New AMD processors can now support following QoS features.
1. Slow Memory Bandwidth Allocation (SMBA)
With this feature, the QOS enforcement policies can be applied
to the external slow memory connected to the host. QOS enforcement
is accomplished by assigning a Class Of Service (COS) to a processor
and specifying allocations or limits for that COS for each resource
to be allocated.
Currently, CXL.memory is the only supported "slow" memory device. With
the support of SMBA feature the hardware enables bandwidth allocation
on the slow memory devices.
2. Bandwidth Monitoring Event Configuration (BMEC)
The bandwidth monitoring events mbm_total_event and mbm_local_event
are set to count all the total and local reads/writes respectively.
With the introduction of slow memory, the two counters are not enough
to count all the different types are memory events. With the feature
BMEC, the users have the option to configure mbm_total_event and
mbm_local_event to count the specific type of events.
Following are the bitmaps of events supported.
Bits Description
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
This series adds support for these features. Also added a minor cleanup(PATCH 1).
Feature description is available in the specification, "AMD64 Technology Platform Quality of Service Extensions, Revision: 1.03 Publication # 56375
Revision: 1.03 Issue Date: February 2022".
Link: https://bugzilla.kernel.org/show_bug.cgi?id=206537
Link: https://www.amd.com/en/support/tech-docs/amd64-technology-platform-quality-service-extensions
---
v12:
Summary of changes:
Refreshed on top of tip/master.
Made the new cpu features SMBA and BMEC invisible from /proc/cpuinfo based on our discussion.
https://lore.kernel.org/lkml/[email protected]/
Updated the Documentation/x86/resctrl.rst with minor changes.
v11:
https://lore.kernel.org/lkml/[email protected]/
Summary of changes:
Removed cpus_read_lock() from the patch 11 and 12 based on our discussion.
https://lore.kernel.org/lkml/[email protected]/
Picked up all the Reviewed-by by Reinette.
Minor text changes on patch 13.
v10:
https://lore.kernel.org/lkml/[email protected]/
Summary of changes:
1. Moved the patch 12 to 1 ("x86/resctrl: Replace smp_call_function_many() with on_each_cpu_mask()")
2. No need to write MSR_IA32_EVT_CFG_BASE on all the CPUs. Replaced on_each_cpu_mask with smp_call_function_any.
3. Updated the subject line of the patches to replace sysfs with resctrl.
4. Added Reviewed-by for the patches which are reviewed by Reinette.
5. Added few more comments suggested by Reinette.
v9:
https://lore.kernel.org/lkml/166990882621.17806.16780480657453071426.stgit@bmoger-ubuntu/
Summary of changes:
1. Rebased on top of lastest tip/master as of 11/30.
2. Most of the changes are result of the comments from Fenghua, Reinette and Peter Newman.
3. Fixed the cpuid dependancy.
4. Added the __init attribute to rdt_get_mon_l3_config and mbm_config_rftype_init.
5. Added new function resctrl_arch_reset_rmid_all to clear all rmid statues.
6. Changed mon_event_config_index_get based on Reinette's comments.
7. Changed mbm_config_rftype_init to take care of few extra error handling.
8. Few other minor changes and text changes.
v8:
https://lore.kernel.org/lkml/166759188265.3281208.11769277079826754455.stgit@bmoger-ubuntu/
Changes:
1. Removed init attribute for rdt_cpu_has to make it available for all the files.
2. Updated the change log for mon_features to correct the names of config files.
3. Changed configuration file name from mbm_total_config to mbm_total_bytes_config.
This is more consistant with other changes.
4. Added lock protection while reading/writing the config file.
5. Other few minor text changes. I have been missing few comments in last couple of
revisions. Hope I have addressed all of them this time.
v7:
https://lore.kernel.org/lkml/166604543832.5345.9696970469830919982.stgit@bmoger-ubuntu/
Changes:
Not much of a change. Missed one comment from Reinette from v5. Corrected it now.
Few format corrections from Sanjaya.
v6:
https://lore.kernel.org/lkml/166543345606.23830.3120625408601531368.stgit@bmoger-ubuntu/
Summary of changes:
1. Rebased on top of lastest tip tree. Fixed few minor conflicts.
2. Fixed format issue with scattered.c.
3. Removed config_name from the structure mon_evt. It is not required.
4. The read/write format for mbm_total_config and mbm_local_config will be same
as schemata format "id0=val0;id1=val1;...". This is comment from Fenghua.
5. Added more comments MSR_IA32_EVT_CFG_BASE writng.
5. Few text changes in resctrl.rst
v5:
https://lore.kernel.org/lkml/166431016617.373387.1968875281081252467.stgit@bmoger-ubuntu/
Summary of changes.
1. Split the series into two. The first two patches are bug fixes. So, sent them separate.
2. The config files mbm_total_config and mbm_local_config are now under
/sys/fs/resctrl/info/L3_MON/. Removed these config files from mon groups.
3. Ran "checkpatch --strict --codespell" on all the patches. Looks good with few known exceptions.
4. Few minor text changes in resctrl.rst file.
v4:
https://lore.kernel.org/lkml/166257348081.1043018.11227924488792315932.stgit@bmoger-ubuntu/
Got numerios of comments from Reinette Chatre. Addressed most of them.
Summary of changes.
1. Removed mon_configurable under /sys/fs/resctrl/info/L3_MON/.
2. Updated mon_features texts if the BMEC is supported.
3. Added more explanation about the slow memory support.
4. Replaced smp_call_function_many with on_each_cpu_mask call.
5. Removed arch_has_empty_bitmaps
6. Few other text changes.
7. Removed Reviewed-by if the patch is modified.
8. Rebased the patches to latest tip.
v3:
https://lore.kernel.org/lkml/166117559756.6695.16047463526634290701.stgit@bmoger-ubuntu/
a. Rebased the patches to latest tip. Resolved some conflicts.
https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git
b. Taken care of feedback from Bagas Sanjaya.
c. Added Reviewed by from Mingo.
Note: I am still looking for comments from Reinette or Fenghua.
v2:
https://lore.kernel.org/lkml/165938717220.724959.10931629283087443782.stgit@bmoger-ubuntu/
a. Rebased the patches to latest stable tree (v5.18.15). Resolved some conflicts.
b. Added the patch to fix CBM issue on AMD. This was originally discussed
https://lore.kernel.org/lkml/[email protected]/
v1:
https://lore.kernel.org/lkml/165757543252.416408.13547339307237713464.stgit@bmoger-ubuntu/
*** BLURB HERE ***
Babu Moger (13):
x86/resctrl: Replace smp_call_function_many() with on_each_cpu_mask()
x86/cpufeatures: Add Slow Memory Bandwidth Allocation feature flag
x86/resctrl: Add a new resource type RDT_RESOURCE_SMBA
x86/cpufeatures: Add Bandwidth Monitoring Event Configuration feature
flag
x86/resctrl: Include new features in command line options
x86/resctrl: Detect and configure Slow Memory Bandwidth Allocation
x86/resctrl: Add __init attribute to rdt_get_mon_l3_config()
x86/resctrl: Support monitor configuration
x86/resctrl: Add interface to read mbm_total_bytes_config
x86/resctrl: Add interface to read mbm_local_bytes_config
x86/resctrl: Add interface to write mbm_total_bytes_config
x86/resctrl: Add interface to write mbm_local_bytes_config
Documentation/x86: Update resctrl.rst for new features
.../admin-guide/kernel-parameters.txt | 2 +-
Documentation/x86/resctrl.rst | 147 ++++++++-
arch/x86/include/asm/cpufeatures.h | 2 +
arch/x86/include/asm/msr-index.h | 2 +
arch/x86/kernel/cpu/cpuid-deps.c | 2 +
arch/x86/kernel/cpu/resctrl/core.c | 54 ++-
arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 13 +-
arch/x86/kernel/cpu/resctrl/internal.h | 28 ++
arch/x86/kernel/cpu/resctrl/monitor.c | 30 +-
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 307 ++++++++++++++++--
arch/x86/kernel/cpu/scattered.c | 2 +
include/linux/resctrl.h | 11 +
12 files changed, 559 insertions(+), 41 deletions(-)
--
2.34.1
In an upcoming change rdt_get_mon_l3_config() needs to call
rdt_cpu_has() to query the monitor related features. It cannot be
called right now because rdt_cpu_has() has the __init attribute but
rdt_get_mon_l3_config() doesn't.
Add the __init attribute to rdt_get_mon_l3_config() that is only called
by get_rdt_mon_resources() that already has the __init attribute. Also
make rdt_cpu_has() available to by rdt_get_mon_l3_config() via
the internal header file.
Reviewed-by: Reinette Chatre <[email protected]>
Signed-off-by: Babu Moger <[email protected]>
---
arch/x86/kernel/cpu/resctrl/core.c | 2 +-
arch/x86/kernel/cpu/resctrl/internal.h | 1 +
arch/x86/kernel/cpu/resctrl/monitor.c | 2 +-
3 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/cpu/resctrl/core.c b/arch/x86/kernel/cpu/resctrl/core.c
index b4fc851f6489..030d3b409768 100644
--- a/arch/x86/kernel/cpu/resctrl/core.c
+++ b/arch/x86/kernel/cpu/resctrl/core.c
@@ -728,7 +728,7 @@ static int __init set_rdt_options(char *str)
}
__setup("rdt", set_rdt_options);
-static bool __init rdt_cpu_has(int flag)
+bool __init rdt_cpu_has(int flag)
{
bool ret = boot_cpu_has(flag);
struct rdt_options *o;
diff --git a/arch/x86/kernel/cpu/resctrl/internal.h b/arch/x86/kernel/cpu/resctrl/internal.h
index fdbbf66312ec..f16b8bc5448c 100644
--- a/arch/x86/kernel/cpu/resctrl/internal.h
+++ b/arch/x86/kernel/cpu/resctrl/internal.h
@@ -512,6 +512,7 @@ void closid_free(int closid);
int alloc_rmid(void);
void free_rmid(u32 rmid);
int rdt_get_mon_l3_config(struct rdt_resource *r);
+bool __init rdt_cpu_has(int flag);
void mon_event_count(void *info);
int rdtgroup_mondata_show(struct seq_file *m, void *arg);
void mon_event_read(struct rmid_read *rr, struct rdt_resource *r,
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index 77538abeb72a..5665ab9725ce 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -763,7 +763,7 @@ static void l3_mon_evt_init(struct rdt_resource *r)
list_add_tail(&mbm_local_event.list, &r->evt_list);
}
-int rdt_get_mon_l3_config(struct rdt_resource *r)
+int __init rdt_get_mon_l3_config(struct rdt_resource *r)
{
unsigned int mbm_offset = boot_cpu_data.x86_cache_mbm_width_offset;
struct rdt_hw_resource *hw_res = resctrl_to_arch_res(r);
--
2.34.1
The event configuration can be viewed by the user by reading the
configuration file /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config.
The event configuration settings are domain specific and will affect all
the CPUs in the domain.
Following are the types of events supported:
==== ===========================================================
Bits Description
==== ===========================================================
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
==== ===========================================================
By default, the mbm_total_bytes_config is set to 0x7f to count all the
event types.
For example:
$cat /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
0=0x7f;1=0x7f;2=0x7f;3=0x7f
In this case, the event mbm_total_bytes is configured with 0x7f on
domains 0 to 3.
Reviewed-by: Reinette Chatre <[email protected]>
Signed-off-by: Babu Moger <[email protected]>
---
arch/x86/include/asm/msr-index.h | 1 +
arch/x86/kernel/cpu/resctrl/internal.h | 24 ++++++
arch/x86/kernel/cpu/resctrl/monitor.c | 4 +-
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 102 +++++++++++++++++++++++++
4 files changed, 130 insertions(+), 1 deletion(-)
diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h
index eaeeb7b415d6..06b64f4050ae 100644
--- a/arch/x86/include/asm/msr-index.h
+++ b/arch/x86/include/asm/msr-index.h
@@ -1069,6 +1069,7 @@
/* - AMD: */
#define MSR_IA32_MBA_BW_BASE 0xc0000200
#define MSR_IA32_SMBA_BW_BASE 0xc0000280
+#define MSR_IA32_EVT_CFG_BASE 0xc0000400
/* MSR_IA32_VMX_MISC bits */
#define MSR_IA32_VMX_MISC_INTEL_PT (1ULL << 14)
diff --git a/arch/x86/kernel/cpu/resctrl/internal.h b/arch/x86/kernel/cpu/resctrl/internal.h
index 0605b04f1b7a..8edecc5763d8 100644
--- a/arch/x86/kernel/cpu/resctrl/internal.h
+++ b/arch/x86/kernel/cpu/resctrl/internal.h
@@ -30,6 +30,29 @@
*/
#define MBM_CNTR_WIDTH_OFFSET_MAX (62 - MBM_CNTR_WIDTH_BASE)
+/* Reads to Local DRAM Memory */
+#define READS_TO_LOCAL_MEM BIT(0)
+
+/* Reads to Remote DRAM Memory */
+#define READS_TO_REMOTE_MEM BIT(1)
+
+/* Non-Temporal Writes to Local Memory */
+#define NON_TEMP_WRITE_TO_LOCAL_MEM BIT(2)
+
+/* Non-Temporal Writes to Remote Memory */
+#define NON_TEMP_WRITE_TO_REMOTE_MEM BIT(3)
+
+/* Reads to Local Memory the system identifies as "Slow Memory" */
+#define READS_TO_LOCAL_S_MEM BIT(4)
+
+/* Reads to Remote Memory the system identifies as "Slow Memory" */
+#define READS_TO_REMOTE_S_MEM BIT(5)
+
+/* Dirty Victims to All Types of Memory */
+#define DIRTY_VICTIMS_TO_ALL_MEM BIT(6)
+
+/* Max event bits supported */
+#define MAX_EVT_CONFIG_BITS GENMASK(6, 0)
struct rdt_fs_context {
struct kernfs_fs_context kfc;
@@ -531,5 +554,6 @@ bool has_busy_rmid(struct rdt_resource *r, struct rdt_domain *d);
void __check_limbo(struct rdt_domain *d, bool force_free);
void rdt_domain_reconfigure_cdp(struct rdt_resource *r);
void __init thread_throttle_mode_init(void);
+void __init mbm_config_rftype_init(const char *config);
#endif /* _ASM_X86_RESCTRL_INTERNAL_H */
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index cf7b6e59fa29..0ff45e34d83c 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -801,8 +801,10 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r)
return ret;
if (rdt_cpu_has(X86_FEATURE_BMEC)) {
- if (rdt_cpu_has(X86_FEATURE_CQM_MBM_TOTAL))
+ if (rdt_cpu_has(X86_FEATURE_CQM_MBM_TOTAL)) {
mbm_total_event.configurable = true;
+ mbm_config_rftype_init("mbm_total_bytes_config");
+ }
if (rdt_cpu_has(X86_FEATURE_CQM_MBM_LOCAL))
mbm_local_event.configurable = true;
}
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index bf0bd4378e13..cd4e668e5019 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1420,6 +1420,93 @@ static int rdtgroup_size_show(struct kernfs_open_file *of,
return ret;
}
+struct mon_config_info {
+ u32 evtid;
+ u32 mon_config;
+};
+
+#define INVALID_CONFIG_INDEX UINT_MAX
+
+/**
+ * mon_event_config_index_get - get the hardware index for the
+ * configurable event
+ * @evtid: event id.
+ *
+ * Return: 0 for evtid == QOS_L3_MBM_TOTAL_EVENT_ID
+ * 1 for evtid == QOS_L3_MBM_LOCAL_EVENT_ID
+ * INVALID_CONFIG_INDEX for invalid evtid
+ */
+static inline unsigned int mon_event_config_index_get(u32 evtid)
+{
+ switch (evtid) {
+ case QOS_L3_MBM_TOTAL_EVENT_ID:
+ return 0;
+ case QOS_L3_MBM_LOCAL_EVENT_ID:
+ return 1;
+ default:
+ /* Should never reach here */
+ return INVALID_CONFIG_INDEX;
+ }
+}
+
+static void mon_event_config_read(void *info)
+{
+ struct mon_config_info *mon_info = info;
+ unsigned int index;
+ u32 h;
+
+ index = mon_event_config_index_get(mon_info->evtid);
+ if (index == INVALID_CONFIG_INDEX) {
+ pr_warn_once("Invalid event id %d\n", mon_info->evtid);
+ return;
+ }
+ rdmsr(MSR_IA32_EVT_CFG_BASE + index, mon_info->mon_config, h);
+
+ /* Report only the valid event configuration bits */
+ mon_info->mon_config &= MAX_EVT_CONFIG_BITS;
+}
+
+static void mondata_config_read(struct rdt_domain *d, struct mon_config_info *mon_info)
+{
+ smp_call_function_any(&d->cpu_mask, mon_event_config_read, mon_info, 1);
+}
+
+static int mbm_config_show(struct seq_file *s, struct rdt_resource *r, u32 evtid)
+{
+ struct mon_config_info mon_info = {0};
+ struct rdt_domain *dom;
+ bool sep = false;
+
+ mutex_lock(&rdtgroup_mutex);
+
+ list_for_each_entry(dom, &r->domains, list) {
+ if (sep)
+ seq_puts(s, ";");
+
+ memset(&mon_info, 0, sizeof(struct mon_config_info));
+ mon_info.evtid = evtid;
+ mondata_config_read(dom, &mon_info);
+
+ seq_printf(s, "%d=0x%02x", dom->id, mon_info.mon_config);
+ sep = true;
+ }
+ seq_puts(s, "\n");
+
+ mutex_unlock(&rdtgroup_mutex);
+
+ return 0;
+}
+
+static int mbm_total_bytes_config_show(struct kernfs_open_file *of,
+ struct seq_file *seq, void *v)
+{
+ struct rdt_resource *r = of->kn->parent->priv;
+
+ mbm_config_show(seq, r, QOS_L3_MBM_TOTAL_EVENT_ID);
+
+ return 0;
+}
+
/* rdtgroup information files for one cache resource. */
static struct rftype res_common_files[] = {
{
@@ -1518,6 +1605,12 @@ static struct rftype res_common_files[] = {
.seq_show = max_threshold_occ_show,
.fflags = RF_MON_INFO | RFTYPE_RES_CACHE,
},
+ {
+ .name = "mbm_total_bytes_config",
+ .mode = 0444,
+ .kf_ops = &rdtgroup_kf_single_ops,
+ .seq_show = mbm_total_bytes_config_show,
+ },
{
.name = "cpus",
.mode = 0644,
@@ -1624,6 +1717,15 @@ void __init thread_throttle_mode_init(void)
rft->fflags = RF_CTRL_INFO | RFTYPE_RES_MB;
}
+void __init mbm_config_rftype_init(const char *config)
+{
+ struct rftype *rft;
+
+ rft = rdtgroup_get_rftype_by_name(config);
+ if (rft)
+ rft->fflags = RF_MON_INFO | RFTYPE_RES_CACHE;
+}
+
/**
* rdtgroup_kn_mode_restrict - Restrict user access to named resctrl file
* @r: The resource group with which the file is associated.
--
2.34.1
The event configuration can be viewed by the user by reading the
configuration file /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config.
The event configuration settings are domain specific and will affect
all the CPUs in the domain.
Following are the types of events supported:
==== ===========================================================
Bits Description
==== ===========================================================
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
==== ===========================================================
By default, the mbm_local_bytes_config is set to 0x15 to count all the
local event types.
For example:
$cat /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
0=0x15;1=0x15;2=0x15;3=0x15
In this case, the event mbm_local_bytes is configured with 0x15 on
domains 0 to 3.
Reviewed-by: Reinette Chatre <[email protected]>
Signed-off-by: Babu Moger <[email protected]>
---
arch/x86/kernel/cpu/resctrl/monitor.c | 4 +++-
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 16 ++++++++++++++++
2 files changed, 19 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index 0ff45e34d83c..28c65008fce1 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -805,8 +805,10 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r)
mbm_total_event.configurable = true;
mbm_config_rftype_init("mbm_total_bytes_config");
}
- if (rdt_cpu_has(X86_FEATURE_CQM_MBM_LOCAL))
+ if (rdt_cpu_has(X86_FEATURE_CQM_MBM_LOCAL)) {
mbm_local_event.configurable = true;
+ mbm_config_rftype_init("mbm_local_bytes_config");
+ }
}
l3_mon_evt_init(r);
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index cd4e668e5019..ad3c7014adf6 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1507,6 +1507,16 @@ static int mbm_total_bytes_config_show(struct kernfs_open_file *of,
return 0;
}
+static int mbm_local_bytes_config_show(struct kernfs_open_file *of,
+ struct seq_file *seq, void *v)
+{
+ struct rdt_resource *r = of->kn->parent->priv;
+
+ mbm_config_show(seq, r, QOS_L3_MBM_LOCAL_EVENT_ID);
+
+ return 0;
+}
+
/* rdtgroup information files for one cache resource. */
static struct rftype res_common_files[] = {
{
@@ -1611,6 +1621,12 @@ static struct rftype res_common_files[] = {
.kf_ops = &rdtgroup_kf_single_ops,
.seq_show = mbm_total_bytes_config_show,
},
+ {
+ .name = "mbm_local_bytes_config",
+ .mode = 0444,
+ .kf_ops = &rdtgroup_kf_single_ops,
+ .seq_show = mbm_local_bytes_config_show,
+ },
{
.name = "cpus",
.mode = 0644,
--
2.34.1
The QoS slow memory configuration details are available via
CPUID_Fn80000020_EDX_x02. Detect the available details and
initialize the rest to defaults.
Reviewed-by: Reinette Chatre <[email protected]>
Signed-off-by: Babu Moger <[email protected]>
---
arch/x86/include/asm/msr-index.h | 1 +
arch/x86/kernel/cpu/resctrl/core.c | 36 +++++++++++++++++++++--
arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 2 +-
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 8 +++--
4 files changed, 41 insertions(+), 6 deletions(-)
diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h
index a85019756003..eaeeb7b415d6 100644
--- a/arch/x86/include/asm/msr-index.h
+++ b/arch/x86/include/asm/msr-index.h
@@ -1068,6 +1068,7 @@
/* - AMD: */
#define MSR_IA32_MBA_BW_BASE 0xc0000200
+#define MSR_IA32_SMBA_BW_BASE 0xc0000280
/* MSR_IA32_VMX_MISC bits */
#define MSR_IA32_VMX_MISC_INTEL_PT (1ULL << 14)
diff --git a/arch/x86/kernel/cpu/resctrl/core.c b/arch/x86/kernel/cpu/resctrl/core.c
index 10a8c9d96f32..b4fc851f6489 100644
--- a/arch/x86/kernel/cpu/resctrl/core.c
+++ b/arch/x86/kernel/cpu/resctrl/core.c
@@ -162,6 +162,13 @@ bool is_mba_sc(struct rdt_resource *r)
if (!r)
return rdt_resources_all[RDT_RESOURCE_MBA].r_resctrl.membw.mba_sc;
+ /*
+ * The software controller support is only applicable to MBA resource.
+ * Make sure to check for resource type.
+ */
+ if (r->rid != RDT_RESOURCE_MBA)
+ return false;
+
return r->membw.mba_sc;
}
@@ -225,9 +232,15 @@ static bool __rdt_get_mem_config_amd(struct rdt_resource *r)
struct rdt_hw_resource *hw_res = resctrl_to_arch_res(r);
union cpuid_0x10_3_eax eax;
union cpuid_0x10_x_edx edx;
- u32 ebx, ecx;
+ u32 ebx, ecx, subleaf;
- cpuid_count(0x80000020, 1, &eax.full, &ebx, &ecx, &edx.full);
+ /*
+ * Query CPUID_Fn80000020_EDX_x01 for MBA and
+ * CPUID_Fn80000020_EDX_x02 for SMBA
+ */
+ subleaf = (r->rid == RDT_RESOURCE_SMBA) ? 2 : 1;
+
+ cpuid_count(0x80000020, subleaf, &eax.full, &ebx, &ecx, &edx.full);
hw_res->num_closid = edx.split.cos_max + 1;
r->default_ctrl = MAX_MBA_BW_AMD;
@@ -750,6 +763,19 @@ static __init bool get_mem_config(void)
return false;
}
+static __init bool get_slow_mem_config(void)
+{
+ struct rdt_hw_resource *hw_res = &rdt_resources_all[RDT_RESOURCE_SMBA];
+
+ if (!rdt_cpu_has(X86_FEATURE_SMBA))
+ return false;
+
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
+ return __rdt_get_mem_config_amd(&hw_res->r_resctrl);
+
+ return false;
+}
+
static __init bool get_rdt_alloc_resources(void)
{
struct rdt_resource *r;
@@ -780,6 +806,9 @@ static __init bool get_rdt_alloc_resources(void)
if (get_mem_config())
ret = true;
+ if (get_slow_mem_config())
+ ret = true;
+
return ret;
}
@@ -869,6 +898,9 @@ static __init void rdt_init_res_defs_amd(void)
} else if (r->rid == RDT_RESOURCE_MBA) {
hw_res->msr_base = MSR_IA32_MBA_BW_BASE;
hw_res->msr_update = mba_wrmsr_amd;
+ } else if (r->rid == RDT_RESOURCE_SMBA) {
+ hw_res->msr_base = MSR_IA32_SMBA_BW_BASE;
+ hw_res->msr_update = mba_wrmsr_amd;
}
}
}
diff --git a/arch/x86/kernel/cpu/resctrl/ctrlmondata.c b/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
index 7eece3d2d0c3..eb07d4435391 100644
--- a/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
+++ b/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
@@ -209,7 +209,7 @@ static int parse_line(char *line, struct resctrl_schema *s,
unsigned long dom_id;
if (rdtgrp->mode == RDT_MODE_PSEUDO_LOCKSETUP &&
- r->rid == RDT_RESOURCE_MBA) {
+ (r->rid == RDT_RESOURCE_MBA || r->rid == RDT_RESOURCE_SMBA)) {
rdt_last_cmd_puts("Cannot pseudo-lock MBA resource\n");
return -EINVAL;
}
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index d23fbc45ecd1..75169680516b 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1210,7 +1210,7 @@ static bool rdtgroup_mode_test_exclusive(struct rdtgroup *rdtgrp)
list_for_each_entry(s, &resctrl_schema_all, list) {
r = s->res;
- if (r->rid == RDT_RESOURCE_MBA)
+ if (r->rid == RDT_RESOURCE_MBA || r->rid == RDT_RESOURCE_SMBA)
continue;
has_cache = true;
list_for_each_entry(d, &r->domains, list) {
@@ -1399,7 +1399,8 @@ static int rdtgroup_size_show(struct kernfs_open_file *of,
ctrl = resctrl_arch_get_config(r, d,
closid,
type);
- if (r->rid == RDT_RESOURCE_MBA)
+ if (r->rid == RDT_RESOURCE_MBA ||
+ r->rid == RDT_RESOURCE_SMBA)
size = ctrl;
else
size = rdtgroup_cbm_to_size(r, d, ctrl);
@@ -2842,7 +2843,8 @@ static int rdtgroup_init_alloc(struct rdtgroup *rdtgrp)
list_for_each_entry(s, &resctrl_schema_all, list) {
r = s->res;
- if (r->rid == RDT_RESOURCE_MBA) {
+ if (r->rid == RDT_RESOURCE_MBA ||
+ r->rid == RDT_RESOURCE_SMBA) {
rdtgroup_init_mba(r, rdtgrp->closid);
if (is_mba_sc(r))
continue;
--
2.34.1
Update the documentation for the new features:
1. Slow Memory Bandwidth allocation (SMBA).
With this feature, the QOS enforcement policies can be applied
to the external slow memory connected to the host. QOS enforcement
is accomplished by assigning a Class Of Service (COS) to a processor
and specifying allocations or limits for that COS for each resource
to be allocated.
2. Bandwidth Monitoring Event Configuration (BMEC).
The bandwidth monitoring events mbm_total_bytes and mbm_local_bytes
are set to count all the total and local reads/writes respectively.
With the introduction of slow memory, the two counters are not
enough to count all the different types of memory events. With the
feature BMEC, the users have the option to configure mbm_total_bytes
and mbm_local_bytes to count the specific type of events.
Also add configuration instructions with examples.
Signed-off-by: Babu Moger <[email protected]>
---
Documentation/x86/resctrl.rst | 147 +++++++++++++++++++++++++++++++++-
1 file changed, 145 insertions(+), 2 deletions(-)
diff --git a/Documentation/x86/resctrl.rst b/Documentation/x86/resctrl.rst
index 71a531061e4e..058257dc56c8 100644
--- a/Documentation/x86/resctrl.rst
+++ b/Documentation/x86/resctrl.rst
@@ -17,14 +17,21 @@ AMD refers to this feature as AMD Platform Quality of Service(AMD QoS).
This feature is enabled by the CONFIG_X86_CPU_RESCTRL and the x86 /proc/cpuinfo
flag bits:
-============================================= ================================
+=============================================== ================================
RDT (Resource Director Technology) Allocation "rdt_a"
CAT (Cache Allocation Technology) "cat_l3", "cat_l2"
CDP (Code and Data Prioritization) "cdp_l3", "cdp_l2"
CQM (Cache QoS Monitoring) "cqm_llc", "cqm_occup_llc"
MBM (Memory Bandwidth Monitoring) "cqm_mbm_total", "cqm_mbm_local"
MBA (Memory Bandwidth Allocation) "mba"
-============================================= ================================
+SMBA (Slow Memory Bandwidth Allocation) ""
+BMEC (Bandwidth Monitoring Event Configuration) ""
+=============================================== ================================
+
+Historically, new features were made visible by default in /proc/cpuinfo. This
+resulted in the feature flags becoming hard to parse by humans. Adding a new
+flag to /proc/cpuinfo should be avoided if user space can obtain information
+about the feature from resctrl's info directory.
To use the feature mount the file system::
@@ -161,6 +168,83 @@ with the following files:
"mon_features":
Lists the monitoring events if
monitoring is enabled for the resource.
+ Example::
+
+ # cat /sys/fs/resctrl/info/L3_MON/mon_features
+ llc_occupancy
+ mbm_total_bytes
+ mbm_local_bytes
+
+ If the system supports Bandwidth Monitoring Event
+ Configuration (BMEC), then the bandwidth events will
+ be configurable. The output will be::
+
+ # cat /sys/fs/resctrl/info/L3_MON/mon_features
+ llc_occupancy
+ mbm_total_bytes
+ mbm_total_bytes_config
+ mbm_local_bytes
+ mbm_local_bytes_config
+
+"mbm_total_bytes_config", "mbm_local_bytes_config":
+ Read/write files containing the configuration for the mbm_total_bytes
+ and mbm_local_bytes events, respectively, when the Bandwidth
+ Monitoring Event Configuration (BMEC) feature is supported.
+ The event configuration settings are domain specific and affect
+ all the CPUs in the domain. When either event configuration is
+ changed, the bandwidth counters for all RMIDs of both events
+ (mbm_total_bytes as well as mbm_local_bytes) are cleared for that
+ domain. The next read for every RMID will report "Unavailable"
+ and subsequent reads will report the valid value.
+
+ Following are the types of events supported:
+
+ ==== ========================================================
+ Bits Description
+ ==== ========================================================
+ 6 Dirty Victims from the QOS domain to all types of memory
+ 5 Reads to slow memory in the non-local NUMA domain
+ 4 Reads to slow memory in the local NUMA domain
+ 3 Non-temporal writes to non-local NUMA domain
+ 2 Non-temporal writes to local NUMA domain
+ 1 Reads to memory in the non-local NUMA domain
+ 0 Reads to memory in the local NUMA domain
+ ==== ========================================================
+
+ By default, the mbm_total_bytes configuration is set to 0x7f to count
+ all the event types and the mbm_local_bytes configuration is set to
+ 0x15 to count all the local memory events.
+
+ Examples:
+
+ * To view the current configuration::
+ ::
+
+ # cat /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
+ 0=0x7f;1=0x7f;2=0x7f;3=0x7f
+
+ # cat /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
+ 0=0x15;1=0x15;3=0x15;4=0x15
+
+ * To change the mbm_total_bytes to count only reads on domain 0,
+ the bits 0, 1, 4 and 5 needs to be set, which is 110011b in binary
+ (in hexadecimal 0x33):
+ ::
+
+ # echo "0=0x33" > /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
+
+ # cat /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
+ 0=0x33;1=0x7f;2=0x7f;3=0x7f
+
+ * To change the mbm_local_bytes to count all the slow memory reads on
+ domain 0 and 1, the bits 4 and 5 needs to be set, which is 110000b
+ in binary (in hexadecimal 0x30):
+ ::
+
+ # echo "0=0x30;1=0x30" > /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
+
+ # cat /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
+ 0=0x30;1=0x30;3=0x15;4=0x15
"max_threshold_occupancy":
Read/write file provides the largest value (in
@@ -464,6 +548,25 @@ Memory bandwidth domain is L3 cache.
MB:<cache_id0>=bw_MBps0;<cache_id1>=bw_MBps1;...
+Slow Memory Bandwidth Allocation (SMBA)
+---------------------------------------
+AMD hardware supports Slow Memory Bandwidth Allocation (SMBA).
+CXL.memory is the only supported "slow" memory device. With the
+support of SMBA, the hardware enables bandwidth allocation on
+the slow memory devices. If there are multiple such devices in
+the system, the throttling logic groups all the slow sources
+together and applies the limit on them as a whole.
+
+The presence of SMBA (with CXL.memory) is independent of slow memory
+devices presence. If there are no such devices on the system, then
+configuring SMBA will have no impact on the performance of the system.
+
+The bandwidth domain for slow memory is L3 cache. Its schemata file
+is formatted as:
+::
+
+ SMBA:<cache_id0>=bandwidth0;<cache_id1>=bandwidth1;...
+
Reading/writing the schemata file
---------------------------------
Reading the schemata file will show the state of all resources
@@ -479,6 +582,46 @@ which you wish to change. E.g.
L3DATA:0=fffff;1=fffff;2=3c0;3=fffff
L3CODE:0=fffff;1=fffff;2=fffff;3=fffff
+Reading/writing the schemata file (on AMD systems)
+--------------------------------------------------
+Reading the schemata file will show the current bandwidth limit on all
+domains. The allocated resources are in multiples of one eighth GB/s.
+When writing to the file, you need to specify what cache id you wish to
+configure the bandwidth limit.
+
+For example, to allocate 2GB/s limit on the first cache id:
+
+::
+
+ # cat schemata
+ MB:0=2048;1=2048;2=2048;3=2048
+ L3:0=ffff;1=ffff;2=ffff;3=ffff
+
+ # echo "MB:1=16" > schemata
+ # cat schemata
+ MB:0=2048;1= 16;2=2048;3=2048
+ L3:0=ffff;1=ffff;2=ffff;3=ffff
+
+Reading/writing the schemata file (on AMD systems) with SMBA feature
+--------------------------------------------------------------------
+Reading and writing the schemata file is the same as without SMBA in
+above section.
+
+For example, to allocate 8GB/s limit on the first cache id:
+
+::
+
+ # cat schemata
+ SMBA:0=2048;1=2048;2=2048;3=2048
+ MB:0=2048;1=2048;2=2048;3=2048
+ L3:0=ffff;1=ffff;2=ffff;3=ffff
+
+ # echo "SMBA:1=64" > schemata
+ # cat schemata
+ SMBA:0=2048;1= 64;2=2048;3=2048
+ MB:0=2048;1=2048;2=2048;3=2048
+ L3:0=ffff;1=ffff;2=ffff;3=ffff
+
Cache Pseudo-Locking
====================
CAT enables a user to specify the amount of cache space that an
--
2.34.1
The event configuration for mbm_local_bytes can be changed by the
user by writing to the configuration file
/sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config.
The event configuration settings are domain specific and will affect all
the CPUs in the domain.
Following are the types of events supported:
==== ===========================================================
Bits Description
==== ===========================================================
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
==== ===========================================================
For example:
To change the mbm_local_bytes_config to count all the non-temporal writes
on domain 0, the bits 2 and 3 needs to be set which is 1100b (in hex 0xc).
Run the command.
$echo 0=0xc > /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
To change the mbm_local_bytes to count only reads to local NUMA domain 1,
the bit 0 needs to be set which 1b (in hex 0x1). Run the command.
$echo 1=0x1 > /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
Reviewed-by: Reinette Chatre <[email protected]>
Signed-off-by: Babu Moger <[email protected]>
---
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 27 +++++++++++++++++++++++++-
1 file changed, 26 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 03284a61c1a0..5990589f8a21 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1638,6 +1638,30 @@ static ssize_t mbm_total_bytes_config_write(struct kernfs_open_file *of,
return ret ?: nbytes;
}
+static ssize_t mbm_local_bytes_config_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes,
+ loff_t off)
+{
+ struct rdt_resource *r = of->kn->parent->priv;
+ int ret;
+
+ /* Valid input requires a trailing newline */
+ if (nbytes == 0 || buf[nbytes - 1] != '\n')
+ return -EINVAL;
+
+ mutex_lock(&rdtgroup_mutex);
+
+ rdt_last_cmd_clear();
+
+ buf[nbytes - 1] = '\0';
+
+ ret = mon_config_write(r, buf, QOS_L3_MBM_LOCAL_EVENT_ID);
+
+ mutex_unlock(&rdtgroup_mutex);
+
+ return ret ?: nbytes;
+}
+
/* rdtgroup information files for one cache resource. */
static struct rftype res_common_files[] = {
{
@@ -1745,9 +1769,10 @@ static struct rftype res_common_files[] = {
},
{
.name = "mbm_local_bytes_config",
- .mode = 0444,
+ .mode = 0644,
.kf_ops = &rdtgroup_kf_single_ops,
.seq_show = mbm_local_bytes_config_show,
+ .write = mbm_local_bytes_config_write,
},
{
.name = "cpus",
--
2.34.1
on_each_cpu_mask() runs the function on each CPU specified by cpumask,
which may include the local processor.
Replace smp_call_function_many() with on_each_cpu_mask() to simplify
the code.
Reviewed-by: Reinette Chatre <[email protected]>
Signed-off-by: Babu Moger <[email protected]>
---
arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 11 +++------
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 29 +++++++----------------
2 files changed, 11 insertions(+), 29 deletions(-)
diff --git a/arch/x86/kernel/cpu/resctrl/ctrlmondata.c b/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
index 1df0e3262bca..7eece3d2d0c3 100644
--- a/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
+++ b/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
@@ -310,7 +310,6 @@ int resctrl_arch_update_domains(struct rdt_resource *r, u32 closid)
enum resctrl_conf_type t;
cpumask_var_t cpu_mask;
struct rdt_domain *d;
- int cpu;
u32 idx;
if (!zalloc_cpumask_var(&cpu_mask, GFP_KERNEL))
@@ -341,13 +340,9 @@ int resctrl_arch_update_domains(struct rdt_resource *r, u32 closid)
if (cpumask_empty(cpu_mask))
goto done;
- cpu = get_cpu();
- /* Update resource control msr on this CPU if it's in cpu_mask. */
- if (cpumask_test_cpu(cpu, cpu_mask))
- rdt_ctrl_update(&msr_param);
- /* Update resource control msr on other CPUs. */
- smp_call_function_many(cpu_mask, rdt_ctrl_update, &msr_param, 1);
- put_cpu();
+
+ /* Update resource control msr on all the CPUs. */
+ on_each_cpu_mask(cpu_mask, rdt_ctrl_update, &msr_param, 1);
done:
free_cpumask_var(cpu_mask);
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 5993da21d822..d23fbc45ecd1 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -325,12 +325,7 @@ static void update_cpu_closid_rmid(void *info)
static void
update_closid_rmid(const struct cpumask *cpu_mask, struct rdtgroup *r)
{
- int cpu = get_cpu();
-
- if (cpumask_test_cpu(cpu, cpu_mask))
- update_cpu_closid_rmid(r);
- smp_call_function_many(cpu_mask, update_cpu_closid_rmid, r, 1);
- put_cpu();
+ on_each_cpu_mask(cpu_mask, update_cpu_closid_rmid, r, 1);
}
static int cpus_mon_write(struct rdtgroup *rdtgrp, cpumask_var_t newmask,
@@ -1866,13 +1861,9 @@ static int set_cache_qos_cfg(int level, bool enable)
/* Pick one CPU from each domain instance to update MSR */
cpumask_set_cpu(cpumask_any(&d->cpu_mask), cpu_mask);
}
- cpu = get_cpu();
- /* Update QOS_CFG MSR on this cpu if it's in cpu_mask. */
- if (cpumask_test_cpu(cpu, cpu_mask))
- update(&enable);
- /* Update QOS_CFG MSR on all other cpus in cpu_mask. */
- smp_call_function_many(cpu_mask, update, &enable, 1);
- put_cpu();
+
+ /* Update QOS_CFG MSR on all the CPUs in cpu_mask */
+ on_each_cpu_mask(cpu_mask, update, &enable, 1);
free_cpumask_var(cpu_mask);
@@ -2349,7 +2340,7 @@ static int reset_all_ctrls(struct rdt_resource *r)
struct msr_param msr_param;
cpumask_var_t cpu_mask;
struct rdt_domain *d;
- int i, cpu;
+ int i;
if (!zalloc_cpumask_var(&cpu_mask, GFP_KERNEL))
return -ENOMEM;
@@ -2370,13 +2361,9 @@ static int reset_all_ctrls(struct rdt_resource *r)
for (i = 0; i < hw_res->num_closid; i++)
hw_dom->ctrl_val[i] = r->default_ctrl;
}
- cpu = get_cpu();
- /* Update CBM on this cpu if it's in cpu_mask. */
- if (cpumask_test_cpu(cpu, cpu_mask))
- rdt_ctrl_update(&msr_param);
- /* Update CBM on all other cpus in cpu_mask. */
- smp_call_function_many(cpu_mask, rdt_ctrl_update, &msr_param, 1);
- put_cpu();
+
+ /* Update CBM on all the CPUs in cpu_mask */
+ on_each_cpu_mask(cpu_mask, rdt_ctrl_update, &msr_param, 1);
free_cpumask_var(cpu_mask);
--
2.34.1
The event configuration for mbm_total_bytes can be changed by the user by
writing to the file /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config.
The event configuration settings are domain specific and affect all the
CPUs in the domain.
Following are the types of events supported:
==== ===========================================================
Bits Description
==== ===========================================================
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
==== ===========================================================
For example:
To change the mbm_total_bytes to count only reads on domain 0, the bits
0, 1, 4 and 5 needs to be set, which is 110011b (in hex 0x33). Run the
command.
$echo 0=0x33 > /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
To change the mbm_total_bytes to count all the slow memory reads on
domain 1, the bits 4 and 5 needs to be set which is 110000b (in hex 0x30).
Run the command.
$echo 1=0x30 > /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
Reviewed-by: Reinette Chatre <[email protected]>
Signed-off-by: Babu Moger <[email protected]>
---
arch/x86/kernel/cpu/resctrl/monitor.c | 17 ++++
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 124 ++++++++++++++++++++++++-
include/linux/resctrl.h | 11 +++
3 files changed, 151 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index 28c65008fce1..7fe51488e136 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -204,6 +204,23 @@ void resctrl_arch_reset_rmid(struct rdt_resource *r, struct rdt_domain *d,
}
}
+/*
+ * Assumes that hardware counters are also reset and thus that there is
+ * no need to record initial non-zero counts.
+ */
+void resctrl_arch_reset_rmid_all(struct rdt_resource *r, struct rdt_domain *d)
+{
+ struct rdt_hw_domain *hw_dom = resctrl_to_arch_dom(d);
+
+ if (is_mbm_total_enabled())
+ memset(hw_dom->arch_mbm_total, 0,
+ sizeof(*hw_dom->arch_mbm_total) * r->num_rmid);
+
+ if (is_mbm_local_enabled())
+ memset(hw_dom->arch_mbm_local, 0,
+ sizeof(*hw_dom->arch_mbm_local) * r->num_rmid);
+}
+
static u64 mbm_overflow_count(u64 prev_msr, u64 cur_msr, unsigned int width)
{
u64 shift = 64 - width, chunks;
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index ad3c7014adf6..03284a61c1a0 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1517,6 +1517,127 @@ static int mbm_local_bytes_config_show(struct kernfs_open_file *of,
return 0;
}
+static void mon_event_config_write(void *info)
+{
+ struct mon_config_info *mon_info = info;
+ unsigned int index;
+
+ index = mon_event_config_index_get(mon_info->evtid);
+ if (index == INVALID_CONFIG_INDEX) {
+ pr_warn_once("Invalid event id %d\n", mon_info->evtid);
+ return;
+ }
+ wrmsr(MSR_IA32_EVT_CFG_BASE + index, mon_info->mon_config, 0);
+}
+
+static int mbm_config_write_domain(struct rdt_resource *r,
+ struct rdt_domain *d, u32 evtid, u32 val)
+{
+ struct mon_config_info mon_info = {0};
+ int ret = 0;
+
+ /* mon_config cannot be more than the supported set of events */
+ if (val > MAX_EVT_CONFIG_BITS) {
+ rdt_last_cmd_puts("Invalid event configuration\n");
+ return -EINVAL;
+ }
+
+ /*
+ * Read the current config value first. If both are the same then
+ * no need to write it again.
+ */
+ mon_info.evtid = evtid;
+ mondata_config_read(d, &mon_info);
+ if (mon_info.mon_config == val)
+ goto out;
+
+ mon_info.mon_config = val;
+
+ /*
+ * Update MSR_IA32_EVT_CFG_BASE MSR on one of the CPUs in the
+ * domain. The MSRs offset from MSR MSR_IA32_EVT_CFG_BASE
+ * are scoped at the domain level. Writing any of these MSRs
+ * on one CPU is observed by all the CPUs in the domain.
+ */
+ smp_call_function_any(&d->cpu_mask, mon_event_config_write,
+ &mon_info, 1);
+
+ /*
+ * When an Event Configuration is changed, the bandwidth counters
+ * for all RMIDs and Events will be cleared by the hardware. The
+ * hardware also sets MSR_IA32_QM_CTR.Unavailable (bit 62) for
+ * every RMID on the next read to any event for every RMID.
+ * Subsequent reads will have MSR_IA32_QM_CTR.Unavailable (bit 62)
+ * cleared while it is tracked by the hardware. Clear the
+ * mbm_local and mbm_total counts for all the RMIDs.
+ */
+ resctrl_arch_reset_rmid_all(r, d);
+
+out:
+ return ret;
+}
+
+static int mon_config_write(struct rdt_resource *r, char *tok, u32 evtid)
+{
+ char *dom_str = NULL, *id_str;
+ unsigned long dom_id, val;
+ struct rdt_domain *d;
+ int ret = 0;
+
+next:
+ if (!tok || tok[0] == '\0')
+ return 0;
+
+ /* Start processing the strings for each domain */
+ dom_str = strim(strsep(&tok, ";"));
+ id_str = strsep(&dom_str, "=");
+
+ if (!id_str || kstrtoul(id_str, 10, &dom_id)) {
+ rdt_last_cmd_puts("Missing '=' or non-numeric domain id\n");
+ return -EINVAL;
+ }
+
+ if (!dom_str || kstrtoul(dom_str, 16, &val)) {
+ rdt_last_cmd_puts("Non-numeric event configuration value\n");
+ return -EINVAL;
+ }
+
+ list_for_each_entry(d, &r->domains, list) {
+ if (d->id == dom_id) {
+ ret = mbm_config_write_domain(r, d, evtid, val);
+ if (ret)
+ return -EINVAL;
+ goto next;
+ }
+ }
+
+ return -EINVAL;
+}
+
+static ssize_t mbm_total_bytes_config_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes,
+ loff_t off)
+{
+ struct rdt_resource *r = of->kn->parent->priv;
+ int ret;
+
+ /* Valid input requires a trailing newline */
+ if (nbytes == 0 || buf[nbytes - 1] != '\n')
+ return -EINVAL;
+
+ mutex_lock(&rdtgroup_mutex);
+
+ rdt_last_cmd_clear();
+
+ buf[nbytes - 1] = '\0';
+
+ ret = mon_config_write(r, buf, QOS_L3_MBM_TOTAL_EVENT_ID);
+
+ mutex_unlock(&rdtgroup_mutex);
+
+ return ret ?: nbytes;
+}
+
/* rdtgroup information files for one cache resource. */
static struct rftype res_common_files[] = {
{
@@ -1617,9 +1738,10 @@ static struct rftype res_common_files[] = {
},
{
.name = "mbm_total_bytes_config",
- .mode = 0444,
+ .mode = 0644,
.kf_ops = &rdtgroup_kf_single_ops,
.seq_show = mbm_total_bytes_config_show,
+ .write = mbm_total_bytes_config_write,
},
{
.name = "mbm_local_bytes_config",
diff --git a/include/linux/resctrl.h b/include/linux/resctrl.h
index 0cee154abc9f..8334eeacfec5 100644
--- a/include/linux/resctrl.h
+++ b/include/linux/resctrl.h
@@ -250,6 +250,17 @@ int resctrl_arch_rmid_read(struct rdt_resource *r, struct rdt_domain *d,
void resctrl_arch_reset_rmid(struct rdt_resource *r, struct rdt_domain *d,
u32 rmid, enum resctrl_event_id eventid);
+/**
+ * resctrl_arch_reset_rmid_all() - Reset all private state associated with
+ * all rmids and eventids.
+ * @r: The resctrl resource.
+ * @d: The domain for which all architectural counter state will
+ * be cleared.
+ *
+ * This can be called from any CPU.
+ */
+void resctrl_arch_reset_rmid_all(struct rdt_resource *r, struct rdt_domain *d);
+
extern unsigned int resctrl_rmid_realloc_threshold;
extern unsigned int resctrl_rmid_realloc_limit;
--
2.34.1
Newer AMD processors support the new feature Bandwidth Monitoring Event
Configuration (BMEC).
The feature support is identified via CPUID Fn8000_0020_EBX_x0 (ECX=0).
Bits Field Name Description
3 EVT_CFG Bandwidth Monitoring Event Configuration (BMEC)
The bandwidth monitoring events mbm_total_bytes and mbm_local_bytes are
set to count all the total and local reads/writes respectively. With the
introduction of slow memory, the two counters are not enough to count
all the different types of memory events. With the feature BMEC, the
users have the option to configure mbm_total_bytes and mbm_local_bytes
to count the specific type of events.
Each BMEC event has a configuration MSR, which contains one field for
each bandwidth type that can be used to configure the bandwidth event
to track any combination of supported bandwidth types. The event will
count requests from every bandwidth type bit that is set in the
corresponding configuration register.
Following are the types of events supported:
==== ========================================================
Bits Description
==== ========================================================
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
==== ========================================================
By default, the mbm_total_bytes configuration is set to 0x7F to count
all the event types and the mbm_local_bytes configuration is set to
0x15 to count all the local memory events.
Feature description is available in the specification, "AMD64
Technology Platform Quality of Service Extensions, Revision: 1.03
Publication".
Reviewed-by: Reinette Chatre <[email protected]>
Signed-off-by: Babu Moger <[email protected]>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=206537
Link: https://www.amd.com/en/support/tech-docs/amd64-technology-platform-quality-service-extensions
---
arch/x86/include/asm/cpufeatures.h | 1 +
arch/x86/kernel/cpu/cpuid-deps.c | 2 ++
arch/x86/kernel/cpu/scattered.c | 1 +
3 files changed, 4 insertions(+)
diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
index fbf243a7937e..123613cdae93 100644
--- a/arch/x86/include/asm/cpufeatures.h
+++ b/arch/x86/include/asm/cpufeatures.h
@@ -308,6 +308,7 @@
#define X86_FEATURE_CALL_DEPTH (11*32+19) /* "" Call depth tracking for RSB stuffing */
#define X86_FEATURE_MSR_TSX_CTRL (11*32+20) /* "" MSR IA32_TSX_CTRL (Intel) implemented */
#define X86_FEATURE_SMBA (11*32+21) /* "" Slow Memory Bandwidth Allocation */
+#define X86_FEATURE_BMEC (11*32+22) /* "" Bandwidth Monitoring Event Configuration */
/* Intel-defined CPU features, CPUID level 0x00000007:1 (EAX), word 12 */
#define X86_FEATURE_AVX_VNNI (12*32+ 4) /* AVX VNNI instructions */
diff --git a/arch/x86/kernel/cpu/cpuid-deps.c b/arch/x86/kernel/cpu/cpuid-deps.c
index d95221117129..f6748c8bd647 100644
--- a/arch/x86/kernel/cpu/cpuid-deps.c
+++ b/arch/x86/kernel/cpu/cpuid-deps.c
@@ -68,6 +68,8 @@ static const struct cpuid_dep cpuid_deps[] = {
{ X86_FEATURE_CQM_OCCUP_LLC, X86_FEATURE_CQM_LLC },
{ X86_FEATURE_CQM_MBM_TOTAL, X86_FEATURE_CQM_LLC },
{ X86_FEATURE_CQM_MBM_LOCAL, X86_FEATURE_CQM_LLC },
+ { X86_FEATURE_BMEC, X86_FEATURE_CQM_MBM_TOTAL },
+ { X86_FEATURE_BMEC, X86_FEATURE_CQM_MBM_LOCAL },
{ X86_FEATURE_AVX512_BF16, X86_FEATURE_AVX512VL },
{ X86_FEATURE_AVX512_FP16, X86_FEATURE_AVX512BW },
{ X86_FEATURE_ENQCMD, X86_FEATURE_XSAVES },
diff --git a/arch/x86/kernel/cpu/scattered.c b/arch/x86/kernel/cpu/scattered.c
index d925753084fb..0dad49a09b7a 100644
--- a/arch/x86/kernel/cpu/scattered.c
+++ b/arch/x86/kernel/cpu/scattered.c
@@ -46,6 +46,7 @@ static const struct cpuid_bit cpuid_bits[] = {
{ X86_FEATURE_PROC_FEEDBACK, CPUID_EDX, 11, 0x80000007, 0 },
{ X86_FEATURE_MBA, CPUID_EBX, 6, 0x80000008, 0 },
{ X86_FEATURE_SMBA, CPUID_EBX, 2, 0x80000020, 0 },
+ { X86_FEATURE_BMEC, CPUID_EBX, 3, 0x80000020, 0 },
{ X86_FEATURE_PERFMON_V2, CPUID_EAX, 0, 0x80000022, 0 },
{ X86_FEATURE_AMD_LBR_V2, CPUID_EAX, 1, 0x80000022, 0 },
{ 0, 0, 0, 0, 0 }
--
2.34.1
Add a new field in struct mon_evt to support Bandwidth Monitoring Event
Configuration(BMEC) and also update the "mon_features" display.
The resctrl file "mon_features" will display the supported events
and files that can be used to configure those events if monitor
configuration is supported.
Before the change.
$cat /sys/fs/resctrl/info/L3_MON/mon_features
llc_occupancy
mbm_total_bytes
mbm_local_bytes
After the change when BMEC is supported.
$cat /sys/fs/resctrl/info/L3_MON/mon_features
llc_occupancy
mbm_total_bytes
mbm_total_bytes_config
mbm_local_bytes
mbm_local_bytes_config
Reviewed-by: Reinette Chatre <[email protected]>
Signed-off-by: Babu Moger <[email protected]>
---
arch/x86/kernel/cpu/resctrl/internal.h | 2 ++
arch/x86/kernel/cpu/resctrl/monitor.c | 7 +++++++
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 5 ++++-
3 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/resctrl/internal.h b/arch/x86/kernel/cpu/resctrl/internal.h
index f16b8bc5448c..0605b04f1b7a 100644
--- a/arch/x86/kernel/cpu/resctrl/internal.h
+++ b/arch/x86/kernel/cpu/resctrl/internal.h
@@ -52,11 +52,13 @@ DECLARE_STATIC_KEY_FALSE(rdt_mon_enable_key);
* struct mon_evt - Entry in the event list of a resource
* @evtid: event id
* @name: name of the event
+ * @configurable: true if the event is configurable
* @list: entry in &rdt_resource->evt_list
*/
struct mon_evt {
enum resctrl_event_id evtid;
char *name;
+ bool configurable;
struct list_head list;
};
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index 5665ab9725ce..cf7b6e59fa29 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -800,6 +800,13 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r)
if (ret)
return ret;
+ if (rdt_cpu_has(X86_FEATURE_BMEC)) {
+ if (rdt_cpu_has(X86_FEATURE_CQM_MBM_TOTAL))
+ mbm_total_event.configurable = true;
+ if (rdt_cpu_has(X86_FEATURE_CQM_MBM_LOCAL))
+ mbm_local_event.configurable = true;
+ }
+
l3_mon_evt_init(r);
r->mon_capable = true;
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 75169680516b..bf0bd4378e13 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -998,8 +998,11 @@ static int rdt_mon_features_show(struct kernfs_open_file *of,
struct rdt_resource *r = of->kn->parent->priv;
struct mon_evt *mevt;
- list_for_each_entry(mevt, &r->evt_list, list)
+ list_for_each_entry(mevt, &r->evt_list, list) {
seq_printf(seq, "%s\n", mevt->name);
+ if (mevt->configurable)
+ seq_printf(seq, "%s_config\n", mevt->name);
+ }
return 0;
}
--
2.34.1
Hi Babu,
On 1/13/2023 7:20 AM, Babu Moger wrote:
> New AMD processors can now support following QoS features.
>
> 1. Slow Memory Bandwidth Allocation (SMBA)
> With this feature, the QOS enforcement policies can be applied
> to the external slow memory connected to the host. QOS enforcement
> is accomplished by assigning a Class Of Service (COS) to a processor
> and specifying allocations or limits for that COS for each resource
> to be allocated.
>
> Currently, CXL.memory is the only supported "slow" memory device. With
> the support of SMBA feature the hardware enables bandwidth allocation
> on the slow memory devices.
>
> 2. Bandwidth Monitoring Event Configuration (BMEC)
> The bandwidth monitoring events mbm_total_event and mbm_local_event
> are set to count all the total and local reads/writes respectively.
> With the introduction of slow memory, the two counters are not enough
> to count all the different types are memory events. With the feature
> BMEC, the users have the option to configure mbm_total_event and
> mbm_local_event to count the specific type of events.
>
> Following are the bitmaps of events supported.
> Bits Description
> 6 Dirty Victims from the QOS domain to all types of memory
> 5 Reads to slow memory in the non-local NUMA domain
> 4 Reads to slow memory in the local NUMA domain
> 3 Non-temporal writes to non-local NUMA domain
> 2 Non-temporal writes to local NUMA domain
> 1 Reads to memory in the non-local NUMA domain
> 0 Reads to memory in the local NUMA domain
>
> This series adds support for these features. Also added a minor cleanup(PATCH 1).
>
> Feature description is available in the specification, "AMD64 Technology Platform Quality of Service Extensions, Revision: 1.03 Publication # 56375
> Revision: 1.03 Issue Date: February 2022".
>
> Link: https://bugzilla.kernel.org/show_bug.cgi?id=206537
> Link: https://www.amd.com/en/support/tech-docs/amd64-technology-platform-quality-service-extensions
This series looks good to me. I added my "Reviewed-by" tag to patch 13/13. All
the other patches in this series already have my "Reviewed-by" tag.
Thank you very much.
Reinette
Hi Babu,
On 1/13/2023 7:20 AM, Babu Moger wrote:
> Update the documentation for the new features:
> 1. Slow Memory Bandwidth allocation (SMBA).
> With this feature, the QOS enforcement policies can be applied
> to the external slow memory connected to the host. QOS enforcement
> is accomplished by assigning a Class Of Service (COS) to a processor
> and specifying allocations or limits for that COS for each resource
> to be allocated.
>
> 2. Bandwidth Monitoring Event Configuration (BMEC).
> The bandwidth monitoring events mbm_total_bytes and mbm_local_bytes
> are set to count all the total and local reads/writes respectively.
> With the introduction of slow memory, the two counters are not
> enough to count all the different types of memory events. With the
> feature BMEC, the users have the option to configure mbm_total_bytes
> and mbm_local_bytes to count the specific type of events.
>
> Also add configuration instructions with examples.
>
> Signed-off-by: Babu Moger <[email protected]>
Thank you.
Reviewed-by: Reinette Chatre <[email protected]>
Reinette
On 1/17/23 16:16, Reinette Chatre wrote:
> Hi Babu,
>
> On 1/13/2023 7:20 AM, Babu Moger wrote:
>> New AMD processors can now support following QoS features.
>>
>> 1. Slow Memory Bandwidth Allocation (SMBA)
>> With this feature, the QOS enforcement policies can be applied
>> to the external slow memory connected to the host. QOS enforcement
>> is accomplished by assigning a Class Of Service (COS) to a processor
>> and specifying allocations or limits for that COS for each resource
>> to be allocated.
>>
>> Currently, CXL.memory is the only supported "slow" memory device. With
>> the support of SMBA feature the hardware enables bandwidth allocation
>> on the slow memory devices.
>>
>> 2. Bandwidth Monitoring Event Configuration (BMEC)
>> The bandwidth monitoring events mbm_total_event and mbm_local_event
>> are set to count all the total and local reads/writes respectively.
>> With the introduction of slow memory, the two counters are not enough
>> to count all the different types are memory events. With the feature
>> BMEC, the users have the option to configure mbm_total_event and
>> mbm_local_event to count the specific type of events.
>>
>> Following are the bitmaps of events supported.
>> Bits Description
>> 6 Dirty Victims from the QOS domain to all types of memory
>> 5 Reads to slow memory in the non-local NUMA domain
>> 4 Reads to slow memory in the local NUMA domain
>> 3 Non-temporal writes to non-local NUMA domain
>> 2 Non-temporal writes to local NUMA domain
>> 1 Reads to memory in the non-local NUMA domain
>> 0 Reads to memory in the local NUMA domain
>>
>> This series adds support for these features. Also added a minor cleanup(PATCH 1).
>>
>> Feature description is available in the specification, "AMD64 Technology Platform Quality of Service Extensions, Revision: 1.03 Publication # 56375
>> Revision: 1.03 Issue Date: February 2022".
>>
>> Link: https://bugzilla.kernel.org/show_bug.cgi?id=206537
>> Link: https://www.amd.com/en/support/tech-docs/amd64-technology-platform-quality-service-extensions
> This series looks good to me. I added my "Reviewed-by" tag to patch 13/13. All
> the other patches in this series already have my "Reviewed-by" tag.
Thank you very much.
Babu Moger
The following commit has been merged into the x86/cache branch of tip:
Commit-ID: 92bd5a1390335bb3cc76bdf1b4356edbc94d408d
Gitweb: https://git.kernel.org/tip/92bd5a1390335bb3cc76bdf1b4356edbc94d408d
Author: Babu Moger <[email protected]>
AuthorDate: Fri, 13 Jan 2023 09:20:37 -06:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 23 Jan 2023 17:40:30 +01:00
x86/resctrl: Add interface to write mbm_total_bytes_config
The event configuration for mbm_total_bytes can be changed by the user by
writing to the file /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config.
The event configuration settings are domain specific and affect all the
CPUs in the domain.
Following are the types of events supported:
==== ===========================================================
Bits Description
==== ===========================================================
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
==== ===========================================================
For example:
To change the mbm_total_bytes to count only reads on domain 0, the bits
0, 1, 4 and 5 needs to be set, which is 110011b (in hex 0x33).
Run the command:
$echo 0=0x33 > /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
To change the mbm_total_bytes to count all the slow memory reads on domain 1,
the bits 4 and 5 needs to be set which is 110000b (in hex 0x30).
Run the command:
$echo 1=0x30 > /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
Signed-off-by: Babu Moger <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Reinette Chatre <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/kernel/cpu/resctrl/monitor.c | 17 +++-
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 124 +++++++++++++++++++++++-
include/linux/resctrl.h | 11 ++-
3 files changed, 151 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index 28c6500..7fe5148 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -204,6 +204,23 @@ void resctrl_arch_reset_rmid(struct rdt_resource *r, struct rdt_domain *d,
}
}
+/*
+ * Assumes that hardware counters are also reset and thus that there is
+ * no need to record initial non-zero counts.
+ */
+void resctrl_arch_reset_rmid_all(struct rdt_resource *r, struct rdt_domain *d)
+{
+ struct rdt_hw_domain *hw_dom = resctrl_to_arch_dom(d);
+
+ if (is_mbm_total_enabled())
+ memset(hw_dom->arch_mbm_total, 0,
+ sizeof(*hw_dom->arch_mbm_total) * r->num_rmid);
+
+ if (is_mbm_local_enabled())
+ memset(hw_dom->arch_mbm_local, 0,
+ sizeof(*hw_dom->arch_mbm_local) * r->num_rmid);
+}
+
static u64 mbm_overflow_count(u64 prev_msr, u64 cur_msr, unsigned int width)
{
u64 shift = 64 - width, chunks;
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index ad3c701..03284a6 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1517,6 +1517,127 @@ static int mbm_local_bytes_config_show(struct kernfs_open_file *of,
return 0;
}
+static void mon_event_config_write(void *info)
+{
+ struct mon_config_info *mon_info = info;
+ unsigned int index;
+
+ index = mon_event_config_index_get(mon_info->evtid);
+ if (index == INVALID_CONFIG_INDEX) {
+ pr_warn_once("Invalid event id %d\n", mon_info->evtid);
+ return;
+ }
+ wrmsr(MSR_IA32_EVT_CFG_BASE + index, mon_info->mon_config, 0);
+}
+
+static int mbm_config_write_domain(struct rdt_resource *r,
+ struct rdt_domain *d, u32 evtid, u32 val)
+{
+ struct mon_config_info mon_info = {0};
+ int ret = 0;
+
+ /* mon_config cannot be more than the supported set of events */
+ if (val > MAX_EVT_CONFIG_BITS) {
+ rdt_last_cmd_puts("Invalid event configuration\n");
+ return -EINVAL;
+ }
+
+ /*
+ * Read the current config value first. If both are the same then
+ * no need to write it again.
+ */
+ mon_info.evtid = evtid;
+ mondata_config_read(d, &mon_info);
+ if (mon_info.mon_config == val)
+ goto out;
+
+ mon_info.mon_config = val;
+
+ /*
+ * Update MSR_IA32_EVT_CFG_BASE MSR on one of the CPUs in the
+ * domain. The MSRs offset from MSR MSR_IA32_EVT_CFG_BASE
+ * are scoped at the domain level. Writing any of these MSRs
+ * on one CPU is observed by all the CPUs in the domain.
+ */
+ smp_call_function_any(&d->cpu_mask, mon_event_config_write,
+ &mon_info, 1);
+
+ /*
+ * When an Event Configuration is changed, the bandwidth counters
+ * for all RMIDs and Events will be cleared by the hardware. The
+ * hardware also sets MSR_IA32_QM_CTR.Unavailable (bit 62) for
+ * every RMID on the next read to any event for every RMID.
+ * Subsequent reads will have MSR_IA32_QM_CTR.Unavailable (bit 62)
+ * cleared while it is tracked by the hardware. Clear the
+ * mbm_local and mbm_total counts for all the RMIDs.
+ */
+ resctrl_arch_reset_rmid_all(r, d);
+
+out:
+ return ret;
+}
+
+static int mon_config_write(struct rdt_resource *r, char *tok, u32 evtid)
+{
+ char *dom_str = NULL, *id_str;
+ unsigned long dom_id, val;
+ struct rdt_domain *d;
+ int ret = 0;
+
+next:
+ if (!tok || tok[0] == '\0')
+ return 0;
+
+ /* Start processing the strings for each domain */
+ dom_str = strim(strsep(&tok, ";"));
+ id_str = strsep(&dom_str, "=");
+
+ if (!id_str || kstrtoul(id_str, 10, &dom_id)) {
+ rdt_last_cmd_puts("Missing '=' or non-numeric domain id\n");
+ return -EINVAL;
+ }
+
+ if (!dom_str || kstrtoul(dom_str, 16, &val)) {
+ rdt_last_cmd_puts("Non-numeric event configuration value\n");
+ return -EINVAL;
+ }
+
+ list_for_each_entry(d, &r->domains, list) {
+ if (d->id == dom_id) {
+ ret = mbm_config_write_domain(r, d, evtid, val);
+ if (ret)
+ return -EINVAL;
+ goto next;
+ }
+ }
+
+ return -EINVAL;
+}
+
+static ssize_t mbm_total_bytes_config_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes,
+ loff_t off)
+{
+ struct rdt_resource *r = of->kn->parent->priv;
+ int ret;
+
+ /* Valid input requires a trailing newline */
+ if (nbytes == 0 || buf[nbytes - 1] != '\n')
+ return -EINVAL;
+
+ mutex_lock(&rdtgroup_mutex);
+
+ rdt_last_cmd_clear();
+
+ buf[nbytes - 1] = '\0';
+
+ ret = mon_config_write(r, buf, QOS_L3_MBM_TOTAL_EVENT_ID);
+
+ mutex_unlock(&rdtgroup_mutex);
+
+ return ret ?: nbytes;
+}
+
/* rdtgroup information files for one cache resource. */
static struct rftype res_common_files[] = {
{
@@ -1617,9 +1738,10 @@ static struct rftype res_common_files[] = {
},
{
.name = "mbm_total_bytes_config",
- .mode = 0444,
+ .mode = 0644,
.kf_ops = &rdtgroup_kf_single_ops,
.seq_show = mbm_total_bytes_config_show,
+ .write = mbm_total_bytes_config_write,
},
{
.name = "mbm_local_bytes_config",
diff --git a/include/linux/resctrl.h b/include/linux/resctrl.h
index 0cee154..8334eea 100644
--- a/include/linux/resctrl.h
+++ b/include/linux/resctrl.h
@@ -250,6 +250,17 @@ int resctrl_arch_rmid_read(struct rdt_resource *r, struct rdt_domain *d,
void resctrl_arch_reset_rmid(struct rdt_resource *r, struct rdt_domain *d,
u32 rmid, enum resctrl_event_id eventid);
+/**
+ * resctrl_arch_reset_rmid_all() - Reset all private state associated with
+ * all rmids and eventids.
+ * @r: The resctrl resource.
+ * @d: The domain for which all architectural counter state will
+ * be cleared.
+ *
+ * This can be called from any CPU.
+ */
+void resctrl_arch_reset_rmid_all(struct rdt_resource *r, struct rdt_domain *d);
+
extern unsigned int resctrl_rmid_realloc_threshold;
extern unsigned int resctrl_rmid_realloc_limit;
The following commit has been merged into the x86/cache branch of tip:
Commit-ID: 73afb2d3ce2d7e0f9eee14e3f2b53d2bbfefd9c8
Gitweb: https://git.kernel.org/tip/73afb2d3ce2d7e0f9eee14e3f2b53d2bbfefd9c8
Author: Babu Moger <[email protected]>
AuthorDate: Fri, 13 Jan 2023 09:20:36 -06:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 23 Jan 2023 17:40:27 +01:00
x86/resctrl: Add interface to read mbm_local_bytes_config
The event configuration can be viewed by the user by reading the configuration
file /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config. The event
configuration settings are domain specific and will affect all the CPUs in the
domain.
Following are the types of events supported:
==== ===========================================================
Bits Description
==== ===========================================================
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
==== ===========================================================
By default, the mbm_local_bytes_config is set to 0x15 to count all the local
event types.
For example:
$cat /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
0=0x15;1=0x15;2=0x15;3=0x15
In this case, the event mbm_local_bytes is configured with 0x15 on
domains 0 to 3.
Signed-off-by: Babu Moger <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Reinette Chatre <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/kernel/cpu/resctrl/monitor.c | 4 +++-
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 16 ++++++++++++++++
2 files changed, 19 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index 0ff45e3..28c6500 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -805,8 +805,10 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r)
mbm_total_event.configurable = true;
mbm_config_rftype_init("mbm_total_bytes_config");
}
- if (rdt_cpu_has(X86_FEATURE_CQM_MBM_LOCAL))
+ if (rdt_cpu_has(X86_FEATURE_CQM_MBM_LOCAL)) {
mbm_local_event.configurable = true;
+ mbm_config_rftype_init("mbm_local_bytes_config");
+ }
}
l3_mon_evt_init(r);
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index cd4e668..ad3c701 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1507,6 +1507,16 @@ static int mbm_total_bytes_config_show(struct kernfs_open_file *of,
return 0;
}
+static int mbm_local_bytes_config_show(struct kernfs_open_file *of,
+ struct seq_file *seq, void *v)
+{
+ struct rdt_resource *r = of->kn->parent->priv;
+
+ mbm_config_show(seq, r, QOS_L3_MBM_LOCAL_EVENT_ID);
+
+ return 0;
+}
+
/* rdtgroup information files for one cache resource. */
static struct rftype res_common_files[] = {
{
@@ -1612,6 +1622,12 @@ static struct rftype res_common_files[] = {
.seq_show = mbm_total_bytes_config_show,
},
{
+ .name = "mbm_local_bytes_config",
+ .mode = 0444,
+ .kf_ops = &rdtgroup_kf_single_ops,
+ .seq_show = mbm_local_bytes_config_show,
+ },
+ {
.name = "cpus",
.mode = 0644,
.kf_ops = &rdtgroup_kf_single_ops,
The following commit has been merged into the x86/cache branch of tip:
Commit-ID: dc2a3e857981f859889933cf66ded117d74edff1
Gitweb: https://git.kernel.org/tip/dc2a3e857981f859889933cf66ded117d74edff1
Author: Babu Moger <[email protected]>
AuthorDate: Fri, 13 Jan 2023 09:20:35 -06:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 23 Jan 2023 17:40:24 +01:00
x86/resctrl: Add interface to read mbm_total_bytes_config
The event configuration can be viewed by the user by reading the
configuration file /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config. The
event configuration settings are domain specific and will affect all the CPUs in
the domain.
Following are the types of events supported:
==== ===========================================================
Bits Description
==== ===========================================================
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
==== ===========================================================
By default, the mbm_total_bytes_config is set to 0x7f to count all the
event types.
For example:
$cat /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
0=0x7f;1=0x7f;2=0x7f;3=0x7f
In this case, the event mbm_total_bytes is configured with 0x7f on
domains 0 to 3.
Signed-off-by: Babu Moger <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Reinette Chatre <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/include/asm/msr-index.h | 1 +-
arch/x86/kernel/cpu/resctrl/internal.h | 24 ++++++-
arch/x86/kernel/cpu/resctrl/monitor.c | 4 +-
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 102 ++++++++++++++++++++++++-
4 files changed, 130 insertions(+), 1 deletion(-)
diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h
index e0a4002..2e5f57c 100644
--- a/arch/x86/include/asm/msr-index.h
+++ b/arch/x86/include/asm/msr-index.h
@@ -1062,6 +1062,7 @@
/* - AMD: */
#define MSR_IA32_MBA_BW_BASE 0xc0000200
#define MSR_IA32_SMBA_BW_BASE 0xc0000280
+#define MSR_IA32_EVT_CFG_BASE 0xc0000400
/* MSR_IA32_VMX_MISC bits */
#define MSR_IA32_VMX_MISC_INTEL_PT (1ULL << 14)
diff --git a/arch/x86/kernel/cpu/resctrl/internal.h b/arch/x86/kernel/cpu/resctrl/internal.h
index 0605b04..8edecc5 100644
--- a/arch/x86/kernel/cpu/resctrl/internal.h
+++ b/arch/x86/kernel/cpu/resctrl/internal.h
@@ -30,6 +30,29 @@
*/
#define MBM_CNTR_WIDTH_OFFSET_MAX (62 - MBM_CNTR_WIDTH_BASE)
+/* Reads to Local DRAM Memory */
+#define READS_TO_LOCAL_MEM BIT(0)
+
+/* Reads to Remote DRAM Memory */
+#define READS_TO_REMOTE_MEM BIT(1)
+
+/* Non-Temporal Writes to Local Memory */
+#define NON_TEMP_WRITE_TO_LOCAL_MEM BIT(2)
+
+/* Non-Temporal Writes to Remote Memory */
+#define NON_TEMP_WRITE_TO_REMOTE_MEM BIT(3)
+
+/* Reads to Local Memory the system identifies as "Slow Memory" */
+#define READS_TO_LOCAL_S_MEM BIT(4)
+
+/* Reads to Remote Memory the system identifies as "Slow Memory" */
+#define READS_TO_REMOTE_S_MEM BIT(5)
+
+/* Dirty Victims to All Types of Memory */
+#define DIRTY_VICTIMS_TO_ALL_MEM BIT(6)
+
+/* Max event bits supported */
+#define MAX_EVT_CONFIG_BITS GENMASK(6, 0)
struct rdt_fs_context {
struct kernfs_fs_context kfc;
@@ -531,5 +554,6 @@ bool has_busy_rmid(struct rdt_resource *r, struct rdt_domain *d);
void __check_limbo(struct rdt_domain *d, bool force_free);
void rdt_domain_reconfigure_cdp(struct rdt_resource *r);
void __init thread_throttle_mode_init(void);
+void __init mbm_config_rftype_init(const char *config);
#endif /* _ASM_X86_RESCTRL_INTERNAL_H */
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index cf7b6e5..0ff45e3 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -801,8 +801,10 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r)
return ret;
if (rdt_cpu_has(X86_FEATURE_BMEC)) {
- if (rdt_cpu_has(X86_FEATURE_CQM_MBM_TOTAL))
+ if (rdt_cpu_has(X86_FEATURE_CQM_MBM_TOTAL)) {
mbm_total_event.configurable = true;
+ mbm_config_rftype_init("mbm_total_bytes_config");
+ }
if (rdt_cpu_has(X86_FEATURE_CQM_MBM_LOCAL))
mbm_local_event.configurable = true;
}
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index bf0bd43..cd4e668 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1420,6 +1420,93 @@ out:
return ret;
}
+struct mon_config_info {
+ u32 evtid;
+ u32 mon_config;
+};
+
+#define INVALID_CONFIG_INDEX UINT_MAX
+
+/**
+ * mon_event_config_index_get - get the hardware index for the
+ * configurable event
+ * @evtid: event id.
+ *
+ * Return: 0 for evtid == QOS_L3_MBM_TOTAL_EVENT_ID
+ * 1 for evtid == QOS_L3_MBM_LOCAL_EVENT_ID
+ * INVALID_CONFIG_INDEX for invalid evtid
+ */
+static inline unsigned int mon_event_config_index_get(u32 evtid)
+{
+ switch (evtid) {
+ case QOS_L3_MBM_TOTAL_EVENT_ID:
+ return 0;
+ case QOS_L3_MBM_LOCAL_EVENT_ID:
+ return 1;
+ default:
+ /* Should never reach here */
+ return INVALID_CONFIG_INDEX;
+ }
+}
+
+static void mon_event_config_read(void *info)
+{
+ struct mon_config_info *mon_info = info;
+ unsigned int index;
+ u32 h;
+
+ index = mon_event_config_index_get(mon_info->evtid);
+ if (index == INVALID_CONFIG_INDEX) {
+ pr_warn_once("Invalid event id %d\n", mon_info->evtid);
+ return;
+ }
+ rdmsr(MSR_IA32_EVT_CFG_BASE + index, mon_info->mon_config, h);
+
+ /* Report only the valid event configuration bits */
+ mon_info->mon_config &= MAX_EVT_CONFIG_BITS;
+}
+
+static void mondata_config_read(struct rdt_domain *d, struct mon_config_info *mon_info)
+{
+ smp_call_function_any(&d->cpu_mask, mon_event_config_read, mon_info, 1);
+}
+
+static int mbm_config_show(struct seq_file *s, struct rdt_resource *r, u32 evtid)
+{
+ struct mon_config_info mon_info = {0};
+ struct rdt_domain *dom;
+ bool sep = false;
+
+ mutex_lock(&rdtgroup_mutex);
+
+ list_for_each_entry(dom, &r->domains, list) {
+ if (sep)
+ seq_puts(s, ";");
+
+ memset(&mon_info, 0, sizeof(struct mon_config_info));
+ mon_info.evtid = evtid;
+ mondata_config_read(dom, &mon_info);
+
+ seq_printf(s, "%d=0x%02x", dom->id, mon_info.mon_config);
+ sep = true;
+ }
+ seq_puts(s, "\n");
+
+ mutex_unlock(&rdtgroup_mutex);
+
+ return 0;
+}
+
+static int mbm_total_bytes_config_show(struct kernfs_open_file *of,
+ struct seq_file *seq, void *v)
+{
+ struct rdt_resource *r = of->kn->parent->priv;
+
+ mbm_config_show(seq, r, QOS_L3_MBM_TOTAL_EVENT_ID);
+
+ return 0;
+}
+
/* rdtgroup information files for one cache resource. */
static struct rftype res_common_files[] = {
{
@@ -1519,6 +1606,12 @@ static struct rftype res_common_files[] = {
.fflags = RF_MON_INFO | RFTYPE_RES_CACHE,
},
{
+ .name = "mbm_total_bytes_config",
+ .mode = 0444,
+ .kf_ops = &rdtgroup_kf_single_ops,
+ .seq_show = mbm_total_bytes_config_show,
+ },
+ {
.name = "cpus",
.mode = 0644,
.kf_ops = &rdtgroup_kf_single_ops,
@@ -1624,6 +1717,15 @@ void __init thread_throttle_mode_init(void)
rft->fflags = RF_CTRL_INFO | RFTYPE_RES_MB;
}
+void __init mbm_config_rftype_init(const char *config)
+{
+ struct rftype *rft;
+
+ rft = rdtgroup_get_rftype_by_name(config);
+ if (rft)
+ rft->fflags = RF_MON_INFO | RFTYPE_RES_CACHE;
+}
+
/**
* rdtgroup_kn_mode_restrict - Restrict user access to named resctrl file
* @r: The resource group with which the file is associated.
The following commit has been merged into the x86/cache branch of tip:
Commit-ID: 4fe61bff5a4100e92f81427dab06b7f3a025f6a2
Gitweb: https://git.kernel.org/tip/4fe61bff5a4100e92f81427dab06b7f3a025f6a2
Author: Babu Moger <[email protected]>
AuthorDate: Fri, 13 Jan 2023 09:20:38 -06:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 23 Jan 2023 17:40:32 +01:00
x86/resctrl: Add interface to write mbm_local_bytes_config
The event configuration for mbm_local_bytes can be changed by the
user by writing to the configuration file
/sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config.
The event configuration settings are domain specific and will affect all
the CPUs in the domain.
Following are the types of events supported:
==== ===========================================================
Bits Description
==== ===========================================================
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
==== ===========================================================
For example, to change the mbm_local_bytes_config to count all the non-temporal
writes on domain 0, the bits 2 and 3 needs to be set which is 1100b (in hex
0xc).
Run the command:
$echo 0=0xc > /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
To change the mbm_local_bytes to count only reads to local NUMA domain 1,
the bit 0 needs to be set which 1b (in hex 0x1). Run the command:
$echo 1=0x1 > /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
Signed-off-by: Babu Moger <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Reinette Chatre <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 27 ++++++++++++++++++++++++-
1 file changed, 26 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 03284a6..5990589 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1638,6 +1638,30 @@ static ssize_t mbm_total_bytes_config_write(struct kernfs_open_file *of,
return ret ?: nbytes;
}
+static ssize_t mbm_local_bytes_config_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes,
+ loff_t off)
+{
+ struct rdt_resource *r = of->kn->parent->priv;
+ int ret;
+
+ /* Valid input requires a trailing newline */
+ if (nbytes == 0 || buf[nbytes - 1] != '\n')
+ return -EINVAL;
+
+ mutex_lock(&rdtgroup_mutex);
+
+ rdt_last_cmd_clear();
+
+ buf[nbytes - 1] = '\0';
+
+ ret = mon_config_write(r, buf, QOS_L3_MBM_LOCAL_EVENT_ID);
+
+ mutex_unlock(&rdtgroup_mutex);
+
+ return ret ?: nbytes;
+}
+
/* rdtgroup information files for one cache resource. */
static struct rftype res_common_files[] = {
{
@@ -1745,9 +1769,10 @@ static struct rftype res_common_files[] = {
},
{
.name = "mbm_local_bytes_config",
- .mode = 0444,
+ .mode = 0644,
.kf_ops = &rdtgroup_kf_single_ops,
.seq_show = mbm_local_bytes_config_show,
+ .write = mbm_local_bytes_config_write,
},
{
.name = "cpus",
The following commit has been merged into the x86/cache branch of tip:
Commit-ID: 0a363fb23ee2f7beb08437ad7db86d195878d79f
Gitweb: https://git.kernel.org/tip/0a363fb23ee2f7beb08437ad7db86d195878d79f
Author: Babu Moger <[email protected]>
AuthorDate: Fri, 13 Jan 2023 09:20:39 -06:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 23 Jan 2023 17:40:37 +01:00
Documentation/x86: Update resctrl.rst for new features
Update the documentation for the new features:
1. Slow Memory Bandwidth allocation (SMBA).
With this feature, the QOS enforcement policies can be applied to the
external slow memory connected to the host. QOS enforcement is accomplished
by assigning a Class Of Service (COS) to a processor and specifying
allocations or limits for that COS for each resource to be allocated.
2. Bandwidth Monitoring Event Configuration (BMEC).
The bandwidth monitoring events mbm_total_bytes and mbm_local_bytes are set
to count all the total and local reads/writes respectively. With the
introduction of slow memory, the two counters are not enough to count all the
different types of memory events. With the feature BMEC, the users have the
option to configure mbm_total_bytes and mbm_local_bytes to count the specific
type of events.
Also add configuration instructions with examples.
Signed-off-by: Babu Moger <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Reinette Chatre <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
Documentation/x86/resctrl.rst | 147 ++++++++++++++++++++++++++++++++-
1 file changed, 145 insertions(+), 2 deletions(-)
diff --git a/Documentation/x86/resctrl.rst b/Documentation/x86/resctrl.rst
index 71a5310..058257d 100644
--- a/Documentation/x86/resctrl.rst
+++ b/Documentation/x86/resctrl.rst
@@ -17,14 +17,21 @@ AMD refers to this feature as AMD Platform Quality of Service(AMD QoS).
This feature is enabled by the CONFIG_X86_CPU_RESCTRL and the x86 /proc/cpuinfo
flag bits:
-============================================= ================================
+=============================================== ================================
RDT (Resource Director Technology) Allocation "rdt_a"
CAT (Cache Allocation Technology) "cat_l3", "cat_l2"
CDP (Code and Data Prioritization) "cdp_l3", "cdp_l2"
CQM (Cache QoS Monitoring) "cqm_llc", "cqm_occup_llc"
MBM (Memory Bandwidth Monitoring) "cqm_mbm_total", "cqm_mbm_local"
MBA (Memory Bandwidth Allocation) "mba"
-============================================= ================================
+SMBA (Slow Memory Bandwidth Allocation) ""
+BMEC (Bandwidth Monitoring Event Configuration) ""
+=============================================== ================================
+
+Historically, new features were made visible by default in /proc/cpuinfo. This
+resulted in the feature flags becoming hard to parse by humans. Adding a new
+flag to /proc/cpuinfo should be avoided if user space can obtain information
+about the feature from resctrl's info directory.
To use the feature mount the file system::
@@ -161,6 +168,83 @@ with the following files:
"mon_features":
Lists the monitoring events if
monitoring is enabled for the resource.
+ Example::
+
+ # cat /sys/fs/resctrl/info/L3_MON/mon_features
+ llc_occupancy
+ mbm_total_bytes
+ mbm_local_bytes
+
+ If the system supports Bandwidth Monitoring Event
+ Configuration (BMEC), then the bandwidth events will
+ be configurable. The output will be::
+
+ # cat /sys/fs/resctrl/info/L3_MON/mon_features
+ llc_occupancy
+ mbm_total_bytes
+ mbm_total_bytes_config
+ mbm_local_bytes
+ mbm_local_bytes_config
+
+"mbm_total_bytes_config", "mbm_local_bytes_config":
+ Read/write files containing the configuration for the mbm_total_bytes
+ and mbm_local_bytes events, respectively, when the Bandwidth
+ Monitoring Event Configuration (BMEC) feature is supported.
+ The event configuration settings are domain specific and affect
+ all the CPUs in the domain. When either event configuration is
+ changed, the bandwidth counters for all RMIDs of both events
+ (mbm_total_bytes as well as mbm_local_bytes) are cleared for that
+ domain. The next read for every RMID will report "Unavailable"
+ and subsequent reads will report the valid value.
+
+ Following are the types of events supported:
+
+ ==== ========================================================
+ Bits Description
+ ==== ========================================================
+ 6 Dirty Victims from the QOS domain to all types of memory
+ 5 Reads to slow memory in the non-local NUMA domain
+ 4 Reads to slow memory in the local NUMA domain
+ 3 Non-temporal writes to non-local NUMA domain
+ 2 Non-temporal writes to local NUMA domain
+ 1 Reads to memory in the non-local NUMA domain
+ 0 Reads to memory in the local NUMA domain
+ ==== ========================================================
+
+ By default, the mbm_total_bytes configuration is set to 0x7f to count
+ all the event types and the mbm_local_bytes configuration is set to
+ 0x15 to count all the local memory events.
+
+ Examples:
+
+ * To view the current configuration::
+ ::
+
+ # cat /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
+ 0=0x7f;1=0x7f;2=0x7f;3=0x7f
+
+ # cat /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
+ 0=0x15;1=0x15;3=0x15;4=0x15
+
+ * To change the mbm_total_bytes to count only reads on domain 0,
+ the bits 0, 1, 4 and 5 needs to be set, which is 110011b in binary
+ (in hexadecimal 0x33):
+ ::
+
+ # echo "0=0x33" > /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
+
+ # cat /sys/fs/resctrl/info/L3_MON/mbm_total_bytes_config
+ 0=0x33;1=0x7f;2=0x7f;3=0x7f
+
+ * To change the mbm_local_bytes to count all the slow memory reads on
+ domain 0 and 1, the bits 4 and 5 needs to be set, which is 110000b
+ in binary (in hexadecimal 0x30):
+ ::
+
+ # echo "0=0x30;1=0x30" > /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
+
+ # cat /sys/fs/resctrl/info/L3_MON/mbm_local_bytes_config
+ 0=0x30;1=0x30;3=0x15;4=0x15
"max_threshold_occupancy":
Read/write file provides the largest value (in
@@ -464,6 +548,25 @@ Memory bandwidth domain is L3 cache.
MB:<cache_id0>=bw_MBps0;<cache_id1>=bw_MBps1;...
+Slow Memory Bandwidth Allocation (SMBA)
+---------------------------------------
+AMD hardware supports Slow Memory Bandwidth Allocation (SMBA).
+CXL.memory is the only supported "slow" memory device. With the
+support of SMBA, the hardware enables bandwidth allocation on
+the slow memory devices. If there are multiple such devices in
+the system, the throttling logic groups all the slow sources
+together and applies the limit on them as a whole.
+
+The presence of SMBA (with CXL.memory) is independent of slow memory
+devices presence. If there are no such devices on the system, then
+configuring SMBA will have no impact on the performance of the system.
+
+The bandwidth domain for slow memory is L3 cache. Its schemata file
+is formatted as:
+::
+
+ SMBA:<cache_id0>=bandwidth0;<cache_id1>=bandwidth1;...
+
Reading/writing the schemata file
---------------------------------
Reading the schemata file will show the state of all resources
@@ -479,6 +582,46 @@ which you wish to change. E.g.
L3DATA:0=fffff;1=fffff;2=3c0;3=fffff
L3CODE:0=fffff;1=fffff;2=fffff;3=fffff
+Reading/writing the schemata file (on AMD systems)
+--------------------------------------------------
+Reading the schemata file will show the current bandwidth limit on all
+domains. The allocated resources are in multiples of one eighth GB/s.
+When writing to the file, you need to specify what cache id you wish to
+configure the bandwidth limit.
+
+For example, to allocate 2GB/s limit on the first cache id:
+
+::
+
+ # cat schemata
+ MB:0=2048;1=2048;2=2048;3=2048
+ L3:0=ffff;1=ffff;2=ffff;3=ffff
+
+ # echo "MB:1=16" > schemata
+ # cat schemata
+ MB:0=2048;1= 16;2=2048;3=2048
+ L3:0=ffff;1=ffff;2=ffff;3=ffff
+
+Reading/writing the schemata file (on AMD systems) with SMBA feature
+--------------------------------------------------------------------
+Reading and writing the schemata file is the same as without SMBA in
+above section.
+
+For example, to allocate 8GB/s limit on the first cache id:
+
+::
+
+ # cat schemata
+ SMBA:0=2048;1=2048;2=2048;3=2048
+ MB:0=2048;1=2048;2=2048;3=2048
+ L3:0=ffff;1=ffff;2=ffff;3=ffff
+
+ # echo "SMBA:1=64" > schemata
+ # cat schemata
+ SMBA:0=2048;1= 64;2=2048;3=2048
+ MB:0=2048;1=2048;2=2048;3=2048
+ L3:0=ffff;1=ffff;2=ffff;3=ffff
+
Cache Pseudo-Locking
====================
CAT enables a user to specify the amount of cache space that an
The following commit has been merged into the x86/cache branch of tip:
Commit-ID: bd334c86b5d70e5d1c6169991802e62c828d6f38
Gitweb: https://git.kernel.org/tip/bd334c86b5d70e5d1c6169991802e62c828d6f38
Author: Babu Moger <[email protected]>
AuthorDate: Fri, 13 Jan 2023 09:20:33 -06:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 23 Jan 2023 17:40:11 +01:00
x86/resctrl: Add __init attribute to rdt_get_mon_l3_config()
In an upcoming change, rdt_get_mon_l3_config() needs to call rdt_cpu_has() to
query the monitor related features. It cannot be called right now because
rdt_cpu_has() has the __init attribute but rdt_get_mon_l3_config() doesn't.
Add the __init attribute to rdt_get_mon_l3_config() that is only called by
get_rdt_mon_resources() that already has the __init attribute. Also make
rdt_cpu_has() available to by rdt_get_mon_l3_config() via the internal header
file.
Signed-off-by: Babu Moger <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Reinette Chatre <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
---
arch/x86/kernel/cpu/resctrl/core.c | 2 +-
arch/x86/kernel/cpu/resctrl/internal.h | 1 +
arch/x86/kernel/cpu/resctrl/monitor.c | 2 +-
3 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/cpu/resctrl/core.c b/arch/x86/kernel/cpu/resctrl/core.c
index b4fc851..030d3b4 100644
--- a/arch/x86/kernel/cpu/resctrl/core.c
+++ b/arch/x86/kernel/cpu/resctrl/core.c
@@ -728,7 +728,7 @@ static int __init set_rdt_options(char *str)
}
__setup("rdt", set_rdt_options);
-static bool __init rdt_cpu_has(int flag)
+bool __init rdt_cpu_has(int flag)
{
bool ret = boot_cpu_has(flag);
struct rdt_options *o;
diff --git a/arch/x86/kernel/cpu/resctrl/internal.h b/arch/x86/kernel/cpu/resctrl/internal.h
index fdbbf66..f16b8bc 100644
--- a/arch/x86/kernel/cpu/resctrl/internal.h
+++ b/arch/x86/kernel/cpu/resctrl/internal.h
@@ -512,6 +512,7 @@ void closid_free(int closid);
int alloc_rmid(void);
void free_rmid(u32 rmid);
int rdt_get_mon_l3_config(struct rdt_resource *r);
+bool __init rdt_cpu_has(int flag);
void mon_event_count(void *info);
int rdtgroup_mondata_show(struct seq_file *m, void *arg);
void mon_event_read(struct rmid_read *rr, struct rdt_resource *r,
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index 77538ab..5665ab9 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -763,7 +763,7 @@ static void l3_mon_evt_init(struct rdt_resource *r)
list_add_tail(&mbm_local_event.list, &r->evt_list);
}
-int rdt_get_mon_l3_config(struct rdt_resource *r)
+int __init rdt_get_mon_l3_config(struct rdt_resource *r)
{
unsigned int mbm_offset = boot_cpu_data.x86_cache_mbm_width_offset;
struct rdt_hw_resource *hw_res = resctrl_to_arch_res(r);
The following commit has been merged into the x86/cache branch of tip:
Commit-ID: d507f83ced55eda848a3f397cf87b07ab11384b9
Gitweb: https://git.kernel.org/tip/d507f83ced55eda848a3f397cf87b07ab11384b9
Author: Babu Moger <[email protected]>
AuthorDate: Fri, 13 Jan 2023 09:20:34 -06:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 23 Jan 2023 17:40:21 +01:00
x86/resctrl: Support monitor configuration
Add a new field in struct mon_evt to support Bandwidth Monitoring Event
Configuration (BMEC) and also update the "mon_features" display.
The resctrl file "mon_features" will display the supported events
and files that can be used to configure those events if monitor
configuration is supported.
Before the change:
$ cat /sys/fs/resctrl/info/L3_MON/mon_features
llc_occupancy
mbm_total_bytes
mbm_local_bytes
After the change when BMEC is supported:
$ cat /sys/fs/resctrl/info/L3_MON/mon_features
llc_occupancy
mbm_total_bytes
mbm_total_bytes_config
mbm_local_bytes
mbm_local_bytes_config
Signed-off-by: Babu Moger <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Reinette Chatre <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/kernel/cpu/resctrl/internal.h | 2 ++
arch/x86/kernel/cpu/resctrl/monitor.c | 7 +++++++
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 5 ++++-
3 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/resctrl/internal.h b/arch/x86/kernel/cpu/resctrl/internal.h
index f16b8bc..0605b04 100644
--- a/arch/x86/kernel/cpu/resctrl/internal.h
+++ b/arch/x86/kernel/cpu/resctrl/internal.h
@@ -52,11 +52,13 @@ DECLARE_STATIC_KEY_FALSE(rdt_mon_enable_key);
* struct mon_evt - Entry in the event list of a resource
* @evtid: event id
* @name: name of the event
+ * @configurable: true if the event is configurable
* @list: entry in &rdt_resource->evt_list
*/
struct mon_evt {
enum resctrl_event_id evtid;
char *name;
+ bool configurable;
struct list_head list;
};
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index 5665ab9..cf7b6e5 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -800,6 +800,13 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r)
if (ret)
return ret;
+ if (rdt_cpu_has(X86_FEATURE_BMEC)) {
+ if (rdt_cpu_has(X86_FEATURE_CQM_MBM_TOTAL))
+ mbm_total_event.configurable = true;
+ if (rdt_cpu_has(X86_FEATURE_CQM_MBM_LOCAL))
+ mbm_local_event.configurable = true;
+ }
+
l3_mon_evt_init(r);
r->mon_capable = true;
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 7516968..bf0bd43 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -998,8 +998,11 @@ static int rdt_mon_features_show(struct kernfs_open_file *of,
struct rdt_resource *r = of->kn->parent->priv;
struct mon_evt *mevt;
- list_for_each_entry(mevt, &r->evt_list, list)
+ list_for_each_entry(mevt, &r->evt_list, list) {
seq_printf(seq, "%s\n", mevt->name);
+ if (mevt->configurable)
+ seq_printf(seq, "%s_config\n", mevt->name);
+ }
return 0;
}
The following commit has been merged into the x86/cache branch of tip:
Commit-ID: 5b6fac3fa44bafee12e0c3d1c5cbae6d058e9c98
Gitweb: https://git.kernel.org/tip/5b6fac3fa44bafee12e0c3d1c5cbae6d058e9c98
Author: Babu Moger <[email protected]>
AuthorDate: Fri, 13 Jan 2023 09:20:32 -06:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 23 Jan 2023 17:38:44 +01:00
x86/resctrl: Detect and configure Slow Memory Bandwidth Allocation
The QoS slow memory configuration details are available via
CPUID_Fn80000020_EDX_x02. Detect the available details and
initialize the rest to defaults.
Signed-off-by: Babu Moger <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Reinette Chatre <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/include/asm/msr-index.h | 1 +-
arch/x86/kernel/cpu/resctrl/core.c | 36 ++++++++++++++++++++--
arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 2 +-
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 8 +++--
4 files changed, 41 insertions(+), 6 deletions(-)
diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h
index 37ff475..e0a4002 100644
--- a/arch/x86/include/asm/msr-index.h
+++ b/arch/x86/include/asm/msr-index.h
@@ -1061,6 +1061,7 @@
/* - AMD: */
#define MSR_IA32_MBA_BW_BASE 0xc0000200
+#define MSR_IA32_SMBA_BW_BASE 0xc0000280
/* MSR_IA32_VMX_MISC bits */
#define MSR_IA32_VMX_MISC_INTEL_PT (1ULL << 14)
diff --git a/arch/x86/kernel/cpu/resctrl/core.c b/arch/x86/kernel/cpu/resctrl/core.c
index 10a8c9d..b4fc851 100644
--- a/arch/x86/kernel/cpu/resctrl/core.c
+++ b/arch/x86/kernel/cpu/resctrl/core.c
@@ -162,6 +162,13 @@ bool is_mba_sc(struct rdt_resource *r)
if (!r)
return rdt_resources_all[RDT_RESOURCE_MBA].r_resctrl.membw.mba_sc;
+ /*
+ * The software controller support is only applicable to MBA resource.
+ * Make sure to check for resource type.
+ */
+ if (r->rid != RDT_RESOURCE_MBA)
+ return false;
+
return r->membw.mba_sc;
}
@@ -225,9 +232,15 @@ static bool __rdt_get_mem_config_amd(struct rdt_resource *r)
struct rdt_hw_resource *hw_res = resctrl_to_arch_res(r);
union cpuid_0x10_3_eax eax;
union cpuid_0x10_x_edx edx;
- u32 ebx, ecx;
+ u32 ebx, ecx, subleaf;
- cpuid_count(0x80000020, 1, &eax.full, &ebx, &ecx, &edx.full);
+ /*
+ * Query CPUID_Fn80000020_EDX_x01 for MBA and
+ * CPUID_Fn80000020_EDX_x02 for SMBA
+ */
+ subleaf = (r->rid == RDT_RESOURCE_SMBA) ? 2 : 1;
+
+ cpuid_count(0x80000020, subleaf, &eax.full, &ebx, &ecx, &edx.full);
hw_res->num_closid = edx.split.cos_max + 1;
r->default_ctrl = MAX_MBA_BW_AMD;
@@ -750,6 +763,19 @@ static __init bool get_mem_config(void)
return false;
}
+static __init bool get_slow_mem_config(void)
+{
+ struct rdt_hw_resource *hw_res = &rdt_resources_all[RDT_RESOURCE_SMBA];
+
+ if (!rdt_cpu_has(X86_FEATURE_SMBA))
+ return false;
+
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
+ return __rdt_get_mem_config_amd(&hw_res->r_resctrl);
+
+ return false;
+}
+
static __init bool get_rdt_alloc_resources(void)
{
struct rdt_resource *r;
@@ -780,6 +806,9 @@ static __init bool get_rdt_alloc_resources(void)
if (get_mem_config())
ret = true;
+ if (get_slow_mem_config())
+ ret = true;
+
return ret;
}
@@ -869,6 +898,9 @@ static __init void rdt_init_res_defs_amd(void)
} else if (r->rid == RDT_RESOURCE_MBA) {
hw_res->msr_base = MSR_IA32_MBA_BW_BASE;
hw_res->msr_update = mba_wrmsr_amd;
+ } else if (r->rid == RDT_RESOURCE_SMBA) {
+ hw_res->msr_base = MSR_IA32_SMBA_BW_BASE;
+ hw_res->msr_update = mba_wrmsr_amd;
}
}
}
diff --git a/arch/x86/kernel/cpu/resctrl/ctrlmondata.c b/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
index 7eece3d..eb07d44 100644
--- a/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
+++ b/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
@@ -209,7 +209,7 @@ static int parse_line(char *line, struct resctrl_schema *s,
unsigned long dom_id;
if (rdtgrp->mode == RDT_MODE_PSEUDO_LOCKSETUP &&
- r->rid == RDT_RESOURCE_MBA) {
+ (r->rid == RDT_RESOURCE_MBA || r->rid == RDT_RESOURCE_SMBA)) {
rdt_last_cmd_puts("Cannot pseudo-lock MBA resource\n");
return -EINVAL;
}
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index d23fbc4..7516968 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1210,7 +1210,7 @@ static bool rdtgroup_mode_test_exclusive(struct rdtgroup *rdtgrp)
list_for_each_entry(s, &resctrl_schema_all, list) {
r = s->res;
- if (r->rid == RDT_RESOURCE_MBA)
+ if (r->rid == RDT_RESOURCE_MBA || r->rid == RDT_RESOURCE_SMBA)
continue;
has_cache = true;
list_for_each_entry(d, &r->domains, list) {
@@ -1399,7 +1399,8 @@ static int rdtgroup_size_show(struct kernfs_open_file *of,
ctrl = resctrl_arch_get_config(r, d,
closid,
type);
- if (r->rid == RDT_RESOURCE_MBA)
+ if (r->rid == RDT_RESOURCE_MBA ||
+ r->rid == RDT_RESOURCE_SMBA)
size = ctrl;
else
size = rdtgroup_cbm_to_size(r, d, ctrl);
@@ -2842,7 +2843,8 @@ static int rdtgroup_init_alloc(struct rdtgroup *rdtgrp)
list_for_each_entry(s, &resctrl_schema_all, list) {
r = s->res;
- if (r->rid == RDT_RESOURCE_MBA) {
+ if (r->rid == RDT_RESOURCE_MBA ||
+ r->rid == RDT_RESOURCE_SMBA) {
rdtgroup_init_mba(r, rdtgrp->closid);
if (is_mba_sc(r))
continue;
The following commit has been merged into the x86/cache branch of tip:
Commit-ID: fc3b618c87833e4d258b66acfca3557c49c4fe97
Gitweb: https://git.kernel.org/tip/fc3b618c87833e4d258b66acfca3557c49c4fe97
Author: Babu Moger <[email protected]>
AuthorDate: Fri, 13 Jan 2023 09:20:27 -06:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 23 Jan 2023 17:38:04 +01:00
x86/resctrl: Replace smp_call_function_many() with on_each_cpu_mask()
on_each_cpu_mask() runs the function on each CPU specified by cpumask,
which may include the local processor.
Replace smp_call_function_many() with on_each_cpu_mask() to simplify
the code.
Signed-off-by: Babu Moger <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Reinette Chatre <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 11 ++------
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 29 ++++++----------------
2 files changed, 11 insertions(+), 29 deletions(-)
diff --git a/arch/x86/kernel/cpu/resctrl/ctrlmondata.c b/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
index 1df0e32..7eece3d 100644
--- a/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
+++ b/arch/x86/kernel/cpu/resctrl/ctrlmondata.c
@@ -310,7 +310,6 @@ int resctrl_arch_update_domains(struct rdt_resource *r, u32 closid)
enum resctrl_conf_type t;
cpumask_var_t cpu_mask;
struct rdt_domain *d;
- int cpu;
u32 idx;
if (!zalloc_cpumask_var(&cpu_mask, GFP_KERNEL))
@@ -341,13 +340,9 @@ int resctrl_arch_update_domains(struct rdt_resource *r, u32 closid)
if (cpumask_empty(cpu_mask))
goto done;
- cpu = get_cpu();
- /* Update resource control msr on this CPU if it's in cpu_mask. */
- if (cpumask_test_cpu(cpu, cpu_mask))
- rdt_ctrl_update(&msr_param);
- /* Update resource control msr on other CPUs. */
- smp_call_function_many(cpu_mask, rdt_ctrl_update, &msr_param, 1);
- put_cpu();
+
+ /* Update resource control msr on all the CPUs. */
+ on_each_cpu_mask(cpu_mask, rdt_ctrl_update, &msr_param, 1);
done:
free_cpumask_var(cpu_mask);
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 5993da2..d23fbc4 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -325,12 +325,7 @@ static void update_cpu_closid_rmid(void *info)
static void
update_closid_rmid(const struct cpumask *cpu_mask, struct rdtgroup *r)
{
- int cpu = get_cpu();
-
- if (cpumask_test_cpu(cpu, cpu_mask))
- update_cpu_closid_rmid(r);
- smp_call_function_many(cpu_mask, update_cpu_closid_rmid, r, 1);
- put_cpu();
+ on_each_cpu_mask(cpu_mask, update_cpu_closid_rmid, r, 1);
}
static int cpus_mon_write(struct rdtgroup *rdtgrp, cpumask_var_t newmask,
@@ -1866,13 +1861,9 @@ static int set_cache_qos_cfg(int level, bool enable)
/* Pick one CPU from each domain instance to update MSR */
cpumask_set_cpu(cpumask_any(&d->cpu_mask), cpu_mask);
}
- cpu = get_cpu();
- /* Update QOS_CFG MSR on this cpu if it's in cpu_mask. */
- if (cpumask_test_cpu(cpu, cpu_mask))
- update(&enable);
- /* Update QOS_CFG MSR on all other cpus in cpu_mask. */
- smp_call_function_many(cpu_mask, update, &enable, 1);
- put_cpu();
+
+ /* Update QOS_CFG MSR on all the CPUs in cpu_mask */
+ on_each_cpu_mask(cpu_mask, update, &enable, 1);
free_cpumask_var(cpu_mask);
@@ -2349,7 +2340,7 @@ static int reset_all_ctrls(struct rdt_resource *r)
struct msr_param msr_param;
cpumask_var_t cpu_mask;
struct rdt_domain *d;
- int i, cpu;
+ int i;
if (!zalloc_cpumask_var(&cpu_mask, GFP_KERNEL))
return -ENOMEM;
@@ -2370,13 +2361,9 @@ static int reset_all_ctrls(struct rdt_resource *r)
for (i = 0; i < hw_res->num_closid; i++)
hw_dom->ctrl_val[i] = r->default_ctrl;
}
- cpu = get_cpu();
- /* Update CBM on this cpu if it's in cpu_mask. */
- if (cpumask_test_cpu(cpu, cpu_mask))
- rdt_ctrl_update(&msr_param);
- /* Update CBM on all other cpus in cpu_mask. */
- smp_call_function_many(cpu_mask, rdt_ctrl_update, &msr_param, 1);
- put_cpu();
+
+ /* Update CBM on all the CPUs in cpu_mask */
+ on_each_cpu_mask(cpu_mask, rdt_ctrl_update, &msr_param, 1);
free_cpumask_var(cpu_mask);
The following commit has been merged into the x86/cache branch of tip:
Commit-ID: 78335aac6156eadad0025ab34469e2adcc60218b
Gitweb: https://git.kernel.org/tip/78335aac6156eadad0025ab34469e2adcc60218b
Author: Babu Moger <[email protected]>
AuthorDate: Fri, 13 Jan 2023 09:20:30 -06:00
Committer: Borislav Petkov (AMD) <[email protected]>
CommitterDate: Mon, 23 Jan 2023 17:38:31 +01:00
x86/cpufeatures: Add Bandwidth Monitoring Event Configuration feature flag
Newer AMD processors support the new feature Bandwidth Monitoring Event
Configuration (BMEC).
The feature support is identified via CPUID Fn8000_0020_EBX_x0[3]: EVT_CFG -
Bandwidth Monitoring Event Configuration (BMEC)
The bandwidth monitoring events mbm_total_bytes and mbm_local_bytes are set to
count all the total and local reads/writes, respectively. With the introduction
of slow memory, the two counters are not enough to count all the different types
of memory events. Therefore, BMEC provides the option to configure
mbm_total_bytes and mbm_local_bytes to count the specific type of events.
Each BMEC event has a configuration MSR which contains one field for each
bandwidth type that can be used to configure the bandwidth event to track any
combination of supported bandwidth types. The event will count requests from
every bandwidth type bit that is set in the corresponding configuration
register.
Following are the types of events supported:
==== ========================================================
Bits Description
==== ========================================================
6 Dirty Victims from the QOS domain to all types of memory
5 Reads to slow memory in the non-local NUMA domain
4 Reads to slow memory in the local NUMA domain
3 Non-temporal writes to non-local NUMA domain
2 Non-temporal writes to local NUMA domain
1 Reads to memory in the non-local NUMA domain
0 Reads to memory in the local NUMA domain
==== ========================================================
By default, the mbm_total_bytes configuration is set to 0x7F to count
all the event types and the mbm_local_bytes configuration is set to 0x15 to
count all the local memory events.
Feature description is available in the specification, "AMD64 Technology
Platform Quality of Service Extensions, Revision: 1.03 Publication" at
https://bugzilla.kernel.org/attachment.cgi?id=301365
Signed-off-by: Babu Moger <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Reinette Chatre <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/include/asm/cpufeatures.h | 1 +
arch/x86/kernel/cpu/cpuid-deps.c | 2 ++
arch/x86/kernel/cpu/scattered.c | 1 +
3 files changed, 4 insertions(+)
diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
index 3f5f64b..f6a41ec 100644
--- a/arch/x86/include/asm/cpufeatures.h
+++ b/arch/x86/include/asm/cpufeatures.h
@@ -308,6 +308,7 @@
#define X86_FEATURE_CALL_DEPTH (11*32+19) /* "" Call depth tracking for RSB stuffing */
#define X86_FEATURE_MSR_TSX_CTRL (11*32+20) /* "" MSR IA32_TSX_CTRL (Intel) implemented */
#define X86_FEATURE_SMBA (11*32+21) /* "" Slow Memory Bandwidth Allocation */
+#define X86_FEATURE_BMEC (11*32+22) /* "" Bandwidth Monitoring Event Configuration */
/* Intel-defined CPU features, CPUID level 0x00000007:1 (EAX), word 12 */
#define X86_FEATURE_AVX_VNNI (12*32+ 4) /* AVX VNNI instructions */
diff --git a/arch/x86/kernel/cpu/cpuid-deps.c b/arch/x86/kernel/cpu/cpuid-deps.c
index d952211..f6748c8 100644
--- a/arch/x86/kernel/cpu/cpuid-deps.c
+++ b/arch/x86/kernel/cpu/cpuid-deps.c
@@ -68,6 +68,8 @@ static const struct cpuid_dep cpuid_deps[] = {
{ X86_FEATURE_CQM_OCCUP_LLC, X86_FEATURE_CQM_LLC },
{ X86_FEATURE_CQM_MBM_TOTAL, X86_FEATURE_CQM_LLC },
{ X86_FEATURE_CQM_MBM_LOCAL, X86_FEATURE_CQM_LLC },
+ { X86_FEATURE_BMEC, X86_FEATURE_CQM_MBM_TOTAL },
+ { X86_FEATURE_BMEC, X86_FEATURE_CQM_MBM_LOCAL },
{ X86_FEATURE_AVX512_BF16, X86_FEATURE_AVX512VL },
{ X86_FEATURE_AVX512_FP16, X86_FEATURE_AVX512BW },
{ X86_FEATURE_ENQCMD, X86_FEATURE_XSAVES },
diff --git a/arch/x86/kernel/cpu/scattered.c b/arch/x86/kernel/cpu/scattered.c
index d925753..0dad49a 100644
--- a/arch/x86/kernel/cpu/scattered.c
+++ b/arch/x86/kernel/cpu/scattered.c
@@ -46,6 +46,7 @@ static const struct cpuid_bit cpuid_bits[] = {
{ X86_FEATURE_PROC_FEEDBACK, CPUID_EDX, 11, 0x80000007, 0 },
{ X86_FEATURE_MBA, CPUID_EBX, 6, 0x80000008, 0 },
{ X86_FEATURE_SMBA, CPUID_EBX, 2, 0x80000020, 0 },
+ { X86_FEATURE_BMEC, CPUID_EBX, 3, 0x80000020, 0 },
{ X86_FEATURE_PERFMON_V2, CPUID_EAX, 0, 0x80000022, 0 },
{ X86_FEATURE_AMD_LBR_V2, CPUID_EAX, 1, 0x80000022, 0 },
{ 0, 0, 0, 0, 0 }