L: linux-kernel@vger.kernel.org
S: Supported
F: Documentation/x86/resctrl*
-F: arch/x86/include/asm/resctrl_sched.h
+F: arch/x86/include/asm/resctrl.h
F: arch/x86/kernel/cpu/resctrl/
F: tools/testing/selftests/resctrl/
/* in KB - valid for CPUS which support this call: */
unsigned int x86_cache_size;
int x86_cache_alignment; /* In bytes */
- /* Cache QoS architectural values: */
+ /* Cache QoS architectural values, valid only on the BSP: */
int x86_cache_max_rmid; /* max index */
int x86_cache_occ_scale; /* scale to bytes */
+ int x86_cache_mbm_width_offset;
int x86_power;
unsigned long loops_per_jiffy;
/* cpuid returned max cores value: */
--- /dev/null
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_X86_RESCTRL_H
+#define _ASM_X86_RESCTRL_H
+
+#ifdef CONFIG_X86_CPU_RESCTRL
+
+#include <linux/sched.h>
+#include <linux/jump_label.h>
+
+#define IA32_PQR_ASSOC 0x0c8f
+
+/**
+ * struct resctrl_pqr_state - State cache for the PQR MSR
+ * @cur_rmid: The cached Resource Monitoring ID
+ * @cur_closid: The cached Class Of Service ID
+ * @default_rmid: The user assigned Resource Monitoring ID
+ * @default_closid: The user assigned cached Class Of Service ID
+ *
+ * The upper 32 bits of IA32_PQR_ASSOC contain closid and the
+ * lower 10 bits rmid. The update to IA32_PQR_ASSOC always
+ * contains both parts, so we need to cache them. This also
+ * stores the user configured per cpu CLOSID and RMID.
+ *
+ * The cache also helps to avoid pointless updates if the value does
+ * not change.
+ */
+struct resctrl_pqr_state {
+ u32 cur_rmid;
+ u32 cur_closid;
+ u32 default_rmid;
+ u32 default_closid;
+};
+
+DECLARE_PER_CPU(struct resctrl_pqr_state, pqr_state);
+
+DECLARE_STATIC_KEY_FALSE(rdt_enable_key);
+DECLARE_STATIC_KEY_FALSE(rdt_alloc_enable_key);
+DECLARE_STATIC_KEY_FALSE(rdt_mon_enable_key);
+
+/*
+ * __resctrl_sched_in() - Writes the task's CLOSid/RMID to IA32_PQR_MSR
+ *
+ * Following considerations are made so that this has minimal impact
+ * on scheduler hot path:
+ * - This will stay as no-op unless we are running on an Intel SKU
+ * which supports resource control or monitoring and we enable by
+ * mounting the resctrl file system.
+ * - Caches the per cpu CLOSid/RMID values and does the MSR write only
+ * when a task with a different CLOSid/RMID is scheduled in.
+ * - We allocate RMIDs/CLOSids globally in order to keep this as
+ * simple as possible.
+ * Must be called with preemption disabled.
+ */
+static void __resctrl_sched_in(void)
+{
+ struct resctrl_pqr_state *state = this_cpu_ptr(&pqr_state);
+ u32 closid = state->default_closid;
+ u32 rmid = state->default_rmid;
+
+ /*
+ * If this task has a closid/rmid assigned, use it.
+ * Else use the closid/rmid assigned to this cpu.
+ */
+ if (static_branch_likely(&rdt_alloc_enable_key)) {
+ if (current->closid)
+ closid = current->closid;
+ }
+
+ if (static_branch_likely(&rdt_mon_enable_key)) {
+ if (current->rmid)
+ rmid = current->rmid;
+ }
+
+ if (closid != state->cur_closid || rmid != state->cur_rmid) {
+ state->cur_closid = closid;
+ state->cur_rmid = rmid;
+ wrmsr(IA32_PQR_ASSOC, rmid, closid);
+ }
+}
+
+static inline void resctrl_sched_in(void)
+{
+ if (static_branch_likely(&rdt_enable_key))
+ __resctrl_sched_in();
+}
+
+void resctrl_cpu_detect(struct cpuinfo_x86 *c);
+
+#else
+
+static inline void resctrl_sched_in(void) {}
+static inline void resctrl_cpu_detect(struct cpuinfo_x86 *c) {}
+
+#endif /* CONFIG_X86_CPU_RESCTRL */
+
+#endif /* _ASM_X86_RESCTRL_H */
+++ /dev/null
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ASM_X86_RESCTRL_SCHED_H
-#define _ASM_X86_RESCTRL_SCHED_H
-
-#ifdef CONFIG_X86_CPU_RESCTRL
-
-#include <linux/sched.h>
-#include <linux/jump_label.h>
-
-#define IA32_PQR_ASSOC 0x0c8f
-
-/**
- * struct resctrl_pqr_state - State cache for the PQR MSR
- * @cur_rmid: The cached Resource Monitoring ID
- * @cur_closid: The cached Class Of Service ID
- * @default_rmid: The user assigned Resource Monitoring ID
- * @default_closid: The user assigned cached Class Of Service ID
- *
- * The upper 32 bits of IA32_PQR_ASSOC contain closid and the
- * lower 10 bits rmid. The update to IA32_PQR_ASSOC always
- * contains both parts, so we need to cache them. This also
- * stores the user configured per cpu CLOSID and RMID.
- *
- * The cache also helps to avoid pointless updates if the value does
- * not change.
- */
-struct resctrl_pqr_state {
- u32 cur_rmid;
- u32 cur_closid;
- u32 default_rmid;
- u32 default_closid;
-};
-
-DECLARE_PER_CPU(struct resctrl_pqr_state, pqr_state);
-
-DECLARE_STATIC_KEY_FALSE(rdt_enable_key);
-DECLARE_STATIC_KEY_FALSE(rdt_alloc_enable_key);
-DECLARE_STATIC_KEY_FALSE(rdt_mon_enable_key);
-
-/*
- * __resctrl_sched_in() - Writes the task's CLOSid/RMID to IA32_PQR_MSR
- *
- * Following considerations are made so that this has minimal impact
- * on scheduler hot path:
- * - This will stay as no-op unless we are running on an Intel SKU
- * which supports resource control or monitoring and we enable by
- * mounting the resctrl file system.
- * - Caches the per cpu CLOSid/RMID values and does the MSR write only
- * when a task with a different CLOSid/RMID is scheduled in.
- * - We allocate RMIDs/CLOSids globally in order to keep this as
- * simple as possible.
- * Must be called with preemption disabled.
- */
-static void __resctrl_sched_in(void)
-{
- struct resctrl_pqr_state *state = this_cpu_ptr(&pqr_state);
- u32 closid = state->default_closid;
- u32 rmid = state->default_rmid;
-
- /*
- * If this task has a closid/rmid assigned, use it.
- * Else use the closid/rmid assigned to this cpu.
- */
- if (static_branch_likely(&rdt_alloc_enable_key)) {
- if (current->closid)
- closid = current->closid;
- }
-
- if (static_branch_likely(&rdt_mon_enable_key)) {
- if (current->rmid)
- rmid = current->rmid;
- }
-
- if (closid != state->cur_closid || rmid != state->cur_rmid) {
- state->cur_closid = closid;
- state->cur_rmid = rmid;
- wrmsr(IA32_PQR_ASSOC, rmid, closid);
- }
-}
-
-static inline void resctrl_sched_in(void)
-{
- if (static_branch_likely(&rdt_enable_key))
- __resctrl_sched_in();
-}
-
-#else
-
-static inline void resctrl_sched_in(void) {}
-
-#endif /* CONFIG_X86_CPU_RESCTRL */
-
-#endif /* _ASM_X86_RESCTRL_SCHED_H */
#include <asm/pci-direct.h>
#include <asm/delay.h>
#include <asm/debugreg.h>
+#include <asm/resctrl.h>
#ifdef CONFIG_X86_64
# include <asm/mmconfig.h>
x86_amd_ls_cfg_ssbd_mask = 1ULL << bit;
}
}
+
+ resctrl_cpu_detect(c);
}
static void early_detect_mem_encrypt(struct cpuinfo_x86 *c)
}
}
-static void init_cqm(struct cpuinfo_x86 *c)
-{
- if (!cpu_has(c, X86_FEATURE_CQM_LLC)) {
- c->x86_cache_max_rmid = -1;
- c->x86_cache_occ_scale = -1;
- return;
- }
-
- /* will be overridden if occupancy monitoring exists */
- c->x86_cache_max_rmid = cpuid_ebx(0xf);
-
- if (cpu_has(c, X86_FEATURE_CQM_OCCUP_LLC) ||
- cpu_has(c, X86_FEATURE_CQM_MBM_TOTAL) ||
- cpu_has(c, X86_FEATURE_CQM_MBM_LOCAL)) {
- u32 eax, ebx, ecx, edx;
-
- /* QoS sub-leaf, EAX=0Fh, ECX=1 */
- cpuid_count(0xf, 1, &eax, &ebx, &ecx, &edx);
-
- c->x86_cache_max_rmid = ecx;
- c->x86_cache_occ_scale = ebx;
- }
-}
-
void get_cpu_cap(struct cpuinfo_x86 *c)
{
u32 eax, ebx, ecx, edx;
init_scattered_cpuid_features(c);
init_speculation_control(c);
- init_cqm(c);
/*
* Clear/Set all flags overridden by options, after probe.
#endif
}
-static void x86_init_cache_qos(struct cpuinfo_x86 *c)
-{
- /*
- * The heavy lifting of max_rmid and cache_occ_scale are handled
- * in get_cpu_cap(). Here we just set the max_rmid for the boot_cpu
- * in case CQM bits really aren't there in this CPU.
- */
- if (c != &boot_cpu_data) {
- boot_cpu_data.x86_cache_max_rmid =
- min(boot_cpu_data.x86_cache_max_rmid,
- c->x86_cache_max_rmid);
- }
-}
-
/*
* Validate that ACPI/mptables have the same information about the
* effective APIC id and update the package map.
#endif
x86_init_rdrand(c);
- x86_init_cache_qos(c);
setup_pku(c);
/*
#include <asm/cpu_device_id.h>
#include <asm/cmdline.h>
#include <asm/traps.h>
+#include <asm/resctrl.h>
#ifdef CONFIG_X86_64
#include <linux/topology.h>
detect_ht_early(c);
}
+static void bsp_init_intel(struct cpuinfo_x86 *c)
+{
+ resctrl_cpu_detect(c);
+}
+
#ifdef CONFIG_X86_32
/*
* Early probe support logic for ppro memory erratum #50
#endif
.c_detect_tlb = intel_detect_tlb,
.c_early_init = early_init_intel,
+ .c_bsp_init = bsp_init_intel,
.c_init = init_intel,
.c_x86_vendor = X86_VENDOR_INTEL,
};
#include <linux/cpuhotplug.h>
#include <asm/intel-family.h>
-#include <asm/resctrl_sched.h>
+#include <asm/resctrl.h>
#include "internal.h"
/* Mutex to protect rdtgroup access. */
static enum cpuhp_state rdt_online;
+/* Runs once on the BSP during boot. */
+void resctrl_cpu_detect(struct cpuinfo_x86 *c)
+{
+ if (!cpu_has(c, X86_FEATURE_CQM_LLC)) {
+ c->x86_cache_max_rmid = -1;
+ c->x86_cache_occ_scale = -1;
+ c->x86_cache_mbm_width_offset = -1;
+ return;
+ }
+
+ /* will be overridden if occupancy monitoring exists */
+ c->x86_cache_max_rmid = cpuid_ebx(0xf);
+
+ if (cpu_has(c, X86_FEATURE_CQM_OCCUP_LLC) ||
+ cpu_has(c, X86_FEATURE_CQM_MBM_TOTAL) ||
+ cpu_has(c, X86_FEATURE_CQM_MBM_LOCAL)) {
+ u32 eax, ebx, ecx, edx;
+
+ /* QoS sub-leaf, EAX=0Fh, ECX=1 */
+ cpuid_count(0xf, 1, &eax, &ebx, &ecx, &edx);
+
+ c->x86_cache_max_rmid = ecx;
+ c->x86_cache_occ_scale = ebx;
+ if (c->x86_vendor == X86_VENDOR_INTEL)
+ c->x86_cache_mbm_width_offset = eax & 0xff;
+ else
+ c->x86_cache_mbm_width_offset = -1;
+ }
+}
+
static int __init resctrl_late_init(void)
{
struct rdt_resource *r;
return ret;
}
-void mon_event_read(struct rmid_read *rr, struct rdt_domain *d,
- struct rdtgroup *rdtgrp, int evtid, int first)
+void mon_event_read(struct rmid_read *rr, struct rdt_resource *r,
+ struct rdt_domain *d, struct rdtgroup *rdtgrp,
+ int evtid, int first)
{
/*
* setup the parameters to send to the IPI to read the data.
*/
rr->rgrp = rdtgrp;
rr->evtid = evtid;
+ rr->r = r;
rr->d = d;
rr->val = 0;
rr->first = first;
goto out;
}
- mon_event_read(&rr, d, rdtgrp, evtid, false);
+ mon_event_read(&rr, r, d, rdtgrp, evtid, false);
if (rr.val & RMID_VAL_ERROR)
seq_puts(m, "Error\n");
#define CQM_LIMBOCHECK_INTERVAL 1000
-#define MBM_CNTR_WIDTH 24
+#define MBM_CNTR_WIDTH_BASE 24
#define MBM_OVERFLOW_INTERVAL 1000
#define MAX_MBA_BW 100u
#define MBA_IS_LINEAR 0x4
#define RMID_VAL_ERROR BIT_ULL(63)
#define RMID_VAL_UNAVAIL BIT_ULL(62)
+/*
+ * With the above fields in use 62 bits remain in MSR_IA32_QM_CTR for
+ * data to be returned. The counter width is discovered from the hardware
+ * as an offset from MBM_CNTR_WIDTH_BASE.
+ */
+#define MBM_CNTR_WIDTH_OFFSET_MAX (62 - MBM_CNTR_WIDTH_BASE)
struct rdt_fs_context {
struct rmid_read {
struct rdtgroup *rgrp;
+ struct rdt_resource *r;
struct rdt_domain *d;
int evtid;
bool first;
struct list_head evt_list;
int num_rmid;
unsigned int mon_scale;
+ unsigned int mbm_width;
unsigned long fflags;
};
unsigned int dom_id);
void mkdir_mondata_subdir_allrdtgrp(struct rdt_resource *r,
struct rdt_domain *d);
-void mon_event_read(struct rmid_read *rr, struct rdt_domain *d,
- struct rdtgroup *rdtgrp, int evtid, int first);
+void mon_event_read(struct rmid_read *rr, struct rdt_resource *r,
+ struct rdt_domain *d, struct rdtgroup *rdtgrp,
+ int evtid, int first);
void mbm_setup_overflow_handler(struct rdt_domain *dom,
unsigned long delay_ms);
void mbm_handle_overflow(struct work_struct *work);
list_add_tail(&entry->list, &rmid_free_lru);
}
-static u64 mbm_overflow_count(u64 prev_msr, u64 cur_msr)
+static u64 mbm_overflow_count(u64 prev_msr, u64 cur_msr, unsigned int width)
{
- u64 shift = 64 - MBM_CNTR_WIDTH, chunks;
+ u64 shift = 64 - width, chunks;
chunks = (cur_msr << shift) - (prev_msr << shift);
return chunks >>= shift;
return 0;
}
- chunks = mbm_overflow_count(m->prev_msr, tval);
+ chunks = mbm_overflow_count(m->prev_msr, tval, rr->r->mbm_width);
m->chunks += chunks;
m->prev_msr = tval;
if (tval & (RMID_VAL_ERROR | RMID_VAL_UNAVAIL))
return;
- chunks = mbm_overflow_count(m->prev_bw_msr, tval);
+ chunks = mbm_overflow_count(m->prev_bw_msr, tval, rr->r->mbm_width);
m->chunks_bw += chunks;
m->chunks = m->chunks_bw;
cur_bw = (chunks * r->mon_scale) >> 20;
}
}
-static void mbm_update(struct rdt_domain *d, int rmid)
+static void mbm_update(struct rdt_resource *r, struct rdt_domain *d, int rmid)
{
struct rmid_read rr;
rr.first = false;
+ rr.r = r;
rr.d = d;
/*
struct rdtgroup *prgrp, *crgrp;
int cpu = smp_processor_id();
struct list_head *head;
+ struct rdt_resource *r;
struct rdt_domain *d;
mutex_lock(&rdtgroup_mutex);
if (!static_branch_likely(&rdt_mon_enable_key))
goto out_unlock;
- d = get_domain_from_cpu(cpu, &rdt_resources_all[RDT_RESOURCE_L3]);
+ r = &rdt_resources_all[RDT_RESOURCE_L3];
+
+ d = get_domain_from_cpu(cpu, r);
if (!d)
goto out_unlock;
list_for_each_entry(prgrp, &rdt_all_groups, rdtgroup_list) {
- mbm_update(d, prgrp->mon.rmid);
+ mbm_update(r, d, prgrp->mon.rmid);
head = &prgrp->mon.crdtgrp_list;
list_for_each_entry(crgrp, head, mon.crdtgrp_list)
- mbm_update(d, crgrp->mon.rmid);
+ mbm_update(r, d, crgrp->mon.rmid);
if (is_mba_sc(NULL))
update_mba_bw(prgrp, d);
int rdt_get_mon_l3_config(struct rdt_resource *r)
{
+ unsigned int mbm_offset = boot_cpu_data.x86_cache_mbm_width_offset;
unsigned int cl_size = boot_cpu_data.x86_cache_size;
int ret;
r->mon_scale = boot_cpu_data.x86_cache_occ_scale;
r->num_rmid = boot_cpu_data.x86_cache_max_rmid + 1;
+ r->mbm_width = MBM_CNTR_WIDTH_BASE;
+
+ if (mbm_offset > 0 && mbm_offset <= MBM_CNTR_WIDTH_OFFSET_MAX)
+ r->mbm_width += mbm_offset;
+ else if (mbm_offset > MBM_CNTR_WIDTH_OFFSET_MAX)
+ pr_warn("Ignoring impossible MBM counter offset\n");
/*
* A reasonable upper limit on the max threshold is the number
#include <asm/cacheflush.h>
#include <asm/intel-family.h>
-#include <asm/resctrl_sched.h>
+#include <asm/resctrl.h>
#include <asm/perf_event.h>
#include "../../events/perf_event.h" /* For X86_CONFIG() */
#include <uapi/linux/magic.h>
-#include <asm/resctrl_sched.h>
+#include <asm/resctrl.h>
#include "internal.h"
DEFINE_STATIC_KEY_FALSE(rdt_enable_key);
goto out_destroy;
if (is_mbm_event(mevt->evtid))
- mon_event_read(&rr, d, prgrp, mevt->evtid, true);
+ mon_event_read(&rr, r, d, prgrp, mevt->evtid, true);
}
kernfs_activate(kn);
return 0;
#include <asm/debugreg.h>
#include <asm/switch_to.h>
#include <asm/vm86.h>
-#include <asm/resctrl_sched.h>
+#include <asm/resctrl.h>
#include <asm/proto.h>
#include "process.h"
#include <asm/switch_to.h>
#include <asm/xen/hypervisor.h>
#include <asm/vdso.h>
-#include <asm/resctrl_sched.h>
+#include <asm/resctrl.h>
#include <asm/unistd.h>
#include <asm/fsgsbase.h>
#ifdef CONFIG_IA32_EMULATION