All of lore.kernel.org
 help / color / mirror / Atom feed
From: Artem Bityutskiy <dedekind1@gmail.com>
To: x86@kernel.org, "Rafael J. Wysocki" <rafael@kernel.org>
Cc: Linux PM Mailing List <linux-pm@vger.kernel.org>,
	Arjan van de Ven <arjan@linux.intel.com>,
	Artem Bityutskiy <dedekind1@gmail.com>
Subject: [PATCH v3 2/2] intel_idle: add C0.2 state for Sapphire Rapids Xeon
Date: Sat, 10 Jun 2023 21:35:18 +0300	[thread overview]
Message-ID: <20230610183518.4061159-3-dedekind1@gmail.com> (raw)
In-Reply-To: <20230610183518.4061159-1-dedekind1@gmail.com>

From: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>

Add Sapphire Rapids Xeon C0.2 state support. This state has a lower exit
latency comparing to C1, and saves energy comparing to POLL.

C0.2 may also improve performance (e.g., as measured by 'hackbench'), because
idle CPU power savings in C0.2 increase busy CPU power budget and therefore,
improve turbo boost of the busy CPU.

Suggested-by: Len Brown <len.brown@intel.com>
Suggested-by: Arjan Van De Ven <arjan.van.de.ven@intel.com>
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
---
 drivers/idle/intel_idle.c | 44 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 43 insertions(+), 1 deletion(-)

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 0bf5e9f5bed8..51f56001e2cd 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -130,6 +130,11 @@ static unsigned int mwait_substates __initdata;
 #define flg2MWAIT(flags) (((flags) >> 24) & 0xFF)
 #define MWAIT2flg(eax) ((eax & 0xFF) << 24)
 
+/*
+ * The maximum possible 'umwait' deadline value.
+ */
+#define UMWAIT_MAX_DEADLINE (~((u64)0))
+
 static __always_inline int __intel_idle(struct cpuidle_device *dev,
 					struct cpuidle_driver *drv, int index)
 {
@@ -263,6 +268,32 @@ static __cpuidle int intel_idle_s2idle(struct cpuidle_device *dev,
 	return 0;
 }
 
+/**
+ * intel_idle_umwait_irq - Request C0.x using the 'umwait' instruction.
+ * @dev: cpuidle device of the target CPU.
+ * @drv: cpuidle driver (assumed to point to intel_idle_driver).
+ * @index: Target idle state index.
+ *
+ * Request C0.1 or C0.2 using 'umwait' instruction with interrupts enabled.
+ */
+static __cpuidle int intel_idle_umwait_irq(struct cpuidle_device *dev,
+					   struct cpuidle_driver *drv,
+					   int index)
+{
+	u32 state = flg2MWAIT(drv->states[index].flags);
+
+	raw_local_irq_enable();
+	/*
+	 * Use the maximum possible deadline value. This means that 'C0.x'
+	 * residency will be limited by the global limit in
+	 * 'IA32_UMWAIT_CONTROL'.
+	 */
+	umwait_idle(UMWAIT_MAX_DEADLINE, state);
+	raw_local_irq_disable();
+
+	return index;
+}
+
 /*
  * States are indexed by the cstate number,
  * which is also the index into the MWAIT hint array.
@@ -1006,6 +1037,13 @@ static struct cpuidle_state adl_n_cstates[] __initdata = {
 };
 
 static struct cpuidle_state spr_cstates[] __initdata = {
+	{
+		.name = "C0.2",
+		.desc = "UMWAIT C0.2",
+		.flags = MWAIT2flg(TPAUSE_C02_STATE) | CPUIDLE_FLAG_IRQ_ENABLE,
+		.exit_latency_ns = 200,
+		.target_residency_ns = 200,
+		.enter = &intel_idle_umwait_irq, },
 	{
 		.name = "C1",
 		.desc = "MWAIT 0x00",
@@ -1904,7 +1942,9 @@ static void state_update_enter_method(struct cpuidle_state *state, int cstate)
 		}
 		return;
 	}
-	if (state->enter == intel_idle_hlt_irq_on)
+
+	if (state->enter == intel_idle_hlt_irq_on ||
+	    state->enter == intel_idle_umwait_irq)
 		return; /* no update scenarios */
 
 	if (state->flags & CPUIDLE_FLAG_INIT_XSTATE) {
@@ -1959,6 +1999,8 @@ static bool should_verify_mwait(struct cpuidle_state *state)
 		return false;
 	if (state->enter == intel_idle_hlt_irq_on)
 		return false;
+	if (state->enter == intel_idle_umwait_irq)
+		return false;
 
 	return true;
 }
-- 
2.40.1


  parent reply	other threads:[~2023-06-10 18:35 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-06-10 18:35 [PATCH v3 0/3] Sapphire Rapids C0.x idle states support Artem Bityutskiy
2023-06-10 18:35 ` [PATCH v3 1/2] x86/mwait: Add support for idle via umwait Artem Bityutskiy
2023-06-29 19:03   ` Rafael J. Wysocki
2023-06-29 22:04   ` Thomas Gleixner
2023-06-29 22:36     ` Thomas Gleixner
2023-06-30 16:54     ` Artem Bityutskiy
2023-07-07 17:13     ` Artem Bityutskiy
2023-06-10 18:35 ` Artem Bityutskiy [this message]
2023-06-29 22:05 ` [PATCH v3 0/3] Sapphire Rapids C0.x idle states support Thomas Gleixner

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=20230610183518.4061159-3-dedekind1@gmail.com \
    --to=dedekind1@gmail.com \
    --cc=arjan@linux.intel.com \
    --cc=linux-pm@vger.kernel.org \
    --cc=rafael@kernel.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.