WSL2-Linux-Kernel/arch/powerpc/include/asm/paravirt.h

123 строки
3.1 KiB
C
Исходник Обычный вид История

/* SPDX-License-Identifier: GPL-2.0-or-later */
#ifndef _ASM_POWERPC_PARAVIRT_H
#define _ASM_POWERPC_PARAVIRT_H
#include <linux/jump_label.h>
#include <asm/smp.h>
#ifdef CONFIG_PPC64
#include <asm/paca.h>
#include <asm/hvcall.h>
#endif
#ifdef CONFIG_PPC_SPLPAR
#include <linux/smp.h>
powerpc/paravirt: Use is_kvm_guest() in vcpu_is_preempted() If its a shared LPAR but not a KVM guest, then see if the vCPU is related to the calling vCPU. On PowerVM, only cores can be preempted. So if one vCPU is a non-preempted state, we can decipher that all other vCPUs sharing the same core are in non-preempted state. Performance results: $ perf stat -r 5 -a perf bench sched pipe -l 10000000 (lesser time is better) powerpc/next 35,107,951.20 msec cpu-clock # 255.898 CPUs utilized ( +- 0.31% ) 23,655,348 context-switches # 0.674 K/sec ( +- 3.72% ) 14,465 cpu-migrations # 0.000 K/sec ( +- 5.37% ) 82,463 page-faults # 0.002 K/sec ( +- 8.40% ) 1,127,182,328,206 cycles # 0.032 GHz ( +- 1.60% ) (66.67%) 78,587,300,622 stalled-cycles-frontend # 6.97% frontend cycles idle ( +- 0.08% ) (50.01%) 654,124,218,432 stalled-cycles-backend # 58.03% backend cycles idle ( +- 1.74% ) (50.01%) 834,013,059,242 instructions # 0.74 insn per cycle # 0.78 stalled cycles per insn ( +- 0.73% ) (66.67%) 132,911,454,387 branches # 3.786 M/sec ( +- 0.59% ) (50.00%) 2,890,882,143 branch-misses # 2.18% of all branches ( +- 0.46% ) (50.00%) 137.195 +- 0.419 seconds time elapsed ( +- 0.31% ) powerpc/next + patchset 29,981,702.64 msec cpu-clock # 255.881 CPUs utilized ( +- 1.30% ) 40,162,456 context-switches # 0.001 M/sec ( +- 0.01% ) 1,110 cpu-migrations # 0.000 K/sec ( +- 5.20% ) 62,616 page-faults # 0.002 K/sec ( +- 3.93% ) 1,430,030,626,037 cycles # 0.048 GHz ( +- 1.41% ) (66.67%) 83,202,707,288 stalled-cycles-frontend # 5.82% frontend cycles idle ( +- 0.75% ) (50.01%) 744,556,088,520 stalled-cycles-backend # 52.07% backend cycles idle ( +- 1.39% ) (50.01%) 940,138,418,674 instructions # 0.66 insn per cycle # 0.79 stalled cycles per insn ( +- 0.51% ) (66.67%) 146,452,852,283 branches # 4.885 M/sec ( +- 0.80% ) (50.00%) 3,237,743,996 branch-misses # 2.21% of all branches ( +- 1.18% ) (50.01%) 117.17 +- 1.52 seconds time elapsed ( +- 1.30% ) This is around 14.6% improvement in performance. Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Waiman Long <longman@redhat.com> [mpe: Fold in performance results from cover letter] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20201202050456.164005-5-srikar@linux.vnet.ibm.com
2020-12-02 08:04:56 +03:00
#include <asm/kvm_guest.h>
#include <asm/cputhreads.h>
DECLARE_STATIC_KEY_FALSE(shared_processor);
static inline bool is_shared_processor(void)
{
return static_branch_unlikely(&shared_processor);
}
/* If bit 0 is set, the cpu has been preempted */
static inline u32 yield_count_of(int cpu)
{
__be32 yield_count = READ_ONCE(lppaca_of(cpu).yield_count);
return be32_to_cpu(yield_count);
}
/*
* Spinlock code confers and prods, so don't trace the hcalls because the
* tracing code takes spinlocks which can cause recursion deadlocks.
*
* These calls are made while the lock is not held: the lock slowpath yields if
* it can not acquire the lock, and unlock slow path might prod if a waiter has
* yielded). So this may not be a problem for simple spin locks because the
* tracing does not technically recurse on the lock, but we avoid it anyway.
*
* However the queued spin lock contended path is more strictly ordered: the
* H_CONFER hcall is made after the task has queued itself on the lock, so then
* recursing on that lock will cause the task to then queue up again behind the
* first instance (or worse: queued spinlocks use tricks that assume a context
* never waits on more than one spinlock, so such recursion may cause random
* corruption in the lock code).
*/
static inline void yield_to_preempted(int cpu, u32 yield_count)
{
plpar_hcall_norets_notrace(H_CONFER, get_hard_smp_processor_id(cpu), yield_count);
}
static inline void prod_cpu(int cpu)
{
plpar_hcall_norets_notrace(H_PROD, get_hard_smp_processor_id(cpu));
}
static inline void yield_to_any(void)
{
plpar_hcall_norets_notrace(H_CONFER, -1, 0);
}
#else
static inline bool is_shared_processor(void)
{
return false;
}
static inline u32 yield_count_of(int cpu)
{
return 0;
}
extern void ___bad_yield_to_preempted(void);
static inline void yield_to_preempted(int cpu, u32 yield_count)
{
___bad_yield_to_preempted(); /* This would be a bug */
}
extern void ___bad_yield_to_any(void);
static inline void yield_to_any(void)
{
___bad_yield_to_any(); /* This would be a bug */
}
extern void ___bad_prod_cpu(void);
static inline void prod_cpu(int cpu)
{
___bad_prod_cpu(); /* This would be a bug */
}
#endif
#define vcpu_is_preempted vcpu_is_preempted
static inline bool vcpu_is_preempted(int cpu)
{
if (!is_shared_processor())
return false;
powerpc/paravirt: Use is_kvm_guest() in vcpu_is_preempted() If its a shared LPAR but not a KVM guest, then see if the vCPU is related to the calling vCPU. On PowerVM, only cores can be preempted. So if one vCPU is a non-preempted state, we can decipher that all other vCPUs sharing the same core are in non-preempted state. Performance results: $ perf stat -r 5 -a perf bench sched pipe -l 10000000 (lesser time is better) powerpc/next 35,107,951.20 msec cpu-clock # 255.898 CPUs utilized ( +- 0.31% ) 23,655,348 context-switches # 0.674 K/sec ( +- 3.72% ) 14,465 cpu-migrations # 0.000 K/sec ( +- 5.37% ) 82,463 page-faults # 0.002 K/sec ( +- 8.40% ) 1,127,182,328,206 cycles # 0.032 GHz ( +- 1.60% ) (66.67%) 78,587,300,622 stalled-cycles-frontend # 6.97% frontend cycles idle ( +- 0.08% ) (50.01%) 654,124,218,432 stalled-cycles-backend # 58.03% backend cycles idle ( +- 1.74% ) (50.01%) 834,013,059,242 instructions # 0.74 insn per cycle # 0.78 stalled cycles per insn ( +- 0.73% ) (66.67%) 132,911,454,387 branches # 3.786 M/sec ( +- 0.59% ) (50.00%) 2,890,882,143 branch-misses # 2.18% of all branches ( +- 0.46% ) (50.00%) 137.195 +- 0.419 seconds time elapsed ( +- 0.31% ) powerpc/next + patchset 29,981,702.64 msec cpu-clock # 255.881 CPUs utilized ( +- 1.30% ) 40,162,456 context-switches # 0.001 M/sec ( +- 0.01% ) 1,110 cpu-migrations # 0.000 K/sec ( +- 5.20% ) 62,616 page-faults # 0.002 K/sec ( +- 3.93% ) 1,430,030,626,037 cycles # 0.048 GHz ( +- 1.41% ) (66.67%) 83,202,707,288 stalled-cycles-frontend # 5.82% frontend cycles idle ( +- 0.75% ) (50.01%) 744,556,088,520 stalled-cycles-backend # 52.07% backend cycles idle ( +- 1.39% ) (50.01%) 940,138,418,674 instructions # 0.66 insn per cycle # 0.79 stalled cycles per insn ( +- 0.51% ) (66.67%) 146,452,852,283 branches # 4.885 M/sec ( +- 0.80% ) (50.00%) 3,237,743,996 branch-misses # 2.21% of all branches ( +- 1.18% ) (50.01%) 117.17 +- 1.52 seconds time elapsed ( +- 1.30% ) This is around 14.6% improvement in performance. Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Waiman Long <longman@redhat.com> [mpe: Fold in performance results from cover letter] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20201202050456.164005-5-srikar@linux.vnet.ibm.com
2020-12-02 08:04:56 +03:00
#ifdef CONFIG_PPC_SPLPAR
if (!is_kvm_guest()) {
int first_cpu = cpu_first_thread_sibling(smp_processor_id());
/*
* Preemption can only happen at core granularity. This CPU
* is not preempted if one of the CPU of this core is not
* preempted.
*/
if (cpu_first_thread_sibling(cpu) == first_cpu)
return false;
}
#endif
if (yield_count_of(cpu) & 1)
return true;
return false;
}
static inline bool pv_is_native_spin_unlock(void)
{
return !is_shared_processor();
}
#endif /* _ASM_POWERPC_PARAVIRT_H */