| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <tip-7ef28739c453bb77dd2dc7a5ddcb534e488e0826@git.kernel.org>
Date: Sat, 23 Jun 2018 05:26:29 -0700
From: tip-bot for Reinette Chatre <tipbot@...or.com>
To: linux-tip-commits@...r.kernel.org
Cc: mingo@...nel.org, linux-kernel@...r.kernel.org,
reinette.chatre@...el.com, tglx@...utronix.de, hpa@...or.com
Subject: [tip:x86/cache] x86/intel_rdt: More precise L2 hit/miss
measurements
Commit-ID: 7ef28739c453bb77dd2dc7a5ddcb534e488e0826
Gitweb: https://git.kernel.org/tip/7ef28739c453bb77dd2dc7a5ddcb534e488e0826
Author: Reinette Chatre <reinette.chatre@...el.com>
AuthorDate: Fri, 22 Jun 2018 15:42:28 -0700
Committer: Thomas Gleixner <tglx@...utronix.de>
CommitDate: Sat, 23 Jun 2018 13:03:52 +0200
x86/intel_rdt: More precise L2 hit/miss measurements
Intel Goldmont processors supports non-architectural precise events that
can be used to give us more insight into the success of L2 cache
pseudo-locking on these platforms.
Introduce a new measurement trigger that will enable two precise events,
MEM_LOAD_UOPS_RETIRED.L2_HIT and MEM_LOAD_UOPS_RETIRED.L2_MISS, while
accessing pseudo-locked data. A new tracepoint, pseudo_lock_l2, is
created to make these results visible to the user.
Signed-off-by: Reinette Chatre <reinette.chatre@...el.com>
Signed-off-by: Thomas Gleixner <tglx@...utronix.de>
Cc: fenghua.yu@...el.com
Cc: tony.luck@...el.com
Cc: vikas.shivappa@...ux.intel.com
Cc: gavin.hindman@...el.com
Cc: jithu.joseph@...el.com
Cc: dave.hansen@...el.com
Cc: hpa@...or.com
Link: https://lkml.kernel.org/r/06b1456da65b543479dac8d9493e41f92f175d6c.1529706536.git.reinette.chatre@intel.com
---
arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c | 145 ++++++++++++++++++++--
arch/x86/kernel/cpu/intel_rdt_pseudo_lock_event.h | 10 ++
2 files changed, 146 insertions(+), 9 deletions(-)
diff --git a/arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c b/arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c
index 5020ac3e83a1..aa9fa8a16875 100644
--- a/arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c
+++ b/arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c
@@ -21,6 +21,7 @@
#include <asm/cacheflush.h>
#include <asm/intel-family.h>
#include <asm/intel_rdt_sched.h>
+#include <asm/perf_event.h>
#include "intel_rdt.h"
#define CREATE_TRACE_POINTS
@@ -60,6 +61,9 @@ static struct class *pseudo_lock_class;
* hardware prefetch disable bits are included here as they are documented
* in the SDM.
*
+ * When adding a platform here also add support for its cache events to
+ * measure_cycles_perf_fn()
+ *
* Return:
* If platform is supported, the bits to disable hardware prefetchers, 0
* if platform is not supported.
@@ -98,6 +102,16 @@ static u64 get_prefetch_disable_bits(void)
return 0;
}
+/*
+ * Helper to write 64bit value to MSR without tracing. Used when
+ * use of the cache should be restricted and use of registers used
+ * for local variables avoided.
+ */
+static inline void pseudo_wrmsrl_notrace(unsigned int msr, u64 val)
+{
+ __wrmsr(msr, (u32)(val & 0xffffffffULL), (u32)(val >> 32));
+}
+
/**
* pseudo_lock_minor_get - Obtain available minor number
* @minor: Pointer to where new minor number will be stored
@@ -831,6 +845,107 @@ static int measure_cycles_lat_fn(void *_plr)
return 0;
}
+static int measure_cycles_perf_fn(void *_plr)
+{
+ struct pseudo_lock_region *plr = _plr;
+ unsigned long long l2_hits, l2_miss;
+ u64 l2_hit_bits, l2_miss_bits;
+ unsigned long i;
+#ifdef CONFIG_KASAN
+ /*
+ * The registers used for local register variables are also used
+ * when KASAN is active. When KASAN is active we use regular variables
+ * at the cost of including cache access latency to these variables
+ * in the measurements.
+ */
+ unsigned int line_size;
+ unsigned int size;
+ void *mem_r;
+#else
+ register unsigned int line_size asm("esi");
+ register unsigned int size asm("edi");
+#ifdef CONFIG_X86_64
+ register void *mem_r asm("rbx");
+#else
+ register void *mem_r asm("ebx");
+#endif /* CONFIG_X86_64 */
+#endif /* CONFIG_KASAN */
+
+ /*
+ * Non-architectural event for the Goldmont Microarchitecture
+ * from Intel x86 Architecture Software Developer Manual (SDM):
+ * MEM_LOAD_UOPS_RETIRED D1H (event number)
+ * Umask values:
+ * L1_HIT 01H
+ * L2_HIT 02H
+ * L1_MISS 08H
+ * L2_MISS 10H
+ */
+
+ /*
+ * Start by setting flags for IA32_PERFEVTSELx:
+ * OS (Operating system mode) 0x2
+ * INT (APIC interrupt enable) 0x10
+ * EN (Enable counter) 0x40
+ *
+ * Then add the Umask value and event number to select performance
+ * event.
+ */
+
+ switch (boot_cpu_data.x86_model) {
+ case INTEL_FAM6_ATOM_GOLDMONT:
+ case INTEL_FAM6_ATOM_GEMINI_LAKE:
+ l2_hit_bits = (0x52ULL << 16) | (0x2 << 8) | 0xd1;
+ l2_miss_bits = (0x52ULL << 16) | (0x10 << 8) | 0xd1;
+ break;
+ default:
+ goto out;
+ }
+
+ local_irq_disable();
+ /*
+ * Call wrmsr direcly to avoid the local register variables from
+ * being overwritten due to reordering of their assignment with
+ * the wrmsr calls.
+ */
+ __wrmsr(MSR_MISC_FEATURE_CONTROL, prefetch_disable_bits, 0x0);
+ /* Disable events and reset counters */
+ pseudo_wrmsrl_notrace(MSR_ARCH_PERFMON_EVENTSEL0, 0x0);
+ pseudo_wrmsrl_notrace(MSR_ARCH_PERFMON_EVENTSEL0 + 1, 0x0);
+ pseudo_wrmsrl_notrace(MSR_ARCH_PERFMON_PERFCTR0, 0x0);
+ pseudo_wrmsrl_notrace(MSR_ARCH_PERFMON_PERFCTR0 + 1, 0x0);
+ /* Set and enable the L2 counters */
+ pseudo_wrmsrl_notrace(MSR_ARCH_PERFMON_EVENTSEL0, l2_hit_bits);
+ pseudo_wrmsrl_notrace(MSR_ARCH_PERFMON_EVENTSEL0 + 1, l2_miss_bits);
+ mem_r = plr->kmem;
+ size = plr->size;
+ line_size = plr->line_size;
+ for (i = 0; i < size; i += line_size) {
+ asm volatile("mov (%0,%1,1), %%eax\n\t"
+ :
+ : "r" (mem_r), "r" (i)
+ : "%eax", "memory");
+ }
+ /*
+ * Call wrmsr directly (no tracing) to not influence
+ * the cache access counters as they are disabled.
+ */
+ pseudo_wrmsrl_notrace(MSR_ARCH_PERFMON_EVENTSEL0,
+ l2_hit_bits & ~(0x40ULL << 16));
+ pseudo_wrmsrl_notrace(MSR_ARCH_PERFMON_EVENTSEL0 + 1,
+ l2_miss_bits & ~(0x40ULL << 16));
+ l2_hits = native_read_pmc(0);
+ l2_miss = native_read_pmc(1);
+ wrmsr(MSR_MISC_FEATURE_CONTROL, 0x0, 0x0);
+ local_irq_enable();
+ trace_pseudo_lock_l2(l2_hits, l2_miss);
+
+out:
+ plr->thread_done = 1;
+ wake_up_interruptible(&plr->lock_thread_wq);
+ return 0;
+}
+
/**
* pseudo_lock_measure_cycles - Trigger latency measure to pseudo-locked region
*
@@ -841,12 +956,12 @@ static int measure_cycles_lat_fn(void *_plr)
*
* Return: 0 on success, <0 on failure
*/
-static int pseudo_lock_measure_cycles(struct rdtgroup *rdtgrp)
+static int pseudo_lock_measure_cycles(struct rdtgroup *rdtgrp, int sel)
{
struct pseudo_lock_region *plr = rdtgrp->plr;
struct task_struct *thread;
unsigned int cpu;
- int ret;
+ int ret = -1;
cpus_read_lock();
mutex_lock(&rdtgroup_mutex);
@@ -863,9 +978,19 @@ static int pseudo_lock_measure_cycles(struct rdtgroup *rdtgrp)
goto out;
}
- thread = kthread_create_on_node(measure_cycles_lat_fn, plr,
- cpu_to_node(cpu),
- "pseudo_lock_measure/%u", cpu);
+ if (sel == 1)
+ thread = kthread_create_on_node(measure_cycles_lat_fn, plr,
+ cpu_to_node(cpu),
+ "pseudo_lock_measure/%u",
+ cpu);
+ else if (sel == 2)
+ thread = kthread_create_on_node(measure_cycles_perf_fn, plr,
+ cpu_to_node(cpu),
+ "pseudo_lock_measure/%u",
+ cpu);
+ else
+ goto out;
+
if (IS_ERR(thread)) {
ret = PTR_ERR(thread);
goto out;
@@ -894,19 +1019,21 @@ static ssize_t pseudo_lock_measure_trigger(struct file *file,
size_t buf_size;
char buf[32];
int ret;
- bool bv;
+ int sel;
buf_size = min(count, (sizeof(buf) - 1));
if (copy_from_user(buf, user_buf, buf_size))
return -EFAULT;
buf[buf_size] = '\0';
- ret = strtobool(buf, &bv);
- if (ret == 0 && bv) {
+ ret = kstrtoint(buf, 10, &sel);
+ if (ret == 0) {
+ if (sel != 1 && sel != 2)
+ return -EINVAL;
ret = debugfs_file_get(file->f_path.dentry);
if (ret)
return ret;
- ret = pseudo_lock_measure_cycles(rdtgrp);
+ ret = pseudo_lock_measure_cycles(rdtgrp, sel);
if (ret == 0)
ret = count;
debugfs_file_put(file->f_path.dentry);
diff --git a/arch/x86/kernel/cpu/intel_rdt_pseudo_lock_event.h b/arch/x86/kernel/cpu/intel_rdt_pseudo_lock_event.h
index 3cd0fa27d5fe..efad50d2ee2f 100644
--- a/arch/x86/kernel/cpu/intel_rdt_pseudo_lock_event.h
+++ b/arch/x86/kernel/cpu/intel_rdt_pseudo_lock_event.h
@@ -15,6 +15,16 @@ TRACE_EVENT(pseudo_lock_mem_latency,
TP_printk("latency=%u", __entry->latency)
);
+TRACE_EVENT(pseudo_lock_l2,
+ TP_PROTO(u64 l2_hits, u64 l2_miss),
+ TP_ARGS(l2_hits, l2_miss),
+ TP_STRUCT__entry(__field(u64, l2_hits)
+ __field(u64, l2_miss)),
+ TP_fast_assign(__entry->l2_hits = l2_hits;
+ __entry->l2_miss = l2_miss;),
+ TP_printk("hits=%llu miss=%llu",
+ __entry->l2_hits, __entry->l2_miss));
+
#endif /* _TRACE_PSEUDO_LOCK_H */
#undef TRACE_INCLUDE_PATH
Powered by blists - more mailing lists