From: dongxiao.xu@intel.com
To: xen-devel@lists.xen.org
Subject: [PATCH v2 8/8] x86: enable CQM monitoring for each domain RMID
Date: Thu, 21 Nov 2013 15:20:44 +0800 [thread overview]
Message-ID: <1385018444-104477-9-git-send-email-dongxiao.xu@intel.com> (raw)
In-Reply-To: <1385018444-104477-1-git-send-email-dongxiao.xu@intel.com>
From: Dongxiao Xu <dongxiao.xu@intel.com>
If the CQM service is attached to a domain, its related RMID will be set
to hardware for monitoring when the domain's vcpu is scheduled in. When
the domain's vcpu is scheduled out, RMID 0 (system reserved) will be set
for monitoring.
Signed-off-by: Jiongxi Li <jiongxi.li@intel.com>
Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
---
xen/arch/x86/domain.c | 5 +++++
xen/arch/x86/pqos.c | 21 ++++++++++++++++++++-
xen/include/asm-x86/msr-index.h | 1 +
xen/include/asm-x86/pqos.h | 1 +
4 files changed, 27 insertions(+), 1 deletion(-)
diff --git a/xen/arch/x86/domain.c b/xen/arch/x86/domain.c
index 9725649..1eda0ab 100644
--- a/xen/arch/x86/domain.c
+++ b/xen/arch/x86/domain.c
@@ -1372,6 +1372,8 @@ static void __context_switch(void)
{
memcpy(&p->arch.user_regs, stack_regs, CTXT_SWITCH_STACK_BYTES);
vcpu_save_fpu(p);
+ if ( system_supports_cqm() )
+ cqm_assoc_rmid(0);
p->arch.ctxt_switch_from(p);
}
@@ -1396,6 +1398,9 @@ static void __context_switch(void)
}
vcpu_restore_fpu_eager(n);
n->arch.ctxt_switch_to(n);
+
+ if ( system_supports_cqm() && n->domain->arch.pqos_cqm_rmid > 0 )
+ cqm_assoc_rmid(n->domain->arch.pqos_cqm_rmid);
}
gdt = !is_pv_32on64_vcpu(n) ? per_cpu(gdt_table, cpu) :
diff --git a/xen/arch/x86/pqos.c b/xen/arch/x86/pqos.c
index d95784c..0cec172 100644
--- a/xen/arch/x86/pqos.c
+++ b/xen/arch/x86/pqos.c
@@ -27,6 +27,7 @@
#include <asm/pqos.h>
static bool_t pqos_enabled = 1;
+static unsigned int rmid_bitwidth;
boolean_param("pqos", pqos_enabled);
static void read_qm_data(void *arg)
@@ -43,6 +44,14 @@ static void get_generic_qm_info(struct qm_element *qm_element)
on_selected_cpus(cpumask_of(cpu), read_qm_data, qm_element, 1);
}
+void cqm_assoc_rmid(uint16_t rmid)
+{
+ uint64_t val;
+ uint64_t rmid_mask = ~(~0ull << rmid_bitwidth);
+ rdmsrl(MSR_IA32_PQR_ASSOC, val);
+ wrmsrl(MSR_IA32_PQR_ASSOC, (val & ~(rmid_mask)) | (rmid & rmid_mask));
+}
+
unsigned int cqm_res_count = 0;
unsigned int cqm_upscaling_factor = 0;
bool_t cqm_enabled = 0;
@@ -78,13 +87,23 @@ static void __init init_cqm(void)
static void __init init_qos_monitor(void)
{
unsigned int qm_features;
- unsigned int eax, ebx, ecx;
+ unsigned int eax, ebx, ecx, i = 0;
if ( !(boot_cpu_has(X86_FEATURE_QOSM)) )
return;
cpuid_count(0xf, 0, &eax, &ebx, &ecx, &qm_features);
+ while (1)
+ {
+ if ( (1 << i) - 1 >= ebx )
+ {
+ rmid_bitwidth = i;
+ break;
+ }
+ i++;
+ }
+
if ( qm_features & QOS_MONITOR_TYPE_L3 )
init_cqm();
}
diff --git a/xen/include/asm-x86/msr-index.h b/xen/include/asm-x86/msr-index.h
index 46ef165..45f4918 100644
--- a/xen/include/asm-x86/msr-index.h
+++ b/xen/include/asm-x86/msr-index.h
@@ -491,5 +491,6 @@
/* Platform QoS register */
#define MSR_IA32_QOSEVTSEL 0x00000c8d
#define MSR_IA32_QMC 0x00000c8e
+#define MSR_IA32_PQR_ASSOC 0x00000c8f
#endif /* __ASM_MSR_INDEX_H */
diff --git a/xen/include/asm-x86/pqos.h b/xen/include/asm-x86/pqos.h
index 5c86c5d..2bb983b 100644
--- a/xen/include/asm-x86/pqos.h
+++ b/xen/include/asm-x86/pqos.h
@@ -51,5 +51,6 @@ unsigned int get_cqm_count(void);
unsigned int get_cqm_avail(void);
void get_cqm_info(uint32_t rmid, cpumask_t cpu_cqmdata_map,
struct xen_domctl_getdomcqminfo *info);
+void cqm_assoc_rmid(uint16_t rmid);
#endif
--
1.7.9.5
next prev parent reply other threads:[~2013-11-21 7:20 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-21 7:20 [PATCH v2 0/8] enable Cache QoS Monitoring (CQM) feature dongxiao.xu
2013-11-21 7:20 ` [PATCH v2 1/8] x86: detect and initialize Cache QoS Monitoring feature dongxiao.xu
2013-11-21 12:14 ` Andrew Cooper
2013-11-21 12:19 ` Andrew Cooper
2013-11-25 3:06 ` Xu, Dongxiao
2013-11-25 15:40 ` Andrew Cooper
2013-11-25 8:57 ` Xu, Dongxiao
2013-11-25 15:58 ` Andrew Cooper
2013-11-21 7:20 ` [PATCH v2 2/8] x86: handle CQM resource when creating/destroying guests dongxiao.xu
2013-11-21 12:33 ` Andrew Cooper
2013-11-25 3:21 ` Xu, Dongxiao
2013-11-25 16:02 ` Andrew Cooper
2013-11-21 7:20 ` [PATCH v2 3/8] tools: " dongxiao.xu
2013-11-21 7:20 ` [PATCH v2 4/8] x86: dynamically attach/detach CQM service for a guest dongxiao.xu
2013-11-21 12:50 ` Andrew Cooper
2013-11-25 3:26 ` Xu, Dongxiao
2013-11-25 16:05 ` Andrew Cooper
2013-11-25 21:06 ` Konrad Rzeszutek Wilk
2013-11-21 7:20 ` [PATCH v2 5/8] tools: " dongxiao.xu
2013-11-25 21:00 ` Konrad Rzeszutek Wilk
2013-11-25 21:01 ` Konrad Rzeszutek Wilk
2013-11-21 7:20 ` [PATCH v2 6/8] x86: get per domain CQM information dongxiao.xu
2013-11-21 14:09 ` Andrew Cooper
2013-11-25 6:20 ` Xu, Dongxiao
2013-11-25 16:28 ` Andrew Cooper
2013-11-21 7:20 ` [PATCH v2 7/8] tools: " dongxiao.xu
2013-11-21 7:20 ` dongxiao.xu [this message]
2013-11-21 14:19 ` [PATCH v2 8/8] x86: enable CQM monitoring for each domain RMID Andrew Cooper
2013-11-25 7:22 ` Xu, Dongxiao
2013-11-25 16:32 ` Andrew Cooper
2013-11-21 14:36 ` [PATCH v2 0/8] enable Cache QoS Monitoring (CQM) feature Andrew Cooper
2013-11-25 7:24 ` Xu, Dongxiao
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=1385018444-104477-9-git-send-email-dongxiao.xu@intel.com \
--to=dongxiao.xu@intel.com \
--cc=xen-devel@lists.xen.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;
as well as URLs for NNTP newsgroup(s).