From: Andi Kleen <andi@firstfloor.org>
To: peterz@infradead.org
Cc: eranian@google.com, linux-kernel@vger.kernel.org,
Andi Kleen <ak@linux.intel.com>
Subject: [PATCH] perf, x86: Widen Haswell OFFCORE mask
Date: Fri, 11 Jul 2014 17:06:15 -0700 [thread overview]
Message-ID: <1405123575-17113-1-git-send-email-andi@firstfloor.org> (raw)
From: Andi Kleen <ak@linux.intel.com>
Haswell supports more bits in the offcore_rsp_* MSRs than Sandy
Bridge. Previously the Haswell code was using the Sandy Bridge
extra register definitions, which prevented users from setting
all of these bits. This in term did not allow to set some valid
SNOOP_* bits, among others.
I allowed all bits the CPU does not #GP on. This is ok because
it's protected by a model check.
Add a new extra_regs table for Haswell and use. Except for the
widened mask it is identical to Sandy Bridge.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
arch/x86/kernel/cpu/perf_event_intel.c | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index adb02aa..5d41e2b 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -189,6 +189,14 @@ static struct extra_reg intel_snbep_extra_regs[] __read_mostly = {
EVENT_EXTRA_END
};
+static struct extra_reg intel_hsw_extra_regs[] __read_mostly = {
+ /* must define OFFCORE_RSP_X first, see intel_fixup_er() */
+ INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x3fffff8fffull, RSP_0),
+ INTEL_UEVENT_EXTRA_REG(0x01bb, MSR_OFFCORE_RSP_1, 0x3fffff8fffull, RSP_1),
+ INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd),
+ EVENT_EXTRA_END
+};
+
EVENT_ATTR_STR(mem-loads, mem_ld_nhm, "event=0x0b,umask=0x10,ldlat=3");
EVENT_ATTR_STR(mem-loads, mem_ld_snb, "event=0xcd,umask=0x1,ldlat=3");
EVENT_ATTR_STR(mem-stores, mem_st_snb, "event=0xcd,umask=0x2");
@@ -2504,7 +2512,7 @@ __init int intel_pmu_init(void)
x86_pmu.event_constraints = intel_hsw_event_constraints;
x86_pmu.pebs_constraints = intel_hsw_pebs_event_constraints;
- x86_pmu.extra_regs = intel_snb_extra_regs;
+ x86_pmu.extra_regs = intel_hsw_extra_regs;
x86_pmu.pebs_aliases = intel_pebs_aliases_snb;
/* all extra regs are per-cpu when HT is on */
x86_pmu.er_flags |= ERF_HAS_RSP_1;
--
1.9.3
next reply other threads:[~2014-07-12 0:06 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-07-12 0:06 Andi Kleen [this message]
2014-07-15 0:41 ` Stephane Eranian
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=1405123575-17113-1-git-send-email-andi@firstfloor.org \
--to=andi@firstfloor.org \
--cc=ak@linux.intel.com \
--cc=eranian@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®