance is doing MM CID operations within
+ * a MM. The common case is uncontended. The rare fixup case adds
+ * some overhead.
+ */
+ scoped_guard(mutex, &mm->mm_cid.mutex) {
+ /* mm_cid::mutex is sufficient to protect mm_cid::users */
+ if (likely(mm->mm_cid.users > 1)) {
+ scoped_guard(raw_spinlock_irq, &mm->mm_cid.lock) {
+ if (!__sched_mm_cid_exit(t))
+ return;
+ /* Mode change required. Transfer currents CID */
+ mm_cid_transit_to_task(current, this_cpu_ptr(mm->mm_cid.pcpu));
+ }
+ mm_cid_fixup_cpus_to_tasks(mm);
+ return;
+ }
+ /* Last user */
+ scoped_guard(raw_spinlock_irq, &mm->mm_cid.lock) {
+ /* Required across execve() */
+ if (t == current)
+ mm_cid_transit_to_task(t, this_cpu_ptr(mm->mm_cid.pcpu));
+ /* Ignore mode change. There is nothing to do. */
+ sched_mm_cid_remove_user(t);
+ }
+ }
- preempt_disable();
- rq = this_rq();
- guard(rq_lock_irqsave)(rq);
- preempt_enable_no_resched(); /* holding spinlock */
- WRITE_ONCE(t->mm_cid_active, 0);
/*
- * Store t->mm_cid_active before loading per-mm/cpu cid.
- * Matches barrier in sched_mm_cid_remote_clear_old().
+ * As this is the last user (execve(), process exit or failed
+ * fork(2)) there is no concurrency anymore.
+ *
+ * Synchronize eventually pending work to ensure that there are no
+ * dangling references left. @t->mm_cid.users is zero so nothing
+ * can queue this work anymore.
*/
- smp_mb();
- mm_cid_put(mm);
- t->last_mm_cid = t->mm_cid = -1;
+ irq_work_sync(&mm->mm_cid.irq_work);
+ cancel_work_sync(&mm->mm_cid.work);
+}
+
+/* Deactivate MM CID allocation across execve() */
+void sched_mm_cid_before_execve(struct task_struct *t)
+{
+ sched_mm_cid_exit(t);
}
+/* Reactivate MM CID after successful execve() */
void sched_mm_cid_after_execve(struct task_struct *t)
{
- struct mm_struct *mm = t->mm;
- struct rq *rq;
+ sched_mm_cid_fork(t);
+}
+
+static void mm_cid_work_fn(struct work_struct *work)
+{
+ struct mm_struct *mm = container_of(work, struct mm_struct, mm_cid.work);
- if (!mm)
+ guard(mutex)(&mm->mm_cid.mutex);
+ /* Did the last user task exit already? */
+ if (!mm->mm_cid.users)
return;
- preempt_disable();
- rq = this_rq();
- scoped_guard (rq_lock_irqsave, rq) {
- preempt_enable_no_resched(); /* holding spinlock */
- WRITE_ONCE(t->mm_cid_active, 1);
- /*
- * Store t->mm_cid_active before loading per-mm/cpu cid.
- * Matches barrier in sched_mm_cid_remote_clear_old().
- */
- smp_mb();
- t->last_mm_cid = t->mm_cid = mm_cid_get(rq, t, mm);
+ scoped_guard(raw_spinlock_irq, &mm->mm_cid.lock) {
+ /* Have fork() or exit() handled it already? */
+ if (!mm->mm_cid.update_deferred)
+ return;
+ /* This clears mm_cid::update_deferred */
+ if (!mm_update_max_cids(mm))
+ return;
+ /* Affinity changes can only switch back to task mode */
+ if (WARN_ON_ONCE(mm->mm_cid.percpu))
+ return;
}
+ mm_cid_fixup_cpus_to_tasks(mm);
}
-void sched_mm_cid_fork(struct task_struct *t)
+static void mm_cid_irq_work(struct irq_work *work)
+{
+ struct mm_struct *mm = container_of(work, struct mm_struct, mm_cid.irq_work);
+
+ /*
+ * Needs to be unconditional because mm_cid::lock cannot be held
+ * when scheduling work as mm_update_cpus_allowed() nests inside
+ * rq::lock and schedule_work() might end up in wakeup...
+ */
+ schedule_work(&mm->mm_cid.work);
+}
+
+void mm_init_cid(struct mm_struct *mm, struct task_struct *p)
{
- WARN_ON_ONCE(!t->mm || t->mm_cid != -1);
- t->mm_cid_active = 1;
+ mm->mm_cid.max_cids = 0;
+ mm->mm_cid.percpu = 0;
+ mm->mm_cid.transit = 0;
+ mm->mm_cid.nr_cpus_allowed = p->nr_cpus_allowed;
+ mm->mm_cid.users = 0;
+ mm->mm_cid.pcpu_thrs = 0;
+ mm->mm_cid.update_deferred = 0;
+ raw_spin_lock_init(&mm->mm_cid.lock);
+ mutex_init(&mm->mm_cid.mutex);
+ mm->mm_cid.irq_work = IRQ_WORK_INIT_HARD(mm_cid_irq_work);
+ INIT_WORK(&mm->mm_cid.work, mm_cid_work_fn);
+ cpumask_copy(mm_cpus_allowed(mm), &p->cpus_mask);
+ bitmap_zero(mm_cidmask(mm), num_possible_cpus());
}
-#endif /* CONFIG_SCHED_MM_CID */
+#else /* CONFIG_SCHED_MM_CID */
+static inline void mm_update_cpus_allowed(struct mm_struct *mm, const struct cpumask *affmsk) { }
+#endif /* !CONFIG_SCHED_MM_CID */
#ifdef CONFIG_SCHED_CLASS_EXT
void sched_deq_and_put_task(struct task_struct *p, int queue_flags,
diff --git a/kernel/sched/membarrier.c b/kernel/sched/membarrier.c
index 62fba83b7bb1..623445603725 100644
--- a/kernel/sched/membarrier.c
+++ b/kernel/sched/membarrier.c
@@ -199,7 +199,7 @@ static void ipi_rseq(void *info)
* is negligible.
*/
smp_mb();
- rseq_preempt(current);
+ rseq_sched_switch_event(current);
}
static void ipi_sync_rq_state(void *info)
@@ -407,9 +407,9 @@ static int membarrier_private_expedited(int flags, int cpu_id)
* membarrier, we will end up with some thread in the mm
* running without a core sync.
*
- * For RSEQ, don't rseq_preempt() the caller. User code
- * is not supposed to issue syscalls at all from inside an
- * rseq critical section.
+ * For RSEQ, don't invoke rseq_sched_switch_event() on the
+ * caller. User code is not supposed to issue syscalls at
+ * all from inside an rseq critical section.
*/
if (flags != MEMBARRIER_FLAG_SYNC_CORE) {
preempt_disable();
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index adfb6e3409d7..f9d0515db130 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -2209,6 +2209,7 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
smp_wmb();
WRITE_ONCE(task_thread_info(p)->cpu, cpu);
p->wake_cpu = cpu;
+ rseq_sched_set_ids_changed(p);
#endif /* CONFIG_SMP */
}
@@ -3540,283 +3541,212 @@ extern const char *preempt_modes[];
#ifdef CONFIG_SCHED_MM_CID
-#define SCHED_MM_CID_PERIOD_NS (100ULL * 1000000) /* 100ms */
-#define MM_CID_SCAN_DELAY 100 /* 100ms */
+static __always_inline bool cid_on_cpu(unsigned int cid)
+{
+ return cid & MM_CID_ONCPU;
+}
-extern raw_spinlock_t cid_lock;
-extern int use_cid_lock;
+static __always_inline bool cid_in_transit(unsigned int cid)
+{
+ return cid & MM_CID_TRANSIT;
+}
-extern void sched_mm_cid_migrate_from(struct task_struct *t);
-extern void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t);
-extern void task_tick_mm_cid(struct rq *rq, struct task_struct *curr);
-extern void init_sched_mm_cid(struct task_struct *t);
+static __always_inline unsigned int cpu_cid_to_cid(unsigned int cid)
+{
+ return cid & ~MM_CID_ONCPU;
+}
-static inline void __mm_cid_put(struct mm_struct *mm, int cid)
+static __always_inline unsigned int cid_to_cpu_cid(unsigned int cid)
{
- if (cid < 0)
- return;
- cpumask_clear_cpu(cid, mm_cidmask(mm));
+ return cid | MM_CID_ONCPU;
}
-/*
- * The per-mm/cpu cid can have the MM_CID_LAZY_PUT flag set or transition to
- * the MM_CID_UNSET state without holding the rq lock, but the rq lock needs to
- * be held to transition to other states.
- *
- * State transitions synchronized with cmpxchg or try_cmpxchg need to be
- * consistent across CPUs, which prevents use of this_cpu_cmpxchg.
- */
-static inline void mm_cid_put_lazy(struct task_struct *t)
+static __always_inline unsigned int cid_to_transit_cid(unsigned int cid)
{
- struct mm_struct *mm = t->mm;
- struct mm_cid __percpu *pcpu_cid = mm->pcpu_cid;
- int cid;
+ return cid | MM_CID_TRANSIT;
+}
- lockdep_assert_irqs_disabled();
- cid = __this_cpu_read(pcpu_cid->cid);
- if (!mm_cid_is_lazy_put(cid) ||
- !try_cmpxchg(&this_cpu_ptr(pcpu_cid)->cid, &cid, MM_CID_UNSET))
- return;
- __mm_cid_put(mm, mm_cid_clear_lazy_put(cid));
+static __always_inline unsigned int cid_from_transit_cid(unsigned int cid)
+{
+ return cid & ~MM_CID_TRANSIT;
}
-static inline int mm_cid_pcpu_unset(struct mm_struct *mm)
+static __always_inline bool cid_on_task(unsigned int cid)
{
- struct mm_cid __percpu *pcpu_cid = mm->pcpu_cid;
- int cid, res;
+ /* True if none of the MM_CID_ONCPU, MM_CID_TRANSIT, MM_CID_UNSET bits is set */
+ return cid < MM_CID_TRANSIT;
+}
- lockdep_assert_irqs_disabled();
- cid = __this_cpu_read(pcpu_cid->cid);
- for (;;) {
- if (mm_cid_is_unset(cid))
- return MM_CID_UNSET;
- /*
- * Attempt transition from valid or lazy-put to unset.
- */
- res = cmpxchg(&this_cpu_ptr(pcpu_cid)->cid, cid, MM_CID_UNSET);
- if (res == cid)
- break;
- cid = res;
- }
- return cid;
+static __always_inline void mm_drop_cid(struct mm_struct *mm, unsigned int cid)
+{
+ clear_bit(cid, mm_cidmask(mm));
}
-static inline void mm_cid_put(struct mm_struct *mm)
+static __always_inline void mm_unset_cid_on_task(struct task_struct *t)
{
- int cid;
+ unsigned int cid = t->mm_cid.cid;
- lockdep_assert_irqs_disabled();
- cid = mm_cid_pcpu_unset(mm);
- if (cid == MM_CID_UNSET)
- return;
- __mm_cid_put(mm, mm_cid_clear_lazy_put(cid));
+ t->mm_cid.cid = MM_CID_UNSET;
+ if (cid_on_task(cid))
+ mm_drop_cid(t->mm, cid);
}
-static inline int __mm_cid_try_get(struct task_struct *t, struct mm_struct *mm)
+static __always_inline void mm_drop_cid_on_cpu(struct mm_struct *mm, struct mm_cid_pcpu *pcp)
{
- struct cpumask *cidmask = mm_cidmask(mm);
- struct mm_cid __percpu *pcpu_cid = mm->pcpu_cid;
- int cid, max_nr_cid, allowed_max_nr_cid;
+ /* Clear the ONCPU bit, but do not set UNSET in the per CPU storage */
+ pcp->cid = cpu_cid_to_cid(pcp->cid);
+ mm_drop_cid(mm, pcp->cid);
+}
- /*
- * After shrinking the number of threads or reducing the number
- * of allowed cpus, reduce the value of max_nr_cid so expansion
- * of cid allocation will preserve cache locality if the number
- * of threads or allowed cpus increase again.
- */
- max_nr_cid = atomic_read(&mm->max_nr_cid);
- while ((allowed_max_nr_cid = min_t(int, READ_ONCE(mm->nr_cpus_allowed),
- atomic_read(&mm->mm_users))),
- max_nr_cid > allowed_max_nr_cid) {
- /* atomic_try_cmpxchg loads previous mm->max_nr_cid into max_nr_cid. */
- if (atomic_try_cmpxchg(&mm->max_nr_cid, &max_nr_cid, allowed_max_nr_cid)) {
- max_nr_cid = allowed_max_nr_cid;
- break;
- }
- }
- /* Try to re-use recent cid. This improves cache locality. */
- cid = __this_cpu_read(pcpu_cid->recent_cid);
- if (!mm_cid_is_unset(cid) && cid < max_nr_cid &&
- !cpumask_test_and_set_cpu(cid, cidmask))
- return cid;
- /*
- * Expand cid allocation if the maximum number of concurrency
- * IDs allocated (max_nr_cid) is below the number cpus allowed
- * and number of threads. Expanding cid allocation as much as
- * possible improves cache locality.
- */
- cid = max_nr_cid;
- while (cid < READ_ONCE(mm->nr_cpus_allowed) && cid < atomic_read(&mm->mm_users)) {
- /* atomic_try_cmpxchg loads previous mm->max_nr_cid into cid. */
- if (!atomic_try_cmpxchg(&mm->max_nr_cid, &cid, cid + 1))
- continue;
- if (!cpumask_test_and_set_cpu(cid, cidmask))
- return cid;
- }
- /*
- * Find the first available concurrency id.
- * Retry finding first zero bit if the mask is temporarily
- * filled. This only happens during concurrent remote-clear
- * which owns a cid without holding a rq lock.
- */
- for (;;) {
- cid = cpumask_first_zero(cidmask);
- if (cid < READ_ONCE(mm->nr_cpus_allowed))
- break;
- cpu_relax();
- }
- if (cpumask_test_and_set_cpu(cid, cidmask))
- return -1;
+static inline unsigned int __mm_get_cid(struct mm_struct *mm, unsigned int max_cids)
+{
+ unsigned int cid = find_first_zero_bit(mm_cidmask(mm), max_cids);
+ if (cid >= max_cids)
+ return MM_CID_UNSET;
+ if (test_and_set_bit(cid, mm_cidmask(mm)))
+ return MM_CID_UNSET;
return cid;
}
-/*
- * Save a snapshot of the current runqueue time of this cpu
- * with the per-cpu cid value, allowing to estimate how recently it was used.
- */
-static inline void mm_cid_snapshot_time(struct rq *rq, struct mm_struct *mm)
+static inline unsigned int mm_get_cid(struct mm_struct *mm)
{
- struct mm_cid *pcpu_cid = per_cpu_ptr(mm->pcpu_cid, cpu_of(rq));
+ unsigned int cid = __mm_get_cid(mm, READ_ONCE(mm->mm_cid.max_cids));
- lockdep_assert_rq_held(rq);
- WRITE_ONCE(pcpu_cid->time, rq->clock);
+ while (cid == MM_CID_UNSET) {
+ cpu_relax();
+ cid = __mm_get_cid(mm, num_possible_cpus());
+ }
+ return cid;
}
-static inline int __mm_cid_get(struct rq *rq, struct task_struct *t,
- struct mm_struct *mm)
+static inline unsigned int mm_cid_converge(struct mm_struct *mm, unsigned int orig_cid,
+ unsigned int max_cids)
{
- int cid;
+ unsigned int new_cid, cid = cpu_cid_to_cid(orig_cid);
- /*
- * All allocations (even those using the cid_lock) are lock-free. If
- * use_cid_lock is set, hold the cid_lock to perform cid allocation to
- * guarantee forward progress.
- */
- if (!READ_ONCE(use_cid_lock)) {
- cid = __mm_cid_try_get(t, mm);
- if (cid >= 0)
- goto end;
- raw_spin_lock(&cid_lock);
- } else {
- raw_spin_lock(&cid_lock);
- cid = __mm_cid_try_get(t, mm);
- if (cid >= 0)
- goto unlock;
+ /* Is it in the optimal CID space? */
+ if (likely(cid < max_cids))
+ return orig_cid;
+
+ /* Try to find one in the optimal space. Otherwise keep the provided. */
+ new_cid = __mm_get_cid(mm, max_cids);
+ if (new_cid != MM_CID_UNSET) {
+ mm_drop_cid(mm, cid);
+ /* Preserve the ONCPU mode of the original CID */
+ return new_cid | (orig_cid & MM_CID_ONCPU);
}
+ return orig_cid;
+}
- /*
- * cid concurrently allocated. Retry while forcing following
- * allocations to use the cid_lock to ensure forward progress.
- */
- WRITE_ONCE(use_cid_lock, 1);
- /*
- * Set use_cid_lock before allocation. Only care about program order
- * because this is only required for forward progress.
- */
- barrier();
- /*
- * Retry until it succeeds. It is guaranteed to eventually succeed once
- * all newcoming allocations observe the use_cid_lock flag set.
- */
- do {
- cid = __mm_cid_try_get(t, mm);
- cpu_relax();
- } while (cid < 0);
- /*
- * Allocate before clearing use_cid_lock. Only care about
- * program order because this is for forward progress.
- */
- barrier();
- WRITE_ONCE(use_cid_lock, 0);
-unlock:
- raw_spin_unlock(&cid_lock);
-end:
- mm_cid_snapshot_time(rq, mm);
+static __always_inline void mm_cid_update_task_cid(struct task_struct *t, unsigned int cid)
+{
+ if (t->mm_cid.cid != cid) {
+ t->mm_cid.cid = cid;
+ rseq_sched_set_ids_changed(t);
+ }
+}
- return cid;
+static __always_inline void mm_cid_update_pcpu_cid(struct mm_struct *mm, unsigned int cid)
+{
+ __this_cpu_write(mm->mm_cid.pcpu->cid, cid);
}
-static inline int mm_cid_get(struct rq *rq, struct task_struct *t,
- struct mm_struct *mm)
+static __always_inline void mm_cid_from_cpu(struct task_struct *t, unsigned int cpu_cid)
{
- struct mm_cid __percpu *pcpu_cid = mm->pcpu_cid;
- int cid;
+ unsigned int max_cids, tcid = t->mm_cid.cid;
+ struct mm_struct *mm = t->mm;
- lockdep_assert_rq_held(rq);
- cid = __this_cpu_read(pcpu_cid->cid);
- if (mm_cid_is_valid(cid)) {
- mm_cid_snapshot_time(rq, mm);
- return cid;
- }
- if (mm_cid_is_lazy_put(cid)) {
- if (try_cmpxchg(&this_cpu_ptr(pcpu_cid)->cid, &cid, MM_CID_UNSET))
- __mm_cid_put(mm, mm_cid_clear_lazy_put(cid));
+ max_cids = READ_ONCE(mm->mm_cid.max_cids);
+ /* Optimize for the common case where both have the ONCPU bit set */
+ if (likely(cid_on_cpu(cpu_cid & tcid))) {
+ if (likely(cpu_cid_to_cid(cpu_cid) < max_cids)) {
+ mm_cid_update_task_cid(t, cpu_cid);
+ return;
+ }
+ /* Try to converge into the optimal CID space */
+ cpu_cid = mm_cid_converge(mm, cpu_cid, max_cids);
+ } else {
+ /* Hand over or drop the task owned CID */
+ if (cid_on_task(tcid)) {
+ if (cid_on_cpu(cpu_cid))
+ mm_unset_cid_on_task(t);
+ else
+ cpu_cid = cid_to_cpu_cid(tcid);
+ }
+ /* Still nothing, allocate a new one */
+ if (!cid_on_cpu(cpu_cid))
+ cpu_cid = cid_to_cpu_cid(mm_get_cid(mm));
}
- cid = __mm_cid_get(rq, t, mm);
- __this_cpu_write(pcpu_cid->cid, cid);
- __this_cpu_write(pcpu_cid->recent_cid, cid);
-
- return cid;
+ mm_cid_update_pcpu_cid(mm, cpu_cid);
+ mm_cid_update_task_cid(t, cpu_cid);
}
-static inline void switch_mm_cid(struct rq *rq,
- struct task_struct *prev,
- struct task_struct *next)
+static __always_inline void mm_cid_from_task(struct task_struct *t, unsigned int cpu_cid)
{
- /*
- * Provide a memory barrier between rq->curr store and load of
- * {prev,next}->mm->pcpu_cid[cpu] on rq->curr->mm transition.
- *
- * Should be adapted if context_switch() is modified.
- */
- if (!next->mm) { // to kernel
- /*
- * user -> kernel transition does not guarantee a barrier, but
- * we can use the fact that it performs an atomic operation in
- * mmgrab().
- */
- if (prev->mm) // from user
- smp_mb__after_mmgrab();
- /*
- * kernel -> kernel transition does not change rq->curr->mm
- * state. It stays NULL.
- */
- } else { // to user
- /*
- * kernel -> user transition does not provide a barrier
- * between rq->curr store and load of {prev,next}->mm->pcpu_cid[cpu].
- * Provide it here.
- */
- if (!prev->mm) { // from kernel
- smp_mb();
- } else { // from user
- /*
- * user->user transition relies on an implicit
- * memory barrier in switch_mm() when
- * current->mm changes. If the architecture
- * switch_mm() does not have an implicit memory
- * barrier, it is emitted here. If current->mm
- * is unchanged, no barrier is needed.
- */
- smp_mb__after_switch_mm();
+ unsigned int max_cids, tcid = t->mm_cid.cid;
+ struct mm_struct *mm = t->mm;
+
+ max_cids = READ_ONCE(mm->mm_cid.max_cids);
+ /* Optimize for the common case, where both have the ONCPU bit clear */
+ if (likely(cid_on_task(tcid | cpu_cid))) {
+ if (likely(tcid < max_cids)) {
+ mm_cid_update_pcpu_cid(mm, tcid);
+ return;
}
+ /* Try to converge into the optimal CID space */
+ tcid = mm_cid_converge(mm, tcid, max_cids);
+ } else {
+ /* Hand over or drop the CPU owned CID */
+ if (cid_on_cpu(cpu_cid)) {
+ if (cid_on_task(tcid))
+ mm_drop_cid_on_cpu(mm, this_cpu_ptr(mm->mm_cid.pcpu));
+ else
+ tcid = cpu_cid_to_cid(cpu_cid);
+ }
+ /* Still nothing, allocate a new one */
+ if (!cid_on_task(tcid))
+ tcid = mm_get_cid(mm);
+ /* Set the transition mode flag if required */
+ tcid |= READ_ONCE(mm->mm_cid.transit);
}
- if (prev->mm_cid_active) {
- mm_cid_snapshot_time(rq, prev->mm);
- mm_cid_put_lazy(prev);
- prev->mm_cid = -1;
- }
- if (next->mm_cid_active)
- next->last_mm_cid = next->mm_cid = mm_cid_get(rq, next, next->mm);
+ mm_cid_update_pcpu_cid(mm, tcid);
+ mm_cid_update_task_cid(t, tcid);
+}
+
+static __always_inline void mm_cid_schedin(struct task_struct *next)
+{
+ struct mm_struct *mm = next->mm;
+ unsigned int cpu_cid;
+
+ if (!next->mm_cid.active)
+ return;
+
+ cpu_cid = __this_cpu_read(mm->mm_cid.pcpu->cid);
+ if (likely(!READ_ONCE(mm->mm_cid.percpu)))
+ mm_cid_from_task(next, cpu_cid);
+ else
+ mm_cid_from_cpu(next, cpu_cid);
+}
+
+static __always_inline void mm_cid_schedout(struct task_struct *prev)
+{
+ /* During mode transitions CIDs are temporary and need to be dropped */
+ if (likely(!cid_in_transit(prev->mm_cid.cid)))
+ return;
+
+ mm_drop_cid(prev->mm, cid_from_transit_cid(prev->mm_cid.cid));
+ prev->mm_cid.cid = MM_CID_UNSET;
+}
+
+static inline void mm_cid_switch_to(struct task_struct *prev, struct task_struct *next)
+{
+ mm_cid_schedout(prev);
+ mm_cid_schedin(next);
}
#else /* !CONFIG_SCHED_MM_CID: */
-static inline void switch_mm_cid(struct rq *rq, struct task_struct *prev, struct task_struct *next) { }
-static inline void sched_mm_cid_migrate_from(struct task_struct *t) { }
-static inline void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t) { }
-static inline void task_tick_mm_cid(struct rq *rq, struct task_struct *curr) { }
-static inline void init_sched_mm_cid(struct task_struct *t) { }
+static inline void mm_cid_switch_to(struct task_struct *prev, struct task_struct *next) { }
#endif /* !CONFIG_SCHED_MM_CID */
extern u64 avg_vruntime(struct cfs_rq *cfs_rq);
diff --git a/kernel/signal.c b/kernel/signal.c
index fe9190d84f28..e42b8bd6922f 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -3125,7 +3125,6 @@ void exit_signals(struct task_struct *tsk)
cgroup_threadgroup_change_begin(tsk);
if (thread_group_empty(tsk) || (tsk->signal->flags & SIGNAL_GROUP_EXIT)) {
- sched_mm_cid_exit_signals(tsk);
tsk->flags |= PF_EXITING;
cgroup_threadgroup_change_end(tsk);
return;
@@ -3136,7 +3135,6 @@ void exit_signals(struct task_struct *tsk)
* From now this task is not visible for group-wide signals,
* see wants_signal(), do_signal_stop().
*/
- sched_mm_cid_exit_signals(tsk);
tsk->flags |= PF_EXITING;
cgroup_threadgroup_change_end(tsk);
diff --git a/lib/bitmap.c b/lib/bitmap.c
index b97692854966..9dc526507875 100644
--- a/lib/bitmap.c
+++ b/lib/bitmap.c
@@ -355,6 +355,12 @@ unsigned int __bitmap_weight_andnot(const unsigned long *bitmap1,
}
EXPORT_SYMBOL(__bitmap_weight_andnot);
+unsigned int __bitmap_weighted_or(unsigned long *dst, const unsigned long *bitmap1,
+ const unsigned long *bitmap2, unsigned int bits)
+{
+ return BITMAP_WEIGHT(({dst[idx] = bitmap1[idx] | bitmap2[idx]; dst[idx]; }), bits);
+}
+
void __bitmap_set(unsigned long *map, unsigned int start, int len)
{
unsigned long *p = map + BIT_WORD(start);
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index b7a0ae2a7b20..4255fcf9c6e5 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -49,6 +49,7 @@
#include <linux/lockdep.h>
#include <linux/kthread.h>
#include <linux/suspend.h>
+#include <linux/rseq.h>
#include <asm/processor.h>
#include <asm/ioctl.h>
@@ -4476,6 +4477,12 @@ static long kvm_vcpu_ioctl(struct file *filp,
r = kvm_arch_vcpu_ioctl_run(vcpu);
vcpu->wants_to_run = false;
+ /*
+ * FIXME: Remove this hack once all KVM architectures
+ * support the generic TIF bits, i.e. a dedicated TIF_RSEQ.
+ */
+ rseq_virt_userspace_exit();
+
trace_kvm_userspace_exit(vcpu->run->exit_reason, r);
break;
}© 2016 - 2025 Red Hat, Inc.