Skip to contentSkip to conversation

arm64: topology: move store_cpu_topology() to shared code

PATCH Applied Conor Dooley 2 participants 5 messages Oct 19, 2022 – Oct 27, 2022

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 arm64: topology: move store_cpu_topology() to shared code

commit 456797da792fa7cbf6698febf275fe9b36691f78 upstream.

arm64's method of defining a default cpu topology requires only minimal changes to apply to RISC-V also. The current arm64 implementation exits early in a uniprocessor configuration by reading MPIDR & claiming that uniprocessor can rely on the default values.

This is appears to be a hangover from prior to '3102bc0e6ac7 ("arm64:
topology: Stop using MPIDR for topology information")', because the current code just assigns default values for multiprocessor systems.

With the MPIDR references removed, store_cpu_topolgy() can be moved to the common arch_topology code.

---
 arch/arm64/kernel/topology.c | 40 ------------------------------------
 drivers/base/arch_topology.c | 19 +++++++++++++++++
 2 files changed, 19 insertions(+), 40 deletions(-)
arch/arm64/kernel/topology.c +0 -40
index 4358bc319306..f35af19b7055 100644
@@ -22,46 +22,6 @@
 #include <asm/cputype.h>
 #include <asm/topology.h>
 
-void store_cpu_topology(unsigned int cpuid)
-{
-	struct cpu_topology *cpuid_topo = &cpu_topology[cpuid];
-	u64 mpidr;
-
-	if (cpuid_topo->package_id != -1)
-		goto topology_populated;
-
-	mpidr = read_cpuid_mpidr();
-
-	/* Uniprocessor systems can rely on default topology values */
-	if (mpidr & MPIDR_UP_BITMASK)
-		return;
-
-	/*
-	 * This would be the place to create cpu topology based on MPIDR.
-	 *
-	 * However, it cannot be trusted to depict the actual topology; some
-	 * pieces of the architecture enforce an artificial cap on Aff0 values
-	 * (e.g. GICv3's ICC_SGI1R_EL1 limits it to 15), leading to an
-	 * artificial cycling of Aff1, Aff2 and Aff3 values. IOW, these end up
-	 * having absolutely no relationship to the actual underlying system
-	 * topology, and cannot be reasonably used as core / package ID.
-	 *
-	 * If the MT bit is set, Aff0 *could* be used to define a thread ID, but
-	 * we still wouldn't be able to obtain a sane core ID. This means we
-	 * need to entirely ignore MPIDR for any topology deduction.
-	 */
-	cpuid_topo->thread_id  = -1;
-	cpuid_topo->core_id    = cpuid;
-	cpuid_topo->package_id = cpu_to_node(cpuid);
-
-	pr_debug("CPU%u: cluster %d core %d thread %d mpidr %#016llx\n",
-		 cpuid, cpuid_topo->package_id, cpuid_topo->core_id,
-		 cpuid_topo->thread_id, mpidr);
-
-topology_populated:
-	update_siblings_masks(cpuid);
-}
-
 #ifdef CONFIG_ACPI
 static bool __init acpi_cpu_is_threaded(int cpu)
 {
drivers/base/arch_topology.c +19 -0
index 8272a3a002a3..51647926e605 100644
@@ -596,4 +596,23 @@ void __init init_cpu_topology(void)
 	else if (of_have_populated_dt() && parse_dt_topology())
 		reset_cpu_topology();
 }
+
+void store_cpu_topology(unsigned int cpuid)
+{
+	struct cpu_topology *cpuid_topo = &cpu_topology[cpuid];
+
+	if (cpuid_topo->package_id != -1)
+		goto topology_populated;
+
+	cpuid_topo->thread_id = -1;
+	cpuid_topo->core_id = cpuid;
+	cpuid_topo->package_id = cpu_to_node(cpuid);
+
+	pr_debug("CPU%u: package %d core %d thread %d\n",
+		 cpuid, cpuid_topo->package_id, cpuid_topo->core_id,
+		 cpuid_topo->thread_id);
+
+topology_populated:
+	update_siblings_masks(cpuid);
+}
 #endif
Signature · 7 lines
2.38.0


_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv
#1 lore raw

PATCH 2/2 riscv: topology: fix default topology reporting

commit fbd92809997a391f28075f1c8b5ee314c225557c upstream.

RISC-V has no sane defaults to fall back on where there is no cpu-map in the devicetree.
Without sane defaults, the package, core and thread IDs are all set to -1. This causes user-visible inaccuracies for tools like hwloc/lstopo which rely on the sysfs cpu topology files to detect a system's topology.

On a PolarFire SoC, which should have 4 harts with a thread each, lstopo currently reports:

Machine (793MB total)
Package L#0

    NUMANode L#0 (P#0 793MB)
    Core L#0
      L1d L#0 (32KB) + L1i L#0 (32KB) + PU L#0 (P#0)
      L1d L#1 (32KB) + L1i L#1 (32KB) + PU L#1 (P#1)
      L1d L#2 (32KB) + L1i L#2 (32KB) + PU L#2 (P#2)
      L1d L#3 (32KB) + L1i L#3 (32KB) + PU L#3 (P#3)

Adding calls to store_cpu_topology() in {boot,smp} hart bringup code results in the correct topolgy being reported:

Machine (793MB total)
Package L#0

    NUMANode L#0 (P#0 793MB)
    L1d L#0 (32KB) + L1i L#0 (32KB) + Core L#0 + PU L#0 (P#0)
    L1d L#1 (32KB) + L1i L#1 (32KB) + Core L#1 + PU L#1 (P#1)
    L1d L#2 (32KB) + L1i L#2 (32KB) + Core L#2 + PU L#2 (P#2)
    L1d L#3 (32KB) + L1i L#3 (32KB) + Core L#3 + PU L#3 (P#3)
CC [email protected] # 456797da792f: arm64: topology: move store_cpu_topology() to shared code
Fixes 03f11f03dbfe ("RISC-V: Parse cpu topology during boot.")

---
I just resolved the conflicts, which was mainly removing mentions of NUMA. Tested in QEMU only.

---
 arch/riscv/Kconfig          | 2 +-
 arch/riscv/kernel/smpboot.c | 4 +++-
 2 files changed, 4 insertions(+), 2 deletions(-)
arch/riscv/Kconfig +1 -1
index 1b894c327578..557c4a8c4087 100644
@@ -35,7 +35,7 @@ config RISCV
 	select CLINT_TIMER if !MMU
 	select COMMON_CLK
 	select EDAC_SUPPORT
-	select GENERIC_ARCH_TOPOLOGY if SMP
+	select GENERIC_ARCH_TOPOLOGY
 	select GENERIC_ATOMIC64 if !64BIT
 	select GENERIC_CLOCKEVENTS
 	select GENERIC_EARLY_IOREMAP
arch/riscv/kernel/smpboot.c +3 -1
index 0b04e0eae3ab..0e0aed380e28 100644
@@ -46,6 +46,8 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	int cpuid;
 	int ret;
 
+	store_cpu_topology(smp_processor_id());
+
 	/* This covers non-smp usecase mandated by "nosmp" option */
 	if (max_cpus == 0)
 		return;
@@ -152,8 +154,8 @@ asmlinkage __visible void smp_callin(void)
 	mmgrab(mm);
 	current->active_mm = mm;
 
+	store_cpu_topology(curr_cpuid);
 	notify_cpu_starting(curr_cpuid);
-	update_siblings_masks(curr_cpuid);
 	set_cpu_online(curr_cpuid, 1);
 
 	/*
Signature · 7 lines
2.38.0


_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv
maintainer Applied
#2 lore raw

Patch "riscv: topology: fix default topology reporting" has been added to the 5.10-stable tree

This is a note to let you know that I've just added the patch titled

riscv: topology: fix default topology reporting

to the 5.10-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
riscv-topology-fix-default-topology-reporting.patch
and it can be found in the queue-5.10 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.

From foo@baz Thu Oct 27 12:15:29 PM CEST 2022
From: Conor Dooley <[email protected]>
Date: Wed, 19 Oct 2022 13:53:03 +0100
Subject: riscv: topology: fix default topology reporting
To: <[email protected]>
Cc: <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, Atish Patra <[email protected]>
Message-ID: <[email protected]>

From: Conor Dooley <[email protected]>

commit fbd92809997a391f28075f1c8b5ee314c225557c upstream.

RISC-V has no sane defaults to fall back on where there is no cpu-map in the devicetree.
Without sane defaults, the package, core and thread IDs are all set to -1. This causes user-visible inaccuracies for tools like hwloc/lstopo which rely on the sysfs cpu topology files to detect a system's topology.

On a PolarFire SoC, which should have 4 harts with a thread each, lstopo currently reports:

Machine (793MB total)
Package L#0

    NUMANode L#0 (P#0 793MB)
    Core L#0
      L1d L#0 (32KB) + L1i L#0 (32KB) + PU L#0 (P#0)
      L1d L#1 (32KB) + L1i L#1 (32KB) + PU L#1 (P#1)
      L1d L#2 (32KB) + L1i L#2 (32KB) + PU L#2 (P#2)
      L1d L#3 (32KB) + L1i L#3 (32KB) + PU L#3 (P#3)

Adding calls to store_cpu_topology() in {boot,smp} hart bringup code results in the correct topolgy being reported:

Machine (793MB total)
Package L#0

    NUMANode L#0 (P#0 793MB)
    L1d L#0 (32KB) + L1i L#0 (32KB) + Core L#0 + PU L#0 (P#0)
    L1d L#1 (32KB) + L1i L#1 (32KB) + Core L#1 + PU L#1 (P#1)
    L1d L#2 (32KB) + L1i L#2 (32KB) + Core L#2 + PU L#2 (P#2)
    L1d L#3 (32KB) + L1i L#3 (32KB) + Core L#3 + PU L#3 (P#3)
CC [email protected] # 456797da792f: arm64: topology: move store_cpu_topology() to shared code
Fixes 03f11f03dbfe ("RISC-V: Parse cpu topology during boot.")
---
 arch/riscv/Kconfig          |    2 +-
 arch/riscv/kernel/smpboot.c |    4 +++-
 2 files changed, 4 insertions(+), 2 deletions(-)
arch/riscv/Kconfig +1 -1
@@ -35,7 +35,7 @@ config RISCV
 	select CLINT_TIMER if !MMU
 	select COMMON_CLK
 	select EDAC_SUPPORT
-	select GENERIC_ARCH_TOPOLOGY if SMP
+	select GENERIC_ARCH_TOPOLOGY
 	select GENERIC_ATOMIC64 if !64BIT
 	select GENERIC_CLOCKEVENTS
 	select GENERIC_EARLY_IOREMAP
arch/riscv/kernel/smpboot.c +3 -1
@@ -46,6 +46,8 @@ void __init smp_prepare_cpus(unsigned in
 	int cpuid;
 	int ret;
 
+	store_cpu_topology(smp_processor_id());
+
 	/* This covers non-smp usecase mandated by "nosmp" option */
 	if (max_cpus == 0)
 		return;
@@ -152,8 +154,8 @@ asmlinkage __visible void smp_callin(voi
 	mmgrab(mm);
 	current->active_mm = mm;
 
+	store_cpu_topology(curr_cpuid);
 	notify_cpu_starting(curr_cpuid);
-	update_siblings_masks(curr_cpuid);
 	set_cpu_online(curr_cpuid, 1);
 
 	/*

Patches currently in stable-queue which might be from [email protected] are

queue-5.10/arm64-topology-move-store_cpu_topology-to-shared-code.patch queue-5.10/riscv-topology-fix-default-topology-reporting.patch queue-5.10/riscv-always-honor-the-config_cmdline_force-when-par.patch

_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv

maintainer Applied
#3 lore raw

Patch "arm64: topology: move store_cpu_topology() to shared code" has been added to the 5.10-stable tree

In reply to Conor Dooley

This is a note to let you know that I've just added the patch titled

arm64: topology: move store_cpu_topology() to shared code

to the 5.10-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
arm64-topology-move-store_cpu_topology-to-shared-code.patch
and it can be found in the queue-5.10 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.

From foo@baz Thu Oct 27 12:15:29 PM CEST 2022
From: Conor Dooley <[email protected]>
Date: Wed, 19 Oct 2022 13:53:02 +0100
Subject: arm64: topology: move store_cpu_topology() to shared code
To: <[email protected]>
Cc: <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, <[email protected]>, Atish Patra <[email protected]>
Message-ID: <[email protected]>

From: Conor Dooley <[email protected]>

commit 456797da792fa7cbf6698febf275fe9b36691f78 upstream.

arm64's method of defining a default cpu topology requires only minimal changes to apply to RISC-V also. The current arm64 implementation exits early in a uniprocessor configuration by reading MPIDR & claiming that uniprocessor can rely on the default values.

This is appears to be a hangover from prior to '3102bc0e6ac7 ("arm64:
topology: Stop using MPIDR for topology information")', because the current code just assigns default values for multiprocessor systems.

With the MPIDR references removed, store_cpu_topolgy() can be moved to the common arch_topology code.

---
 arch/arm64/kernel/topology.c |   40 ----------------------------------------
 drivers/base/arch_topology.c |   19 +++++++++++++++++++
 2 files changed, 19 insertions(+), 40 deletions(-)
arch/arm64/kernel/topology.c +0 -40
@@ -22,46 +22,6 @@
 #include <asm/cputype.h>
 #include <asm/topology.h>
 
-void store_cpu_topology(unsigned int cpuid)
-{
-	struct cpu_topology *cpuid_topo = &cpu_topology[cpuid];
-	u64 mpidr;
-
-	if (cpuid_topo->package_id != -1)
-		goto topology_populated;
-
-	mpidr = read_cpuid_mpidr();
-
-	/* Uniprocessor systems can rely on default topology values */
-	if (mpidr & MPIDR_UP_BITMASK)
-		return;
-
-	/*
-	 * This would be the place to create cpu topology based on MPIDR.
-	 *
-	 * However, it cannot be trusted to depict the actual topology; some
-	 * pieces of the architecture enforce an artificial cap on Aff0 values
-	 * (e.g. GICv3's ICC_SGI1R_EL1 limits it to 15), leading to an
-	 * artificial cycling of Aff1, Aff2 and Aff3 values. IOW, these end up
-	 * having absolutely no relationship to the actual underlying system
-	 * topology, and cannot be reasonably used as core / package ID.
-	 *
-	 * If the MT bit is set, Aff0 *could* be used to define a thread ID, but
-	 * we still wouldn't be able to obtain a sane core ID. This means we
-	 * need to entirely ignore MPIDR for any topology deduction.
-	 */
-	cpuid_topo->thread_id  = -1;
-	cpuid_topo->core_id    = cpuid;
-	cpuid_topo->package_id = cpu_to_node(cpuid);
-
-	pr_debug("CPU%u: cluster %d core %d thread %d mpidr %#016llx\n",
-		 cpuid, cpuid_topo->package_id, cpuid_topo->core_id,
-		 cpuid_topo->thread_id, mpidr);
-
-topology_populated:
-	update_siblings_masks(cpuid);
-}
-
 #ifdef CONFIG_ACPI
 static bool __init acpi_cpu_is_threaded(int cpu)
 {
drivers/base/arch_topology.c +19 -0
@@ -596,4 +596,23 @@ void __init init_cpu_topology(void)
 	else if (of_have_populated_dt() && parse_dt_topology())
 		reset_cpu_topology();
 }
+
+void store_cpu_topology(unsigned int cpuid)
+{
+	struct cpu_topology *cpuid_topo = &cpu_topology[cpuid];
+
+	if (cpuid_topo->package_id != -1)
+		goto topology_populated;
+
+	cpuid_topo->thread_id = -1;
+	cpuid_topo->core_id = cpuid;
+	cpuid_topo->package_id = cpu_to_node(cpuid);
+
+	pr_debug("CPU%u: package %d core %d thread %d\n",
+		 cpuid, cpuid_topo->package_id, cpuid_topo->core_id,
+		 cpuid_topo->thread_id);
+
+topology_populated:
+	update_siblings_masks(cpuid);
+}
 #endif

Patches currently in stable-queue which might be from [email protected] are

queue-5.10/arm64-topology-move-store_cpu_topology-to-shared-code.patch queue-5.10/riscv-topology-fix-default-topology-reporting.patch queue-5.10/riscv-always-honor-the-config_cmdline_force-when-par.patch

_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv

maintainer
#4 lore raw
In reply to Conor Dooley
Quoted 1 line · Conor Dooley

commit 456797da792fa7cbf6698febf275fe9b36691f78 upstream.

All now queued up, thanks.

greg k-h