arch/resctrl: AMD QoS support
Subsystems (5)
Moger, Babu
reviewerv9 cover arch/resctrl: AMD QoS support
This series adds support for AMD64 architectural extensions for Platform Quality of Service. These extensions are intended to provide for the monitoring of the usage of certain system resources by one or more processors and for the separate allocation and enforcement of limits on the use of certain system resources by one or more processors.
The monitoring and enforcement are not necessarily applied across the entire system, but in general apply to a QOS domain which corresponds to some shared system resource. The set of resources which are monitored and the set for which the enforcement of limits is provided are implementation dependent. Platform QOS features are implemented on a logical processor basis. Therefore, multiple hardware threads of a single physical CPU core may have independent resource monitoring and enforcement configurations.
AMD's next generation of processors support following QoS sub-features.
- L3 Cache allocation enforcement
- L3 Cache occupancy monitoring
- L3 Code-Data Prioritization support
- Memory Bandwidth Enforcement(Allocation)
The public specification for this feature is available at
https://developer.amd.com/wp-content/resources/56375.pdf
Obviously, there are multiple ways we can go about these changes. We felt it is appropriate to rename and re-organize the code little bit before making the functional changes. The first few patches(1-10) renames and re-organizes the sources in preparation. Rest of the patches(7-11) adds support for AMD QoS features.
Please review.
Changes from v8 -> v9:
https://lore.kernel.org/lkml/[email protected]/
a. Addressed numerous comments from Borislav Petkov.
1. Moved all the resctrl code to new directory under arch/x86/kernel/cpu/resctrl. Updated makefies(patch #1)
2. Little bit rework and rename of RDT initialization code(patch #3)
3. Merged the patch 5 and 9.
4. Few corrections and code indentation work on patch 11.
5. Some more text corrections.
b. Addressed few text comments from Randy Dunlap.
Changes from v7 -> v8:
https://lore.kernel.org/lkml/[email protected]/
a. Addressed comments from Borislav Petkov.
Changed the subject line to arch/resctrl from arch/x86. Few code cleanup related changes on patch 2 and 3.
Changes from v6 -> v7:
a. Mainly addressed the comments from Mattew Wilcox. Removed both
CONFIG_INTEL_RDT and CONFIG_AMD_QOS config parameter. We dont need to use two separate config parameter to enable this feature. The final code will be same in both the cases. So, used the new vendor neutral parameter CONFIG_RESCTRL to enable resource control feature for both Intel and AMD
Changes from v5 -> v6:
https://lore.kernel.org/lkml/[email protected]/
a. Addressed comments from Fenghua Yu. Added vendor check while
detecting MBA software controller support.
a. Rebased again on top of
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86/cache
Changes from v4 -> v5:
https://lore.kernel.org/lkml/[email protected]/
a. Addressed comments from Fenghua Yu.
b. The functions update_mba_bw and set_mba_sc is not required for AMD.
Removed all the changes related to these functions.
Changes from v3 -> v4:
https://lore.kernel.org/lkml/[email protected]/
a. Addressed comments from Reinette Chatre and Borislav Petkov. b. Removed X86 dependency for CONFIG_AMD_QOS. Implicitly is it already
dependent on X86.
c. Updated the MAINTAINER file for name changes.
d. Addressed most of "checkpatch.pl --strict" issues.
d. Updated Documentation/x86/resctrl_ui.txt(previously
intel_rdt_ui.txt) file with AMD specific details. Changed few names to resctrl from intel_rdt.
Changes from v2 -> v3:
https://lore.kernel.org/lkml/[email protected]/
a. Rebased the patches on top of below branch as suggested by Thomas Gleixner.
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86/cache
b. Addressed comments from Reinette Chatre, Fenghua Yu and Borislav Petkov. c. Main changes are related to renaming the files and functions.
Renamed from intel_rdt to more generic resctrl(patches 1 to 3).
d. Config parameter changed from PLATFORM_QOS to more generic RESCTRL. e. Fixed minor indentation issues.
Changes from v1 -> v2:
https://lore.kernel.org/lkml/[email protected]/
a. Removed RFC from subject header. Based on the discussion so far,
plan is to go ahead with these patches and eventually re-structure the code to make arch and non-arch separate.
b. Addressed comments from Reinette Chatre and Fenghua Yu. c. Separated quirks and MBA from rdt init code. Kept the rest of the
code as is.
d. Added _intel suffixes all the Intel only code just like AMD code. e. Added one more patch to bring the macros into header file. f. Few minor text changes.
v1:
https://lore.kernel.org/lkml/[email protected]/
Babu Moger (12):
x86/resctrl: Rename and move rdt files to new directory x86/resctrl: Rename the RDT functions and definitions x86/resctrl: Re-arrange RDT init code
x86/resctrl: Bring all the macros to resctrl/internal.h x86/resctrl: Initialize the resource functions that are different x86/resctrl: Bring cbm_validate function into the resource structure x86/resctrl: Add vendor check for MBA software controller x86/resctrl: Rename config parameter INTEL_RDT to RESCTRL x86/resctrl: Fix the messages in rdt_last_cmd_printf and
rdt_last_cmd_puts
x86/resctrl: Introduce QOS feature for AMD
Documentation: Rename and update intel_rdt_ui.txt to resctrl_ui.txt
MAINTAINERS: Update the file and documentation names in arch/x86
Sherry Hurwitz (1):
x86/resctrl: Add AMD feature bit X86_FEATURE_MBA in cpuid bits array
.../x86/{intel_rdt_ui.txt => resctrl_ui.txt} | 9 +-
MAINTAINERS | 6 +-
arch/x86/Kconfig | 23 ++-
.../{intel_rdt_sched.h => resctrl_sched.h} | 28 +--
arch/x86/kernel/cpu/Makefile | 5 +-
arch/x86/kernel/cpu/resctrl/Makefile | 7 +
.../cpu/{intel_rdt.c => resctrl/core.c} | 184 ++++++++++++++----
.../ctrlmondata.c} | 102 ++++++++--
.../cpu/{intel_rdt.h => resctrl/internal.h} | 47 +++--
.../monitor.c} | 16 +-
.../pseudo_lock.c} | 40 ++--
.../pseudo_lock_event.h} | 2 +-
.../rdtgroup.c} | 59 +++---
arch/x86/kernel/cpu/scattered.c | 7 +-
arch/x86/kernel/process_32.c | 4 +-
arch/x86/kernel/process_64.c | 4 +-
include/linux/sched.h | 2 +-
17 files changed, 380 insertions(+), 165 deletions(-)
rename Documentation/x86/{intel_rdt_ui.txt => resctrl_ui.txt} (99%)
rename arch/x86/include/asm/{intel_rdt_sched.h => resctrl_sched.h} (77%)
create mode 100644 arch/x86/kernel/cpu/resctrl/Makefile
rename arch/x86/kernel/cpu/{intel_rdt.c => resctrl/core.c} (83%)
rename arch/x86/kernel/cpu/{intel_rdt_ctrlmondata.c => resctrl/ctrlmondata.c} (81%)
rename arch/x86/kernel/cpu/{intel_rdt.h => resctrl/internal.h} (92%)
rename arch/x86/kernel/cpu/{intel_rdt_monitor.c => resctrl/monitor.c} (98%)
rename arch/x86/kernel/cpu/{intel_rdt_pseudo_lock.c => resctrl/pseudo_lock.c} (97%)
rename arch/x86/kernel/cpu/{intel_rdt_pseudo_lock_event.h => resctrl/pseudo_lock_event.h} (95%)
rename arch/x86/kernel/cpu/{intel_rdt_rdtgroup.c => resctrl/rdtgroup.c} (98%)Signature · 1 line
2.17.1
Moger, Babu
reviewerPATCH v9 1/13 x86/resctrl: Rename and move rdt files to new directory
New generation of AMD processors start supporting RDT(or QOS) features. Together these features will be called as RESCTRL. With more than one vendors supporting these features, it seems more appropriate to rename these files.
Create a new directory with the name 'resctrl' and move all the intel_rdt files to the new directory. This way all the resctrl related code resides inside one directory.
---
.../x86/include/asm/{intel_rdt_sched.h => resctrl_sched.h} | 0
arch/x86/kernel/cpu/Makefile | 5 +----
arch/x86/kernel/cpu/resctrl/Makefile | 7 +++++++
arch/x86/kernel/cpu/{intel_rdt.c => resctrl/core.c} | 4 ++--
.../cpu/{intel_rdt_ctrlmondata.c => resctrl/ctrlmondata.c} | 2 +-
arch/x86/kernel/cpu/{intel_rdt.h => resctrl/internal.h} | 6 +++---
.../kernel/cpu/{intel_rdt_monitor.c => resctrl/monitor.c} | 2 +-
.../cpu/{intel_rdt_pseudo_lock.c => resctrl/pseudo_lock.c} | 6 +++---
.../pseudo_lock_event.h} | 2 +-
.../cpu/{intel_rdt_rdtgroup.c => resctrl/rdtgroup.c} | 4 ++--
arch/x86/kernel/process_32.c | 2 +-
arch/x86/kernel/process_64.c | 2 +-
12 files changed, 23 insertions(+), 19 deletions(-)
rename arch/x86/include/asm/{intel_rdt_sched.h => resctrl_sched.h} (100%)
create mode 100644 arch/x86/kernel/cpu/resctrl/Makefile
rename arch/x86/kernel/cpu/{intel_rdt.c => resctrl/core.c} (99%)
rename arch/x86/kernel/cpu/{intel_rdt_ctrlmondata.c => resctrl/ctrlmondata.c} (99%)
rename arch/x86/kernel/cpu/{intel_rdt.h => resctrl/internal.h} (99%)
rename arch/x86/kernel/cpu/{intel_rdt_monitor.c => resctrl/monitor.c} (99%)
rename arch/x86/kernel/cpu/{intel_rdt_pseudo_lock.c => resctrl/pseudo_lock.c} (99%)
rename arch/x86/kernel/cpu/{intel_rdt_pseudo_lock_event.h => resctrl/pseudo_lock_event.h} (95%)
rename arch/x86/kernel/cpu/{intel_rdt_rdtgroup.c => resctrl/rdtgroup.c} (99%)arch/x86/include/asm/resctrl_sched.h renamed from arch/x86/include/asm/intel_rdt_sched.h +0 -0
arch/x86/kernel/cpu/Makefile +1 -4
@@ -35,13 +35,10 @@ obj-$(CONFIG_CPU_SUP_CENTAUR) += centaur.o obj-$(CONFIG_CPU_SUP_TRANSMETA_32) += transmeta.o obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o -obj-$(CONFIG_INTEL_RDT) += intel_rdt.o intel_rdt_rdtgroup.o intel_rdt_monitor.o -obj-$(CONFIG_INTEL_RDT) += intel_rdt_ctrlmondata.o intel_rdt_pseudo_lock.o -CFLAGS_intel_rdt_pseudo_lock.o = -I$(src) - obj-$(CONFIG_X86_MCE) += mcheck/ obj-$(CONFIG_MTRR) += mtrr/ obj-$(CONFIG_MICROCODE) += microcode/ +obj-$(CONFIG_INTEL_RDT) += resctrl/ obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o
arch/x86/kernel/cpu/resctrl/Makefile new file +7 -0
@@ -0,0 +1,7 @@ +# +# Makefile for resource control feature code +# + +obj-$(CONFIG_INTEL_RDT) += core.o rdtgroup.o monitor.o +obj-$(CONFIG_INTEL_RDT) += ctrlmondata.o pseudo_lock.o +CFLAGS_pseudo_lock.o = -I$(src)
arch/x86/kernel/cpu/resctrl/core.c renamed from arch/x86/kernel/cpu/intel_rdt.c +2 -2
@@ -30,8 +30,8 @@#include <linux/cpuhotplug.h> #include <asm/intel-family.h> -#include <asm/intel_rdt_sched.h> -#include "intel_rdt.h" +#include <asm/resctrl_sched.h> +#include "internal.h" #define MBA_IS_LINEAR 0x4#define MBA_MAX_MBPS U32_MAX
arch/x86/kernel/cpu/resctrl/ctrlmondata.c renamed from arch/x86/kernel/cpu/intel_rdt_ctrlmondata.c +1 -1
@@ -26,7 +26,7 @@#include <linux/kernfs.h> #include <linux/seq_file.h> #include <linux/slab.h> -#include "intel_rdt.h" +#include "internal.h" /* * Check whether MBA bandwidth percentage value is correct. The value is
arch/x86/kernel/cpu/resctrl/internal.h renamed from arch/x86/kernel/cpu/intel_rdt.h +3 -3
@@ -1,6 +1,6 @@/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _ASM_X86_INTEL_RDT_H -#define _ASM_X86_INTEL_RDT_H +#ifndef _ASM_X86_RESCTRL_INTERNAL_H +#define _ASM_X86_RESCTRL_INTERNAL_H#include <linux/sched.h> #include <linux/kernfs.h>
@@ -568,4 +568,4 @@ void cqm_handle_limbo(struct work_struct *work);bool has_busy_rmid (struct rdt_resource *r,struct rdt_domain *d);void __check_limbo (struct rdt_domain *d,bool force_free); -#endif /* _ASM_X86_INTEL_RDT_H */ +#endif /* _ASM_X86_RESCTRL_INTERNAL_H */
arch/x86/kernel/cpu/resctrl/monitor.c renamed from arch/x86/kernel/cpu/intel_rdt_monitor.c +1 -1
@@ -26,7 +26,7 @@#include <linux/module.h> #include <linux/slab.h> #include <asm/cpu_device_id.h> -#include "intel_rdt.h" +#include "internal.h" #define MSR_IA32_QM_CTR 0x0c8e#define MSR_IA32_QM_EVTSEL 0x0c8d
arch/x86/kernel/cpu/resctrl/pseudo_lock.c renamed from arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c +3 -3
@@ -24,14 +24,14 @@#include <asm/cacheflush.h> #include <asm/intel-family.h> -#include <asm/intel_rdt_sched.h> +#include <asm/resctrl_sched.h> #include <asm/perf_event.h> #include "../../events/perf_event.h" /* For X86_CONFIG() */ -#include "intel_rdt.h" +#include "internal.h" #define CREATE_TRACE_POINTS -#include "intel_rdt_pseudo_lock_event.h" +#include "pseudo_lock_event.h" /* * MSR_MISC_FEATURE_CONTROL register enables the modification of hardware
arch/x86/kernel/cpu/resctrl/pseudo_lock_event.h renamed from arch/x86/kernel/cpu/intel_rdt_pseudo_lock_event.h +1 -1
@@ -39,5 +39,5 @@ TRACE_EVENT(pseudo_lock_l3,#undef TRACE_INCLUDE_PATH#define TRACE_INCLUDE_PATH . -#define TRACE_INCLUDE_FILE intel_rdt_pseudo_lock_event +#define TRACE_INCLUDE_FILE pseudo_lock_event#include <trace/define_trace.h>
arch/x86/kernel/cpu/resctrl/rdtgroup.c renamed from arch/x86/kernel/cpu/intel_rdt_rdtgroup.c +2 -2
@@ -35,8 +35,8 @@#include <uapi/linux/magic.h> -#include <asm/intel_rdt_sched.h> -#include "intel_rdt.h" +#include <asm/resctrl_sched.h> +#include "internal.h" DEFINE_STATIC_KEY_FALSE (rdt_enable_key);DEFINE_STATIC_KEY_FALSE (rdt_mon_enable_key);
arch/x86/kernel/process_32.c +1 -1
@@ -56,7 +56,7 @@#include <asm/debugreg.h> #include <asm/switch_to.h> #include <asm/vm86.h> -#include <asm/intel_rdt_sched.h> +#include <asm/resctrl_sched.h> #include <asm/proto.h> void __show_regs (struct pt_regs *regs,enum show_regs_mode mode)
arch/x86/kernel/process_64.c +1 -1
@@ -52,7 +52,7 @@#include <asm/switch_to.h> #include <asm/xen/hypervisor.h> #include <asm/vdso.h> -#include <asm/intel_rdt_sched.h> +#include <asm/resctrl_sched.h> #include <asm/unistd.h> #ifdef CONFIG_IA32_EMULATION /* Not included via unistd.h */
Signature · 1 line
2.17.1
Borislav Petkov
maintainerRe: [PATCH v9 01/13] x86/resctrl: Rename and move rdt files to new directory
Quoted 12 lines · Babu
New generation of AMD processors start supporting RDT(or QOS) features. Together these features will be called as RESCTRL. With more than one vendors supporting these features, it seems more appropriate to rename these files.
Create a new directory with the name 'resctrl' and move all the
intel_rdtfiles to the new directory. This way all the resctrl related code resides inside one directory.Suggested-by Borislav Petkov <[email protected]>Signed-off-by Babu Moger <[email protected]>---
...
arch/x86/kernel/cpu/resctrl/Makefile line 3
Quoted 9 lines
arch/x86/kernel/cpu/resctrl/Makefile new file +3
-0@@ -0,0 +1,7 @@ +# +# Makefile for resource control feature code +#
I've replaced that comment with
# SPDX-License-Identifier: GPL-2.0
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.tip-bot for Babu Moger
merge bot[tip:x86/cache] x86/resctrl: Rename and move rdt files to a separate directory
In reply to Moger, Babux86/resctrl: Rename and move rdt files to a separate directory
Commit-ID: fa7d949337ccad32c76740c88e0e0351c349053b Gitweb: https://git.kernel.org/tip/fa7d949337ccad32c76740c88e0e0351c349053b Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:25 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:18 +0100
x86/resctrl: Rename and move rdt files to a separate directory
New generation of AMD processors add support for RDT (or QOS) features. Together, these features will be called RESCTRL. With more than one vendors supporting these features, it seems more appropriate to rename these files.
Create a new directory with the name 'resctrl' and move all the intel_rdt files to the new directory. This way all the resctrl related code resides inside one directory.
[ bp: Add SPDX identifier to the Makefile ]
---
arch/x86/include/asm/{intel_rdt_sched.h => resctrl_sched.h} | 0
arch/x86/kernel/cpu/Makefile | 5 +----
arch/x86/kernel/cpu/resctrl/Makefile | 4 ++++
arch/x86/kernel/cpu/{intel_rdt.c => resctrl/core.c} | 4 ++--
.../kernel/cpu/{intel_rdt_ctrlmondata.c => resctrl/ctrlmondata.c} | 2 +-
arch/x86/kernel/cpu/{intel_rdt.h => resctrl/internal.h} | 6 +++---
arch/x86/kernel/cpu/{intel_rdt_monitor.c => resctrl/monitor.c} | 2 +-
.../kernel/cpu/{intel_rdt_pseudo_lock.c => resctrl/pseudo_lock.c} | 6 +++---
.../{intel_rdt_pseudo_lock_event.h => resctrl/pseudo_lock_event.h} | 2 +-
arch/x86/kernel/cpu/{intel_rdt_rdtgroup.c => resctrl/rdtgroup.c} | 4 ++--
arch/x86/kernel/process_32.c | 2 +-
arch/x86/kernel/process_64.c | 2 +-
12 files changed, 20 insertions(+), 19 deletions(-)arch/x86/include/asm/resctrl_sched.h renamed from arch/x86/include/asm/intel_rdt_sched.h +0 -0
arch/x86/kernel/cpu/Makefile +1 -4
@@ -36,13 +36,10 @@ obj-$(CONFIG_CPU_SUP_CENTAUR) += centaur.o obj-$(CONFIG_CPU_SUP_TRANSMETA_32) += transmeta.o obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o -obj-$(CONFIG_INTEL_RDT) += intel_rdt.o intel_rdt_rdtgroup.o intel_rdt_monitor.o -obj-$(CONFIG_INTEL_RDT) += intel_rdt_ctrlmondata.o intel_rdt_pseudo_lock.o -CFLAGS_intel_rdt_pseudo_lock.o = -I$(src) - obj-$(CONFIG_X86_MCE) += mcheck/ obj-$(CONFIG_MTRR) += mtrr/ obj-$(CONFIG_MICROCODE) += microcode/ +obj-$(CONFIG_INTEL_RDT) += resctrl/ obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o
arch/x86/kernel/cpu/resctrl/Makefile new file +4 -0
@@ -0,0 +1,4 @@ +# SPDX-License-Identifier: GPL-2.0 +obj-$(CONFIG_INTEL_RDT) += core.o rdtgroup.o monitor.o +obj-$(CONFIG_INTEL_RDT) += ctrlmondata.o pseudo_lock.o +CFLAGS_pseudo_lock.o = -I$(src)
arch/x86/kernel/cpu/resctrl/core.c renamed from arch/x86/kernel/cpu/intel_rdt.c +2 -2
@@ -30,8 +30,8 @@#include <linux/cpuhotplug.h> #include <asm/intel-family.h> -#include <asm/intel_rdt_sched.h> -#include "intel_rdt.h" +#include <asm/resctrl_sched.h> +#include "internal.h" #define MBA_IS_LINEAR 0x4#define MBA_MAX_MBPS U32_MAX
arch/x86/kernel/cpu/resctrl/ctrlmondata.c renamed from arch/x86/kernel/cpu/intel_rdt_ctrlmondata.c +1 -1
@@ -26,7 +26,7 @@#include <linux/kernfs.h> #include <linux/seq_file.h> #include <linux/slab.h> -#include "intel_rdt.h" +#include "internal.h" /* * Check whether MBA bandwidth percentage value is correct. The value is
arch/x86/kernel/cpu/resctrl/internal.h renamed from arch/x86/kernel/cpu/intel_rdt.h +3 -3
@@ -1,6 +1,6 @@/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _ASM_X86_INTEL_RDT_H -#define _ASM_X86_INTEL_RDT_H +#ifndef _ASM_X86_RESCTRL_INTERNAL_H +#define _ASM_X86_RESCTRL_INTERNAL_H#include <linux/sched.h> #include <linux/kernfs.h>
@@ -568,4 +568,4 @@ void cqm_handle_limbo(struct work_struct *work);bool has_busy_rmid (struct rdt_resource *r,struct rdt_domain *d);void __check_limbo (struct rdt_domain *d,bool force_free); -#endif /* _ASM_X86_INTEL_RDT_H */ +#endif /* _ASM_X86_RESCTRL_INTERNAL_H */
arch/x86/kernel/cpu/resctrl/monitor.c renamed from arch/x86/kernel/cpu/intel_rdt_monitor.c +1 -1
@@ -26,7 +26,7 @@#include <linux/module.h> #include <linux/slab.h> #include <asm/cpu_device_id.h> -#include "intel_rdt.h" +#include "internal.h" #define MSR_IA32_QM_CTR 0x0c8e#define MSR_IA32_QM_EVTSEL 0x0c8d
arch/x86/kernel/cpu/resctrl/pseudo_lock.c renamed from arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c +3 -3
@@ -24,14 +24,14 @@#include <asm/cacheflush.h> #include <asm/intel-family.h> -#include <asm/intel_rdt_sched.h> +#include <asm/resctrl_sched.h> #include <asm/perf_event.h> #include "../../events/perf_event.h" /* For X86_CONFIG() */ -#include "intel_rdt.h" +#include "internal.h" #define CREATE_TRACE_POINTS -#include "intel_rdt_pseudo_lock_event.h" +#include "pseudo_lock_event.h" /* * MSR_MISC_FEATURE_CONTROL register enables the modification of hardware
arch/x86/kernel/cpu/resctrl/pseudo_lock_event.h renamed from arch/x86/kernel/cpu/intel_rdt_pseudo_lock_event.h +1 -1
@@ -39,5 +39,5 @@ TRACE_EVENT(pseudo_lock_l3,#undef TRACE_INCLUDE_PATH#define TRACE_INCLUDE_PATH . -#define TRACE_INCLUDE_FILE intel_rdt_pseudo_lock_event +#define TRACE_INCLUDE_FILE pseudo_lock_event#include <trace/define_trace.h>
arch/x86/kernel/cpu/resctrl/rdtgroup.c renamed from arch/x86/kernel/cpu/intel_rdt_rdtgroup.c +2 -2
@@ -35,8 +35,8 @@#include <uapi/linux/magic.h> -#include <asm/intel_rdt_sched.h> -#include "intel_rdt.h" +#include <asm/resctrl_sched.h> +#include "internal.h" DEFINE_STATIC_KEY_FALSE (rdt_enable_key);DEFINE_STATIC_KEY_FALSE (rdt_mon_enable_key);
arch/x86/kernel/process_32.c +1 -1
@@ -56,7 +56,7 @@#include <asm/debugreg.h> #include <asm/switch_to.h> #include <asm/vm86.h> -#include <asm/intel_rdt_sched.h> +#include <asm/resctrl_sched.h> #include <asm/proto.h> void __show_regs (struct pt_regs *regs,enum show_regs_mode mode)
arch/x86/kernel/process_64.c +1 -1
@@ -52,7 +52,7 @@#include <asm/switch_to.h> #include <asm/xen/hypervisor.h> #include <asm/vdso.h> -#include <asm/intel_rdt_sched.h> +#include <asm/resctrl_sched.h> #include <asm/unistd.h> #include <asm/fsgsbase.h> #ifdef CONFIG_IA32_EMULATION
Re: [PATCH v9 01/13] x86/resctrl: Rename and move rdt files to new directory
In reply to Moger, BabuQuoted 26 lines · Moger, Babu
New generation of AMD processors start supporting RDT(or QOS) features. Together these features will be called as RESCTRL. With more than one vendors supporting these features, it seems more appropriate to rename these files.
Create a new directory with the name 'resctrl' and move all the
intel_rdtfiles to the new directory. This way all the resctrl related code resides inside one directory.Suggested-by Borislav Petkov <[email protected]>Signed-off-by Babu Moger <[email protected]>--- .../x86/include/asm/{intel_rdt_sched.h => resctrl_sched.h} | 0 arch/x86/kernel/cpu/Makefile | 5 +---- arch/x86/kernel/cpu/resctrl/Makefile | 7 +++++++ arch/x86/kernel/cpu/{intel_rdt.c => resctrl/core.c} | 4 ++-- .../cpu/{intel_rdt_ctrlmondata.c => resctrl/ctrlmondata.c} | 2 +- arch/x86/kernel/cpu/{intel_rdt.h => resctrl/internal.h} | 6 +++--- .../kernel/cpu/{intel_rdt_monitor.c => resctrl/monitor.c} | 2 +- .../cpu/{intel_rdt_pseudo_lock.c => resctrl/pseudo_lock.c} | 6 +++--- .../pseudo_lock_event.h} | 2 +- .../cpu/{intel_rdt_rdtgroup.c => resctrl/rdtgroup.c} | 4 ++-- arch/x86/kernel/process_32.c | 2 +- arch/x86/kernel/process_64.c | 2 +- 12 files changed, 23 insertions(+), 19 deletions(-) rename arch/x86/include/asm/{intel_rdt_sched.h => resctrl_sched.h} (100%)
Quoted 8 lines
create mode 100644 arch/x86/kernel/cpu/resctrl/Makefile
rename arch/x86/kernel/cpu/{intel_rdt.c => resctrl/core.c} (99%)
rename arch/x86/kernel/cpu/{intel_rdt_ctrlmondata.c => resctrl/ctrlmondata.c} (99%)
rename arch/x86/kernel/cpu/{intel_rdt.h => resctrl/internal.h} (99%)
rename arch/x86/kernel/cpu/{intel_rdt_monitor.c => resctrl/monitor.c} (99%)
rename arch/x86/kernel/cpu/{intel_rdt_pseudo_lock.c => resctrl/pseudo_lock.c} (99%)
rename arch/x86/kernel/cpu/{intel_rdt_pseudo_lock_event.h => resctrl/pseudo_lock_event.h} (95%)
rename arch/x86/kernel/cpu/{intel_rdt_rdtgroup.c => resctrl/rdtgroup.c} (99%)Ugh, violent NAK on this unreadable directory naming: 'resctrl' is an ugly double/triple abbreviation that nobody recognizes for what it is to begin with, and even the long form 'resource control' is an overly generic naming - *everything* the kernel does is in essence 'resource control' ...
So please find some better name and standardize the namespace around it. A couple of suggestions:
- 'Hardware Quality of Service', i.e. HW_QOS, hw_qos
- or 'CPU bandwidth control', i.e. CPU_BW, cpu_bw
- or 'Hardware Bandwidth Control', i.e. HW_BW, hw_bw
etc.
Thanks,
ngo
Borislav Petkov
maintainerRe: [PATCH v9 01/13] x86/resctrl: Rename and move rdt files to new directory
Quoted 5 lines · Ingo Molnar
Ugh, violent NAK on this unreadable directory naming: 'resctrl' is an ugly double/triple abbreviation that nobody recognizes for what it is to begin with, and even the long form 'resource control' is an overly generic naming - *everything* the kernel does is in essence 'resource control' ...
Well, the fs this thing uses is called "resctrl".
Documentation/x86/resctrl_ui.txt:1075:the resctrl will still mount but cannot create CTRL_MON directories. Documentation/x86/resctrl_ui.txt:1082:# mount -t resctrl resctrl /sys/fs/resctrl Documentation/x86/resctrl_ui.txt:1083:# cd /sys/fs/resctrl
Are you saying that the fs should be renamed now too?
Quoted 6 lines
So please find some better name and standardize the namespace around it. A couple of suggestions:
- 'Hardware Quality of Service', i.e.
HW_QOS,hw_qos
- or 'CPU bandwidth control', i.e.CPU_BW,cpu_bw
- or 'Hardware Bandwidth Control', i.e.HW_BW,hw_bw
How are those *abbreviations* better? "hw_bw" is especially cryptic and the others are no better.
"resctrl" to mean "resource control" is much better IMO. And it is different from the "other" resource controlling the kernel does because it is under arch/x86/kernel/cpu/ which tells you it is a *CPU* resource control.
And also matches the user-visible "resctrl" filesystem.
But I don't have the energy to bikeshed this morning so whatever, as long as it is short...
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Re: [PATCH v9 01/13] x86/resctrl: Rename and move rdt files to new directory
Quoted 7 lines · Borislav Petkov
Quoted 5 lines · Ingo Molnar
Ugh, violent NAK on this unreadable directory naming: 'resctrl' is an ugly double/triple abbreviation that nobody recognizes for what it is to begin with, and even the long form 'resource control' is an overly generic naming - *everything* the kernel does is in essence 'resource control' ...
Quoted 7 lines
Well, the fs this thing uses is called "resctrl".
Documentation/x86/resctrl_ui.txt:1075:the resctrl will still mount but cannot create
CTRL_MONdirectories. Documentation/x86/resctrl_ui.txt:1082:# mount -t resctrl resctrl/sys/fs/resctrlDocumentation/x86/resctrl_ui.txt:1083:# cd/sys/fs/resctrlAre you saying that the fs should be renamed now too?
Sigh, probably not. I only noticed this naming snafu with the renaming commit. The high level name was always RDT-ish - which while an acronym is at least is a familiar high level name now with no obvious generic namespace collision, while 'resctrl' less so.
Quoted 2 lines
How are those *abbreviations* better? "
hw_bw" is especially cryptic and the others are no better.
Those were suggestions - but I'd be fine with 'resource_control':
Quoted 1 line
"resctrl" to mean "resource control" is much better IMO.
Then at least make the directory name resource_control/, which is only marginally longer and a lot more readable.
We really don't have to fit directly names into the 8 character DOS limit anymore. ;-)
Quoted 3 lines
[...] And it is different from the "other" resource controlling the kernel does because it is under
arch/x86/kernel/cpu/which tells you it is a *CPU* resource control.
Yeah, so this is not obvious from the filesystem name, nor does it excuse the pointless abbreviation.
High level names matter.
Also, the Kconfig space, when it gets extended with the AMD bits, should probably follow the same nomenclature: CONFIG_X86_CPU_RESOURCE_CONTROL=y or such.
Ingo
Borislav Petkov
maintainerRe: [PATCH v9 01/13] x86/resctrl: Rename and move rdt files to new directory
Quoted 5 lines · Ingo Molnar
Then at least make the directory name resource_control/, which is only marginally longer and a lot more readable.
We really don't have to fit directly names into the 8 character DOS limit anymore. ;-)
How about
resource_ctl
?
resource_control/ is kinda long-ish and the other names we have there are nice and short, see below.
BTW, while we're talking renaming, I have a patch which renames the MCE pile and am planning to slap it in around -rc6 timeframe since we don't have a lot of RAS commits this time around, see also the end of this mail. It makes the naming there all nicely regular. :)
Quoted 3 lines
Also, the
Kconfigspace, when it gets extended with the AMD bits, should probably follow the same nomenclature:CONFIG_X86_CPU_RESOURCE_CONTROL=yor such.
Sure, I can do that together with the directory rename once we've agreed on the name.
Thx.
arch/x86/kernel/cpu/
|-- amd.c |-- aperfmperf.c |-- bugs.c |-- cacheinfo.c |-- centaur.c |-- common.c |-- cpu.h |-- cpuid-deps.c |-- cyrix.c |-- hygon.c |-- hypervisor.c |-- intel.c |-- intel_pconfig.c |-- Makefile |-- match.c |-- mcheck | |-- dev-mcelog.c | |-- Makefile | |-- mce_amd.c | |-- mce-apei.c | |-- mce.c | |-- mce-genpool.c | |-- mce-inject.c | |-- mce_intel.c | |-- mce-internal.h | |-- mce-severity.c | |-- p5.c | |-- therm_throt.c | |-- threshold.c | `-- winchip.c |-- microcode | |-- amd.c | |-- core.c | |-- intel.c | `-- Makefile |-- mkcapflags.sh |-- mshyperv.c |-- mtrr | |-- amd.c | |-- centaur.c | |-- cleanup.c | |-- cyrix.c | |-- generic.c | |-- if.c | |-- Makefile | |-- mtrr.c | `-- mtrr.h |-- perfctr-watchdog.c |-- powerflags.c |-- proc.c |-- rdrand.c |-- resctrl | |-- core.c | |-- ctrlmondata.c | |-- internal.h | |-- Makefile | |-- monitor.c | |-- pseudo_lock.c | |-- pseudo_lock_event.h | `-- rdtgroup.c |-- scattered.c |-- topology.c |-- transmeta.c |-- umc.c
`-- vmware.c
4 directories, 61 files
---
From: Borislav Petkov <[email protected]>
Date: Sun, 18 Nov 2018 15:15:05 +0100
Subject: [PATCH] x86/mce: Streamline MCE subsystem's naming
Rename the containing folder to "mce" which is the most widespread name. Drop the "mce[-_]" filename prefix of some compilation units (while others don't have it).
This unifies the file naming in the MCE subsystem:
mce/
|-- amd.c |-- apei.c |-- core.c |-- dev-mcelog.c |-- genpool.c |-- inject.c |-- intel.c |-- internal.h |-- Makefile |-- p5.c |-- severity.c |-- therm_throt.c |-- threshold.c
`-- winchip.c
No functional changes.
---
arch/x86/kernel/cpu/Makefile | 2 +-
arch/x86/kernel/cpu/{mcheck => mce}/Makefile | 10 +++++-----
arch/x86/kernel/cpu/{mcheck/mce_amd.c => mce/amd.c} | 2 +-
arch/x86/kernel/cpu/{mcheck/mce-apei.c => mce/apei.c} | 2 +-
arch/x86/kernel/cpu/{mcheck/mce.c => mce/core.c} | 2 +-
arch/x86/kernel/cpu/{mcheck => mce}/dev-mcelog.c | 2 +-
.../kernel/cpu/{mcheck/mce-genpool.c => mce/genpool.c} | 2 +-
.../kernel/cpu/{mcheck/mce-inject.c => mce/inject.c} | 2 +-
.../x86/kernel/cpu/{mcheck/mce_intel.c => mce/intel.c} | 2 +-
.../cpu/{mcheck/mce-internal.h => mce/internal.h} | 0
arch/x86/kernel/cpu/{mcheck => mce}/p5.c | 0
.../cpu/{mcheck/mce-severity.c => mce/severity.c} | 2 +-
arch/x86/kernel/cpu/{mcheck => mce}/therm_throt.c | 0
arch/x86/kernel/cpu/{mcheck => mce}/threshold.c | 0
arch/x86/kernel/cpu/{mcheck => mce}/winchip.c | 0
15 files changed, 14 insertions(+), 14 deletions(-)
rename arch/x86/kernel/cpu/{mcheck => mce}/Makefile (52%)
rename arch/x86/kernel/cpu/{mcheck/mce_amd.c => mce/amd.c} (99%)
rename arch/x86/kernel/cpu/{mcheck/mce-apei.c => mce/apei.c} (99%)
rename arch/x86/kernel/cpu/{mcheck/mce.c => mce/core.c} (99%)
rename arch/x86/kernel/cpu/{mcheck => mce}/dev-mcelog.c (99%)
rename arch/x86/kernel/cpu/{mcheck/mce-genpool.c => mce/genpool.c} (99%)
rename arch/x86/kernel/cpu/{mcheck/mce-inject.c => mce/inject.c} (99%)
rename arch/x86/kernel/cpu/{mcheck/mce_intel.c => mce/intel.c} (99%)
rename arch/x86/kernel/cpu/{mcheck/mce-internal.h => mce/internal.h} (100%)
rename arch/x86/kernel/cpu/{mcheck => mce}/p5.c (100%)
rename arch/x86/kernel/cpu/{mcheck/mce-severity.c => mce/severity.c} (99%)
rename arch/x86/kernel/cpu/{mcheck => mce}/therm_throt.c (100%)
rename arch/x86/kernel/cpu/{mcheck => mce}/threshold.c (100%)
rename arch/x86/kernel/cpu/{mcheck => mce}/winchip.c (100%)arch/x86/kernel/cpu/Makefile +1 -1
@@ -40,7 +40,7 @@ obj-$(CONFIG_INTEL_RDT) += intel_rdt.o intel_rdt_rdtgroup.o intel_rdt_monitor.o obj-$(CONFIG_INTEL_RDT) += intel_rdt_ctrlmondata.o intel_rdt_pseudo_lock.o CFLAGS_intel_rdt_pseudo_lock.o = -I$(src) -obj-$(CONFIG_X86_MCE) += mcheck/ +obj-$(CONFIG_X86_MCE) += mce/ obj-$(CONFIG_MTRR) += mtrr/ obj-$(CONFIG_MICROCODE) += microcode/
arch/x86/kernel/cpu/mce/Makefile renamed from arch/x86/kernel/cpu/mcheck/Makefile +5 -5
@@ -1,14 +1,14 @@ # SPDX-License-Identifier: GPL-2.0 -obj-y = mce.o mce-severity.o mce-genpool.o +obj-y = core.o severity.o genpool.o obj-$(CONFIG_X86_ANCIENT_MCE) += winchip.o p5.o -obj-$(CONFIG_X86_MCE_INTEL) += mce_intel.o -obj-$(CONFIG_X86_MCE_AMD) += mce_amd.o +obj-$(CONFIG_X86_MCE_INTEL) += intel.o +obj-$(CONFIG_X86_MCE_AMD) += amd.o obj-$(CONFIG_X86_MCE_THRESHOLD) += threshold.o -obj-$(CONFIG_X86_MCE_INJECT) += mce-inject.o +obj-$(CONFIG_X86_MCE_INJECT) += inject.o obj-$(CONFIG_X86_THERMAL_VECTOR) += therm_throt.o -obj-$(CONFIG_ACPI_APEI) += mce-apei.o +obj-$(CONFIG_ACPI_APEI) += apei.o obj-$(CONFIG_X86_MCELOG_LEGACY) += dev-mcelog.o
arch/x86/kernel/cpu/mce/amd.c renamed from arch/x86/kernel/cpu/mcheck/mce_amd.c +1 -1
@@ -28,7 +28,7 @@#include <asm/msr.h> #include <asm/trace/irq_vectors.h> -#include "mce-internal.h" +#include "internal.h" #define NR_BLOCKS 5#define THRESHOLD_MAX 0xFFF
arch/x86/kernel/cpu/mce/apei.c renamed from arch/x86/kernel/cpu/mcheck/mce-apei.c +1 -1
@@ -36,7 +36,7 @@#include <acpi/ghes.h> #include <asm/mce.h> -#include "mce-internal.h" +#include "internal.h" void apei_mce_report_mem_error (int severity,struct cper_sec_mem_err *mem_err) {
arch/x86/kernel/cpu/mce/core.c renamed from arch/x86/kernel/cpu/mcheck/mce.c +1 -1
@@ -52,7 +52,7 @@#include <asm/msr.h> #include <asm/reboot.h> -#include "mce-internal.h" +#include "internal.h" static DEFINE_MUTEX (mce_log_mutex);
arch/x86/kernel/cpu/mce/dev-mcelog.c renamed from arch/x86/kernel/cpu/mcheck/dev-mcelog.c +1 -1
@@ -15,7 +15,7 @@#include <linux/kmod.h> #include <linux/poll.h> -#include "mce-internal.h" +#include "internal.h" static BLOCKING_NOTIFIER_HEAD (mce_injector_chain);
arch/x86/kernel/cpu/mce/genpool.c renamed from arch/x86/kernel/cpu/mcheck/mce-genpool.c +1 -1
@@ -10,7 +10,7 @@#include <linux/mm.h> #include <linux/genalloc.h> #include <linux/llist.h> -#include "mce-internal.h" +#include "internal.h" /* * printk() is not safe in MCE context. This is a lock-less memory allocator
arch/x86/kernel/cpu/mce/inject.c renamed from arch/x86/kernel/cpu/mcheck/mce-inject.c +1 -1
@@ -38,7 +38,7 @@#include <asm/nmi.h> #include <asm/smp.h> -#include "mce-internal.h" +#include "internal.h" /* * Collect all the MCi_XXX settings
arch/x86/kernel/cpu/mce/intel.c renamed from arch/x86/kernel/cpu/mcheck/mce_intel.c +1 -1
@@ -18,7 +18,7 @@#include <asm/msr.h> #include <asm/mce.h> -#include "mce-internal.h" +#include "internal.h" /* * Support for Intel Correct Machine Check Interrupts. This allows
arch/x86/kernel/cpu/mce/internal.h renamed from arch/x86/kernel/cpu/mcheck/mce-internal.h +0 -0
arch/x86/kernel/cpu/mce/p5.c renamed from arch/x86/kernel/cpu/mcheck/p5.c +0 -0
arch/x86/kernel/cpu/mce/severity.c renamed from arch/x86/kernel/cpu/mcheck/mce-severity.c +1 -1
@@ -16,7 +16,7 @@#include <asm/mce.h> #include <linux/uaccess.h> -#include "mce-internal.h" +#include "internal.h" /* * Grade an mce by severity. In general the most severe ones are processed
arch/x86/kernel/cpu/mce/therm_throt.c renamed from arch/x86/kernel/cpu/mcheck/therm_throt.c +0 -0
arch/x86/kernel/cpu/mce/threshold.c renamed from arch/x86/kernel/cpu/mcheck/threshold.c +0 -0
arch/x86/kernel/cpu/mce/winchip.c renamed from arch/x86/kernel/cpu/mcheck/winchip.c +0 -0
Signature · 8 lines
2.19.1
--
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Re: [PATCH v9 01/13] x86/resctrl: Rename and move rdt files to new directory
Quoted 4 lines · Borislav Petkov
Quoted 3 lines · Ingo Molnar
Then at least make the directory name resource_control/, which is only marginally longer and a lot more readable.
Quoted 8 lines
Quoted 2 lines
We really don't have to fit directly names into the 8 character DOS limit anymore. ;-)
How about
resource_ctl?
The thing is, thinking about this as a 'CPU resource' is really a misnomer on the conceptual level, which is why it's bothering me: RDT is not really about 'CPU resources', because registers are CPU resources, ioports and iomem are CPU resources, APICs are CPU resources and PMU events are resources - none of which is part of RDT.
The key difference in RDT is that they are *shared* resources - caches really - where the ad-hoc cache sharing might be causing security and scalability problems so there's partitioning and throttling (bandwidth control) support in the hardware.
Is there any other resource handled than caches by RDT or by the AMD variant?
So how about "cache_control"? It's shorter and a lot closer to what the code actually does.
Quoted 7 lines
resource_control/ is kinda long-ish and the other names we have there are nice and short, see below.
BTW, while we're talking renaming, I have a patch which renames the MCE pile and am planning to slap it in around -rc6 timeframe since we don't have a lot of RAS commits this time around, see also the end of this mail. It makes the naming there all nicely regular. :)
That's cool - these IMHO need to be done periodically to keep overall namespace complexity low enough. (As long as can be done without breaking any ABI that is.)
Quoted 7 lines
15 files changed, 14 insertions(+), 14 deletions(-)
rename arch/x86/kernel/cpu/{mcheck => mce}/Makefile (52%)
rename arch/x86/kernel/cpu/{mcheck/mce_amd.c => mce/amd.c} (99%)
rename arch/x86/kernel/cpu/{mcheck/mce-apei.c => mce/apei.c} (99%)
rename arch/x86/kernel/cpu/{mcheck/mce.c => mce/core.c} (99%)
rename arch/x86/kernel/cpu/{mcheck => mce}/dev-mcelog.c (99%)
rename arch/x86/kernel/cpu/{mcheck/mce-genpool.c => mce/genpool.c} (99%)Quoted 8 lines
rename arch/x86/kernel/cpu/{mcheck/mce-inject.c => mce/inject.c} (99%)
rename arch/x86/kernel/cpu/{mcheck/mce_intel.c => mce/intel.c} (99%)
rename arch/x86/kernel/cpu/{mcheck/mce-internal.h => mce/internal.h} (100%)
rename arch/x86/kernel/cpu/{mcheck => mce}/p5.c (100%)
rename arch/x86/kernel/cpu/{mcheck/mce-severity.c => mce/severity.c} (99%)
rename arch/x86/kernel/cpu/{mcheck => mce}/therm_throt.c (100%)
rename arch/x86/kernel/cpu/{mcheck => mce}/threshold.c (100%)
rename arch/x86/kernel/cpu/{mcheck => mce}/winchip.c (100%)Standardizing around 'MCE' sounds good to me!
Thanks,
Ingo
Re: [PATCH v9 01/13] x86/resctrl: Rename and move rdt files to new directory
Quoted 3 lines · Ingo Molnar
* Borislav Petkov <[email protected]> wrote:
Is there any other resource handled than caches by RDT or by the AMD variant?
Memory bandwidth.
Quoted 5 lines
So how about "
cache_control"? It's shorter and a lot closer to what the code actually does.Quoted 2 lines
resource_control/ is kinda long-ish and the other names we have there are nice and short, see below.
resctrl is short and good enough. Can we move on an care about technical problems rather than wasting time with bike shed painting?
Thanks,
tglx
Luck, Tony
maintainerRe: [PATCH v9 01/13] x86/resctrl: Rename and move rdt files to new directory
In reply to Borislav PetkovQuoted 2 lines · Borislav Petkov
Sure, I can do that together with the directory rename once we've agreed on the name.
Is this talking about renaming /sys/fs/resctrl?
If so NAK to that. It is ABI now. Lots of scripts depend on that name.
-Tony
Borislav Petkov
maintainerRe: [PATCH v9 01/13] x86/resctrl: Rename and move rdt files to new directory
Quoted 4 lines · Tony
Is this talking about renaming
/sys/fs/resctrl?If so NAK to that. It is ABI now. Lots of scripts depend on that name.
No no, that is cast in stone. The kernel source dir is called "resctrl" now too:
arch/x86/kernel/cpu/resctrl
so even less confusion.
:)
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Quoted 3 lines · Babu
This series adds support for AMD64 architectural extensions for Platform Quality of Service.
The term "QoS" is already used for net/sched/. It will be bad naming to have QoS - and then an AMD QoS / Platform QoS as well. Preexisting, decade-old howtos may speak of enabling "QoS", and with the ambiguity, the unsuspecting kernel builder apprentice may enable the wrong thing in his config.
So either the QoS from networking gets correspondingly renamed ("packet QoS"?.. or something), or else AMD QoS should be changed.
Moger, Babu
reviewerHi Jan,
Quoted 26 lines
-----Original Message-----
From: Jan Engelhardt <[email protected]>
Sent: Sunday, December 23, 2018 12:27 PM
To: Moger, Babu <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; akpm@linux- foundation.org; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; Singh, Brijesh <[email protected]>; Hurwitz, Sherry <[email protected]>; [email protected]; Lendacky, Thomas <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; linux- [email protected]
Subject: Re: [PATCH v9 00/13]arch/resctrl: AMD QoS supportQuoted 3 lines · Babu
This series adds support for AMD64 architectural extensions for Platform Quality of Service.
Quoted 8 lines
The term "QoS" is already used for
net/sched/. It will be bad naming to have QoS - and then an AMD QoS / Platform QoS as well. Preexisting, decade-old howtos may speak of enabling "QoS", and with the ambiguity, the unsuspecting kernel builder apprentice may enable the wrong thing in his config.So either the QoS from networking gets correspondingly renamed ("packet QoS"?.. or something), or else AMD QoS should be changed.
The QoS name is not visible to the user. This feature is referred as QoS in AMD documents. But for the kernel user it is called as RESCTRL which we agreed upon in our earlier discussions. So there is no ambiguity or confusion here. Let me know if you have any questions.
Moger, Babu
reviewerPATCH v9 2/13 x86/resctrl: Rename the RDT functions and definitions
In reply to Moger, BabuAs AMD is starting to support RESCTRL features, rename the RDT functions and definitions to more generic names.
Replace intel_rdt to resctrl where applicable.
--- arch/x86/include/asm/resctrl_sched.h | 24 +++++++++++----------- arch/x86/kernel/cpu/resctrl/core.c | 28 +++++++++++++------------- arch/x86/kernel/cpu/resctrl/internal.h | 2 +- arch/x86/kernel/cpu/resctrl/monitor.c | 11 +++++----- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 10 ++++----- arch/x86/kernel/process_32.c | 2 +- arch/x86/kernel/process_64.c | 2 +- 7 files changed, 40 insertions(+), 39 deletions(-)
arch/x86/include/asm/resctrl_sched.h +12 -12
@@ -1,6 +1,6 @@/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _ASM_X86_INTEL_RDT_SCHED_H -#define _ASM_X86_INTEL_RDT_SCHED_H +#ifndef _ASM_X86_RESCTRL_SCHED_H +#define _ASM_X86_RESCTRL_SCHED_H#ifdef CONFIG_INTEL_RDT
@@ -10,7 +10,7 @@#define IA32_PQR_ASSOC 0x0c8f/** -* struct intel_pqr_state - State cache for the PQR MSR +* struct resctrl_pqr_state - State cache for the PQR MSR * @cur_rmid: The cached Resource Monitoring ID * @cur_closid: The cached Class Of Service ID * @default_rmid: The user assigned Resource Monitoring ID
@@ -24,21 +24,21 @@* The cache also helps to avoid pointless updates if the value does * not change. */ -struct intel_pqr_state { +struct resctrl_pqr_state {u32 cur_rmid;u32 cur_closid;u32 default_rmid;u32 default_closid; }; -DECLARE_PER_CPU(struct intel_pqr_state, pqr_state); +DECLARE_PER_CPU(struct resctrl_pqr_state, pqr_state);DECLARE_STATIC_KEY_FALSE (rdt_enable_key);DECLARE_STATIC_KEY_FALSE (rdt_alloc_enable_key);DECLARE_STATIC_KEY_FALSE (rdt_mon_enable_key);/* -* __intel_rdt_sched_in() - Writes the task's CLOSid/RMID to IA32_PQR_MSR +* __resctrl_sched_in() - Writes the task's CLOSid/RMID to IA32_PQR_MSR * * Following considerations are made so that this has minimal impact * on scheduler hot path:
@@ -51,9 +51,9 @@ DECLARE_STATIC_KEY_FALSE(rdt_mon_enable_key);* simple as possible. * Must be called with preemption disabled. */ -static void __intel_rdt_sched_in (void ) +static void __resctrl_sched_in (void ) { -struct intel_pqr_state *state =this_cpu_ptr (&pqr_state); +struct resctrl_pqr_state *state =this_cpu_ptr (&pqr_state);u32 closid = state->default_closid;u32 rmid = state->default_rmid;
@@ -78,16 +78,16 @@ static void __intel_rdt_sched_in(void) } } -static inline void intel_rdt_sched_in (void ) +static inline void resctrl_sched_in (void ) {if (static_branch_likely (&rdt_enable_key)) -__intel_rdt_sched_in (); +__resctrl_sched_in (); }#else -static inline void intel_rdt_sched_in (void ) {} +static inline void resctrl_sched_in (void ) {}#endif /* CONFIG_INTEL_RDT */ -#endif /* _ASM_X86_INTEL_RDT_SCHED_H */ +#endif /* _ASM_X86_RESCTRL_SCHED_H */
arch/x86/kernel/cpu/resctrl/core.c +14 -14
@@ -22,7 +22,7 @@* Software Developer Manual June 2016, volume 3, section 17.17. */ -#define pr_fmt (fmt) KBUILD_MODNAME ": " fmt +#define pr_fmt (fmt) "resctrl: " fmt#include <linux/slab.h> #include <linux/err.h>
@@ -40,12 +40,12 @@DEFINE_MUTEX (rdtgroup_mutex);/* -* The cached intel_pqr_state is strictly per CPU and can never be +* The cached resctrl_pqr_state is strictly per CPU and can never be * updated from a remote CPU. Functions which modify the state * are called with interrupts disabled and no preemption, which * is sufficient for the protection. */ -DEFINE_PER_CPU(struct intel_pqr_state, pqr_state); +DEFINE_PER_CPU(struct resctrl_pqr_state, pqr_state);/* * Used to store the max resource name width and max resource data width
@@ -639,7 +639,7 @@ static void domain_remove_cpu(int cpu, struct rdt_resource *r)static void clear_closid_rmid (int cpu) { -struct intel_pqr_state *state =this_cpu_ptr (&pqr_state); +struct resctrl_pqr_state *state =this_cpu_ptr (&pqr_state); state->default_closid =0 ; state->default_rmid =0 ;
@@ -648,7 +648,7 @@ static void clear_closid_rmid(int cpu)wrmsr (IA32_PQR_ASSOC ,0 ,0 ); } -static int intel_rdt_online_cpu (unsigned int cpu) +static int resctrl_online_cpu (unsigned int cpu) {struct rdt_resource *r;
@@ -674,7 +674,7 @@ static void clear_childcpus(struct rdtgroup *r, unsigned int cpu) } } -static int intel_rdt_offline_cpu (unsigned int cpu) +static int resctrl_offline_cpu (unsigned int cpu) {struct rdtgroup *rdtgrp;struct rdt_resource *r;
@@ -866,7 +866,7 @@ static __init bool get_rdt_resources(void)static enum cpuhp_state rdt_online; -static int __init intel_rdt_late_init (void ) +static int __init resctrl_late_init (void ) {struct rdt_resource *r;int state, ret;
@@ -877,8 +877,8 @@ static int __init intel_rdt_late_init(void)rdt_init_padding (); state =cpuhp_setup_state (CPUHP_AP_ONLINE_DYN , -"x86/rdt/cat:online:" , - intel_rdt_online_cpu, intel_rdt_offline_cpu); +"x86/resctrl/cat:online:" , + resctrl_online_cpu, resctrl_offline_cpu);if (state <0 )return state;
@@ -890,20 +890,20 @@ static int __init intel_rdt_late_init(void) rdt_online = state;for_each_alloc_capable_rdt_resource (r) -pr_info ("Intel RDT %s allocation detected\n" , r->name); +pr_info ("%s allocation detected\n" , r->name);for_each_mon_capable_rdt_resource (r) -pr_info ("Intel RDT %s monitoring detected\n" , r->name); +pr_info ("%s monitoring detected\n" , r->name);return 0 ; } -late_initcall (intel_rdt_late_init); +late_initcall (resctrl_late_init); -static void __exit intel_rdt_exit(void) +static void __exit resctrl_exit(void) {cpuhp_remove_state (rdt_online);rdtgroup_exit (); } -__exitcall (intel_rdt_exit); +__exitcall (resctrl_exit);
arch/x86/kernel/cpu/resctrl/internal.h +1 -1
@@ -69,7 +69,7 @@ struct rmid_read {u64 val; }; -extern unsigned int intel_cqm_threshold; +extern unsigned int resctrl_cqm_threshold;extern bool rdt_alloc_capable;extern bool rdt_mon_capable;extern unsigned int rdt_mon_features;
arch/x86/kernel/cpu/resctrl/monitor.c +6 -5
@@ -73,7 +73,7 @@ unsigned int rdt_mon_features;* This is the threshold cache occupancy at which we will consider an * RMID available for re-allocation. */ -unsigned int intel_cqm_threshold; +unsigned int resctrl_cqm_threshold;static inline struct rmid_entry *__rmid_entry (u32 rmid) {
@@ -107,7 +107,7 @@ static bool rmid_dirty(struct rmid_entry *entry) {u64 val =__rmid_read (entry->rmid,QOS_L3_OCCUP_EVENT_ID ); -return val >= intel_cqm_threshold; +return val >= resctrl_cqm_threshold; }/*
@@ -187,7 +187,7 @@ static void add_rmid_to_limbo(struct rmid_entry *entry)list_for_each_entry (d, &r->domains, list) {if (cpumask_test_cpu (cpu, &d->cpu_mask)) { val =__rmid_read (entry->rmid,QOS_L3_OCCUP_EVENT_ID ); -if (val <= intel_cqm_threshold) +if (val <= resctrl_cqm_threshold)continue ; }
@@ -625,6 +625,7 @@ static void l3_mon_evt_init(struct rdt_resource *r)int rdt_get_mon_l3_config (struct rdt_resource *r) { +unsigned int cl_size = boot_cpu_data.x86_cache_size;int ret; r->mon_scale = boot_cpu_data.x86_cache_occ_scale;
@@ -637,10 +638,10 @@ int rdt_get_mon_l3_config(struct rdt_resource *r)* * For a 35MB LLC and 56 RMIDs, this is ~1.8% of the LLC. */ - intel_cqm_threshold = boot_cpu_data.x86_cache_size *1024 / r->num_rmid; + resctrl_cqm_threshold = cl_size *1024 / r->num_rmid;/* h/w works in units of "boot_cpu_data.x86_cache_occ_scale" */ - intel_cqm_threshold /= r->mon_scale; + resctrl_cqm_threshold /= r->mon_scale; ret =dom_data_init (r);if (ret)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +5 -5
@@ -298,7 +298,7 @@ static int rdtgroup_cpus_show(struct kernfs_open_file *of, }/* -* This is safe against intel_rdt_sched_in() called from __switch_to() +* This is safe against resctrl_sched_in() called from __switch_to() * because __switch_to() is executed with interrupts disabled. A local call * from update_closid_rmid() is proteced against __switch_to() because * preemption is disabled.
@@ -317,7 +317,7 @@ static void update_cpu_closid_rmid(void *info)* executing task might have its own closid selected. Just reuse * the context switch code. */ -intel_rdt_sched_in (); +resctrl_sched_in (); }/*
@@ -542,7 +542,7 @@ static void move_myself(struct callback_head *head)preempt_disable ();/* update PQR_ASSOC MSR to make resource group go into effect */ -intel_rdt_sched_in (); +resctrl_sched_in ();preempt_enable ();kfree (callback);
@@ -926,7 +926,7 @@ static int max_threshold_occ_show(struct kernfs_open_file *of, {struct rdt_resource *r = of->kn->parent->priv; -seq_printf (seq,"%u\n" , intel_cqm_threshold * r->mon_scale); +seq_printf (seq,"%u\n" , resctrl_cqm_threshold * r->mon_scale);return 0 ; }
@@ -945,7 +945,7 @@ static ssize_t max_threshold_occ_write(struct kernfs_open_file *of,if (bytes > (boot_cpu_data.x86_cache_size *1024 ))return -EINVAL ; - intel_cqm_threshold = bytes / r->mon_scale; + resctrl_cqm_threshold = bytes / r->mon_scale;return nbytes; }
arch/x86/kernel/process_32.c +1 -1
@@ -302,7 +302,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)this_cpu_write (current_task, next_p);/* Load the Intel cache allocation PQR MSR. */ -intel_rdt_sched_in (); +resctrl_sched_in ();return prev_p; }
arch/x86/kernel/process_64.c +1 -1
@@ -536,7 +536,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) }/* Load the Intel cache allocation PQR MSR. */ -intel_rdt_sched_in (); +resctrl_sched_in ();return prev_p; }
Signature · 1 line
2.17.1
tip-bot for Babu Moger
merge bot[tip:x86/cache] x86/resctrl: Rename the RDT functions and definitions
x86/resctrl: Rename the RDT functions and definitions
Commit-ID: 352940ececaca58536a7fc4ff6b41d181156fd65 Gitweb: https://git.kernel.org/tip/352940ececaca58536a7fc4ff6b41d181156fd65 Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:27 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:18 +0100
x86/resctrl: Rename the RDT functions and definitions
As AMD is starting to support RESCTRL features, rename the RDT functions and definitions to more generic names.
Replace "intel_rdt" with "resctrl" where applicable.
--- arch/x86/include/asm/resctrl_sched.h | 24 ++++++++++++------------ arch/x86/kernel/cpu/resctrl/core.c | 28 ++++++++++++++-------------- arch/x86/kernel/cpu/resctrl/internal.h | 2 +- arch/x86/kernel/cpu/resctrl/monitor.c | 11 ++++++----- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 10 +++++----- arch/x86/kernel/process_32.c | 2 +- arch/x86/kernel/process_64.c | 2 +- 7 files changed, 40 insertions(+), 39 deletions(-)
arch/x86/include/asm/resctrl_sched.h +12 -12
@@ -1,6 +1,6 @@/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _ASM_X86_INTEL_RDT_SCHED_H -#define _ASM_X86_INTEL_RDT_SCHED_H +#ifndef _ASM_X86_RESCTRL_SCHED_H +#define _ASM_X86_RESCTRL_SCHED_H#ifdef CONFIG_INTEL_RDT
@@ -10,7 +10,7 @@#define IA32_PQR_ASSOC 0x0c8f/** -* struct intel_pqr_state - State cache for the PQR MSR +* struct resctrl_pqr_state - State cache for the PQR MSR * @cur_rmid: The cached Resource Monitoring ID * @cur_closid: The cached Class Of Service ID * @default_rmid: The user assigned Resource Monitoring ID
@@ -24,21 +24,21 @@* The cache also helps to avoid pointless updates if the value does * not change. */ -struct intel_pqr_state { +struct resctrl_pqr_state {u32 cur_rmid;u32 cur_closid;u32 default_rmid;u32 default_closid; }; -DECLARE_PER_CPU(struct intel_pqr_state, pqr_state); +DECLARE_PER_CPU(struct resctrl_pqr_state, pqr_state);DECLARE_STATIC_KEY_FALSE (rdt_enable_key);DECLARE_STATIC_KEY_FALSE (rdt_alloc_enable_key);DECLARE_STATIC_KEY_FALSE (rdt_mon_enable_key);/* -* __intel_rdt_sched_in() - Writes the task's CLOSid/RMID to IA32_PQR_MSR +* __resctrl_sched_in() - Writes the task's CLOSid/RMID to IA32_PQR_MSR * * Following considerations are made so that this has minimal impact * on scheduler hot path:
@@ -51,9 +51,9 @@ DECLARE_STATIC_KEY_FALSE(rdt_mon_enable_key);* simple as possible. * Must be called with preemption disabled. */ -static void __intel_rdt_sched_in (void ) +static void __resctrl_sched_in (void ) { -struct intel_pqr_state *state =this_cpu_ptr (&pqr_state); +struct resctrl_pqr_state *state =this_cpu_ptr (&pqr_state);u32 closid = state->default_closid;u32 rmid = state->default_rmid;
@@ -78,16 +78,16 @@ static void __intel_rdt_sched_in(void) } } -static inline void intel_rdt_sched_in (void ) +static inline void resctrl_sched_in (void ) {if (static_branch_likely (&rdt_enable_key)) -__intel_rdt_sched_in (); +__resctrl_sched_in (); }#else -static inline void intel_rdt_sched_in (void ) {} +static inline void resctrl_sched_in (void ) {}#endif /* CONFIG_INTEL_RDT */ -#endif /* _ASM_X86_INTEL_RDT_SCHED_H */ +#endif /* _ASM_X86_RESCTRL_SCHED_H */
arch/x86/kernel/cpu/resctrl/core.c +14 -14
@@ -22,7 +22,7 @@* Software Developer Manual June 2016, volume 3, section 17.17. */ -#define pr_fmt (fmt) KBUILD_MODNAME ": " fmt +#define pr_fmt (fmt) "resctrl: " fmt#include <linux/slab.h> #include <linux/err.h>
@@ -40,12 +40,12 @@DEFINE_MUTEX (rdtgroup_mutex);/* -* The cached intel_pqr_state is strictly per CPU and can never be +* The cached resctrl_pqr_state is strictly per CPU and can never be * updated from a remote CPU. Functions which modify the state * are called with interrupts disabled and no preemption, which * is sufficient for the protection. */ -DEFINE_PER_CPU(struct intel_pqr_state, pqr_state); +DEFINE_PER_CPU(struct resctrl_pqr_state, pqr_state);/* * Used to store the max resource name width and max resource data width
@@ -639,7 +639,7 @@ static void domain_remove_cpu(int cpu, struct rdt_resource *r)static void clear_closid_rmid (int cpu) { -struct intel_pqr_state *state =this_cpu_ptr (&pqr_state); +struct resctrl_pqr_state *state =this_cpu_ptr (&pqr_state); state->default_closid =0 ; state->default_rmid =0 ;
@@ -648,7 +648,7 @@ static void clear_closid_rmid(int cpu)wrmsr (IA32_PQR_ASSOC ,0 ,0 ); } -static int intel_rdt_online_cpu (unsigned int cpu) +static int resctrl_online_cpu (unsigned int cpu) {struct rdt_resource *r;
@@ -674,7 +674,7 @@ static void clear_childcpus(struct rdtgroup *r, unsigned int cpu) } } -static int intel_rdt_offline_cpu (unsigned int cpu) +static int resctrl_offline_cpu (unsigned int cpu) {struct rdtgroup *rdtgrp;struct rdt_resource *r;
@@ -866,7 +866,7 @@ static __init bool get_rdt_resources(void)static enum cpuhp_state rdt_online; -static int __init intel_rdt_late_init (void ) +static int __init resctrl_late_init (void ) {struct rdt_resource *r;int state, ret;
@@ -877,8 +877,8 @@ static int __init intel_rdt_late_init(void)rdt_init_padding (); state =cpuhp_setup_state (CPUHP_AP_ONLINE_DYN , -"x86/rdt/cat:online:" , - intel_rdt_online_cpu, intel_rdt_offline_cpu); +"x86/resctrl/cat:online:" , + resctrl_online_cpu, resctrl_offline_cpu);if (state <0 )return state;
@@ -890,20 +890,20 @@ static int __init intel_rdt_late_init(void) rdt_online = state;for_each_alloc_capable_rdt_resource (r) -pr_info ("Intel RDT %s allocation detected\n" , r->name); +pr_info ("%s allocation detected\n" , r->name);for_each_mon_capable_rdt_resource (r) -pr_info ("Intel RDT %s monitoring detected\n" , r->name); +pr_info ("%s monitoring detected\n" , r->name);return 0 ; } -late_initcall (intel_rdt_late_init); +late_initcall (resctrl_late_init); -static void __exit intel_rdt_exit(void) +static void __exit resctrl_exit(void) {cpuhp_remove_state (rdt_online);rdtgroup_exit (); } -__exitcall (intel_rdt_exit); +__exitcall (resctrl_exit);
arch/x86/kernel/cpu/resctrl/internal.h +1 -1
@@ -69,7 +69,7 @@ struct rmid_read {u64 val; }; -extern unsigned int intel_cqm_threshold; +extern unsigned int resctrl_cqm_threshold;extern bool rdt_alloc_capable;extern bool rdt_mon_capable;extern unsigned int rdt_mon_features;
arch/x86/kernel/cpu/resctrl/monitor.c +6 -5
@@ -73,7 +73,7 @@ unsigned int rdt_mon_features;* This is the threshold cache occupancy at which we will consider an * RMID available for re-allocation. */ -unsigned int intel_cqm_threshold; +unsigned int resctrl_cqm_threshold;static inline struct rmid_entry *__rmid_entry (u32 rmid) {
@@ -107,7 +107,7 @@ static bool rmid_dirty(struct rmid_entry *entry) {u64 val =__rmid_read (entry->rmid,QOS_L3_OCCUP_EVENT_ID ); -return val >= intel_cqm_threshold; +return val >= resctrl_cqm_threshold; }/*
@@ -187,7 +187,7 @@ static void add_rmid_to_limbo(struct rmid_entry *entry)list_for_each_entry (d, &r->domains, list) {if (cpumask_test_cpu (cpu, &d->cpu_mask)) { val =__rmid_read (entry->rmid,QOS_L3_OCCUP_EVENT_ID ); -if (val <= intel_cqm_threshold) +if (val <= resctrl_cqm_threshold)continue ; }
@@ -625,6 +625,7 @@ static void l3_mon_evt_init(struct rdt_resource *r)int rdt_get_mon_l3_config (struct rdt_resource *r) { +unsigned int cl_size = boot_cpu_data.x86_cache_size;int ret; r->mon_scale = boot_cpu_data.x86_cache_occ_scale;
@@ -637,10 +638,10 @@ int rdt_get_mon_l3_config(struct rdt_resource *r)* * For a 35MB LLC and 56 RMIDs, this is ~1.8% of the LLC. */ - intel_cqm_threshold = boot_cpu_data.x86_cache_size *1024 / r->num_rmid; + resctrl_cqm_threshold = cl_size *1024 / r->num_rmid;/* h/w works in units of "boot_cpu_data.x86_cache_occ_scale" */ - intel_cqm_threshold /= r->mon_scale; + resctrl_cqm_threshold /= r->mon_scale; ret =dom_data_init (r);if (ret)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +5 -5
@@ -298,7 +298,7 @@ static int rdtgroup_cpus_show(struct kernfs_open_file *of, }/* -* This is safe against intel_rdt_sched_in() called from __switch_to() +* This is safe against resctrl_sched_in() called from __switch_to() * because __switch_to() is executed with interrupts disabled. A local call * from update_closid_rmid() is proteced against __switch_to() because * preemption is disabled.
@@ -317,7 +317,7 @@ static void update_cpu_closid_rmid(void *info)* executing task might have its own closid selected. Just reuse * the context switch code. */ -intel_rdt_sched_in (); +resctrl_sched_in (); }/*
@@ -542,7 +542,7 @@ static void move_myself(struct callback_head *head)preempt_disable ();/* update PQR_ASSOC MSR to make resource group go into effect */ -intel_rdt_sched_in (); +resctrl_sched_in ();preempt_enable ();kfree (callback);
@@ -926,7 +926,7 @@ static int max_threshold_occ_show(struct kernfs_open_file *of, {struct rdt_resource *r = of->kn->parent->priv; -seq_printf (seq,"%u\n" , intel_cqm_threshold * r->mon_scale); +seq_printf (seq,"%u\n" , resctrl_cqm_threshold * r->mon_scale);return 0 ; }
@@ -945,7 +945,7 @@ static ssize_t max_threshold_occ_write(struct kernfs_open_file *of,if (bytes > (boot_cpu_data.x86_cache_size *1024 ))return -EINVAL ; - intel_cqm_threshold = bytes / r->mon_scale; + resctrl_cqm_threshold = bytes / r->mon_scale;return nbytes; }
arch/x86/kernel/process_32.c +1 -1
@@ -302,7 +302,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)this_cpu_write (current_task, next_p);/* Load the Intel cache allocation PQR MSR. */ -intel_rdt_sched_in (); +resctrl_sched_in ();return prev_p; }
arch/x86/kernel/process_64.c +1 -1
@@ -664,7 +664,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) }/* Load the Intel cache allocation PQR MSR. */ -intel_rdt_sched_in (); +resctrl_sched_in ();return prev_p; }
Moger, Babu
reviewerPATCH v9 3/13 x86/resctrl: Re-arrange RDT init code
In reply to Moger, BabuSeparate the call sequence for rdt_quirks and MBA feature. This is in preparation to handle vendor differences in these call sequences. Rename the functions to make it bit more meaningful.
--- arch/x86/kernel/cpu/resctrl/core.c | 28 +++++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +21 -7
@@ -259,7 +259,7 @@ static inline bool rdt_get_mb_table(struct rdt_resource *r)return false; } -static bool rdt_get_mem_config (struct rdt_resource *r) +static bool __get_mem_config (struct rdt_resource *r) {union cpuid_0x10_3_eax eax;union cpuid_0x10_x_edx edx;
@@ -794,6 +794,14 @@ static bool __init rdt_cpu_has(int flag)return ret; } +static __init bool get_mem_config (void ) +{ +if (rdt_cpu_has (X86_FEATURE_MBA )) +return __get_mem_config (&rdt_resources_all[RDT_RESOURCE_MBA ]); + +return false; +} +static __init bool get_rdt_alloc_resources (void ) {bool ret = false;
@@ -818,10 +826,9 @@ static __init bool get_rdt_alloc_resources(void) ret = true; } -if (rdt_cpu_has (X86_FEATURE_MBA )) { -if (rdt_get_mem_config (&rdt_resources_all[RDT_RESOURCE_MBA ])) - ret = true; - } +if (get_mem_config ()) + ret = true; +return ret; }
@@ -840,7 +847,7 @@ static __init bool get_rdt_mon_resources(void)return !rdt_get_mon_l3_config (&rdt_resources_all[RDT_RESOURCE_L3 ]); } -static __init void rdt_quirks (void ) +static __init void __check_quirks_intel (void ) {switch (boot_cpu_data.x86_model) {case INTEL_FAM6_HASWELL_X :
@@ -855,9 +862,14 @@ static __init void rdt_quirks(void) } } +static __init void check_quirks (void ) +{ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) +__check_quirks_intel (); +} +static __init bool get_rdt_resources (void ) { -rdt_quirks (); rdt_alloc_capable =get_rdt_alloc_resources (); rdt_mon_capable =get_rdt_mon_resources ();
@@ -871,6 +883,8 @@ static int __init resctrl_late_init(void)struct rdt_resource *r;int state, ret; +check_quirks (); +if (!get_rdt_resources ())return -ENODEV ;
Signature · 1 line
2.17.1
tip-bot for Babu Moger
merge bot[tip:x86/cache] x86/resctrl: Re-arrange the RDT init code
x86/resctrl: Re-arrange the RDT init code
Commit-ID: 0f00717ecce4b366b3f530153e982cb2eeadafd1 Gitweb: https://git.kernel.org/tip/0f00717ecce4b366b3f530153e982cb2eeadafd1 Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:29 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:19 +0100
x86/resctrl: Re-arrange the RDT init code
Separate the call sequence for rdt_quirks and MBA feature. This is in preparation to handle vendor differences in these call sequences. Rename the functions to make the flow a bit more meaningful.
--- arch/x86/kernel/cpu/resctrl/core.c | 28 +++++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +21 -7
@@ -259,7 +259,7 @@ static inline bool rdt_get_mb_table(struct rdt_resource *r)return false; } -static bool rdt_get_mem_config (struct rdt_resource *r) +static bool __get_mem_config (struct rdt_resource *r) {union cpuid_0x10_3_eax eax;union cpuid_0x10_x_edx edx;
@@ -794,6 +794,14 @@ static bool __init rdt_cpu_has(int flag)return ret; } +static __init bool get_mem_config (void ) +{ +if (rdt_cpu_has (X86_FEATURE_MBA )) +return __get_mem_config (&rdt_resources_all[RDT_RESOURCE_MBA ]); + +return false; +} +static __init bool get_rdt_alloc_resources (void ) {bool ret = false;
@@ -818,10 +826,9 @@ static __init bool get_rdt_alloc_resources(void) ret = true; } -if (rdt_cpu_has (X86_FEATURE_MBA )) { -if (rdt_get_mem_config (&rdt_resources_all[RDT_RESOURCE_MBA ])) - ret = true; - } +if (get_mem_config ()) + ret = true; +return ret; }
@@ -840,7 +847,7 @@ static __init bool get_rdt_mon_resources(void)return !rdt_get_mon_l3_config (&rdt_resources_all[RDT_RESOURCE_L3 ]); } -static __init void rdt_quirks (void ) +static __init void __check_quirks_intel (void ) {switch (boot_cpu_data.x86_model) {case INTEL_FAM6_HASWELL_X :
@@ -855,9 +862,14 @@ static __init void rdt_quirks(void) } } +static __init void check_quirks (void ) +{ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) +__check_quirks_intel (); +} +static __init bool get_rdt_resources (void ) { -rdt_quirks (); rdt_alloc_capable =get_rdt_alloc_resources (); rdt_mon_capable =get_rdt_mon_resources ();
@@ -871,6 +883,8 @@ static int __init resctrl_late_init(void)struct rdt_resource *r;int state, ret; +check_quirks (); +if (!get_rdt_resources ())return -ENODEV ;
Moger, Babu
reviewerPATCH v9 4/13 x86/resctrl: Bring all the macros to resctrl/internal.h
In reply to Moger, BabuBring all the macros to resctrl/internal.h and rename the registers with MSR_ prefix for consistency.
--- arch/x86/kernel/cpu/resctrl/core.c | 22 ++++++++++------------ arch/x86/kernel/cpu/resctrl/internal.h | 15 ++++++++++----- arch/x86/kernel/cpu/resctrl/monitor.c | 3 --- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 4 ++-- 4 files changed, 22 insertions(+), 22 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +10 -12
@@ -33,9 +33,6 @@#include <asm/resctrl_sched.h> #include "internal.h" -#define MBA_IS_LINEAR 0x4 -#define MBA_MAX_MBPS U32_MAX -/* Mutex to protect rdtgroup access. */ DEFINE_MUTEX (rdtgroup_mutex);
@@ -72,7 +69,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L3, .name = "L3", .domains = domain_init(RDT_RESOURCE_L3), - .msr_base = IA32_L3_CBM_BASE, + .msr_base = MSR_IA32_L3_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 3, .cache = {
@@ -89,7 +86,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L3DATA, .name = "L3DATA", .domains = domain_init(RDT_RESOURCE_L3DATA), - .msr_base = IA32_L3_CBM_BASE, + .msr_base = MSR_IA32_L3_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 3, .cache = {
@@ -106,7 +103,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L3CODE, .name = "L3CODE", .domains = domain_init(RDT_RESOURCE_L3CODE), - .msr_base = IA32_L3_CBM_BASE, + .msr_base = MSR_IA32_L3_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 3, .cache = {
@@ -123,7 +120,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L2, .name = "L2", .domains = domain_init(RDT_RESOURCE_L2), - .msr_base = IA32_L2_CBM_BASE, + .msr_base = MSR_IA32_L2_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 2, .cache = {
@@ -140,7 +137,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L2DATA, .name = "L2DATA", .domains = domain_init(RDT_RESOURCE_L2DATA), - .msr_base = IA32_L2_CBM_BASE, + .msr_base = MSR_IA32_L2_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 2, .cache = {
@@ -157,7 +154,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L2CODE, .name = "L2CODE", .domains = domain_init(RDT_RESOURCE_L2CODE), - .msr_base = IA32_L2_CBM_BASE, + .msr_base = MSR_IA32_L2_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 2, .cache = {
@@ -174,7 +171,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_MBA, .name = "MB", .domains = domain_init(RDT_RESOURCE_MBA), - .msr_base = IA32_MBA_THRTL_BASE, + .msr_base = MSR_IA32_MBA_THRTL_BASE, .msr_update = mba_wrmsr, .cache_level = 3, .parse_ctrlval = parse_bw,
@@ -211,9 +208,10 @@ static inline void cache_alloc_hsw_probe(void)struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_L3 ];u32 l, h, max_cbm =BIT_MASK (20 ) -1 ; -if (wrmsr_safe (IA32_L3_CBM_BASE , max_cbm,0 )) +if (wrmsr_safe (MSR_IA32_L3_CBM_BASE , max_cbm,0 ))return ; -rdmsr (IA32_L3_CBM_BASE , l, h); + +rdmsr (MSR_IA32_L3_CBM_BASE , l, h);/* If all the bits were set in MSR, return success */ if (l != max_cbm)
arch/x86/kernel/cpu/resctrl/internal.h +10 -5
@@ -6,11 +6,14 @@#include <linux/kernfs.h> #include <linux/jump_label.h> -#define IA32_L3_QOS_CFG 0xc81 -#define IA32_L2_QOS_CFG 0xc82 -#define IA32_L3_CBM_BASE 0xc90 -#define IA32_L2_CBM_BASE 0xd10 -#define IA32_MBA_THRTL_BASE 0xd50 +#define MSR_IA32_L3_QOS_CFG 0xc81 +#define MSR_IA32_L2_QOS_CFG 0xc82 +#define MSR_IA32_L3_CBM_BASE 0xc90 +#define MSR_IA32_L2_CBM_BASE 0xd10 +#define MSR_IA32_MBA_THRTL_BASE 0xd50 + +#define MSR_IA32_QM_CTR 0x0c8e +#define MSR_IA32_QM_EVTSEL 0x0c8d#define L3_QOS_CDP_ENABLE 0x01ULL
@@ -29,6 +32,8 @@#define MBM_CNTR_WIDTH 24#define MBM_OVERFLOW_INTERVAL 1000#define MAX_MBA_BW 100u +#define MBA_IS_LINEAR 0x4 +#define MBA_MAX_MBPS U32_MAX#define RMID_VAL_ERROR BIT_ULL(63)#define RMID_VAL_UNAVAIL BIT_ULL(62)
arch/x86/kernel/cpu/resctrl/monitor.c +0 -3
@@ -28,9 +28,6 @@#include <asm/cpu_device_id.h> #include "internal.h" -#define MSR_IA32_QM_CTR 0x0c8e -#define MSR_IA32_QM_EVTSEL 0x0c8d -struct rmid_entry {u32 rmid;int busy;
arch/x86/kernel/cpu/resctrl/rdtgroup.c +2 -2
@@ -1722,14 +1722,14 @@ static void l3_qos_cfg_update(void *arg) {bool *enable = arg; -wrmsrl (IA32_L3_QOS_CFG , *enable ?L3_QOS_CDP_ENABLE :0ULL ); +wrmsrl (MSR_IA32_L3_QOS_CFG , *enable ?L3_QOS_CDP_ENABLE :0ULL ); }static void l2_qos_cfg_update (void *arg) {bool *enable = arg; -wrmsrl (IA32_L2_QOS_CFG , *enable ?L2_QOS_CDP_ENABLE :0ULL ); +wrmsrl (MSR_IA32_L2_QOS_CFG , *enable ?L2_QOS_CDP_ENABLE :0ULL ); }static inline bool is_mba_linear (void )
Signature · 1 line
2.17.1
tip-bot for Babu Moger
merge bot[tip:x86/cache] x86/resctrl: Move all the macros to resctrl/internal.h
x86/resctrl: Move all the macros to resctrl/internal.h
Commit-ID: aa50453a448ad645ea05788505680aa403934aa8 Gitweb: https://git.kernel.org/tip/aa50453a448ad645ea05788505680aa403934aa8 Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:31 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:19 +0100
x86/resctrl: Move all the macros to resctrl/internal.h
Move all the macros to resctrl/internal.h and rename the registers with MSR_ prefix for consistency.
[bp: align MSR definitions vertically ]
--- arch/x86/kernel/cpu/resctrl/core.c | 22 ++++++++++------------ arch/x86/kernel/cpu/resctrl/internal.h | 19 ++++++++++++------- arch/x86/kernel/cpu/resctrl/monitor.c | 3 --- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 4 ++-- 4 files changed, 24 insertions(+), 24 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +10 -12
@@ -33,9 +33,6 @@#include <asm/resctrl_sched.h> #include "internal.h" -#define MBA_IS_LINEAR 0x4 -#define MBA_MAX_MBPS U32_MAX -/* Mutex to protect rdtgroup access. */ DEFINE_MUTEX (rdtgroup_mutex);
@@ -72,7 +69,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L3, .name = "L3", .domains = domain_init(RDT_RESOURCE_L3), - .msr_base = IA32_L3_CBM_BASE, + .msr_base = MSR_IA32_L3_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 3, .cache = {
@@ -89,7 +86,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L3DATA, .name = "L3DATA", .domains = domain_init(RDT_RESOURCE_L3DATA), - .msr_base = IA32_L3_CBM_BASE, + .msr_base = MSR_IA32_L3_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 3, .cache = {
@@ -106,7 +103,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L3CODE, .name = "L3CODE", .domains = domain_init(RDT_RESOURCE_L3CODE), - .msr_base = IA32_L3_CBM_BASE, + .msr_base = MSR_IA32_L3_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 3, .cache = {
@@ -123,7 +120,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L2, .name = "L2", .domains = domain_init(RDT_RESOURCE_L2), - .msr_base = IA32_L2_CBM_BASE, + .msr_base = MSR_IA32_L2_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 2, .cache = {
@@ -140,7 +137,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L2DATA, .name = "L2DATA", .domains = domain_init(RDT_RESOURCE_L2DATA), - .msr_base = IA32_L2_CBM_BASE, + .msr_base = MSR_IA32_L2_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 2, .cache = {
@@ -157,7 +154,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_L2CODE, .name = "L2CODE", .domains = domain_init(RDT_RESOURCE_L2CODE), - .msr_base = IA32_L2_CBM_BASE, + .msr_base = MSR_IA32_L2_CBM_BASE, .msr_update = cat_wrmsr, .cache_level = 2, .cache = {
@@ -174,7 +171,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_MBA, .name = "MB", .domains = domain_init(RDT_RESOURCE_MBA), - .msr_base = IA32_MBA_THRTL_BASE, + .msr_base = MSR_IA32_MBA_THRTL_BASE, .msr_update = mba_wrmsr, .cache_level = 3, .parse_ctrlval = parse_bw,
@@ -211,9 +208,10 @@ static inline void cache_alloc_hsw_probe(void)struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_L3 ];u32 l, h, max_cbm =BIT_MASK (20 ) -1 ; -if (wrmsr_safe (IA32_L3_CBM_BASE , max_cbm,0 )) +if (wrmsr_safe (MSR_IA32_L3_CBM_BASE , max_cbm,0 ))return ; -rdmsr (IA32_L3_CBM_BASE , l, h); + +rdmsr (MSR_IA32_L3_CBM_BASE , l, h);/* If all the bits were set in MSR, return success */ if (l != max_cbm)
arch/x86/kernel/cpu/resctrl/internal.h +12 -7
@@ -6,15 +6,18 @@#include <linux/kernfs.h> #include <linux/jump_label.h> -#define IA32_L3_QOS_CFG 0xc81 -#define IA32_L2_QOS_CFG 0xc82 -#define IA32_L3_CBM_BASE 0xc90 -#define IA32_L2_CBM_BASE 0xd10 -#define IA32_MBA_THRTL_BASE 0xd50 +#define MSR_IA32_L3_QOS_CFG 0xc81 +#define MSR_IA32_L2_QOS_CFG 0xc82 +#define MSR_IA32_L3_CBM_BASE 0xc90 +#define MSR_IA32_L2_CBM_BASE 0xd10 +#define MSR_IA32_MBA_THRTL_BASE 0xd50 -#define L3_QOS_CDP_ENABLE 0x01ULL +#define MSR_IA32_QM_CTR 0x0c8e +#define MSR_IA32_QM_EVTSEL 0x0c8d -#define L2_QOS_CDP_ENABLE 0x01ULL +#define L3_QOS_CDP_ENABLE 0x01ULL + +#define L2_QOS_CDP_ENABLE 0x01ULL/* * Event IDs are used to program IA32_QM_EVTSEL before reading event
@@ -29,6 +32,8 @@#define MBM_CNTR_WIDTH 24#define MBM_OVERFLOW_INTERVAL 1000#define MAX_MBA_BW 100u +#define MBA_IS_LINEAR 0x4 +#define MBA_MAX_MBPS U32_MAX#define RMID_VAL_ERROR BIT_ULL(63)#define RMID_VAL_UNAVAIL BIT_ULL(62)
arch/x86/kernel/cpu/resctrl/monitor.c +0 -3
@@ -28,9 +28,6 @@#include <asm/cpu_device_id.h> #include "internal.h" -#define MSR_IA32_QM_CTR 0x0c8e -#define MSR_IA32_QM_EVTSEL 0x0c8d -struct rmid_entry {u32 rmid;int busy;
arch/x86/kernel/cpu/resctrl/rdtgroup.c +2 -2
@@ -1722,14 +1722,14 @@ static void l3_qos_cfg_update(void *arg) {bool *enable = arg; -wrmsrl (IA32_L3_QOS_CFG , *enable ?L3_QOS_CDP_ENABLE :0ULL ); +wrmsrl (MSR_IA32_L3_QOS_CFG , *enable ?L3_QOS_CDP_ENABLE :0ULL ); }static void l2_qos_cfg_update (void *arg) {bool *enable = arg; -wrmsrl (IA32_L2_QOS_CFG , *enable ?L2_QOS_CDP_ENABLE :0ULL ); +wrmsrl (MSR_IA32_L2_QOS_CFG , *enable ?L2_QOS_CDP_ENABLE :0ULL ); }static inline bool is_mba_linear (void )
Moger, Babu
reviewerPATCH v9 5/13 x86/resctrl: Initialize the resource functions that are different
In reply to Moger, BabuInitialize the resource functions that are different between the vendors. Some features are initialized differently between the vendors. Add _intel suffix to Intel specific functions.
For example, MBA feature varies significantly between Intel and AMD. Separate the initialization of these resource functions. That way we can easily add AMD's functions later.
--- arch/x86/kernel/cpu/resctrl/core.c | 34 +++++++++++++++++++---- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 4 +-- arch/x86/kernel/cpu/resctrl/internal.h | 8 ++++-- 3 files changed, 37 insertions(+), 9 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +29 -5
@@ -57,7 +57,8 @@ int max_name_width, max_data_width;bool rdt_alloc_capable;static void -mba_wrmsr (struct rdt_domain *d,struct msr_param *m,struct rdt_resource *r); +mba_wrmsr_intel (struct rdt_domain *d,struct msr_param *m, +struct rdt_resource *r);static void cat_wrmsr (struct rdt_domain *d,struct msr_param *m,struct rdt_resource *r);
@@ -171,10 +172,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_MBA, .name = "MB", .domains = domain_init(RDT_RESOURCE_MBA), - .msr_base = MSR_IA32_MBA_THRTL_BASE, - .msr_update = mba_wrmsr, .cache_level = 3, - .parse_ctrlval = parse_bw, .format_str = "%d=%*u", .fflags = RFTYPE_RES_MB, },
@@ -357,7 +355,8 @@ u32 delay_bw_map(unsigned long bw, struct rdt_resource *r) }static void -mba_wrmsr (struct rdt_domain *d,struct msr_param *m,struct rdt_resource *r) +mba_wrmsr_intel (struct rdt_domain *d,struct msr_param *m, +struct rdt_resource *r) {unsigned int i;
@@ -874,6 +873,25 @@ static __init bool get_rdt_resources(void)return (rdt_mon_capable || rdt_alloc_capable); } +static __init void rdt_init_res_defs_intel (void ) +{ +struct rdt_resource *r; + +for_each_rdt_resource (r) { +if (r->rid ==RDT_RESOURCE_MBA ) { + r->msr_base =MSR_IA32_MBA_THRTL_BASE ; + r->msr_update = mba_wrmsr_intel; + r->parse_ctrlval = parse_bw_intel; + } + } +} + +static __init void rdt_init_res_defs (void ) +{ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) +rdt_init_res_defs_intel (); +} +static enum cpuhp_state rdt_online;static int __init resctrl_late_init (void )
@@ -881,6 +899,12 @@ static int __init resctrl_late_init(void)struct rdt_resource *r;int state, ret; +/* +* Initialize functions(or definitions) that are different +* between vendors here. +*/ +rdt_init_res_defs (); +check_quirks ();if (!get_rdt_resources ())
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +2 -2
@@ -64,8 +64,8 @@ static bool bw_validate(char *buf, unsigned long *data, struct rdt_resource *r)return true; } -int parse_bw (struct rdt_parse_data *data,struct rdt_resource *r, -struct rdt_domain *d) +int parse_bw_intel (struct rdt_parse_data *data,struct rdt_resource *r, +struct rdt_domain *d) {unsigned long bw_val;
arch/x86/kernel/cpu/resctrl/internal.h +6 -2
@@ -444,8 +444,8 @@ struct rdt_resource {int parse_cbm (struct rdt_parse_data *data,struct rdt_resource *r,struct rdt_domain *d); -int parse_bw (struct rdt_parse_data *data,struct rdt_resource *r, -struct rdt_domain *d); +int parse_bw_intel (struct rdt_parse_data *data,struct rdt_resource *r, +struct rdt_domain *d);extern struct mutex rdtgroup_mutex;
@@ -468,6 +468,10 @@ enum { RDT_NUM_RESOURCES, }; +#define for_each_rdt_resource (r) \ + for (r = rdt_resources_all; r < rdt_resources_all + RDT_NUM_RESOURCES;\ + r++) +#define for_each_capable_rdt_resource (r) \ for (r = rdt_resources_all; r < rdt_resources_all + RDT_NUM_RESOURCES;\ r++) \
Signature · 1 line
2.17.1
tip-bot for Babu Moger
merge bot[tip:x86/cache] x86/resctrl: Initialize the vendor-specific resource functions
x86/resctrl: Initialize the vendor-specific resource functions
Commit-ID: 1ad4fa41d99f606098fd7b90c1f5deb2b3dfae5d Gitweb: https://git.kernel.org/tip/1ad4fa41d99f606098fd7b90c1f5deb2b3dfae5d Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:33 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:19 +0100
x86/resctrl: Initialize the vendor-specific resource functions
Initialize the resource functions that are different between the vendors. Some features are initialized differently between the vendors. Add _intel suffix to Intel-specific functions.
For example, the MBA feature varies significantly between Intel and AMD. Separate the initialization of these resource functions. That way we can easily add AMD's functions later.
--- arch/x86/kernel/cpu/resctrl/core.c | 34 ++++++++++++++++++++++++++----- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 4 ++-- arch/x86/kernel/cpu/resctrl/internal.h | 8 ++++++-- 3 files changed, 37 insertions(+), 9 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +29 -5
@@ -57,7 +57,8 @@ int max_name_width, max_data_width;bool rdt_alloc_capable;static void -mba_wrmsr (struct rdt_domain *d,struct msr_param *m,struct rdt_resource *r); +mba_wrmsr_intel (struct rdt_domain *d,struct msr_param *m, +struct rdt_resource *r);static void cat_wrmsr (struct rdt_domain *d,struct msr_param *m,struct rdt_resource *r);
@@ -171,10 +172,7 @@ struct rdt_resource rdt_resources_all[] = { .rid = RDT_RESOURCE_MBA, .name = "MB", .domains = domain_init(RDT_RESOURCE_MBA), - .msr_base = MSR_IA32_MBA_THRTL_BASE, - .msr_update = mba_wrmsr, .cache_level = 3, - .parse_ctrlval = parse_bw, .format_str = "%d=%*u", .fflags = RFTYPE_RES_MB, },
@@ -357,7 +355,8 @@ u32 delay_bw_map(unsigned long bw, struct rdt_resource *r) }static void -mba_wrmsr (struct rdt_domain *d,struct msr_param *m,struct rdt_resource *r) +mba_wrmsr_intel (struct rdt_domain *d,struct msr_param *m, +struct rdt_resource *r) {unsigned int i;
@@ -874,6 +873,25 @@ static __init bool get_rdt_resources(void)return (rdt_mon_capable || rdt_alloc_capable); } +static __init void rdt_init_res_defs_intel (void ) +{ +struct rdt_resource *r; + +for_each_rdt_resource (r) { +if (r->rid ==RDT_RESOURCE_MBA ) { + r->msr_base =MSR_IA32_MBA_THRTL_BASE ; + r->msr_update = mba_wrmsr_intel; + r->parse_ctrlval = parse_bw_intel; + } + } +} + +static __init void rdt_init_res_defs (void ) +{ +if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) +rdt_init_res_defs_intel (); +} +static enum cpuhp_state rdt_online;static int __init resctrl_late_init (void )
@@ -881,6 +899,12 @@ static int __init resctrl_late_init(void)struct rdt_resource *r;int state, ret; +/* +* Initialize functions(or definitions) that are different +* between vendors here. +*/ +rdt_init_res_defs (); +check_quirks ();if (!get_rdt_resources ())
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +2 -2
@@ -64,8 +64,8 @@ static bool bw_validate(char *buf, unsigned long *data, struct rdt_resource *r)return true; } -int parse_bw (struct rdt_parse_data *data,struct rdt_resource *r, -struct rdt_domain *d) +int parse_bw_intel (struct rdt_parse_data *data,struct rdt_resource *r, +struct rdt_domain *d) {unsigned long bw_val;
arch/x86/kernel/cpu/resctrl/internal.h +6 -2
@@ -444,8 +444,8 @@ struct rdt_resource {int parse_cbm (struct rdt_parse_data *data,struct rdt_resource *r,struct rdt_domain *d); -int parse_bw (struct rdt_parse_data *data,struct rdt_resource *r, -struct rdt_domain *d); +int parse_bw_intel (struct rdt_parse_data *data,struct rdt_resource *r, +struct rdt_domain *d);extern struct mutex rdtgroup_mutex;
@@ -468,6 +468,10 @@ enum { RDT_NUM_RESOURCES, }; +#define for_each_rdt_resource (r) \ + for (r = rdt_resources_all; r < rdt_resources_all + RDT_NUM_RESOURCES;\ + r++) +#define for_each_capable_rdt_resource (r) \ for (r = rdt_resources_all; r < rdt_resources_all + RDT_NUM_RESOURCES;\ r++) \
Moger, Babu
reviewerPATCH v9 6/13 x86/resctrl: Bring cbm_validate function into the resource structure
In reply to Moger, BabuIdea is to bring all the functions that are different between the vendors into resource structure and initialize them dynamically. Add _intel suffix to Intel specific functions.
Following function is implemented separately for each vendors. cbm_validate : Cache bitmask validate function. AMD allows
non-contiguous masks. So, use separate functions for Intel and AMD.
--- arch/x86/kernel/cpu/resctrl/core.c | 9 ++++++++- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 4 ++-- arch/x86/kernel/cpu/resctrl/internal.h | 11 +++++++---- 3 files changed, 17 insertions(+), 7 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +8 -1
@@ -878,7 +878,14 @@ static __init void rdt_init_res_defs_intel(void)struct rdt_resource *r;for_each_rdt_resource (r) { -if (r->rid ==RDT_RESOURCE_MBA ) { +if (r->rid ==RDT_RESOURCE_L3 || + r->rid ==RDT_RESOURCE_L3DATA || + r->rid ==RDT_RESOURCE_L3CODE || + r->rid ==RDT_RESOURCE_L2 || + r->rid ==RDT_RESOURCE_L2DATA || + r->rid ==RDT_RESOURCE_L2CODE ) + r->cbm_validate = cbm_validate_intel; +else if (r->rid ==RDT_RESOURCE_MBA ) { r->msr_base =MSR_IA32_MBA_THRTL_BASE ; r->msr_update = mba_wrmsr_intel; r->parse_ctrlval = parse_bw_intel;
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +2 -2
@@ -88,7 +88,7 @@ int parse_bw_intel(struct rdt_parse_data *data, struct rdt_resource *r,* are allowed (e.g. FFFFH, 0FF0H, 003CH, etc.). * Additionally Haswell requires at least two bits set. */ -static bool cbm_validate (char *buf,u32 *data,struct rdt_resource *r) +bool cbm_validate_intel (char *buf,u32 *data,struct rdt_resource *r) {unsigned long first_bit, zero_bit, val;unsigned int cbm_len = r->cache.cbm_len;
@@ -148,7 +148,7 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,return -EINVAL ; } -if (!cbm_validate (data->buf, &cbm_val, r)) +if (r->cbm_validate && !r->cbm_validate (data->buf, &cbm_val, r))return -EINVAL ; if ((rdtgrp->mode == RDT_MODE_EXCLUSIVE ||
arch/x86/kernel/cpu/resctrl/internal.h +7 -4
@@ -410,10 +410,11 @@ struct rdt_parse_data {* @cache: Cache allocation related data * @format_str: Per resource format string to show domain value * @parse_ctrlval: Per resource function pointer to parse control values -* @evt_list: List of monitoring events -* @num_rmid: Number of RMIDs available -* @mon_scale: cqm counter * mon_scale = occupancy in bytes -* @fflags: flags to choose base and info files +* @cbm_validate Cache bitmask validate function +* @evt_list: List of monitoring events +* @num_rmid: Number of RMIDs available +* @mon_scale: cqm counter * mon_scale = occupancy in bytes +* @fflags: flags to choose base and info files */ struct rdt_resource {int rid;
@@ -436,6 +437,7 @@ struct rdt_resource {int (*parse_ctrlval)(struct rdt_parse_data *data,struct rdt_resource *r,struct rdt_domain *d); +bool (*cbm_validate)(char *buf,u32 *data,struct rdt_resource *r);struct list_head evt_list;int num_rmid;unsigned int mon_scale;
@@ -576,5 +578,6 @@ void cqm_setup_limbo_handler(struct rdt_domain *dom, unsigned long delay_ms);void cqm_handle_limbo (struct work_struct *work);bool has_busy_rmid (struct rdt_resource *r,struct rdt_domain *d);void __check_limbo (struct rdt_domain *d,bool force_free); +bool cbm_validate_intel (char *buf,u32 *data,struct rdt_resource *r);#endif /* _ASM_X86_RESCTRL_INTERNAL_H */
Signature · 1 line
2.17.1
tip-bot for Babu Moger
merge bot[tip:x86/cache] x86/resctrl: Bring cbm_validate() into the resource structure
x86/resctrl: Bring cbm_validate() into the resource structure
Commit-ID: a36c5ff560fb32d17c020fa3e7f4c70f753d8834 Gitweb: https://git.kernel.org/tip/a36c5ff560fb32d17c020fa3e7f4c70f753d8834 Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:35 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:19 +0100
x86/resctrl: Bring cbm_validate() into the resource structure
Bring all the functions that are different between the vendors into the resource structure and initialize them dynamically. Add _intel suffix to the Intel-specific functions.
cbm_validate() which does cache bitmask validation, differs between the vendors as AMD allows non-contiguous masks. So, use separate functions for Intel and AMD.
[ bp: Massage commit message and fixup rdt_resource members' vertical
alignment. ]
--- arch/x86/kernel/cpu/resctrl/core.c | 9 ++++++++- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 4 ++-- arch/x86/kernel/cpu/resctrl/internal.h | 15 +++++++++------ 3 files changed, 19 insertions(+), 9 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +8 -1
@@ -878,7 +878,14 @@ static __init void rdt_init_res_defs_intel(void)struct rdt_resource *r;for_each_rdt_resource (r) { -if (r->rid ==RDT_RESOURCE_MBA ) { +if (r->rid ==RDT_RESOURCE_L3 || + r->rid ==RDT_RESOURCE_L3DATA || + r->rid ==RDT_RESOURCE_L3CODE || + r->rid ==RDT_RESOURCE_L2 || + r->rid ==RDT_RESOURCE_L2DATA || + r->rid ==RDT_RESOURCE_L2CODE ) + r->cbm_validate = cbm_validate_intel; +else if (r->rid ==RDT_RESOURCE_MBA ) { r->msr_base =MSR_IA32_MBA_THRTL_BASE ; r->msr_update = mba_wrmsr_intel; r->parse_ctrlval = parse_bw_intel;
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +2 -2
@@ -88,7 +88,7 @@ int parse_bw_intel(struct rdt_parse_data *data, struct rdt_resource *r,* are allowed (e.g. FFFFH, 0FF0H, 003CH, etc.). * Additionally Haswell requires at least two bits set. */ -static bool cbm_validate (char *buf,u32 *data,struct rdt_resource *r) +bool cbm_validate_intel (char *buf,u32 *data,struct rdt_resource *r) {unsigned long first_bit, zero_bit, val;unsigned int cbm_len = r->cache.cbm_len;
@@ -148,7 +148,7 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,return -EINVAL ; } -if (!cbm_validate (data->buf, &cbm_val, r)) +if (r->cbm_validate && !r->cbm_validate (data->buf, &cbm_val, r))return -EINVAL ; if ((rdtgrp->mode == RDT_MODE_EXCLUSIVE ||
arch/x86/kernel/cpu/resctrl/internal.h +9 -6
@@ -396,9 +396,9 @@ struct rdt_parse_data {* struct rdt_resource - attributes of an RDT resource * @rid: The index of the resource * @alloc_enabled: Is allocation enabled on this machine -* @mon_enabled: Is monitoring enabled for this feature +* @mon_enabled: Is monitoring enabled for this feature * @alloc_capable: Is allocation available on this machine -* @mon_capable: Is monitor feature available on this machine +* @mon_capable: Is monitor feature available on this machine * @name: Name to use in "schemata" file * @num_closid: Number of CLOSIDs available * @cache_level: Which cache level defines scope of this resource
@@ -410,10 +410,11 @@ struct rdt_parse_data {* @cache: Cache allocation related data * @format_str: Per resource format string to show domain value * @parse_ctrlval: Per resource function pointer to parse control values -* @evt_list: List of monitoring events -* @num_rmid: Number of RMIDs available -* @mon_scale: cqm counter * mon_scale = occupancy in bytes -* @fflags: flags to choose base and info files +* @cbm_validate Cache bitmask validate function +* @evt_list: List of monitoring events +* @num_rmid: Number of RMIDs available +* @mon_scale: cqm counter * mon_scale = occupancy in bytes +* @fflags: flags to choose base and info files */ struct rdt_resource {int rid;
@@ -436,6 +437,7 @@ struct rdt_resource {int (*parse_ctrlval)(struct rdt_parse_data *data,struct rdt_resource *r,struct rdt_domain *d); +bool (*cbm_validate)(char *buf,u32 *data,struct rdt_resource *r);struct list_head evt_list;int num_rmid;unsigned int mon_scale;
@@ -576,5 +578,6 @@ void cqm_setup_limbo_handler(struct rdt_domain *dom, unsigned long delay_ms);void cqm_handle_limbo (struct work_struct *work);bool has_busy_rmid (struct rdt_resource *r,struct rdt_domain *d);void __check_limbo (struct rdt_domain *d,bool force_free); +bool cbm_validate_intel (char *buf,u32 *data,struct rdt_resource *r);#endif /* _ASM_X86_RESCTRL_INTERNAL_H */
Moger, Babu
reviewerPATCH v9 7/13 x86/resctrl: Add vendor check for MBA software controller
In reply to Moger, BabuMBA software controller support is available only on Intel.
--- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +4 -1
@@ -1878,7 +1878,10 @@ static int parse_rdtgroupfs_options(char *data)if (ret) goto out; }else if (!strcmp (token,"mba_MBps" )) { - ret =set_mba_sc (true); +if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) + ret =set_mba_sc (true); +else + ret = -EINVAL ;if (ret) goto out; }else {
Signature · 1 line
2.17.1
tip-bot for Babu Moger
merge bot[tip:x86/cache] x86/resctrl: Add vendor check for the MBA software controller
x86/resctrl: Add vendor check for the MBA software controller
Commit-ID: 580ebb66cbb3de979adb4becbad7bd383746f162 Gitweb: https://git.kernel.org/tip/580ebb66cbb3de979adb4becbad7bd383746f162 Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:37 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:19 +0100
x86/resctrl: Add vendor check for the MBA software controller
MBA software controller support is available only on Intel.
--- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +4 -1
@@ -1878,7 +1878,10 @@ static int parse_rdtgroupfs_options(char *data)if (ret) goto out; }else if (!strcmp (token,"mba_MBps" )) { - ret =set_mba_sc (true); +if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) + ret =set_mba_sc (true); +else + ret = -EINVAL ;if (ret) goto out; }else {
Moger, Babu
reviewerPATCH v9 8/13 x86/resctrl: Rename config parameter INTEL_RDT to RESCTRL
In reply to Moger, BabuResource control feature is supported by both Intel and AMD. So, rename the INTEL_RDT to vendor neutral RESCTRL.
Now CONFIG_RESCTRL will be used for both Intel and AMD to enable Resource Control support. Update the texts in config and condition accordingly.
--- arch/x86/Kconfig | 23 ++++++++++++++++------- arch/x86/include/asm/resctrl_sched.h | 4 ++-- arch/x86/kernel/cpu/Makefile | 2 +- arch/x86/kernel/cpu/resctrl/Makefile | 4 ++-- include/linux/sched.h | 2 +- 5 files changed, 22 insertions(+), 13 deletions(-)
arch/x86/Kconfig +16 -7
@@ -445,16 +445,25 @@ config RETPOLINE code are eliminated. Since this includes the syscall entry path, it is not entirely pointless. -config INTEL_RDT - bool "Intel Resource Director Technology support" +config RESCTRL + bool "Resource Control support" default n - depends on X86 && CPU_SUP_INTEL + depends on X86 && (CPU_SUP_INTEL || CPU_SUP_AMD) select KERNFS help - Select to enable resource allocation and monitoring which are - sub-features of Intel Resource Director Technology(RDT). More - information about RDT can be found in the Intel x86 - Architecture Software Developer Manual. + Select to enable Resource Control feature support. + + These features are intended to provide the support for the allocation + and monitoring of the usage of certain system resources by one or more + processors. + + Intel refers to this feature as Intel Resource Director Technology + (Intel(R) RDT). More information about RDT can be found in the + Intel x86 Architecture Software Developer Manual. + + AMD refers to this feature as AMD Platform Quality of Service(AMD QoS). + More information about AMD QoS can be found in AMD64 Technology + Platform Quality of Service Extensions manual. Say N if unsure.
arch/x86/include/asm/resctrl_sched.h +2 -2
@@ -2,7 +2,7 @@#ifndef _ASM_X86_RESCTRL_SCHED_H#define _ASM_X86_RESCTRL_SCHED_H -#ifdef CONFIG_INTEL_RDT +#ifdef CONFIG_RESCTRL #include <linux/sched.h> #include <linux/jump_label.h>
@@ -88,6 +88,6 @@ static inline void resctrl_sched_in(void)static inline void resctrl_sched_in (void ) {} -#endif /* CONFIG_INTEL_RDT */ +#endif /* CONFIG_RESCTRL */ #endif /* _ASM_X86_RESCTRL_SCHED_H */
arch/x86/kernel/cpu/Makefile +1 -1
@@ -38,7 +38,7 @@ obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o obj-$(CONFIG_X86_MCE) += mcheck/ obj-$(CONFIG_MTRR) += mtrr/ obj-$(CONFIG_MICROCODE) += microcode/ -obj-$(CONFIG_INTEL_RDT) += resctrl/ +obj-$(CONFIG_RESCTRL) += resctrl/ obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o
arch/x86/kernel/cpu/resctrl/Makefile +2 -2
@@ -2,6 +2,6 @@ # Makefile for resource control feature code # -obj-$(CONFIG_INTEL_RDT) += core.o rdtgroup.o monitor.o -obj-$(CONFIG_INTEL_RDT) += ctrlmondata.o pseudo_lock.o +obj-$(CONFIG_RESCTRL) += core.o rdtgroup.o monitor.o +obj-$(CONFIG_RESCTRL) += ctrlmondata.o pseudo_lock.o CFLAGS_pseudo_lock.o = -I$(src)
include/linux/sched.h +1 -1
@@ -985,7 +985,7 @@ struct task_struct {/* cg_list protected by css_set_lock and tsk->alloc_lock: */ struct list_head cg_list;#endif -#ifdef CONFIG_INTEL_RDT +#ifdef CONFIG_RESCTRL u32 closid;u32 rmid;#endif
Signature · 1 line
2.17.1
Borislav Petkov
maintainerRe: [PATCH v9 08/13] x86/resctrl: Rename config parameter INTEL_RDT to RESCTRL
Quoted 40 lines · Babu
Resource control feature is supported by both Intel and AMD. So, rename the
INTEL_RDTto vendor neutral RESCTRL.Now
CONFIG_RESCTRLwill be used for both Intel and AMD to enable Resource Control support. Update the texts in config and condition accordingly.Signed-off-by Babu Moger <[email protected]>--- arch/x86/Kconfig | 23 ++++++++++++++++------- arch/x86/include/asm/resctrl_sched.h | 4 ++-- arch/x86/kernel/cpu/Makefile | 2 +- arch/x86/kernel/cpu/resctrl/Makefile | 4 ++-- include/linux/sched.h | 2 +- 5 files changed, 22 insertions(+), 13 deletions(-)arch/x86/Kconfig +6
-7@@ -445,16 +445,25 @@ config RETPOLINE code are eliminated. Since this includes the syscall entry path, it is not entirely pointless. -config INTEL_RDT - bool "Intel Resource Director Technology support" +config RESCTRL + bool "Resource Control support" default n - depends on X86 && CPU_SUP_INTEL + depends on X86 && (CPU_SUP_INTEL || CPU_SUP_AMD) select KERNFS help - Select to enable resource allocation and monitoring which are - sub-features of Intel Resource Director Technology(RDT). More - information about RDT can be found in the Intel x86 - Architecture Software Developer Manual. + Select to enable Resource Control feature support. + + These features are intended to provide the support for the allocation
arch/x86/Kconfig config RETPOLINE · line 466
Quoted 10 lines
arch/x86/Kconfig +10
-0+ and monitoring of the usage of certain system resources by one or more + processors. + + Intel refers to this feature as Intel Resource Director Technology + (Intel(R) RDT). More information about RDT can be found in the + Intel x86 Architecture Software Developer Manual. + + AMD refers to this feature as AMD Platform Quality of Service(AMD QoS). + More information about AMD QoS can be found in AMD64 Technology + Platform Quality of Service Extensions manual.
I've simplified this to:
Enable Resource Control support.
Provide support for the allocation and monitoring of system resources usage by the CPU.
Intel calls this Intel Resource Director Technology (Intel(R) RDT). More information about RDT can be found in the Intel x86 Architecture Software Developer Manual.
AMD calls this AMD Platform Quality of Service (AMD QoS). More information about AMD QoS can be found in the AMD64 Technology Platform Quality of Service Extensions manual.
Say N if unsure.
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.tip-bot for Babu Moger
merge bot[tip:x86/cache] x86/resctrl: Rename the config option INTEL_RDT to RESCTRL
In reply to Moger, Babux86/resctrl: Rename the config option INTEL_RDT to RESCTRL
Commit-ID: 6fe07ce35e8ad870ba1cf82e0481e0fc0f526eff Gitweb: https://git.kernel.org/tip/6fe07ce35e8ad870ba1cf82e0481e0fc0f526eff Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:39 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:19 +0100
x86/resctrl: Rename the config option INTEL_RDT to RESCTRL
The resource control feature is supported by both Intel and AMD. So, rename CONFIG_INTEL_RDT to the vendor-neutral CONFIG_RESCTRL.
Now CONFIG_RESCTRL will be used for both Intel and AMD to enable Resource Control support. Update the texts in config and condition accordingly.
[ bp: Simplify Kconfig text. ]
--- arch/x86/Kconfig | 22 +++++++++++++++------- arch/x86/include/asm/resctrl_sched.h | 4 ++-- arch/x86/kernel/cpu/Makefile | 2 +- arch/x86/kernel/cpu/resctrl/Makefile | 4 ++-- include/linux/sched.h | 2 +- 5 files changed, 21 insertions(+), 13 deletions(-)
arch/x86/Kconfig +15 -7
@@ -448,15 +448,23 @@ config RETPOLINE code are eliminated. Since this includes the syscall entry path, it is not entirely pointless. -config INTEL_RDT - bool "Intel Resource Director Technology support" - depends on X86 && CPU_SUP_INTEL +config RESCTRL + bool "Resource Control support" + depends on X86 && (CPU_SUP_INTEL || CPU_SUP_AMD) select KERNFS help - Select to enable resource allocation and monitoring which are - sub-features of Intel Resource Director Technology(RDT). More - information about RDT can be found in the Intel x86 - Architecture Software Developer Manual. + Enable Resource Control support. + + Provide support for the allocation and monitoring of system resources + usage by the CPU. + + Intel calls this Intel Resource Director Technology + (Intel(R) RDT). More information about RDT can be found in the + Intel x86 Architecture Software Developer Manual. + + AMD calls this AMD Platform Quality of Service (AMD QoS). + More information about AMD QoS can be found in the AMD64 Technology + Platform Quality of Service Extensions manual. Say N if unsure.
arch/x86/include/asm/resctrl_sched.h +2 -2
@@ -2,7 +2,7 @@#ifndef _ASM_X86_RESCTRL_SCHED_H#define _ASM_X86_RESCTRL_SCHED_H -#ifdef CONFIG_INTEL_RDT +#ifdef CONFIG_RESCTRL #include <linux/sched.h> #include <linux/jump_label.h>
@@ -88,6 +88,6 @@ static inline void resctrl_sched_in(void)static inline void resctrl_sched_in (void ) {} -#endif /* CONFIG_INTEL_RDT */ +#endif /* CONFIG_RESCTRL */ #endif /* _ASM_X86_RESCTRL_SCHED_H */
arch/x86/kernel/cpu/Makefile +1 -1
@@ -39,7 +39,7 @@ obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o obj-$(CONFIG_X86_MCE) += mcheck/ obj-$(CONFIG_MTRR) += mtrr/ obj-$(CONFIG_MICROCODE) += microcode/ -obj-$(CONFIG_INTEL_RDT) += resctrl/ +obj-$(CONFIG_RESCTRL) += resctrl/ obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o
arch/x86/kernel/cpu/resctrl/Makefile +2 -2
@@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0 -obj-$(CONFIG_INTEL_RDT) += core.o rdtgroup.o monitor.o -obj-$(CONFIG_INTEL_RDT) += ctrlmondata.o pseudo_lock.o +obj-$(CONFIG_RESCTRL) += core.o rdtgroup.o monitor.o +obj-$(CONFIG_RESCTRL) += ctrlmondata.o pseudo_lock.o CFLAGS_pseudo_lock.o = -I$(src)
include/linux/sched.h +1 -1
@@ -993,7 +993,7 @@ struct task_struct {/* cg_list protected by css_set_lock and tsk->alloc_lock: */ struct list_head cg_list;#endif -#ifdef CONFIG_INTEL_RDT +#ifdef CONFIG_RESCTRL u32 closid;u32 rmid;#endif
Moger, Babu
reviewerPATCH v9 9/13 x86/resctrl: Add AMD feature bit X86_FEATURE_MBA in cpuid bits array
In reply to Moger, BabuFrom: Sherry Hurwitz <[email protected]>
The feature bit X86_FEATURE_MBA is detected via CPUID leaf 0x80000008 EBX Bit 06. This bit indicates the support of AMD's MBA feature.
This feature is supported by both Intel and AMD. But they are detected in different CPUID leaves.
--- arch/x86/kernel/cpu/scattered.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
arch/x86/kernel/cpu/scattered.c +6 -1
@@ -17,7 +17,11 @@ struct cpuid_bit {u32 sub_leaf; }; -/* Please keep the leaf sorted by cpuid_bit.level for faster search. */ +/* +* Please keep the leaf sorted by cpuid_bit.level for faster search. +* X86_FEATURE_MBA supported by both Intel and AMD. But the cpuid +* levels are different. Add a separate enty for each. +*/ static const struct cpuid_bit cpuid_bits[] = { {X86_FEATURE_APERFMPERF ,CPUID_ECX ,0 ,0x00000006 ,0 }, {X86_FEATURE_EPB ,CPUID_ECX ,3 ,0x00000006 ,0 },
@@ -29,6 +33,7 @@ static const struct cpuid_bit cpuid_bits[] = { { X86_FEATURE_HW_PSTATE, CPUID_EDX, 7, 0x80000007, 0 }, { X86_FEATURE_CPB, CPUID_EDX, 9, 0x80000007, 0 }, { X86_FEATURE_PROC_FEEDBACK, CPUID_EDX, 11, 0x80000007, 0 }, + { X86_FEATURE_MBA, CPUID_EBX, 6, 0x80000008, 0 }, { X86_FEATURE_SME, CPUID_EAX, 0, 0x8000001f, 0 }, { X86_FEATURE_SEV, CPUID_EAX, 1, 0x8000001f, 0 }, { 0, 0, 0, 0, 0 }
Signature · 1 line
2.17.1
tip-bot for Sherry Hurwitz
merge bot[tip:x86/cache] x86/resctrl: Add AMD's X86_FEATURE_MBA to the scattered CPUID features
x86/resctrl: Add AMD's X86_FEATURE_MBA to the scattered CPUID features
Commit-ID: 9f72f855a6cdbb5313787145a69b474cd9f55f28 Gitweb: https://git.kernel.org/tip/9f72f855a6cdbb5313787145a69b474cd9f55f28 Author: Sherry Hurwitz <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:41 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:19 +0100
x86/resctrl: Add AMD's X86_FEATURE_MBA to the scattered CPUID features
The feature bit X86_FEATURE_MBA is detected via CPUID leaf 0x80000008 EBX Bit 06. This bit indicates the support of AMD's MBA feature.
This feature is supported by both Intel and AMD. But they are detected in different CPUID leaves.
[ bp: s/cpuid/CPUID/g ]
--- arch/x86/kernel/cpu/scattered.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
arch/x86/kernel/cpu/scattered.c +6 -1
@@ -17,7 +17,11 @@ struct cpuid_bit {u32 sub_leaf; }; -/* Please keep the leaf sorted by cpuid_bit.level for faster search. */ +/* +* Please keep the leaf sorted by cpuid_bit.level for faster search. +* X86_FEATURE_MBA is supported by both Intel and AMD. But the CPUID +* levels are different and there is a separate entry for each. +*/ static const struct cpuid_bit cpuid_bits[] = { {X86_FEATURE_APERFMPERF ,CPUID_ECX ,0 ,0x00000006 ,0 }, {X86_FEATURE_EPB ,CPUID_ECX ,3 ,0x00000006 ,0 },
@@ -29,6 +33,7 @@ static const struct cpuid_bit cpuid_bits[] = { { X86_FEATURE_HW_PSTATE, CPUID_EDX, 7, 0x80000007, 0 }, { X86_FEATURE_CPB, CPUID_EDX, 9, 0x80000007, 0 }, { X86_FEATURE_PROC_FEEDBACK, CPUID_EDX, 11, 0x80000007, 0 }, + { X86_FEATURE_MBA, CPUID_EBX, 6, 0x80000008, 0 }, { X86_FEATURE_SME, CPUID_EAX, 0, 0x8000001f, 0 }, { X86_FEATURE_SEV, CPUID_EAX, 1, 0x8000001f, 0 }, { 0, 0, 0, 0, 0 }
Moger, Babu
reviewerPATCH v9 10/13 x86/resctrl: Fix the messages in rdt_last_cmd_printf and rdt_last_cmd_puts
In reply to Moger, BabuFix the messages in rdt_last_cmd_printf and rdt_last_cmd_puts to make it more meaningful and consistent.
--- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 22 +++++++------- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 34 ++++++++++----------- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 36 +++++++++++------------ 3 files changed, 46 insertions(+), 46 deletions(-)
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +11 -11
@@ -70,7 +70,7 @@ int parse_bw_intel(struct rdt_parse_data *data, struct rdt_resource *r,unsigned long bw_val;if (d->have_new_ctrl) { -rdt_last_cmd_printf ("duplicate domain %d\n" , d->id); +rdt_last_cmd_printf ("Duplicate domain %d\n" , d->id);return -EINVAL ; }
@@ -96,12 +96,12 @@ bool cbm_validate_intel(char *buf, u32 *data, struct rdt_resource *r) ret =kstrtoul (buf,16 , &val);if (ret) { -rdt_last_cmd_printf ("non-hex character in mask %s\n" , buf); +rdt_last_cmd_printf ("Non-hex character in the mask %s\n" , buf);return false; }if (val ==0 || val > r->default_ctrl) { -rdt_last_cmd_puts ("mask out of range\n" ); +rdt_last_cmd_puts ("Mask out of range\n" );return false; }
@@ -109,12 +109,12 @@ bool cbm_validate_intel(char *buf, u32 *data, struct rdt_resource *r) zero_bit =find_next_zero_bit (&val, cbm_len, first_bit);if (find_next_bit (&val, cbm_len, zero_bit) < cbm_len) { -rdt_last_cmd_printf ("mask %lx has non-consecutive 1-bits\n" , val); +rdt_last_cmd_printf ("The mask %lx has non-consecutive 1-bits\n" , val);return false; }if ((zero_bit - first_bit) < r->cache.min_cbm_bits) { -rdt_last_cmd_printf ("Need at least %d bits in mask\n" , +rdt_last_cmd_printf ("Need at least %d bits in the mask\n" , r->cache.min_cbm_bits);return false; }
@@ -134,7 +134,7 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,u32 cbm_val;if (d->have_new_ctrl) { -rdt_last_cmd_printf ("duplicate domain %d\n" , d->id); +rdt_last_cmd_printf ("Duplicate domain %d\n" , d->id);return -EINVAL ; }
@@ -144,7 +144,7 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,*/ if (rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP &&rdtgroup_pseudo_locked_in_hierarchy (d)) { -rdt_last_cmd_printf ("pseudo-locked region in hierarchy\n" ); +rdt_last_cmd_printf ("Pseudo-locked region in hierarchy\n" );return -EINVAL ; }
@@ -163,14 +163,14 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,* either is exclusive. */ if (rdtgroup_cbm_overlaps (r, d, cbm_val, rdtgrp->closid, true)) { -rdt_last_cmd_printf ("overlaps with exclusive group\n" ); +rdt_last_cmd_printf ("Overlaps with exclusive group\n" );return -EINVAL ; }if (rdtgroup_cbm_overlaps (r, d, cbm_val, rdtgrp->closid, false)) {if (rdtgrp->mode ==RDT_MODE_EXCLUSIVE || rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP ) { -rdt_last_cmd_printf ("overlaps with other group\n" ); +rdt_last_cmd_printf ("0verlaps with other group\n" );return -EINVAL ; } }
@@ -292,7 +292,7 @@ static int rdtgroup_parse_resource(char *resname, char *tok,if (!strcmp (resname, r->name) && rdtgrp->closid < r->num_closid)return parse_line (tok, r, rdtgrp); } -rdt_last_cmd_printf ("unknown/unsupported resource name '%s'\n" , resname); +rdt_last_cmd_printf ("Unknown or unsupported resource name '%s'\n" , resname);return -EINVAL ; }
@@ -323,7 +323,7 @@ ssize_t rdtgroup_schemata_write(struct kernfs_open_file *of,*/ if (rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKED ) { ret = -EINVAL ; -rdt_last_cmd_puts ("resource group is pseudo-locked\n" ); +rdt_last_cmd_puts ("Resource group is pseudo-locked\n" ); goto out; }
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +17 -17
@@ -213,7 +213,7 @@ static int pseudo_lock_cstates_constrain(struct pseudo_lock_region *plr)for_each_cpu (cpu, &plr->d->cpu_mask) { pm_req =kzalloc (sizeof (*pm_req),GFP_KERNEL );if (!pm_req) { -rdt_last_cmd_puts ("fail allocating mem for PM QoS\n" ); +rdt_last_cmd_puts ("Failure to allocate mem for PM QoS\n" ); ret = -ENOMEM ; goto out_err; }
@@ -222,7 +222,7 @@ static int pseudo_lock_cstates_constrain(struct pseudo_lock_region *plr)DEV_PM_QOS_RESUME_LATENCY , 30);if (ret <0 ) { -rdt_last_cmd_printf ("fail to add latency req cpu%d\n" , +rdt_last_cmd_printf ("Failed to add latency req cpu%d\n" , cpu);kfree (pm_req); ret =-1 ;
@@ -289,7 +289,7 @@ static int pseudo_lock_region_init(struct pseudo_lock_region *plr) plr->cpu =cpumask_first (&plr->d->cpu_mask);if (!cpu_online (plr->cpu)) { -rdt_last_cmd_printf ("cpu %u associated with cache not online\n" , +rdt_last_cmd_printf ("CPU %u associated with cache not online\n" , plr->cpu); ret = -ENODEV ; goto out_region;
@@ -307,7 +307,7 @@ static int pseudo_lock_region_init(struct pseudo_lock_region *plr) } ret =-1 ; -rdt_last_cmd_puts ("unable to determine cache line size\n" ); +rdt_last_cmd_puts ("Unable to determine cache line size\n" ); out_region:pseudo_lock_region_clear (plr);return ret;
@@ -361,14 +361,14 @@ static int pseudo_lock_region_alloc(struct pseudo_lock_region *plr)* KMALLOC_MAX_SIZE. */ if (plr->size >KMALLOC_MAX_SIZE ) { -rdt_last_cmd_puts ("requested region exceeds maximum size\n" ); +rdt_last_cmd_puts ("Requested region exceeds maximum size\n" ); ret = -E2BIG ; goto out_region; } plr->kmem =kzalloc (plr->size,GFP_KERNEL );if (!plr->kmem) { -rdt_last_cmd_puts ("unable to allocate memory\n" ); +rdt_last_cmd_puts ("Unable to allocate memory\n" ); ret = -ENOMEM ; goto out_region; }
@@ -665,7 +665,7 @@ int rdtgroup_locksetup_enter(struct rdtgroup *rdtgrp)* default closid associated with it. */ if (rdtgrp == &rdtgroup_default) { -rdt_last_cmd_puts ("cannot pseudo-lock default group\n" ); +rdt_last_cmd_puts ("Cannot pseudo-lock default group\n" );return -EINVAL ; }
@@ -707,17 +707,17 @@ int rdtgroup_locksetup_enter(struct rdtgroup *rdtgrp)*/ prefetch_disable_bits =get_prefetch_disable_bits ();if (prefetch_disable_bits ==0 ) { -rdt_last_cmd_puts ("pseudo-locking not supported\n" ); +rdt_last_cmd_puts ("Pseudo-locking not supported\n" );return -EINVAL ; }if (rdtgroup_monitor_in_progress (rdtgrp)) { -rdt_last_cmd_puts ("monitoring in progress\n" ); +rdt_last_cmd_puts ("Monitoring in progress\n" );return -EINVAL ; }if (rdtgroup_tasks_assigned (rdtgrp)) { -rdt_last_cmd_puts ("tasks assigned to resource group\n" ); +rdt_last_cmd_puts ("Tasks assigned to resource group\n" );return -EINVAL ; }
@@ -727,13 +727,13 @@ int rdtgroup_locksetup_enter(struct rdtgroup *rdtgrp) }if (rdtgroup_locksetup_user_restrict (rdtgrp)) { -rdt_last_cmd_puts ("unable to modify resctrl permissions\n" ); +rdt_last_cmd_puts ("Unable to modify resctrl permissions\n" );return -EIO ; } ret =pseudo_lock_init (rdtgrp);if (ret) { -rdt_last_cmd_puts ("unable to init pseudo-lock region\n" ); +rdt_last_cmd_puts ("Unable to init pseudo-lock region\n" ); goto out_release; }
@@ -770,7 +770,7 @@ int rdtgroup_locksetup_exit(struct rdtgroup *rdtgrp)if (rdt_mon_capable) { ret =alloc_rmid ();if (ret <0 ) { -rdt_last_cmd_puts ("out of RMIDs\n" ); +rdt_last_cmd_puts ("Out of RMIDs\n" );return ret; } rdtgrp->mon.rmid = ret;
@@ -1304,7 +1304,7 @@ int rdtgroup_pseudo_lock_create(struct rdtgroup *rdtgrp) "pseudo_lock/%u", plr->cpu);if (IS_ERR (thread)) { ret =PTR_ERR (thread); -rdt_last_cmd_printf ("locking thread returned error %d\n" , ret); +rdt_last_cmd_printf ("Locking thread returned error %d\n" , ret); goto out_cstates; }
@@ -1322,13 +1322,13 @@ int rdtgroup_pseudo_lock_create(struct rdtgroup *rdtgrp)* the cleared, but not freed, plr struct resulting in an * empty pseudo-locking loop. */ -rdt_last_cmd_puts ("locking thread interrupted\n" ); +rdt_last_cmd_puts ("Locking thread interrupted\n" ); goto out_cstates; } ret =pseudo_lock_minor_get (&new_minor);if (ret <0 ) { -rdt_last_cmd_puts ("unable to obtain a new minor number\n" ); +rdt_last_cmd_puts ("Unable to obtain a new minor number\n" ); goto out_cstates; }
@@ -1360,7 +1360,7 @@ int rdtgroup_pseudo_lock_create(struct rdtgroup *rdtgrp)if (IS_ERR (dev)) { ret =PTR_ERR (dev); -rdt_last_cmd_printf ("failed to create character device: %d\n" , +rdt_last_cmd_printf ("Failed to create character device: %d\n" , ret); goto out_debugfs; }
arch/x86/kernel/cpu/resctrl/rdtgroup.c +18 -18
@@ -345,7 +345,7 @@ static int cpus_mon_write(struct rdtgroup *rdtgrp, cpumask_var_t newmask,/* Check whether cpus belong to parent ctrl group */ cpumask_andnot (tmpmask, newmask, &prgrp->cpu_mask);if (cpumask_weight (tmpmask)) { -rdt_last_cmd_puts ("can only add CPUs to mongroup that belong to parent\n" ); +rdt_last_cmd_puts ("Can only add CPUs to mongroup that belong to parent\n" );return -EINVAL ; }
@@ -470,14 +470,14 @@ static ssize_t rdtgroup_cpus_write(struct kernfs_open_file *of,rdt_last_cmd_clear ();if (!rdtgrp) { ret = -ENOENT ; -rdt_last_cmd_puts ("directory was removed\n" ); +rdt_last_cmd_puts ("Directory was removed\n" ); goto unlock; }if (rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKED || rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP ) { ret = -EINVAL ; -rdt_last_cmd_puts ("pseudo-locking in progress\n" ); +rdt_last_cmd_puts ("Pseudo-locking in progress\n" ); goto unlock; }
@@ -487,7 +487,7 @@ static ssize_t rdtgroup_cpus_write(struct kernfs_open_file *of, ret =cpumask_parse (buf, newmask);if (ret) { -rdt_last_cmd_puts ("bad cpu list/mask\n" ); +rdt_last_cmd_puts ("Bad cpu list/mask\n" ); goto unlock; }
@@ -495,7 +495,7 @@ static ssize_t rdtgroup_cpus_write(struct kernfs_open_file *of,cpumask_andnot (tmpmask, newmask, cpu_online_mask);if (cpumask_weight (tmpmask)) { ret = -EINVAL ; -rdt_last_cmd_puts ("can only assign online cpus\n" ); +rdt_last_cmd_puts ("Can only assign online cpus\n" ); goto unlock; }
@@ -574,7 +574,7 @@ static int __rdtgroup_move_task(struct task_struct *tsk,*/ atomic_dec (&rdtgrp->waitcount);kfree (callback); -rdt_last_cmd_puts ("task exited\n" ); +rdt_last_cmd_puts ("Task exited\n" ); }else {/* * For ctrl_mon groups move both closid and rmid.
@@ -692,7 +692,7 @@ static ssize_t rdtgroup_tasks_write(struct kernfs_open_file *of,if (rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKED || rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP ) { ret = -EINVAL ; -rdt_last_cmd_puts ("pseudo-locking in progress\n" ); +rdt_last_cmd_puts ("Pseudo-locking in progress\n" ); goto unlock; }
@@ -1158,14 +1158,14 @@ static bool rdtgroup_mode_test_exclusive(struct rdtgroup *rdtgrp)list_for_each_entry (d, &r->domains, list) {if (rdtgroup_cbm_overlaps (r, d, d->ctrl_val[closid], rdtgrp->closid, false)) { -rdt_last_cmd_puts ("schemata overlaps\n" ); +rdt_last_cmd_puts ("Schemata overlaps\n" );return false; } } }if (!has_cache) { -rdt_last_cmd_puts ("cannot be exclusive without CAT/CDP\n" ); +rdt_last_cmd_puts ("Cannot be exclusive without CAT/CDP\n" );return false; }
@@ -1206,7 +1206,7 @@ static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, goto out;if (mode ==RDT_MODE_PSEUDO_LOCKED ) { -rdt_last_cmd_printf ("cannot change pseudo-locked group\n" ); +rdt_last_cmd_printf ("Cannot change pseudo-locked group\n" ); ret = -EINVAL ; goto out; }
@@ -1235,7 +1235,7 @@ static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, goto out; rdtgrp->mode =RDT_MODE_PSEUDO_LOCKSETUP ; }else { -rdt_last_cmd_printf ("unknown/unsupported mode\n" ); +rdt_last_cmd_printf ("Unknown orunsupported mode\n" ); ret = -EINVAL ; }
@@ -2543,7 +2543,7 @@ static int rdtgroup_init_alloc(struct rdtgroup *rdtgrp) tmp_cbm = d->new_ctrl;if (bitmap_weight (&tmp_cbm, r->cache.cbm_len) < r->cache.min_cbm_bits) { -rdt_last_cmd_printf ("no space on %s:%d\n" , +rdt_last_cmd_printf ("No space on %s:%d\n" , r->name, d->id);return -ENOSPC ; }
@@ -2560,7 +2560,7 @@ static int rdtgroup_init_alloc(struct rdtgroup *rdtgrp)continue ; ret =update_domains (r, rdtgrp->closid);if (ret <0 ) { -rdt_last_cmd_puts ("failed to initialize allocations\n" ); +rdt_last_cmd_puts ("Failed to initialize allocations\n" );return ret; } rdtgrp->mode =RDT_MODE_SHAREABLE ;
@@ -2583,7 +2583,7 @@ static int mkdir_rdt_prepare(struct kernfs_node *parent_kn,rdt_last_cmd_clear ();if (!prdtgrp) { ret = -ENODEV ; -rdt_last_cmd_puts ("directory was removed\n" ); +rdt_last_cmd_puts ("Directory was removed\n" ); goto out_unlock; }
@@ -2591,7 +2591,7 @@ static int mkdir_rdt_prepare(struct kernfs_node *parent_kn, (prdtgrp->mode == RDT_MODE_PSEUDO_LOCKSETUP || prdtgrp->mode == RDT_MODE_PSEUDO_LOCKED)) { ret = -EINVAL ; -rdt_last_cmd_puts ("pseudo-locking in progress\n" ); +rdt_last_cmd_puts ("Pseudo-locking in progress\n" ); goto out_unlock; }
@@ -2599,7 +2599,7 @@ static int mkdir_rdt_prepare(struct kernfs_node *parent_kn, rdtgrp =kzalloc (sizeof (*rdtgrp),GFP_KERNEL );if (!rdtgrp) { ret = -ENOSPC ; -rdt_last_cmd_puts ("kernel out of memory\n" ); +rdt_last_cmd_puts ("Kernel out of memory\n" ); goto out_unlock; } *r = rdtgrp;
@@ -2640,7 +2640,7 @@ static int mkdir_rdt_prepare(struct kernfs_node *parent_kn,if (rdt_mon_capable) { ret =alloc_rmid ();if (ret <0 ) { -rdt_last_cmd_puts ("out of RMIDs\n" ); +rdt_last_cmd_puts ("Out of RMIDs\n" ); goto out_destroy; } rdtgrp->mon.rmid = ret;
@@ -2728,7 +2728,7 @@ static int rdtgroup_mkdir_ctrl_mon(struct kernfs_node *parent_kn, kn = rdtgrp->kn; ret =closid_alloc ();if (ret <0 ) { -rdt_last_cmd_puts ("out of CLOSIDs\n" ); +rdt_last_cmd_puts ("Out of CLOSIDs\n" ); goto out_common_fail; } closid = ret;
Signature · 1 line
2.17.1
tip-bot for Babu Moger
merge bot[tip:x86/cache] x86/resctrl: Fixup the user-visible strings
x86/resctrl: Fixup the user-visible strings
Commit-ID: 723f1a0dd8e26a7523ba068204bee11c95ded38d Gitweb: https://git.kernel.org/tip/723f1a0dd8e26a7523ba068204bee11c95ded38d Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:43 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:20 +0100
x86/resctrl: Fixup the user-visible strings
Fix the messages in rdt_last_cmd_printf() and rdt_last_cmd_puts() to make them more meaningful and consistent.
[ bp: s/cpu/CPU/; s/mem\W/memory ]
--- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 22 +++++++++---------- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 34 ++++++++++++++--------------- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 36 +++++++++++++++---------------- 3 files changed, 46 insertions(+), 46 deletions(-)
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +11 -11
@@ -70,7 +70,7 @@ int parse_bw_intel(struct rdt_parse_data *data, struct rdt_resource *r,unsigned long bw_val;if (d->have_new_ctrl) { -rdt_last_cmd_printf ("duplicate domain %d\n" , d->id); +rdt_last_cmd_printf ("Duplicate domain %d\n" , d->id);return -EINVAL ; }
@@ -96,12 +96,12 @@ bool cbm_validate_intel(char *buf, u32 *data, struct rdt_resource *r) ret =kstrtoul (buf,16 , &val);if (ret) { -rdt_last_cmd_printf ("non-hex character in mask %s\n" , buf); +rdt_last_cmd_printf ("Non-hex character in the mask %s\n" , buf);return false; }if (val ==0 || val > r->default_ctrl) { -rdt_last_cmd_puts ("mask out of range\n" ); +rdt_last_cmd_puts ("Mask out of range\n" );return false; }
@@ -109,12 +109,12 @@ bool cbm_validate_intel(char *buf, u32 *data, struct rdt_resource *r) zero_bit =find_next_zero_bit (&val, cbm_len, first_bit);if (find_next_bit (&val, cbm_len, zero_bit) < cbm_len) { -rdt_last_cmd_printf ("mask %lx has non-consecutive 1-bits\n" , val); +rdt_last_cmd_printf ("The mask %lx has non-consecutive 1-bits\n" , val);return false; }if ((zero_bit - first_bit) < r->cache.min_cbm_bits) { -rdt_last_cmd_printf ("Need at least %d bits in mask\n" , +rdt_last_cmd_printf ("Need at least %d bits in the mask\n" , r->cache.min_cbm_bits);return false; }
@@ -134,7 +134,7 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,u32 cbm_val;if (d->have_new_ctrl) { -rdt_last_cmd_printf ("duplicate domain %d\n" , d->id); +rdt_last_cmd_printf ("Duplicate domain %d\n" , d->id);return -EINVAL ; }
@@ -144,7 +144,7 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,*/ if (rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP &&rdtgroup_pseudo_locked_in_hierarchy (d)) { -rdt_last_cmd_printf ("pseudo-locked region in hierarchy\n" ); +rdt_last_cmd_printf ("Pseudo-locked region in hierarchy\n" );return -EINVAL ; }
@@ -163,14 +163,14 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,* either is exclusive. */ if (rdtgroup_cbm_overlaps (r, d, cbm_val, rdtgrp->closid, true)) { -rdt_last_cmd_printf ("overlaps with exclusive group\n" ); +rdt_last_cmd_printf ("Overlaps with exclusive group\n" );return -EINVAL ; }if (rdtgroup_cbm_overlaps (r, d, cbm_val, rdtgrp->closid, false)) {if (rdtgrp->mode ==RDT_MODE_EXCLUSIVE || rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP ) { -rdt_last_cmd_printf ("overlaps with other group\n" ); +rdt_last_cmd_printf ("0verlaps with other group\n" );return -EINVAL ; } }
@@ -292,7 +292,7 @@ static int rdtgroup_parse_resource(char *resname, char *tok,if (!strcmp (resname, r->name) && rdtgrp->closid < r->num_closid)return parse_line (tok, r, rdtgrp); } -rdt_last_cmd_printf ("unknown/unsupported resource name '%s'\n" , resname); +rdt_last_cmd_printf ("Unknown or unsupported resource name '%s'\n" , resname);return -EINVAL ; }
@@ -323,7 +323,7 @@ ssize_t rdtgroup_schemata_write(struct kernfs_open_file *of,*/ if (rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKED ) { ret = -EINVAL ; -rdt_last_cmd_puts ("resource group is pseudo-locked\n" ); +rdt_last_cmd_puts ("Resource group is pseudo-locked\n" ); goto out; }
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +17 -17
@@ -213,7 +213,7 @@ static int pseudo_lock_cstates_constrain(struct pseudo_lock_region *plr)for_each_cpu (cpu, &plr->d->cpu_mask) { pm_req =kzalloc (sizeof (*pm_req),GFP_KERNEL );if (!pm_req) { -rdt_last_cmd_puts ("fail allocating mem for PM QoS\n" ); +rdt_last_cmd_puts ("Failure to allocate memory for PM QoS\n" ); ret = -ENOMEM ; goto out_err; }
@@ -222,7 +222,7 @@ static int pseudo_lock_cstates_constrain(struct pseudo_lock_region *plr)DEV_PM_QOS_RESUME_LATENCY , 30);if (ret <0 ) { -rdt_last_cmd_printf ("fail to add latency req cpu%d\n" , +rdt_last_cmd_printf ("Failed to add latency req CPU%d\n" , cpu);kfree (pm_req); ret =-1 ;
@@ -289,7 +289,7 @@ static int pseudo_lock_region_init(struct pseudo_lock_region *plr) plr->cpu =cpumask_first (&plr->d->cpu_mask);if (!cpu_online (plr->cpu)) { -rdt_last_cmd_printf ("cpu %u associated with cache not online\n" , +rdt_last_cmd_printf ("CPU %u associated with cache not online\n" , plr->cpu); ret = -ENODEV ; goto out_region;
@@ -307,7 +307,7 @@ static int pseudo_lock_region_init(struct pseudo_lock_region *plr) } ret =-1 ; -rdt_last_cmd_puts ("unable to determine cache line size\n" ); +rdt_last_cmd_puts ("Unable to determine cache line size\n" ); out_region:pseudo_lock_region_clear (plr);return ret;
@@ -361,14 +361,14 @@ static int pseudo_lock_region_alloc(struct pseudo_lock_region *plr)* KMALLOC_MAX_SIZE. */ if (plr->size >KMALLOC_MAX_SIZE ) { -rdt_last_cmd_puts ("requested region exceeds maximum size\n" ); +rdt_last_cmd_puts ("Requested region exceeds maximum size\n" ); ret = -E2BIG ; goto out_region; } plr->kmem =kzalloc (plr->size,GFP_KERNEL );if (!plr->kmem) { -rdt_last_cmd_puts ("unable to allocate memory\n" ); +rdt_last_cmd_puts ("Unable to allocate memory\n" ); ret = -ENOMEM ; goto out_region; }
@@ -665,7 +665,7 @@ int rdtgroup_locksetup_enter(struct rdtgroup *rdtgrp)* default closid associated with it. */ if (rdtgrp == &rdtgroup_default) { -rdt_last_cmd_puts ("cannot pseudo-lock default group\n" ); +rdt_last_cmd_puts ("Cannot pseudo-lock default group\n" );return -EINVAL ; }
@@ -707,17 +707,17 @@ int rdtgroup_locksetup_enter(struct rdtgroup *rdtgrp)*/ prefetch_disable_bits =get_prefetch_disable_bits ();if (prefetch_disable_bits ==0 ) { -rdt_last_cmd_puts ("pseudo-locking not supported\n" ); +rdt_last_cmd_puts ("Pseudo-locking not supported\n" );return -EINVAL ; }if (rdtgroup_monitor_in_progress (rdtgrp)) { -rdt_last_cmd_puts ("monitoring in progress\n" ); +rdt_last_cmd_puts ("Monitoring in progress\n" );return -EINVAL ; }if (rdtgroup_tasks_assigned (rdtgrp)) { -rdt_last_cmd_puts ("tasks assigned to resource group\n" ); +rdt_last_cmd_puts ("Tasks assigned to resource group\n" );return -EINVAL ; }
@@ -727,13 +727,13 @@ int rdtgroup_locksetup_enter(struct rdtgroup *rdtgrp) }if (rdtgroup_locksetup_user_restrict (rdtgrp)) { -rdt_last_cmd_puts ("unable to modify resctrl permissions\n" ); +rdt_last_cmd_puts ("Unable to modify resctrl permissions\n" );return -EIO ; } ret =pseudo_lock_init (rdtgrp);if (ret) { -rdt_last_cmd_puts ("unable to init pseudo-lock region\n" ); +rdt_last_cmd_puts ("Unable to init pseudo-lock region\n" ); goto out_release; }
@@ -770,7 +770,7 @@ int rdtgroup_locksetup_exit(struct rdtgroup *rdtgrp)if (rdt_mon_capable) { ret =alloc_rmid ();if (ret <0 ) { -rdt_last_cmd_puts ("out of RMIDs\n" ); +rdt_last_cmd_puts ("Out of RMIDs\n" );return ret; } rdtgrp->mon.rmid = ret;
@@ -1304,7 +1304,7 @@ int rdtgroup_pseudo_lock_create(struct rdtgroup *rdtgrp) "pseudo_lock/%u", plr->cpu);if (IS_ERR (thread)) { ret =PTR_ERR (thread); -rdt_last_cmd_printf ("locking thread returned error %d\n" , ret); +rdt_last_cmd_printf ("Locking thread returned error %d\n" , ret); goto out_cstates; }
@@ -1322,13 +1322,13 @@ int rdtgroup_pseudo_lock_create(struct rdtgroup *rdtgrp)* the cleared, but not freed, plr struct resulting in an * empty pseudo-locking loop. */ -rdt_last_cmd_puts ("locking thread interrupted\n" ); +rdt_last_cmd_puts ("Locking thread interrupted\n" ); goto out_cstates; } ret =pseudo_lock_minor_get (&new_minor);if (ret <0 ) { -rdt_last_cmd_puts ("unable to obtain a new minor number\n" ); +rdt_last_cmd_puts ("Unable to obtain a new minor number\n" ); goto out_cstates; }
@@ -1360,7 +1360,7 @@ int rdtgroup_pseudo_lock_create(struct rdtgroup *rdtgrp)if (IS_ERR (dev)) { ret =PTR_ERR (dev); -rdt_last_cmd_printf ("failed to create character device: %d\n" , +rdt_last_cmd_printf ("Failed to create character device: %d\n" , ret); goto out_debugfs; }
arch/x86/kernel/cpu/resctrl/rdtgroup.c +18 -18
@@ -345,7 +345,7 @@ static int cpus_mon_write(struct rdtgroup *rdtgrp, cpumask_var_t newmask,/* Check whether cpus belong to parent ctrl group */ cpumask_andnot (tmpmask, newmask, &prgrp->cpu_mask);if (cpumask_weight (tmpmask)) { -rdt_last_cmd_puts ("can only add CPUs to mongroup that belong to parent\n" ); +rdt_last_cmd_puts ("Can only add CPUs to mongroup that belong to parent\n" );return -EINVAL ; }
@@ -470,14 +470,14 @@ static ssize_t rdtgroup_cpus_write(struct kernfs_open_file *of,rdt_last_cmd_clear ();if (!rdtgrp) { ret = -ENOENT ; -rdt_last_cmd_puts ("directory was removed\n" ); +rdt_last_cmd_puts ("Directory was removed\n" ); goto unlock; }if (rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKED || rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP ) { ret = -EINVAL ; -rdt_last_cmd_puts ("pseudo-locking in progress\n" ); +rdt_last_cmd_puts ("Pseudo-locking in progress\n" ); goto unlock; }
@@ -487,7 +487,7 @@ static ssize_t rdtgroup_cpus_write(struct kernfs_open_file *of, ret =cpumask_parse (buf, newmask);if (ret) { -rdt_last_cmd_puts ("bad cpu list/mask\n" ); +rdt_last_cmd_puts ("Bad CPU list/mask\n" ); goto unlock; }
@@ -495,7 +495,7 @@ static ssize_t rdtgroup_cpus_write(struct kernfs_open_file *of,cpumask_andnot (tmpmask, newmask, cpu_online_mask);if (cpumask_weight (tmpmask)) { ret = -EINVAL ; -rdt_last_cmd_puts ("can only assign online cpus\n" ); +rdt_last_cmd_puts ("Can only assign online CPUs\n" ); goto unlock; }
@@ -574,7 +574,7 @@ static int __rdtgroup_move_task(struct task_struct *tsk,*/ atomic_dec (&rdtgrp->waitcount);kfree (callback); -rdt_last_cmd_puts ("task exited\n" ); +rdt_last_cmd_puts ("Task exited\n" ); }else {/* * For ctrl_mon groups move both closid and rmid.
@@ -692,7 +692,7 @@ static ssize_t rdtgroup_tasks_write(struct kernfs_open_file *of,if (rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKED || rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP ) { ret = -EINVAL ; -rdt_last_cmd_puts ("pseudo-locking in progress\n" ); +rdt_last_cmd_puts ("Pseudo-locking in progress\n" ); goto unlock; }
@@ -1158,14 +1158,14 @@ static bool rdtgroup_mode_test_exclusive(struct rdtgroup *rdtgrp)list_for_each_entry (d, &r->domains, list) {if (rdtgroup_cbm_overlaps (r, d, d->ctrl_val[closid], rdtgrp->closid, false)) { -rdt_last_cmd_puts ("schemata overlaps\n" ); +rdt_last_cmd_puts ("Schemata overlaps\n" );return false; } } }if (!has_cache) { -rdt_last_cmd_puts ("cannot be exclusive without CAT/CDP\n" ); +rdt_last_cmd_puts ("Cannot be exclusive without CAT/CDP\n" );return false; }
@@ -1206,7 +1206,7 @@ static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, goto out;if (mode ==RDT_MODE_PSEUDO_LOCKED ) { -rdt_last_cmd_printf ("cannot change pseudo-locked group\n" ); +rdt_last_cmd_printf ("Cannot change pseudo-locked group\n" ); ret = -EINVAL ; goto out; }
@@ -1235,7 +1235,7 @@ static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, goto out; rdtgrp->mode =RDT_MODE_PSEUDO_LOCKSETUP ; }else { -rdt_last_cmd_printf ("unknown/unsupported mode\n" ); +rdt_last_cmd_printf ("Unknown orunsupported mode\n" ); ret = -EINVAL ; }
@@ -2543,7 +2543,7 @@ static int rdtgroup_init_alloc(struct rdtgroup *rdtgrp) tmp_cbm = d->new_ctrl;if (bitmap_weight (&tmp_cbm, r->cache.cbm_len) < r->cache.min_cbm_bits) { -rdt_last_cmd_printf ("no space on %s:%d\n" , +rdt_last_cmd_printf ("No space on %s:%d\n" , r->name, d->id);return -ENOSPC ; }
@@ -2560,7 +2560,7 @@ static int rdtgroup_init_alloc(struct rdtgroup *rdtgrp)continue ; ret =update_domains (r, rdtgrp->closid);if (ret <0 ) { -rdt_last_cmd_puts ("failed to initialize allocations\n" ); +rdt_last_cmd_puts ("Failed to initialize allocations\n" );return ret; } rdtgrp->mode =RDT_MODE_SHAREABLE ;
@@ -2583,7 +2583,7 @@ static int mkdir_rdt_prepare(struct kernfs_node *parent_kn,rdt_last_cmd_clear ();if (!prdtgrp) { ret = -ENODEV ; -rdt_last_cmd_puts ("directory was removed\n" ); +rdt_last_cmd_puts ("Directory was removed\n" ); goto out_unlock; }
@@ -2591,7 +2591,7 @@ static int mkdir_rdt_prepare(struct kernfs_node *parent_kn, (prdtgrp->mode == RDT_MODE_PSEUDO_LOCKSETUP || prdtgrp->mode == RDT_MODE_PSEUDO_LOCKED)) { ret = -EINVAL ; -rdt_last_cmd_puts ("pseudo-locking in progress\n" ); +rdt_last_cmd_puts ("Pseudo-locking in progress\n" ); goto out_unlock; }
@@ -2599,7 +2599,7 @@ static int mkdir_rdt_prepare(struct kernfs_node *parent_kn, rdtgrp =kzalloc (sizeof (*rdtgrp),GFP_KERNEL );if (!rdtgrp) { ret = -ENOSPC ; -rdt_last_cmd_puts ("kernel out of memory\n" ); +rdt_last_cmd_puts ("Kernel out of memory\n" ); goto out_unlock; } *r = rdtgrp;
@@ -2640,7 +2640,7 @@ static int mkdir_rdt_prepare(struct kernfs_node *parent_kn,if (rdt_mon_capable) { ret =alloc_rmid ();if (ret <0 ) { -rdt_last_cmd_puts ("out of RMIDs\n" ); +rdt_last_cmd_puts ("Out of RMIDs\n" ); goto out_destroy; } rdtgrp->mon.rmid = ret;
@@ -2728,7 +2728,7 @@ static int rdtgroup_mkdir_ctrl_mon(struct kernfs_node *parent_kn, kn = rdtgrp->kn; ret =closid_alloc ();if (ret <0 ) { -rdt_last_cmd_puts ("out of CLOSIDs\n" ); +rdt_last_cmd_puts ("Out of CLOSIDs\n" ); goto out_common_fail; } closid = ret;
Reinette Chatre
maintainerRe: [PATCH v9 10/13] x86/resctrl: Fix the messages in rdt_last_cmd_printf and rdt_last_cmd_puts
In reply to Moger, BabuHi Babu and Borislav,
Two typos seemed to have slipped through into the merged commit ...
Quoted 7 lines · Babu
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +1
-1@@ -163,14 +163,14 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,* either is exclusive. */ if (rdtgroup_cbm_overlaps (r, d, cbm_val, rdtgrp->closid, true)) { -rdt_last_cmd_printf ("overlaps with exclusive group\n" ); +rdt_last_cmd_printf ("Overlaps with exclusive group\n" );return -EINVAL ;
arch/x86/kernel/cpu/resctrl/ctrlmondata.c int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r, · line 176
Quoted 10 lines
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +1
-1}if (rdtgroup_cbm_overlaps (r, d, cbm_val, rdtgrp->closid, false)) {if (rdtgrp->mode ==RDT_MODE_EXCLUSIVE || rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP ) { -rdt_last_cmd_printf ("overlaps with other group\n" ); +rdt_last_cmd_printf ("0verlaps with other group\n" );return -EINVAL ; } }
There is a zero instead of O used in second "Overlaps".
arch/x86/kernel/cpu/resctrl/rdtgroup.c static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, · line 1240
Quoted 8 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1
-1@@ -1235,7 +1235,7 @@ static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, goto out; rdtgrp->mode =RDT_MODE_PSEUDO_LOCKSETUP ; }else { -rdt_last_cmd_printf ("unknown/unsupported mode\n" ); +rdt_last_cmd_printf ("Unknown orunsupported mode\n" ); ret = -EINVAL ; }
Missing a space.
Reinette
Borislav Petkov
maintainerRe: [PATCH v9 10/13] x86/resctrl: Fix the messages in rdt_last_cmd_printf and rdt_last_cmd_puts
Quoted 14 lines · Reinette Chatre
Hi Babu and Borislav,
Two typos seemed to have slipped through into the merged commit ...
Quoted 9 lines · Babu
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +1
-1@@ -163,14 +163,14 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,* either is exclusive. */ if (rdtgroup_cbm_overlaps (r, d, cbm_val, rdtgrp->closid, true)) { -rdt_last_cmd_printf ("overlaps with exclusive group\n" ); +rdt_last_cmd_printf ("Overlaps with exclusive group\n" );return -EINVAL ; }
Quoted 10 lines
arch/x86/kernel/cpu/resctrl/ctrlmondata.c int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r, · line 176
Quoted 8 lines
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +1
-1if (rdtgroup_cbm_overlaps (r, d, cbm_val, rdtgrp->closid, false)) {if (rdtgrp->mode ==RDT_MODE_EXCLUSIVE || rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP ) { -rdt_last_cmd_printf ("overlaps with other group\n" ); +rdt_last_cmd_printf ("0verlaps with other group\n" );return -EINVAL ; } }There is a zero instead of O used in second "Overlaps".
LOL!
That's so l337 :-)
Quoted 10 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, · line 1240
Quoted 8 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1
-1@@ -1235,7 +1235,7 @@ static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, goto out; rdtgrp->mode =RDT_MODE_PSEUDO_LOCKSETUP ; }else { -rdt_last_cmd_printf ("unknown/unsupported mode\n" ); +rdt_last_cmd_printf ("Unknown orunsupported mode\n" ); ret = -EINVAL ; }Missing a space.
Whoops, nice catch. Can you send a fix pls?
Thx.
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Reinette Chatre
maintainerRe: [PATCH v9 10/13] x86/resctrl: Fix the messages in rdt_last_cmd_printf and rdt_last_cmd_puts
Hi Borislav,
Quoted 41 lines · Borislav Petkov
Quoted 14 lines · Reinette Chatre
Hi Babu and Borislav,
Two typos seemed to have slipped through into the merged commit ...
Quoted 9 lines · Babu
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +1
-1@@ -163,14 +163,14 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r,* either is exclusive. */ if (rdtgroup_cbm_overlaps (r, d, cbm_val, rdtgrp->closid, true)) { -rdt_last_cmd_printf ("overlaps with exclusive group\n" ); +rdt_last_cmd_printf ("Overlaps with exclusive group\n" );return -EINVAL ; }Quoted 10 lines
arch/x86/kernel/cpu/resctrl/ctrlmondata.c int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r, · line 176
Quoted 8 lines
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +1
-1if (rdtgroup_cbm_overlaps (r, d, cbm_val, rdtgrp->closid, false)) {if (rdtgrp->mode ==RDT_MODE_EXCLUSIVE || rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP ) { -rdt_last_cmd_printf ("overlaps with other group\n" ); +rdt_last_cmd_printf ("0verlaps with other group\n" );return -EINVAL ; } }There is a zero instead of O used in second "Overlaps".
LOL!
That's so l337 :-)
Quoted 10 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, · line 1240
Quoted 8 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1
-1@@ -1235,7 +1235,7 @@ static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, goto out; rdtgrp->mode =RDT_MODE_PSEUDO_LOCKSETUP ; }else { -rdt_last_cmd_printf ("unknown/unsupported mode\n" ); +rdt_last_cmd_printf ("Unknown orunsupported mode\n" ); ret = -EINVAL ; }Missing a space.
Quoted 1 line
Whoops, nice catch. Can you send a fix pls?
Will do. Since there is overlap here I am planning for the fix to also include changing rdt_last_cmd_printf() instances like the above to use rdt_last_cmd_puts() instead.
Reinette
Moger, Babu
reviewerPATCH v9 11/13 x86/resctrl: Introduce QOS feature for AMD
In reply to Moger, BabuEnable QOS feature on AMD.
Following QoS sub-features are supported in AMD if the underlying hardware supports it.
- L3 Cache allocation enforcement
- L3 Cache occupancy monitoring
- L3 Code-Data Prioritization support
- Memory Bandwidth Enforcement(Allocation)
The specification for this feature is available at
https://developer.amd.com/wp-content/resources/56375.pdf
There are differences in the way some of the features are implemented. Separate those functions and add those as vendor specific functions. The major difference is in MBA feature.
- AMD uses CPUID leaf 0x80000020 to initialize the MBA features.
- AMD uses direct bandwidth value instead of delay based on bandwidth values.
- MSR register base addresses are different for MBA.
- Also AMD allows non-contiguous L3 cache bit masks.
--- arch/x86/kernel/cpu/resctrl/core.c | 69 +++++++++++++++++++++- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 70 +++++++++++++++++++++++ arch/x86/kernel/cpu/resctrl/internal.h | 5 ++ 3 files changed, 141 insertions(+), 3 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +66 -3
@@ -61,6 +61,9 @@ mba_wrmsr_intel(struct rdt_domain *d, struct msr_param *m, struct rdt_resource *r);static void cat_wrmsr (struct rdt_domain *d,struct msr_param *m,struct rdt_resource *r); +static void +mba_wrmsr_amd (struct rdt_domain *d,struct msr_param *m, +struct rdt_resource *r);#define domain_init (id) LIST_HEAD_INIT(rdt_resources_all[id].domains)
@@ -255,7 +258,7 @@ static inline bool rdt_get_mb_table(struct rdt_resource *r)return false; } -static bool __get_mem_config (struct rdt_resource *r) +static bool __get_mem_config_intel (struct rdt_resource *r) {union cpuid_0x10_3_eax eax;union cpuid_0x10_x_edx edx;
@@ -281,6 +284,30 @@ static bool __get_mem_config(struct rdt_resource *r)return true; } +static bool __rdt_get_mem_config_amd (struct rdt_resource *r) +{ +union cpuid_0x10_3_eax eax; +union cpuid_0x10_x_edx edx; +u32 ebx, ecx; + +cpuid_count (0x80000020 ,1 , &eax.full, &ebx, &ecx, &edx.full); + r->num_closid = edx.split.cos_max +1 ; + r->default_ctrl =MAX_MBA_BW_AMD ; + +/* AMD does not use delay */ + r->membw.delay_linear = false; + + r->membw.min_bw =0 ; + r->membw.bw_gran =1 ; +/* Max value is 2048, Data width should be 4 in decimal */ + r->data_width =4 ; + + r->alloc_capable = true; + r->alloc_enabled = true; + +return true; +} +static void rdt_get_cache_alloc_cfg (int idx,struct rdt_resource *r) {union cpuid_0x10_1_eax eax;
@@ -340,6 +367,15 @@ static int get_cache_id(int cpu, int level)return -1 ; } +static void +mba_wrmsr_amd (struct rdt_domain *d,struct msr_param *m,struct rdt_resource *r) +{ +unsigned int i; + +for (i = m->low; i < m->high; i++) +wrmsrl (r->msr_base + i, d->ctrl_val[i]); +} +/* * Map the memory b/w percentage value to delay values * that can be written to QOS_MSRs.
@@ -793,8 +829,13 @@ static bool __init rdt_cpu_has(int flag)static __init bool get_mem_config (void ) { -if (rdt_cpu_has (X86_FEATURE_MBA )) -return __get_mem_config (&rdt_resources_all[RDT_RESOURCE_MBA ]); +if (!rdt_cpu_has (X86_FEATURE_MBA )) +return false; + +if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) +return __get_mem_config_intel (&rdt_resources_all[RDT_RESOURCE_MBA ]); +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +return __rdt_get_mem_config_amd (&rdt_resources_all[RDT_RESOURCE_MBA ]);return false; }
@@ -893,10 +934,32 @@ static __init void rdt_init_res_defs_intel(void) } } +static __init void rdt_init_res_defs_amd (void ) +{ +struct rdt_resource *r; + +for_each_rdt_resource (r) { +if (r->rid ==RDT_RESOURCE_L3 || + r->rid ==RDT_RESOURCE_L3DATA || + r->rid ==RDT_RESOURCE_L3CODE || + r->rid ==RDT_RESOURCE_L2 || + r->rid ==RDT_RESOURCE_L2DATA || + r->rid ==RDT_RESOURCE_L2CODE ) + r->cbm_validate = cbm_validate_amd; +else if (r->rid ==RDT_RESOURCE_MBA ) { + r->msr_base =MSR_IA32_MBA_BW_BASE ; + r->msr_update = mba_wrmsr_amd; + r->parse_ctrlval = parse_bw_amd; + } + } +} +static __init void rdt_init_res_defs (void ) {if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL )rdt_init_res_defs_intel (); +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +rdt_init_res_defs_amd (); }static enum cpuhp_state rdt_online;
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +70 -0
82 lines not on this page · raw message
arch/x86/kernel/cpu/resctrl/internal.h +5 -0
@@ -11,6 +11,7 @@#define MSR_IA32_L3_CBM_BASE 0xc90#define MSR_IA32_L2_CBM_BASE 0xd10#define MSR_IA32_MBA_THRTL_BASE 0xd50 +#define MSR_IA32_MBA_BW_BASE 0xc0000200#define MSR_IA32_QM_CTR 0x0c8e#define MSR_IA32_QM_EVTSEL 0x0c8d
@@ -34,6 +35,7 @@#define MAX_MBA_BW 100u#define MBA_IS_LINEAR 0x4#define MBA_MAX_MBPS U32_MAX +#define MAX_MBA_BW_AMD 0x800#define RMID_VAL_ERROR BIT_ULL(63)#define RMID_VAL_UNAVAIL BIT_ULL(62)
@@ -448,6 +450,8 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r, struct rdt_domain *d);int parse_bw_intel (struct rdt_parse_data *data,struct rdt_resource *r,struct rdt_domain *d); +int parse_bw_amd (struct rdt_parse_data *data,struct rdt_resource *r, +struct rdt_domain *d);extern struct mutex rdtgroup_mutex;
@@ -579,5 +583,6 @@ void cqm_handle_limbo(struct work_struct *work);bool has_busy_rmid (struct rdt_resource *r,struct rdt_domain *d);void __check_limbo (struct rdt_domain *d,bool force_free);bool cbm_validate_intel (char *buf,u32 *data,struct rdt_resource *r); +bool cbm_validate_amd (char *buf,u32 *data,struct rdt_resource *r);#endif /* _ASM_X86_RESCTRL_INTERNAL_H */
Signature · 1 line
2.17.1
tip-bot for Babu Moger
merge bot[tip:x86/cache] x86/resctrl: Introduce AMD QOS feature
x86/resctrl: Introduce AMD QOS feature
Commit-ID: 4d05bf71f157d756932e77cdee16dc99e235d636 Gitweb: https://git.kernel.org/tip/4d05bf71f157d756932e77cdee16dc99e235d636 Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:45 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:20 +0100
x86/resctrl: Introduce AMD QOS feature
Enable QOS feature on AMD.
Following QoS sub-features are supported on AMD if the underlying hardware supports it:
- L3 Cache allocation enforcement
- L3 Cache occupancy monitoring
- L3 Code-Data Prioritization support
- Memory Bandwidth Enforcement (Allocation)
The specification is available at:
https://developer.amd.com/wp-content/resources/56375.pdf
There are differences in the way some of the features are implemented. Separate those functions and add those as vendor specific functions.
The major difference is in MBA feature:
- AMD uses CPUID leaf 0x80000020 to initialize the MBA features.
- AMD uses direct bandwidth value instead of delay based on bandwidth values.
- MSR register base addresses are different for MBA.
- AMD allows non-contiguous L3 cache bit masks.
--- arch/x86/kernel/cpu/resctrl/core.c | 69 ++++++++++++++++++++++++++++-- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 71 +++++++++++++++++++++++++++++++ arch/x86/kernel/cpu/resctrl/internal.h | 5 +++ 3 files changed, 142 insertions(+), 3 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +66 -3
@@ -61,6 +61,9 @@ mba_wrmsr_intel(struct rdt_domain *d, struct msr_param *m, struct rdt_resource *r);static void cat_wrmsr (struct rdt_domain *d,struct msr_param *m,struct rdt_resource *r); +static void +mba_wrmsr_amd (struct rdt_domain *d,struct msr_param *m, +struct rdt_resource *r);#define domain_init (id) LIST_HEAD_INIT(rdt_resources_all[id].domains)
@@ -255,7 +258,7 @@ static inline bool rdt_get_mb_table(struct rdt_resource *r)return false; } -static bool __get_mem_config (struct rdt_resource *r) +static bool __get_mem_config_intel (struct rdt_resource *r) {union cpuid_0x10_3_eax eax;union cpuid_0x10_x_edx edx;
@@ -281,6 +284,30 @@ static bool __get_mem_config(struct rdt_resource *r)return true; } +static bool __rdt_get_mem_config_amd (struct rdt_resource *r) +{ +union cpuid_0x10_3_eax eax; +union cpuid_0x10_x_edx edx; +u32 ebx, ecx; + +cpuid_count (0x80000020 ,1 , &eax.full, &ebx, &ecx, &edx.full); + r->num_closid = edx.split.cos_max +1 ; + r->default_ctrl =MAX_MBA_BW_AMD ; + +/* AMD does not use delay */ + r->membw.delay_linear = false; + + r->membw.min_bw =0 ; + r->membw.bw_gran =1 ; +/* Max value is 2048, Data width should be 4 in decimal */ + r->data_width =4 ; + + r->alloc_capable = true; + r->alloc_enabled = true; + +return true; +} +static void rdt_get_cache_alloc_cfg (int idx,struct rdt_resource *r) {union cpuid_0x10_1_eax eax;
@@ -340,6 +367,15 @@ static int get_cache_id(int cpu, int level)return -1 ; } +static void +mba_wrmsr_amd (struct rdt_domain *d,struct msr_param *m,struct rdt_resource *r) +{ +unsigned int i; + +for (i = m->low; i < m->high; i++) +wrmsrl (r->msr_base + i, d->ctrl_val[i]); +} +/* * Map the memory b/w percentage value to delay values * that can be written to QOS_MSRs.
@@ -793,8 +829,13 @@ static bool __init rdt_cpu_has(int flag)static __init bool get_mem_config (void ) { -if (rdt_cpu_has (X86_FEATURE_MBA )) -return __get_mem_config (&rdt_resources_all[RDT_RESOURCE_MBA ]); +if (!rdt_cpu_has (X86_FEATURE_MBA )) +return false; + +if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) +return __get_mem_config_intel (&rdt_resources_all[RDT_RESOURCE_MBA ]); +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +return __rdt_get_mem_config_amd (&rdt_resources_all[RDT_RESOURCE_MBA ]);return false; }
@@ -893,10 +934,32 @@ static __init void rdt_init_res_defs_intel(void) } } +static __init void rdt_init_res_defs_amd (void ) +{ +struct rdt_resource *r; + +for_each_rdt_resource (r) { +if (r->rid ==RDT_RESOURCE_L3 || + r->rid ==RDT_RESOURCE_L3DATA || + r->rid ==RDT_RESOURCE_L3CODE || + r->rid ==RDT_RESOURCE_L2 || + r->rid ==RDT_RESOURCE_L2DATA || + r->rid ==RDT_RESOURCE_L2CODE ) + r->cbm_validate = cbm_validate_amd; +else if (r->rid ==RDT_RESOURCE_MBA ) { + r->msr_base =MSR_IA32_MBA_BW_BASE ; + r->msr_update = mba_wrmsr_amd; + r->parse_ctrlval = parse_bw_amd; + } + } +} +static __init void rdt_init_res_defs (void ) {if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL )rdt_init_res_defs_intel (); +else if (boot_cpu_data.x86_vendor ==X86_VENDOR_AMD ) +rdt_init_res_defs_amd (); }static enum cpuhp_state rdt_online;
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +71 -0
83 lines not on this page · raw message
arch/x86/kernel/cpu/resctrl/internal.h +5 -0
@@ -11,6 +11,7 @@#define MSR_IA32_L3_CBM_BASE 0xc90#define MSR_IA32_L2_CBM_BASE 0xd10#define MSR_IA32_MBA_THRTL_BASE 0xd50 +#define MSR_IA32_MBA_BW_BASE 0xc0000200#define MSR_IA32_QM_CTR 0x0c8e#define MSR_IA32_QM_EVTSEL 0x0c8d
@@ -34,6 +35,7 @@#define MAX_MBA_BW 100u#define MBA_IS_LINEAR 0x4#define MBA_MAX_MBPS U32_MAX +#define MAX_MBA_BW_AMD 0x800#define RMID_VAL_ERROR BIT_ULL(63)#define RMID_VAL_UNAVAIL BIT_ULL(62)
@@ -448,6 +450,8 @@ int parse_cbm(struct rdt_parse_data *data, struct rdt_resource *r, struct rdt_domain *d);int parse_bw_intel (struct rdt_parse_data *data,struct rdt_resource *r,struct rdt_domain *d); +int parse_bw_amd (struct rdt_parse_data *data,struct rdt_resource *r, +struct rdt_domain *d);extern struct mutex rdtgroup_mutex;
@@ -579,5 +583,6 @@ void cqm_handle_limbo(struct work_struct *work);bool has_busy_rmid (struct rdt_resource *r,struct rdt_domain *d);void __check_limbo (struct rdt_domain *d,bool force_free);bool cbm_validate_intel (char *buf,u32 *data,struct rdt_resource *r); +bool cbm_validate_amd (char *buf,u32 *data,struct rdt_resource *r);#endif /* _ASM_X86_RESCTRL_INTERNAL_H */
Moger, Babu
reviewerPATCH v9 12/13 Documentation: Rename and update intel_rdt_ui.txt to resctrl_ui.txt
In reply to Moger, BabuRename intel_rdt_ui.txt to generic resctrl_ui.txt and update the documentation for AMD.
---
Documentation/x86/{intel_rdt_ui.txt => resctrl_ui.txt} | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
rename Documentation/x86/{intel_rdt_ui.txt => resctrl_ui.txt} (99%)Documentation/x86/resctrl_ui.txt renamed from Documentation/x86/intel_rdt_ui.txt +6 -3
@@ -1,4 +1,7 @@ -User Interface for Resource Allocation in Intel Resource Director Technology +User Interface for Resource Control feature + +Intel refers to this feature as Intel Resource Director Technology(Intel(R) RDT). +AMD refers to this feature as AMD Platform Quality of Service(AMD QoS). Copyright (C) 2016 Intel Corporation
@@ -6,8 +9,8 @@ Fenghua Yu <[email protected]> Tony Luck <[email protected]> Vikas Shivappa <[email protected]> -This feature is enabled by the CONFIG_INTEL_RDT Kconfig and the -X86 /proc/cpuinfo flag bits: +This feature is enabled by the CONFIG_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"
Signature · 1 line
2.17.1
tip-bot for Babu Moger
merge bot[tip:x86/cache] Documentation: Rename and update intel_rdt_ui.txt to resctrl_ui.txt
Documentation: Rename and update intel_rdt_ui.txt to resctrl_ui.txt
Commit-ID: a6f771c9bf4eea2da1516e70c283ede61a7d666f Gitweb: https://git.kernel.org/tip/a6f771c9bf4eea2da1516e70c283ede61a7d666f Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:47 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:20 +0100
Documentation: Rename and update intel_rdt_ui.txt to resctrl_ui.txt
Rename intel_rdt_ui.txt to generic resctrl_ui.txt and update the documentation for AMD.
---
Documentation/x86/{intel_rdt_ui.txt => resctrl_ui.txt} | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)Documentation/x86/resctrl_ui.txt renamed from Documentation/x86/intel_rdt_ui.txt +6 -3
@@ -1,4 +1,7 @@ -User Interface for Resource Allocation in Intel Resource Director Technology +User Interface for Resource Control feature + +Intel refers to this feature as Intel Resource Director Technology(Intel(R) RDT). +AMD refers to this feature as AMD Platform Quality of Service(AMD QoS). Copyright (C) 2016 Intel Corporation
@@ -6,8 +9,8 @@ Fenghua Yu <[email protected]> Tony Luck <[email protected]> Vikas Shivappa <[email protected]> -This feature is enabled by the CONFIG_INTEL_RDT Kconfig and the -X86 /proc/cpuinfo flag bits: +This feature is enabled by the CONFIG_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"
Re: [PATCH v9 12/13] Documentation: Rename and update intel_rdt_ui.txt to resctrl_ui.txt
In reply to Moger, BabuQuoted 6 lines · Babu
Rename intel_rdt_ui.txt to generic resctrl_ui.txt and update the documentation for AMD.
Signed-off-by Babu Moger <[email protected]>--- Documentation/x86/{intel_rdt_ui.txt => resctrl_ui.txt} | 9
Other filenames in the directory use "-".
Plus, this is not really about _user_interface.
resctl.txt would be a better name.
Pavel
Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Moger, Babu
reviewerPATCH v9 13/13 MAINTAINERS: Update the file and documentation names in arch/x86
In reply to Moger, BabuThe file names with intel_rdt* are moved to the new directory arch/x86/kernel/cpu/resctrl/. Update the MAINTAINERS to reflect the changed files(and documentation).
--- MAINTAINERS | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
MAINTAINERS +3 -3
@@ -12267,9 +12267,9 @@ M: Fenghua Yu <[email protected]> M: Reinette Chatre <[email protected]> L: [email protected] S: Supported -F: arch/x86/kernel/cpu/intel_rdt* -F: arch/x86/include/asm/intel_rdt_sched.h -F: Documentation/x86/intel_rdt* +F: arch/x86/kernel/cpu/resctrl/ +F: arch/x86/include/asm/resctrl_sched.h +F: Documentation/x86/resctrl* READ-COPY UPDATE (RCU) M: "Paul E. McKenney" <[email protected]>
Signature · 1 line
2.17.1
tip-bot for Babu Moger
merge bot[tip:x86/cache] MAINTAINERS: Update resctrl filename patterns
MAINTAINERS: Update resctrl filename patterns
Commit-ID: 1f8251d3bfadf42357cf6c3eebb5cda6cd3987b5 Gitweb: https://git.kernel.org/tip/1f8251d3bfadf42357cf6c3eebb5cda6cd3987b5 Author: Babu Moger <[email protected]> AuthorDate: Wed, 21 Nov 2018 20:28:49 +0000 Committer: Borislav Petkov <[email protected]> CommitDate: Thu, 22 Nov 2018 20:16:20 +0100
MAINTAINERS: Update resctrl filename patterns
The file nameswith intel_rdt* have been moved to the new directory arch/x86/kernel/cpu/resctrl/. Update to reflect the changed files and documentation.
--- MAINTAINERS | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
MAINTAINERS +3 -3
@@ -12545,9 +12545,9 @@ M: Fenghua Yu <[email protected]> M: Reinette Chatre <[email protected]> L: [email protected] S: Supported -F: arch/x86/kernel/cpu/intel_rdt* -F: arch/x86/include/asm/intel_rdt_sched.h -F: Documentation/x86/intel_rdt* +F: arch/x86/kernel/cpu/resctrl/ +F: arch/x86/include/asm/resctrl_sched.h +F: Documentation/x86/resctrl* READ-COPY UPDATE (RCU) M: "Paul E. McKenney" <[email protected]>