RE: x86: Support for lazy preemption in Debian Linux 4.14.20 AMD64
Niltze [Hello], Thomas-
FYI
Starting from Linux kernel 4.14.18, arch/x86/include/asm/thread_info.h
adds an extra line of code that affects your patch --included in
Debian kernel packaging for Linux 4.14.17 / 4.15.4:
debian/patches/features/all/rt/x86-preempt-lazy.patch
i.e.,
...
struct task_struct;
#include <asm/cpufeature.h>
#include <linux/atomic.h>
struct thread_info {
unsigned long flags; /* low level flags */
u32 status; /* thread synchronous flags */
...
When Debian packaging applies the RT patch with fuzz=0, it fails.
Accordingly, I edited your patch (attached) for it to apply (fuzz=0)
to my subject Debian kernel -- *assuming* the RT kernel version will
want to keep the new line of code.
Best Professional Regards.
--
Jose R R
http://metztli.it
---------------------------------------------------------------------------------------------
Download Metztli Reiser4: Debian Stretch w/ Linux 4.14 AMD64
---------------------------------------------------------------------------------------------
feats ZSTD compression https://sf.net/projects/metztli-reiser4/
Subject: x86: Support for lazy preemption
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 01 Nov 2012 11:03:47 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.14/older/patches-4.14.15-rt13.tar.xz
Implement the x86 pieces for lazy preempt.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/x86/Kconfig | 1 +
arch/x86/entry/common.c | 4 ++--
arch/x86/entry/entry_32.S | 17 +++++++++++++++++
arch/x86/entry/entry_64.S | 16 ++++++++++++++++
arch/x86/include/asm/preempt.h | 31 ++++++++++++++++++++++++++++++-
arch/x86/include/asm/thread_info.h | 11 +++++++++++
arch/x86/kernel/asm-offsets.c | 2 ++
7 files changed, 79 insertions(+), 3 deletions(-)
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -170,6 +170,7 @@ config X86
select HAVE_HARDLOCKUP_DETECTOR_PERF if PERF_EVENTS && HAVE_PERF_EVENTS_NMI
select HAVE_PERF_REGS
select HAVE_PERF_USER_STACK_DUMP
+ select HAVE_PREEMPT_LAZY
select HAVE_RCU_TABLE_FREE
select HAVE_REGS_AND_STACK_ACCESS_API
select HAVE_RELIABLE_STACKTRACE if X86_64 && UNWINDER_FRAME_POINTER && STACK_VALIDATION
--- a/arch/x86/entry/common.c
+++ b/arch/x86/entry/common.c
@@ -132,7 +132,7 @@ static long syscall_trace_enter(struct p
#define EXIT_TO_USERMODE_LOOP_FLAGS \
(_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_UPROBE | \
- _TIF_NEED_RESCHED | _TIF_USER_RETURN_NOTIFY | _TIF_PATCH_PENDING)
+ _TIF_NEED_RESCHED_MASK | _TIF_USER_RETURN_NOTIFY | _TIF_PATCH_PENDING)
static void exit_to_usermode_loop(struct pt_regs *regs, u32 cached_flags)
{
@@ -147,7 +147,7 @@ static void exit_to_usermode_loop(struct
/* We have work to do. */
local_irq_enable();
- if (cached_flags & _TIF_NEED_RESCHED)
+ if (cached_flags & _TIF_NEED_RESCHED_MASK)
schedule();
#ifdef ARCH_RT_DELAYS_SIGNAL_SEND
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -350,8 +350,25 @@ END(ret_from_exception)
ENTRY(resume_kernel)
DISABLE_INTERRUPTS(CLBR_ANY)
.Lneed_resched:
+ # preempt count == 0 + NEED_RS set?
cmpl $0, PER_CPU_VAR(__preempt_count)
+#ifndef CONFIG_PREEMPT_LAZY
jnz restore_all
+#else
+ jz test_int_off
+
+ # atleast preempt count == 0 ?
+ cmpl $_PREEMPT_ENABLED,PER_CPU_VAR(__preempt_count)
+ jne restore_all
+
+ movl PER_CPU_VAR(current_task), %ebp
+ cmpl $0,TASK_TI_preempt_lazy_count(%ebp) # non-zero preempt_lazy_count ?
+ jnz restore_all
+
+ testl $_TIF_NEED_RESCHED_LAZY, TASK_TI_flags(%ebp)
+ jz restore_all
+test_int_off:
+#endif
testl $X86_EFLAGS_IF, PT_EFLAGS(%esp) # interrupts off (exception path) ?
jz restore_all
call preempt_schedule_irq
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -778,7 +778,23 @@ GLOBAL(swapgs_restore_regs_and_return_to
bt $9, EFLAGS(%rsp) /* were interrupts off? */
jnc 1f
0: cmpl $0, PER_CPU_VAR(__preempt_count)
+#ifndef CONFIG_PREEMPT_LAZY
jnz 1f
+#else
+ jz do_preempt_schedule_irq
+
+ # atleast preempt count == 0 ?
+ cmpl $_PREEMPT_ENABLED,PER_CPU_VAR(__preempt_count)
+ jnz 1f
+
+ movq PER_CPU_VAR(current_task), %rcx
+ cmpl $0, TASK_TI_preempt_lazy_count(%rcx)
+ jnz 1f
+
+ bt $TIF_NEED_RESCHED_LAZY,TASK_TI_flags(%rcx)
+ jnc 1f
+do_preempt_schedule_irq:
+#endif
call preempt_schedule_irq
jmp 0b
1:
--- a/arch/x86/include/asm/preempt.h
+++ b/arch/x86/include/asm/preempt.h
@@ -86,17 +86,46 @@ static __always_inline void __preempt_co
* a decrement which hits zero means we have no preempt_count and should
* reschedule.
*/
-static __always_inline bool __preempt_count_dec_and_test(void)
+static __always_inline bool ____preempt_count_dec_and_test(void)
{
GEN_UNARY_RMWcc("decl", __preempt_count, __percpu_arg(0), e);
}
+static __always_inline bool __preempt_count_dec_and_test(void)
+{
+ if (____preempt_count_dec_and_test())
+ return true;
+#ifdef CONFIG_PREEMPT_LAZY
+ if (current_thread_info()->preempt_lazy_count)
+ return false;
+ return test_thread_flag(TIF_NEED_RESCHED_LAZY);
+#else
+ return false;
+#endif
+}
+
/*
* Returns true when we need to resched and can (barring IRQ state).
*/
static __always_inline bool should_resched(int preempt_offset)
{
+#ifdef CONFIG_PREEMPT_LAZY
+ u32 tmp;
+
+ tmp = raw_cpu_read_4(__preempt_count);
+ if (tmp == preempt_offset)
+ return true;
+
+ /* preempt count == 0 ? */
+ tmp &= ~PREEMPT_NEED_RESCHED;
+ if (tmp)
+ return false;
+ if (current_thread_info()->preempt_lazy_count)
+ return false;
+ return test_thread_flag(TIF_NEED_RESCHED_LAZY);
+#else
return unlikely(raw_cpu_read_4(__preempt_count) == preempt_offset);
+#endif
}
#ifdef CONFIG_PREEMPT
--- a/arch/x86/include/asm/thread_info.h
+++ b/arch/x86/include/asm/thread_info.h
@@ -55,11 +55,14 @@ struct task_struct;
struct thread_info {
unsigned long flags; /* low level flags */
u32 status; /* thread synchronous flags */
+ int preempt_lazy_count; /* 0 => lazy preemptable
+ <0 => BUG */
};
#define INIT_THREAD_INFO(tsk) \
{ \
.flags = 0, \
+ .preempt_lazy_count = 0, \
}
#define init_stack (init_thread_union.stack)
@@ -68,6 +71,10 @@ struct thread_info {
#include <asm/asm-offsets.h>
+#define GET_THREAD_INFO(reg) \
+ _ASM_MOV PER_CPU_VAR(cpu_current_top_of_stack),reg ; \
+ _ASM_SUB $(THREAD_SIZE),reg ;
+
#endif
/*
@@ -83,6 +90,7 @@ struct thread_info {
#define TIF_SYSCALL_EMU 6 /* syscall emulation active */
#define TIF_SYSCALL_AUDIT 7 /* syscall auditing active */
#define TIF_SECCOMP 8 /* secure computing */
+#define TIF_NEED_RESCHED_LAZY 9 /* lazy rescheduling necessary */
#define TIF_USER_RETURN_NOTIFY 11 /* notify kernel of userspace return */
#define TIF_UPROBE 12 /* breakpointed or singlestepping */
#define TIF_PATCH_PENDING 13 /* pending live patching update */
@@ -109,6 +117,7 @@ struct thread_info {
#define _TIF_SYSCALL_EMU (1 << TIF_SYSCALL_EMU)
#define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT)
#define _TIF_SECCOMP (1 << TIF_SECCOMP)
+#define _TIF_NEED_RESCHED_LAZY (1 << TIF_NEED_RESCHED_LAZY)
#define _TIF_USER_RETURN_NOTIFY (1 << TIF_USER_RETURN_NOTIFY)
#define _TIF_UPROBE (1 << TIF_UPROBE)
#define _TIF_PATCH_PENDING (1 << TIF_PATCH_PENDING)
@@ -150,6 +159,8 @@ struct thread_info {
#define _TIF_WORK_CTXSW_PREV (_TIF_WORK_CTXSW|_TIF_USER_RETURN_NOTIFY)
#define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW)
+#define _TIF_NEED_RESCHED_MASK (_TIF_NEED_RESCHED | _TIF_NEED_RESCHED_LAZY)
+
#define STACK_WARN (THREAD_SIZE/8)
/*
--- a/arch/x86/kernel/asm-offsets.c
+++ b/arch/x86/kernel/asm-offsets.c
@@ -38,6 +38,7 @@ void common(void) {
BLANK();
OFFSET(TASK_TI_flags, task_struct, thread_info.flags);
+ OFFSET(TASK_TI_preempt_lazy_count, task_struct, thread_info.preempt_lazy_count);
OFFSET(TASK_addr_limit, task_struct, thread.addr_limit);
BLANK();
@@ -94,6 +95,7 @@ void common(void) {
BLANK();
DEFINE(PTREGS_SIZE, sizeof(struct pt_regs));
+ DEFINE(_PREEMPT_ENABLED, PREEMPT_ENABLED);
/* TLB state for the entry code */
OFFSET(TLB_STATE_user_pcid_flush_mask, tlb_state, user_pcid_flush_mask);
Reply to: