From: Andrew Cooper <andrew.cooper3@citrix.com>
To: dongxiao.xu@intel.com
Cc: xen-devel@lists.xen.org
Subject: Re: [PATCH v2 1/8] x86: detect and initialize Cache QoS Monitoring feature
Date: Thu, 21 Nov 2013 12:14:28 +0000 [thread overview]
Message-ID: <528DF924.4030702@citrix.com> (raw)
In-Reply-To: <1385018444-104477-2-git-send-email-dongxiao.xu@intel.com>
On 21/11/13 07:20, dongxiao.xu@intel.com wrote:
> From: Dongxiao Xu <dongxiao.xu@intel.com>
>
> Detect platform QoS feature status and enumerate the resource types,
> one of which is to monitor the L3 cache occupancy.
>
> Also introduce a Xen grub command line parameter to control the
> QoS feature status globally.
>
> Signed-off-by: Jiongxi Li <jiongxi.li@intel.com>
> Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
> ---
> xen/arch/x86/Makefile | 1 +
> xen/arch/x86/cpu/intel.c | 6 +++
> xen/arch/x86/pqos.c | 89 ++++++++++++++++++++++++++++++++++++++
> xen/arch/x86/setup.c | 3 ++
> xen/include/asm-x86/cpufeature.h | 1 +
> xen/include/asm-x86/pqos.h | 37 ++++++++++++++++
> 6 files changed, 137 insertions(+)
> create mode 100644 xen/arch/x86/pqos.c
> create mode 100644 xen/include/asm-x86/pqos.h
>
> diff --git a/xen/arch/x86/Makefile b/xen/arch/x86/Makefile
> index d502bdf..54962e0 100644
> --- a/xen/arch/x86/Makefile
> +++ b/xen/arch/x86/Makefile
> @@ -58,6 +58,7 @@ obj-y += crash.o
> obj-y += tboot.o
> obj-y += hpet.o
> obj-y += xstate.o
> +obj-y += pqos.o
>
> obj-$(crash_debug) += gdbstub.o
>
> diff --git a/xen/arch/x86/cpu/intel.c b/xen/arch/x86/cpu/intel.c
> index 27fe762..f0d83ea 100644
> --- a/xen/arch/x86/cpu/intel.c
> +++ b/xen/arch/x86/cpu/intel.c
> @@ -230,6 +230,12 @@ static void __devinit init_intel(struct cpuinfo_x86 *c)
> ( c->cpuid_level >= 0x00000006 ) &&
> ( cpuid_eax(0x00000006) & (1u<<2) ) )
> set_bit(X86_FEATURE_ARAT, c->x86_capability);
> +
> + /* Check platform QoS monitoring capability */
> + if ((c->cpuid_level >= 0x00000007) &&
> + (cpuid_ebx(0x00000007) & (1u<<12)))
> + set_bit(X86_FEATURE_QOSM, c->x86_capability);
> +
> }
>
> static struct cpu_dev intel_cpu_dev __cpuinitdata = {
> diff --git a/xen/arch/x86/pqos.c b/xen/arch/x86/pqos.c
> new file mode 100644
> index 0000000..e6ab416
> --- /dev/null
> +++ b/xen/arch/x86/pqos.c
> @@ -0,0 +1,89 @@
> +/*
> + * pqos.c: Platform QoS related service for guest.
> + *
> + * Copyright (c) 2013, Intel Corporation
> + * Author: Jiongxi Li <jiongxi.li@intel.com>
> + * Author: Dongxiao Xu <dongxiao.xu@intel.com>
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License along with
> + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
> + * Place - Suite 330, Boston, MA 02111-1307 USA.
> + */
> +#include <asm/processor.h>
> +#include <xen/init.h>
> +#include <asm/pqos.h>
> +
> +static bool_t pqos_enabled = 1;
I cant see anywhere in the series where this is used outside of an init
function, in which case it should be __initdata.
> +boolean_param("pqos", pqos_enabled);
> +
> +unsigned int cqm_res_count = 0;
> +unsigned int cqm_upscaling_factor = 0;
> +bool_t cqm_enabled = 0;
> +struct cqm_res_struct *cqm_res_array = NULL;
> +
> +static void __init init_cqm(void)
> +{
> + unsigned int eax, edx;
> + unsigned int max_cqm_rmid;
> +
> + cpuid_count(0xf, 1, &eax, &cqm_upscaling_factor, &max_cqm_rmid, &edx);
> + if ( !(edx & QOS_MONITOR_EVTID_L3) )
> + return;
> +
> + cqm_res_count = max_cqm_rmid + 1;
> +
Range check on cqm_res_count ? If max_cqm_rmid ends up as -1 from the
cpuid, we will allocate a 0 length array and crash later when reserving
RMID 0
> + cqm_res_array = xzalloc_array(struct cqm_res_struct, cqm_res_count);
> + if ( !cqm_res_array )
> + {
> + cqm_res_count = 0;
> + return;
> + }
> +
> + cqm_enabled = 1;
> +
> + /* Reserve RMID 0 for all domains not being monitored */
> + cqm_res_array[0].inuse = 1;
> +
> + printk(XENLOG_INFO "Cache QoS Monitoring Enabled.\n");
> +}
> +
> +static void __init init_qos_monitor(void)
> +{
> + unsigned int qm_features;
> + unsigned int eax, ebx, ecx;
> +
> + if ( !(boot_cpu_has(X86_FEATURE_QOSM)) )
> + return;
> +
> + cpuid_count(0xf, 0, &eax, &ebx, &ecx, &qm_features);
> +
> + if ( qm_features & QOS_MONITOR_TYPE_L3 )
> + init_cqm();
> +}
> +
> +void __init init_platform_qos(void)
> +{
> + if ( !pqos_enabled )
> + return;
> +
> + init_qos_monitor();
> +}
> +
> +/*
> + * Local variables:
> + * mode: C
> + * c-file-style: "BSD"
> + * c-basic-offset: 4
> + * tab-width: 4
> + * indent-tabs-mode: nil
> + * End:
> + */
> diff --git a/xen/arch/x86/setup.c b/xen/arch/x86/setup.c
> index 5bf4ee0..95418e4 100644
> --- a/xen/arch/x86/setup.c
> +++ b/xen/arch/x86/setup.c
> @@ -48,6 +48,7 @@
> #include <asm/setup.h>
> #include <xen/cpu.h>
> #include <asm/nmi.h>
> +#include <asm/pqos.h>
>
> /* opt_nosmp: If true, secondary processors are ignored. */
> static bool_t __initdata opt_nosmp;
> @@ -1402,6 +1403,8 @@ void __init __start_xen(unsigned long mbi_p)
>
> domain_unpause_by_systemcontroller(dom0);
>
> + init_platform_qos();
> +
Does this want to be slightly earlier during startup? Does it make sense
to have dom0 monitored? Is it possible to monitor Xen's L3 usage as well?
> reset_stack_and_jump(init_done);
> }
>
> diff --git a/xen/include/asm-x86/cpufeature.h b/xen/include/asm-x86/cpufeature.h
> index 1cfaf94..ca59668 100644
> --- a/xen/include/asm-x86/cpufeature.h
> +++ b/xen/include/asm-x86/cpufeature.h
> @@ -147,6 +147,7 @@
> #define X86_FEATURE_ERMS (7*32+ 9) /* Enhanced REP MOVSB/STOSB */
> #define X86_FEATURE_INVPCID (7*32+10) /* Invalidate Process Context ID */
> #define X86_FEATURE_RTM (7*32+11) /* Restricted Transactional Memory */
> +#define X86_FEATURE_QOSM (7*32+12) /* Platform QoS monitoring capability */
> #define X86_FEATURE_NO_FPU_SEL (7*32+13) /* FPU CS/DS stored as zero */
> #define X86_FEATURE_SMAP (7*32+20) /* Supervisor Mode Access Prevention */
>
> diff --git a/xen/include/asm-x86/pqos.h b/xen/include/asm-x86/pqos.h
> new file mode 100644
> index 0000000..934d68a
> --- /dev/null
> +++ b/xen/include/asm-x86/pqos.h
It might make sense to split this file. The QOS_* defines could be
argued as asm-x86, (perhaps asm-x86/qpos-defines.h)
However, the cqm_res_struct and function declaration are very certainly
not asm code, and would probably better live in arch/x86/qpos.h
> @@ -0,0 +1,37 @@
> +/*
> + * pqos.h: Platform QoS related service for guest.
> + *
> + * Copyright (c) 2013, Intel Corporation
> + * Author: Jiongxi Li <jiongxi.li@intel.com>
> + * Author: Dongxiao Xu <dongxiao.xu@intel.com>
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License along with
> + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
> + * Place - Suite 330, Boston, MA 02111-1307 USA.
> + */
> +#ifndef ASM_PQOS_H
> +#define ASM_PQOS_H
> +
> +/* QoS Resource Type Enumeration */
> +#define QOS_MONITOR_TYPE_L3 0x2
> +
> +/* QoS Monitoring Event ID */
> +#define QOS_MONITOR_EVTID_L3 0x1
> +
> +struct cqm_res_struct {
This name is redundant with the extra "_struct" at the end. How about
cqm_res or cqm_resource ?
> + domid_t domain_id;
You will need to include one of the types headers.
> + bool_t inuse;
> +};
> +
> +void init_platform_qos(void);
> +
> +#endif
next prev parent reply other threads:[~2013-11-21 12:14 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 [this message]
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 ` [PATCH v2 8/8] x86: enable CQM monitoring for each domain RMID dongxiao.xu
2013-11-21 14:19 ` 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=528DF924.4030702@citrix.com \
--to=andrew.cooper3@citrix.com \
--cc=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).