All of lore.kernel.org
 help / color / mirror / Atom feed
From: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
To: Michael Ellerman <mpe@ellerman.id.au>
Cc: linuxppc-dev <linuxppc-dev@lists.ozlabs.org>,
	Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
	Aboorva Devarajan <aboorvad@linux.vnet.ibm.com>,
	Shrikanth Hegde <sshegde@linux.vnet.ibm.com>,
	Ajay Kaher <akaher@vmware.com>,
	Alexey Makhalov <amakhalov@vmware.com>,
	VMware PV-Drivers Reviewers <pv-drivers@vmware.com>,
	Nicholas Piggin <npiggin@gmail.com>,
	Christophe Leroy <christophe.leroy@csgroup.eu>,
	virtualization@lists.linux-foundation.org, x86@kernel.org,
	linux-kernel@vger.kernel.org
Subject: [PATCH v4] powerpc/paravirt: Improve vcpu_is_preempted
Date: Thu, 19 Oct 2023 14:44:52 +0530	[thread overview]
Message-ID: <20231019091452.95260-1-srikar@linux.vnet.ibm.com> (raw)

PowerVM Hypervisor dispatches on a whole core basis. In a shared LPAR, a
CPU from a core that is CEDED or preempted may have a larger latency. In
such a scenario, its preferable to choose a different CPU to run.

If one of the CPUs in the core is active, i.e neither CEDED nor
preempted, then consider this CPU as not preempted.

Also if any of the CPUs in the core has yielded but OS has not requested
CEDE or CONFER, then consider this CPU to be preempted.

Correct detection of preempted CPUs is important for detecting idle
CPUs/cores in task scheduler.

Tested-by: Aboorva Devarajan <aboorvad@linux.vnet.ibm.com>
Reviewed-by: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
---
Changelog:
v2 (http://lkml.kernel.org/r/20231018155838.2332822-1-srikar@linux.vnet.ibm.com) -> v4:
Resolved comments from Michael Ellerman

v1 (http://lkml.kernel.org/r/20231009051740.17683-1-srikar@linux.vnet.ibm.com) -> v2:
Handle lppaca_of(cpu) in !PPC_SPLPAR case.
1. Fixed some compilation issues reported by kernelbot
	a. https://lore.kernel.org/oe-kbuild-all/202310102341.K0sgoqQL-lkp@intel.com/
	b.  https://lore.kernel.org/oe-kbuild-all/202310091636.lElmJkYV-lkp@intel.com/
2. Resolved comments from Shrikanth

---
 arch/powerpc/include/asm/paravirt.h | 47 +++++++++++++++++++++++++++--
 1 file changed, 44 insertions(+), 3 deletions(-)

diff --git a/arch/powerpc/include/asm/paravirt.h b/arch/powerpc/include/asm/paravirt.h
index e08513d73119..ac4279208d63 100644
--- a/arch/powerpc/include/asm/paravirt.h
+++ b/arch/powerpc/include/asm/paravirt.h
@@ -71,6 +71,11 @@ static inline void yield_to_any(void)
 {
 	plpar_hcall_norets_notrace(H_CONFER, -1, 0);
 }
+
+static inline bool is_vcpu_idle(int vcpu)
+{
+	return lppaca_of(vcpu).idle;
+}
 #else
 static inline bool is_shared_processor(void)
 {
@@ -100,6 +105,10 @@ static inline void prod_cpu(int cpu)
 	___bad_prod_cpu(); /* This would be a bug */
 }
 
+static inline bool is_vcpu_idle(int vcpu)
+{
+	return false;
+}
 #endif
 
 #define vcpu_is_preempted vcpu_is_preempted
@@ -121,9 +130,23 @@ static inline bool vcpu_is_preempted(int cpu)
 	if (!is_shared_processor())
 		return false;
 
+	/*
+	 * If the hypervisor has dispatched the target CPU on a physical
+	 * processor, then the target CPU is definitely not preempted.
+	 */
+	if (!(yield_count_of(cpu) & 1))
+		return false;
+
+	/*
+	 * If the target CPU has yielded to Hypervisor but OS has not
+	 * requested idle then the target CPU is definitely preempted.
+	 */
+	if (!is_vcpu_idle(cpu))
+		return true;
+
 #ifdef CONFIG_PPC_SPLPAR
 	if (!is_kvm_guest()) {
-		int first_cpu;
+		int first_cpu, i;
 
 		/*
 		 * The result of vcpu_is_preempted() is used in a
@@ -149,11 +172,29 @@ static inline bool vcpu_is_preempted(int cpu)
 		 */
 		if (cpu_first_thread_sibling(cpu) == first_cpu)
 			return false;
+
+		/*
+		 * If any of the threads of the target CPU's core are not
+		 * preempted or ceded, then consider target CPU to be
+		 * non-preempted.
+		 */
+		first_cpu = cpu_first_thread_sibling(cpu);
+		for (i = first_cpu; i < first_cpu + threads_per_core; i++) {
+			if (i == cpu)
+				continue;
+			if (!(yield_count_of(i) & 1))
+				return false;
+			if (!is_vcpu_idle(i))
+				return true;
+		}
 	}
 #endif
 
-	if (yield_count_of(cpu) & 1)
-		return true;
+	/*
+	 * None of the threads in target CPU's core are running but none of
+	 * them were preempted too. Hence assume the target CPU to be
+	 * non-preempted.
+	 */
 	return false;
 }
 

base-commit: eddc90ea2af5933249ea1a78119f2c8ef8d07156
-- 
2.31.1


WARNING: multiple messages have this Message-ID (diff)
From: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
To: Michael Ellerman <mpe@ellerman.id.au>
Cc: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>,
	Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
	VMware PV-Drivers Reviewers <pv-drivers@vmware.com>,
	Aboorva Devarajan <aboorvad@linux.vnet.ibm.com>,
	x86@kernel.org, linux-kernel@vger.kernel.org,
	Nicholas Piggin <npiggin@gmail.com>,
	virtualization@lists.linux-foundation.org,
	Ajay Kaher <akaher@vmware.com>,
	Alexey Makhalov <amakhalov@vmware.com>,
	linuxppc-dev <linuxppc-dev@lists.ozlabs.org>
Subject: [PATCH v4] powerpc/paravirt: Improve vcpu_is_preempted
Date: Thu, 19 Oct 2023 14:44:52 +0530	[thread overview]
Message-ID: <20231019091452.95260-1-srikar@linux.vnet.ibm.com> (raw)

PowerVM Hypervisor dispatches on a whole core basis. In a shared LPAR, a
CPU from a core that is CEDED or preempted may have a larger latency. In
such a scenario, its preferable to choose a different CPU to run.

If one of the CPUs in the core is active, i.e neither CEDED nor
preempted, then consider this CPU as not preempted.

Also if any of the CPUs in the core has yielded but OS has not requested
CEDE or CONFER, then consider this CPU to be preempted.

Correct detection of preempted CPUs is important for detecting idle
CPUs/cores in task scheduler.

Tested-by: Aboorva Devarajan <aboorvad@linux.vnet.ibm.com>
Reviewed-by: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
---
Changelog:
v2 (http://lkml.kernel.org/r/20231018155838.2332822-1-srikar@linux.vnet.ibm.com) -> v4:
Resolved comments from Michael Ellerman

v1 (http://lkml.kernel.org/r/20231009051740.17683-1-srikar@linux.vnet.ibm.com) -> v2:
Handle lppaca_of(cpu) in !PPC_SPLPAR case.
1. Fixed some compilation issues reported by kernelbot
	a. https://lore.kernel.org/oe-kbuild-all/202310102341.K0sgoqQL-lkp@intel.com/
	b.  https://lore.kernel.org/oe-kbuild-all/202310091636.lElmJkYV-lkp@intel.com/
2. Resolved comments from Shrikanth

---
 arch/powerpc/include/asm/paravirt.h | 47 +++++++++++++++++++++++++++--
 1 file changed, 44 insertions(+), 3 deletions(-)

diff --git a/arch/powerpc/include/asm/paravirt.h b/arch/powerpc/include/asm/paravirt.h
index e08513d73119..ac4279208d63 100644
--- a/arch/powerpc/include/asm/paravirt.h
+++ b/arch/powerpc/include/asm/paravirt.h
@@ -71,6 +71,11 @@ static inline void yield_to_any(void)
 {
 	plpar_hcall_norets_notrace(H_CONFER, -1, 0);
 }
+
+static inline bool is_vcpu_idle(int vcpu)
+{
+	return lppaca_of(vcpu).idle;
+}
 #else
 static inline bool is_shared_processor(void)
 {
@@ -100,6 +105,10 @@ static inline void prod_cpu(int cpu)
 	___bad_prod_cpu(); /* This would be a bug */
 }
 
+static inline bool is_vcpu_idle(int vcpu)
+{
+	return false;
+}
 #endif
 
 #define vcpu_is_preempted vcpu_is_preempted
@@ -121,9 +130,23 @@ static inline bool vcpu_is_preempted(int cpu)
 	if (!is_shared_processor())
 		return false;
 
+	/*
+	 * If the hypervisor has dispatched the target CPU on a physical
+	 * processor, then the target CPU is definitely not preempted.
+	 */
+	if (!(yield_count_of(cpu) & 1))
+		return false;
+
+	/*
+	 * If the target CPU has yielded to Hypervisor but OS has not
+	 * requested idle then the target CPU is definitely preempted.
+	 */
+	if (!is_vcpu_idle(cpu))
+		return true;
+
 #ifdef CONFIG_PPC_SPLPAR
 	if (!is_kvm_guest()) {
-		int first_cpu;
+		int first_cpu, i;
 
 		/*
 		 * The result of vcpu_is_preempted() is used in a
@@ -149,11 +172,29 @@ static inline bool vcpu_is_preempted(int cpu)
 		 */
 		if (cpu_first_thread_sibling(cpu) == first_cpu)
 			return false;
+
+		/*
+		 * If any of the threads of the target CPU's core are not
+		 * preempted or ceded, then consider target CPU to be
+		 * non-preempted.
+		 */
+		first_cpu = cpu_first_thread_sibling(cpu);
+		for (i = first_cpu; i < first_cpu + threads_per_core; i++) {
+			if (i == cpu)
+				continue;
+			if (!(yield_count_of(i) & 1))
+				return false;
+			if (!is_vcpu_idle(i))
+				return true;
+		}
 	}
 #endif
 
-	if (yield_count_of(cpu) & 1)
-		return true;
+	/*
+	 * None of the threads in target CPU's core are running but none of
+	 * them were preempted too. Hence assume the target CPU to be
+	 * non-preempted.
+	 */
 	return false;
 }
 

base-commit: eddc90ea2af5933249ea1a78119f2c8ef8d07156
-- 
2.31.1


             reply	other threads:[~2023-10-19  9:15 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-10-19  9:14 Srikar Dronamraju [this message]
2023-10-19  9:14 ` [PATCH v4] powerpc/paravirt: Improve vcpu_is_preempted Srikar Dronamraju
2023-10-27  9:59 ` Michael Ellerman
2023-10-27  9:59   ` Michael Ellerman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20231019091452.95260-1-srikar@linux.vnet.ibm.com \
    --to=srikar@linux.vnet.ibm.com \
    --cc=aboorvad@linux.vnet.ibm.com \
    --cc=akaher@vmware.com \
    --cc=amakhalov@vmware.com \
    --cc=christophe.leroy@csgroup.eu \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=mpe@ellerman.id.au \
    --cc=npiggin@gmail.com \
    --cc=pv-drivers@vmware.com \
    --cc=sshegde@linux.vnet.ibm.com \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.