From: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
To: Ingo Molnar <mingo@kernel.org>
Cc: Stephane Eranian <eranian@google.com>,
Michael Ellerman <michael@ellerman.id.au>,
Paul Mackerras <paulus@samba.org>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Maynard Johnson <mpjohn@us.ibm.com>,
Anton Blanchard <anton@samba.org>,
linux-kernel@vger.kernel.org, peterz@infradead.org,
mingo@elte.hu, ak@linux.intel.com, acme@redhat.com,
jolsa@redhat.com, namhyung.kim@lge.com
Subject: Re: [PATCH v7 07/18] perf: add generic memory sampling interface
Date: Fri, 15 Feb 2013 11:46:22 -0800 [thread overview]
Message-ID: <20130215194622.GA6396@us.ibm.com> (raw)
In-Reply-To: <20130125090158.GB31478@gmail.com>
>
> * Stephane Eranian <eranian@google.com> wrote:
>
> > This patch adds PERF_SAMPLE_DSRC.
> >
> > PERF_SAMPLE_DSRC collects the data source, i.e., where
> > did the data associated with the sampled instruction
> > come from. Information is stored in a perf_mem_dsrc
> > structure. It contains opcode, mem level, tlb, snoop,
> > lock information, subject to availability in hardware.
> >
> > Signed-off-by: Stephane Eranian <eranian@google.com>
> > ---
> > include/linux/perf_event.h | 2 ++
> > include/uapi/linux/perf_event.h | 68 +++++++++++++++++++++++++++++++++++++--
> > kernel/events/core.c | 6 ++++
> > 3 files changed, 74 insertions(+), 2 deletions(-)
> >
> > diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
> > index bb2429d..8fe4610 100644
> > --- a/include/linux/perf_event.h
> > +++ b/include/linux/perf_event.h
> > @@ -579,6 +579,7 @@ struct perf_sample_data {
> > u32 reserved;
> > } cpu_entry;
> > u64 period;
> > + union perf_mem_dsrc dsrc;
> > struct perf_callchain_entry *callchain;
> > struct perf_raw_record *raw;
> > struct perf_branch_stack *br_stack;
> > @@ -599,6 +600,7 @@ static inline void perf_sample_data_init(struct perf_sample_data *data,
> > data->regs_user.regs = NULL;
> > data->stack_user_size = 0;
> > data->weight = 0;
> > + data->dsrc.val = 0;
> > }
> >
> > extern void perf_output_sample(struct perf_output_handle *handle,
> > diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
> > index 3e6c394..3e4844c 100644
> > --- a/include/uapi/linux/perf_event.h
> > +++ b/include/uapi/linux/perf_event.h
> > @@ -133,9 +133,9 @@ enum perf_event_sample_format {
> > PERF_SAMPLE_REGS_USER = 1U << 12,
> > PERF_SAMPLE_STACK_USER = 1U << 13,
> > PERF_SAMPLE_WEIGHT = 1U << 14,
> > + PERF_SAMPLE_DSRC = 1U << 15,
> >
> > - PERF_SAMPLE_MAX = 1U << 15, /* non-ABI */
> > -
> > + PERF_SAMPLE_MAX = 1U << 16, /* non-ABI */
> > };
> >
> > /*
> > @@ -591,6 +591,7 @@ enum perf_event_type {
> > * u64 dyn_size; } && PERF_SAMPLE_STACK_USER
> > *
> > * { u64 weight; } && PERF_SAMPLE_WEIGHT
> > + * { u64 dsrc; } && PERF_SAMPLE_DSRC
> > * };
> > */
> > PERF_RECORD_SAMPLE = 9,
> > @@ -616,4 +617,67 @@ enum perf_callchain_context {
> > #define PERF_FLAG_FD_OUTPUT (1U << 1)
> > #define PERF_FLAG_PID_CGROUP (1U << 2) /* pid=cgroup id, per-cpu mode only */
> >
> > +union perf_mem_dsrc {
> > + __u64 val;
> > + struct {
> > + __u64 mem_op:5, /* type of opcode */
> > + mem_lvl:14, /* memory hierarchy level */
> > + mem_snoop:5, /* snoop mode */
> > + mem_lock:2, /* lock instr */
> > + mem_dtlb:7, /* tlb access */
> > + mem_rsvd:31;
> > + };
POWER could use an additional field:
mem_deratmiss:1
AFAICT, POWER does not currently save the mem_op, snoop or lock info
for the sampled instruction. I guess we can leave them set to 0.
> > +};
> > +
> > +/* type of opcode (load/store/prefetch,code) */
> > +#define PERF_MEM_OP_NA 0x01 /* not available */
> > +#define PERF_MEM_OP_LOAD 0x02 /* load instruction */
> > +#define PERF_MEM_OP_STORE 0x04 /* store instruction */
> > +#define PERF_MEM_OP_PFETCH 0x08 /* prefetch */
> > +#define PERF_MEM_OP_EXEC 0x10 /* code (execution) */
> > +#define PERF_MEM_OP_SHIFT 0
> > +
> > +/* memory hierarchy (memory level, hit or miss) */
> > +#define PERF_MEM_LVL_NA 0x01 /* not available */
> > +#define PERF_MEM_LVL_HIT 0x02 /* hit level */
> > +#define PERF_MEM_LVL_MISS 0x04 /* miss level */
> > +#define PERF_MEM_LVL_L1 0x08 /* L1 */
> > +#define PERF_MEM_LVL_LFB 0x10 /* Line Fill Buffer */
> > +#define PERF_MEM_LVL_L2 0x20 /* L2 hit */
> > +#define PERF_MEM_LVL_L3 0x40 /* L3 hit */
> > +#define PERF_MEM_LVL_LOC_RAM 0x80 /* Local DRAM */
> > +#define PERF_MEM_LVL_REM_RAM1 0x100 /* Remote DRAM (1 hop) */
> > +#define PERF_MEM_LVL_REM_RAM2 0x200 /* Remote DRAM (2 hops) */
> > +#define PERF_MEM_LVL_REM_CCE1 0x400 /* Remote Cache (1 hop) */
> > +#define PERF_MEM_LVL_REM_CCE2 0x800 /* Remote Cache (2 hops) */
> > +#define PERF_MEM_LVL_IO 0x1000 /* I/O memory */
> > +#define PERF_MEM_LVL_UNC 0x2000 /* Uncached memory */
> > +#define PERF_MEM_LVL_SHIFT 5
POWER saves following information to describe where the data was loaded from
after a Dcache or DTLB miss.
FROM_L2
FROM_L3
FROM_L2.1_SHR From another L2 or L3 on same chip, shared
FROM_L2.1_MOD From another L2 or L3 on same chip, modified
FROM_L3.1_SHR From remote L2 or L3, shared
FROM_L3.1_MOD From remote L2 or L3, modified
FROM_RL2L3_SHR From remote L2 or L3, shared
FROM_RL2L3_MOD From remote L2 or L3, modified
FROM_DL2L3_SHR From distant L2 or L3, shared
FROM_DL2L3_MOD From distant L2 or L3, modified
POWER uses 4 bits and a running count for its (currently) 13 possible values.
The macros in the patch use a separate bit for each level - is that to allow
selecting more than one level at the same time ? If so, we will need to reserve
a few more bits to allow for Power's memory levels that don't map to the above.
> > +
> > +/* snoop mode */
> > +#define PERF_MEM_SNOOP_NA 0x01 /* not available */
> > +#define PERF_MEM_SNOOP_NONE 0x02 /* no snoop */
> > +#define PERF_MEM_SNOOP_HIT 0x04 /* snoop hit */
> > +#define PERF_MEM_SNOOP_MISS 0x08 /* snoop miss */
> > +#define PERF_MEM_SNOOP_HITM 0x10 /* snoop hit modified */
> > +#define PERF_MEM_SNOOP_SHIFT 19
> > +
> > +/* locked instruction */
> > +#define PERF_MEM_LOCK_NA 0x01 /* not available */
> > +#define PERF_MEM_LOCK_LOCKED 0x02 /* locked transaction */
> > +#define PERF_MEM_LOCK_SHIFT 24
> > +
> > +/* TLB access */
> > +#define PERF_MEM_TLB_NA 0x01 /* not available */
> > +#define PERF_MEM_TLB_HIT 0x02 /* hit level */
> > +#define PERF_MEM_TLB_MISS 0x04 /* miss level */
> > +#define PERF_MEM_TLB_L1 0x08 /* L1 */
> > +#define PERF_MEM_TLB_L2 0x10 /* L2 */
> > +#define PERF_MEM_TLB_WK 0x20 /* Hardware Walker*/
> > +#define PERF_MEM_TLB_OS 0x40 /* OS fault handler */
> > +#define PERF_MEM_TLB_SHIFT 26
On POWER, like with the Dcache source above, we have 4 bits to describe where
the DTLB was loaded from after a dTLB miss.
We would probably need to allow more bits to for the memory level of the dTLB
load source.
> > +
> > +#define PERF_MEM_S(a, s) \
> > + (((u64)PERF_MEM_##a##_##s) << PERF_MEM_##a##_SHIFT)
> > +
>
> Would be nice to get feedback from PowerPC folks to see how well
> this matches their memory profiling hw capabilities?
>
> I suspect there's a lot of differences, but one can always hope
> ...
>
> If there's some hope for unification we could at least shape it
> in a way that they could pick up and extend.
Thanks for Ccing.
While on the topic of sampled instructions, POWER saves following information
(in addition to the above memory info) for sampled instructions.
- whether the sampled instruction encountered a stall
- the reasons for the stall.
- whether the instruction was from hypervisor
- there was a branch mis-predict,
- thresholding information
These are clubbed into an "event vector" that is saved for sampled
instructions. We have been meaning to find ways to present that to
to user space. Are there plans to retreive and present these too.
Sukadev
next prev parent reply other threads:[~2013-02-15 19:46 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-01-24 15:10 [PATCH v7 00/18] perf: add memory access sampling support Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 01/18] perf, x86: Support CPU specific sysfs events Stephane Eranian
2013-01-25 12:16 ` [tip:perf/x86] perf/x86: " tip-bot for Andi Kleen
2013-04-02 9:38 ` [tip:perf/core] " tip-bot for Andi Kleen
2013-01-24 15:10 ` [PATCH v7 02/18] perf/x86: improve sysfs event mapping with event string Stephane Eranian
2013-01-25 12:17 ` [tip:perf/x86] perf/x86: Improve " tip-bot for Stephane Eranian
2013-04-02 9:39 ` [tip:perf/core] " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 03/18] perf/x86: add flags to event constraints Stephane Eranian
2013-01-25 12:18 ` [tip:perf/x86] perf/x86: Add " tip-bot for Stephane Eranian
2013-04-02 9:40 ` [tip:perf/core] " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 04/18] perf, core: Add a concept of a weightened sample v2 Stephane Eranian
2013-01-25 12:20 ` [tip:perf/x86] perf/core: Add weighted samples tip-bot for Andi Kleen
2013-04-02 9:42 ` [tip:perf/core] " tip-bot for Andi Kleen
2013-01-24 15:10 ` [PATCH v7 05/18] perf, tools: Add support for weight v7 (modified) Stephane Eranian
2013-04-02 9:49 ` [tip:perf/core] perf " tip-bot for Andi Kleen
2013-01-24 15:10 ` [PATCH v7 06/18] perf: add support for PERF_SAMPLE_ADDR in dump_sampple() Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 07/18] perf: add generic memory sampling interface Stephane Eranian
2013-01-25 9:01 ` Ingo Molnar
2013-01-25 15:30 ` Stephane Eranian
2013-01-29 10:37 ` Michael Ellerman
2013-02-15 19:46 ` Sukadev Bhattiprolu [this message]
2013-02-16 2:45 ` Benjamin Herrenschmidt
2013-02-16 8:41 ` Ingo Molnar
2013-02-16 14:14 ` Stephane Eranian
2013-01-25 12:21 ` [tip:perf/x86] perf: Add " tip-bot for Stephane Eranian
2013-04-02 9:43 ` [tip:perf/core] " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 08/18] perf/x86: add memory profiling via PEBS Load Latency Stephane Eranian
2013-01-25 12:22 ` [tip:perf/x86] perf/x86: Add " tip-bot for Stephane Eranian
2013-04-02 9:44 ` [tip:perf/core] " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 09/18] perf/x86: export PEBS load latency threshold register to sysfs Stephane Eranian
2013-01-25 12:23 ` [tip:perf/x86] perf/x86: Export " tip-bot for Stephane Eranian
2013-04-02 9:45 ` [tip:perf/core] " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 10/18] perf/x86: add support for PEBS Precise Store Stephane Eranian
2013-01-25 12:24 ` [tip:perf/x86] perf/x86: Add " tip-bot for Stephane Eranian
2013-04-02 9:47 ` [tip:perf/core] " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 11/18] perf tools: add mem access sampling core support Stephane Eranian
2013-03-27 14:14 ` Jiri Olsa
2013-03-27 14:20 ` Peter Zijlstra
2013-03-27 14:34 ` Jiri Olsa
2013-03-27 14:48 ` Stephane Eranian
2013-03-27 16:56 ` Arnaldo Carvalho de Melo
2013-03-28 14:24 ` Stephane Eranian
2013-03-28 15:00 ` Arnaldo Carvalho de Melo
2013-03-28 15:06 ` Stephane Eranian
2013-03-28 15:12 ` Arnaldo Carvalho de Melo
2013-03-28 15:15 ` Stephane Eranian
2013-03-27 14:23 ` Jiri Olsa
2013-04-02 9:50 ` [tip:perf/core] perf tools: Add " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 12/18] perf report: add support for mem access profiling Stephane Eranian
2013-04-02 9:53 ` [tip:perf/core] perf report: Add " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 13/18] perf record: add " Stephane Eranian
2013-04-02 9:51 ` [tip:perf/core] perf record: Add " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 14/18] perf tools: add new mem command for memory " Stephane Eranian
2013-04-02 9:55 ` [tip:perf/core] perf tools: Add " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 15/18] perf: add PERF_RECORD_MISC_MMAP_DATA to RECORD_MMAP Stephane Eranian
2013-01-25 12:25 ` [tip:perf/x86] perf: Add " tip-bot for Stephane Eranian
2013-04-02 9:48 ` [tip:perf/core] " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 16/18] perf tools: detect data vs. text mappings Stephane Eranian
2013-04-02 9:57 ` [tip:perf/core] perf machine: Detect " tip-bot for Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 17/18] perf tools: Ignore ABS symbols when loading data maps Stephane Eranian
2013-01-24 15:10 ` [PATCH v7 18/18] perf tools: Fix output of symbol_daddr offset Stephane Eranian
2013-04-02 9:58 ` [tip:perf/core] " tip-bot for Namhyung Kim
2013-01-25 8:55 ` [PATCH v7 00/18] perf: add memory access sampling support Ingo Molnar
2013-01-25 15:28 ` Stephane Eranian
2013-01-25 10:38 ` Ingo Molnar
2013-02-05 13:03 ` Stephane Eranian
2013-02-05 15:35 ` Arnaldo Carvalho de Melo
2013-02-06 13:24 ` Ingo Molnar
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=20130215194622.GA6396@us.ibm.com \
--to=sukadev@linux.vnet.ibm.com \
--cc=acme@redhat.com \
--cc=ak@linux.intel.com \
--cc=anton@samba.org \
--cc=benh@kernel.crashing.org \
--cc=eranian@google.com \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=michael@ellerman.id.au \
--cc=mingo@elte.hu \
--cc=mingo@kernel.org \
--cc=mpjohn@us.ibm.com \
--cc=namhyung.kim@lge.com \
--cc=paulus@samba.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 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.