From: Len Brown <lenb@kernel.org>
To: Matthew Garrett <mjg59@srcf.ucam.org>
Cc: "Yu, Luming" <luming.yu@intel.com>,
Philip Langdale <philipl@overt.org>,
Jeff Garrett <jeff@jgarrett.org>,
Andi Kleen <andi@firstfloor.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
"linux-acpi@vger.kernel.org" <linux-acpi@vger.kernel.org>,
"venki@google.com" <venki@google.com>
Subject: [PATCH] ACPI: make acpi_idle Nehalem-aware
Date: Wed, 21 Jul 2010 17:31:27 -0400 (EDT) [thread overview]
Message-ID: <alpine.LFD.2.00.1007211729550.409@localhost.localdomain> (raw)
In-Reply-To: <20100525185507.GA15997@srcf.ucam.org>
From: Len Brown <len.brown@intel.com>
The BIOS exports deep C-states on modern Intel processors
as "C3-type" to satisfy various legacy Operating Systems.
However, the hardware actually supports C2-type, and does
not require the extra costs of C3-type.
One of the costs is to check the BM_STS (Bus Master Status)
bit before entering C3, and instead choose a shallower C-state
if there was "recent bus master activity".
We have found a number of systems in the field that erroneously
set BM_STS and prevent entry into deep C-states.
Re-define BIOS presented C3-type states as C2-type states
on modern processors to avoid this issue.
If a device in the system really does want to prevent use
of a deep C-state, its Linux driver should register its
constraints via pm_qos_add_request().
https://bugzilla.kernel.org/show_bug.cgi?id=15886
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/processor_idle.c | 38 ++++++++++++++++++++++++++++++++++++++
1 files changed, 38 insertions(+), 0 deletions(-)
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index b1b3856..14d1a0c 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -607,6 +607,38 @@ static void acpi_processor_power_verify_c3(struct acpi_processor *pr,
return;
}
+/*
+ * Modern Intel processors support only ACPI C2-type C-states.
+ * But the BIOS tends to report its deepest C-state as C3-type
+ * to satisfy various old operating systems. We can skip
+ * C3 OS overhead by treating the deep-states as C2-type.
+ * Also, we can avoid checking BM_STS, which on some systems
+ * erroneously prevents entry into C3-type states.
+ */
+static int acpi_c3type_is_really_c2type(void) {
+
+ if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL)
+ return 0;
+
+ if (boot_cpu_data.x86 != 6)
+ return 0;
+
+ switch(boot_cpu_data.x86_model) {
+ case 0x1A: /* Core i7, Xeon 5500 series */
+ case 0x1E: /* Core i7 and i5 Processor */
+ case 0x1F: /* Core i7 and i5 Processor */
+ case 0x2E: /* NHM-EX Xeon */
+ case 0x2F: /* WSM-EX Xeon */
+ case 0x25: /* WSM */
+ case 0x2C: /* WSM */
+ case 0x2A: /* SNB */
+ case 0x2D: /* SNB Xeon */
+ return 1;
+ default:
+ return 0;
+ }
+}
+
static int acpi_processor_power_verify(struct acpi_processor *pr)
{
unsigned int i;
@@ -617,6 +649,12 @@ static int acpi_processor_power_verify(struct acpi_processor *pr)
for (i = 1; i < ACPI_PROCESSOR_MAX_POWER && i <= max_cstate; i++) {
struct acpi_processor_cx *cx = &pr->power.states[i];
+ if ((cx->type == ACPI_STATE_C3)
+ && acpi_c3type_is_really_c2type()) {
+ ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Redefining C3-type to C2\n"));
+ cx->type = ACPI_STATE_C2;
+ }
+
switch (cx->type) {
case ACPI_STATE_C1:
cx->valid = 1;
--
1.7.2.rc3.43.g24e7a
next prev parent reply other threads:[~2010-07-21 21:31 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-26 8:47 acpi_idle: Very idle Core i7 machine never enters C3 Jeff Garrett
2010-01-26 12:41 ` peng huang
2010-01-26 14:59 ` Jeff Garrett
2010-01-27 13:27 ` peng huang
2010-02-05 16:22 ` Jeff Garrett
2010-01-26 21:45 ` Andi Kleen
2010-02-05 16:09 ` Jeff Garrett
2010-02-05 17:45 ` Len Brown
2010-02-05 20:53 ` Jeff Garrett
2010-04-27 2:40 ` Philip Langdale
2010-04-27 7:26 ` Len Brown
2010-04-27 15:41 ` Philip Langdale
2010-04-27 12:47 ` Jeff Garrett
2010-04-30 14:57 ` Philip Langdale
2010-04-30 16:25 ` Len Brown
2010-04-30 17:44 ` Matthew Garrett
2010-04-30 18:35 ` Philip Langdale
2010-05-25 5:43 ` Len Brown
2010-05-25 5:59 ` Yu, Luming
2010-05-25 12:39 ` Matthew Garrett
2010-05-25 12:43 ` Matthew Garrett
2010-05-25 15:33 ` Len Brown
2010-05-25 18:55 ` Matthew Garrett
2010-07-21 21:31 ` Len Brown [this message]
2010-07-22 0:53 ` [PATCH] ACPI: make acpi_idle Nehalem-aware Venkatesh Pallipadi
2010-07-22 7:47 ` Andi Kleen
2010-07-22 15:57 ` Len Brown
2010-07-22 21:21 ` [PATCH] ACPI: skip checking BM_STS if the BIOS doesn't ask for it Len Brown
2010-07-22 21:40 ` [PATCH] ACPI: create "processor.bm_check_disable" boot param Len Brown
2010-07-26 7:24 ` Andi Kleen
2010-07-27 0:19 ` Len Brown
2010-07-27 11:28 ` Andi Kleen
2010-07-28 18:58 ` Len Brown
2010-07-22 21:25 ` [PATCH] ACPI: make acpi_idle Nehalem-aware Iain
2010-07-22 21:53 ` Iain
2010-07-22 22:01 ` Len Brown
2010-07-23 12:40 ` Iain
2010-08-03 6:55 ` Pavel Machek
2010-08-03 7:05 ` Andi Kleen
2010-05-25 12:37 ` acpi_idle: Very idle Core i7 machine never enters C3 Matthew Garrett
2010-05-25 15:40 ` Len Brown
2010-07-22 5:34 ` Len Brown
2010-02-01 14:10 ` Pavel Machek
2010-02-05 16:30 ` Jeff Garrett
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=alpine.LFD.2.00.1007211729550.409@localhost.localdomain \
--to=lenb@kernel.org \
--cc=andi@firstfloor.org \
--cc=jeff@jgarrett.org \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luming.yu@intel.com \
--cc=mjg59@srcf.ucam.org \
--cc=philipl@overt.org \
--cc=venki@google.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox