[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1355871266.21420.17.camel@sos-dev03.amd.com>
Date: Tue, 18 Dec 2012 16:54:26 -0600
From: Suravee Suthikulpanit <suravee.suthikulpanit@....com>
To: Robert Richter <rric@...nel.org>, Ingo Molnar <mingo@...nel.org>
CC: <a.p.zijlstra@...llo.nl>, <paulus@...ba.org>, <mingo@...hat.com>,
<acme@...stprotocols.net>, <tglx@...utronix.de>, <hpa@...or.com>,
<x86@...nel.org>, <linux-kernel@...r.kernel.org>,
<jacob.shin@....com>, <suravee.suthikulpanit@....com>
Subject: Re: [PATCH 1/1] Change IBS PMU to use perf_hw_context
Ingo, Robert
I am including a set of output from "perf report" to help validating IBS in per-process mode.
In this experiment I ran a couple test cases:
case 1. perf record -e cycles (baseline per-process mode w/ regular counter)
case 2. perf record -a -e cycles:p (baseline system-wide mode w/ IBS)
case 3. perf record -e cycles:p (the proposed per-process mode w/IBS)
In all 3 test cases, the target application (classic) are showing about 27K samples.
I am also including the IBS OP MSRs (0xc00110[33-3a]) snapshots on all 32 cores
(using rdmsr tools) from case 2 and 3 above.
------------------------------------------------------------
CASE1:
# ========
# captured on: Tue Dec 18 16:32:43 2012
# hostname : sos-dev02
# os release : 3.7.0-IBS+
# perf version : 3.7.rc8.g805f38
# arch : x86_64
# nrcpus online : 32
# nrcpus avail : 32
# cpudesc : AMD Eng Sample, 1S228145TGG54_31/22/20_2/16
# cpuid : AuthenticAMD,21,2,0
# total memory : 32863836 kB
# cmdline : /sandbox/kernels/suravee/tools/perf/perf record -e cycles taskset -c 31 src/classic
# event : name = cycles, type = 0, config = 0x0, config1 = 0x0, config2 = 0x0, excl_usr = 0, excl_kern = 0, excl_host = 0, excl_guest = 1, precise_ip = 0, id = { 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229 }
# HEADER_CPU_TOPOLOGY info available, use -I to display
# HEADER_NUMA_TOPOLOGY info available, use -I to display
# pmu mappings: cpu = 4, software = 1, tracepoint = 2, ibs_fetch = 6, ibs_op = 7, breakpoint = 5
# ========
#
# Samples: 27K of event 'cycles'
# Event count (approx.): 20938245323
#
# Overhead Samples Command Shared Object Symbol
# ........ ........... ....... ................. .......................................
#
99.16% 26927 classic classic [.] multiply_matrices() <--- TARGET APP
0.32% 78 classic libc-2.15.so [.] random
0.10% 23 classic libc-2.15.so [.] random_r
0.07% 16 classic classic [.] initialize_matrices()
0.04% 10 classic [kernel.kallsyms] [k] ttwu_do_wakeup
0.03% 9 classic [kernel.kallsyms] [k] clear_page_c
0.02% 11 classic [kernel.kallsyms] [k] native_write_msr_safe
0.02% 5 classic libc-2.15.so [.] rand
0.02% 2 classic ld-2.15.so [.] 0x000000000000a456
------------------------------------------------------------
CASE 2:
# ========
# captured on: Tue Dec 18 16:11:35 2012
# hostname : sos-dev02
# os release : 3.7.0-IBS+
# perf version : 3.7.rc8.g805f38
# arch : x86_64
# nrcpus online : 32
# nrcpus avail : 32
# cpudesc : AMD Eng Sample, 1S228145TGG54_31/22/20_2/16
# cpuid : AuthenticAMD,21,2,0
# total memory : 32863836 kB
# cmdline : /sandbox/kernels/suravee/tools/perf/perf record -a -e cycles:p taskset -c 31 src/classic
# event : name = cycles:p, type = 0, config = 0x0, config1 = 0x0, config2 = 0x0, excl_usr = 0, excl_kern = 0, excl_host = 0, excl_guest = 0, precise_ip = 1, id = { 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98 }
# HEADER_CPU_TOPOLOGY info available, use -I to display
# HEADER_NUMA_TOPOLOGY info available, use -I to display
# pmu mappings: cpu = 4, software = 1, tracepoint = 2, ibs_fetch = 6, ibs_op = 7, breakpoint = 5
# ========
#
# Samples: 189K of event 'cycles:p'
# Event count (approx.): 40504131338
#
# Overhead Samples Command Shared Object Symbol
# ........ ........... ............... ................................ ...........................................
#
51.07% 26959 classic classic [.] multiply_matrices() <------ TARGET APP
35.39% 131620 swapper [kernel.kallsyms] [k] acpi_idle_do_entry
2.10% 4673 swapper [kernel.kallsyms] [k] native_safe_halt
0.71% 1303 rdmsr ld-2.15.so [.] 0x0000000000002a44
0.33% 639 rdmsr [kernel.kallsyms] [k] irq_return
0.26% 499 rdmsr libc-2.15.so [.] 0x0000000000131d80
0.25% 440 rdmsr [kernel.kallsyms] [k] generic_exec_single
0.25% 470 rdmsr [kernel.kallsyms] [k] __do_fault
0.24% 478 rdmsr [kernel.kallsyms] [k] unmap_single_vma
------------------------------------------------------------
CASE 3:
# ========
# captured on: Tue Dec 18 16:13:53 2012
# hostname : sos-dev02
# os release : 3.7.0-IBS+
# perf version : 3.7.rc8.g805f38
# arch : x86_64
# nrcpus online : 32
# nrcpus avail : 32
# cpudesc : AMD Eng Sample, 1S228145TGG54_31/22/20_2/16
# cpuid : AuthenticAMD,21,2,0
# total memory : 32863836 kB
# cmdline : /sandbox/kernels/suravee/tools/perf/perf record -e cycles:p taskset -c 31 src/classic
# event : name = cycles:p, type = 0, config = 0x0, config1 = 0x0, config2 = 0x0, excl_usr = 0, excl_kern = 0, excl_host = 0, excl_guest = 0, precise_ip = 1, id = { 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131 }
# HEADER_CPU_TOPOLOGY info available, use -I to display
# HEADER_NUMA_TOPOLOGY info available, use -I to display
# pmu mappings: cpu = 4, software = 1, tracepoint = 2, ibs_fetch = 6, ibs_op = 7, breakpoint = 5
# ========
#
# Samples: 27K of event 'cycles:p'
# Event count (approx.): 20851884446
#
# Overhead Samples Command Shared Object Symbol
# ........ ........... ....... ................. ..............................
#
99.37% 27020 classic classic [.] multiply_matrices() <--- TARGET APP
0.22% 58 classic libc-2.15.so [.] random_r
0.13% 32 classic classic [.] initialize_matrices()
0.10% 26 classic libc-2.15.so [.] random
0.03% 8 classic libc-2.15.so [.] rand
0.03% 7 classic [kernel.kallsyms] [k] clear_page_c
0.01% 2 classic ld-2.15.so [.] 0x000000000000a423
0.01% 2 classic [kernel.kallsyms] [k] retint_swapgs
0.01% 2 classic [kernel.kallsyms] [k] ttwu_do_wakeup
------------------------------------------------------------
IBS MSR VALUES FROM CASE 2:
core : 0xc0011033 0xc0011034 0xc0011035 0xc0011036 0xc0011037 0xc0011038 0xc0011039 0xc001103a
0 : 0000002300040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000040002 0000000000000000 000000fdfd300000 0000000000000100
1 : 0000006200040000 ffffffff813d8c6d 00000058000b0002 0000000000000000 0000000000000000 0000000000000400 000000fdfd300400 0000000000000100
2 : 0000006000040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000040002 0000000000000000 000000fdfd300000 0000000000000100
3 : 0000005000040000 ffffffff813d8c6d 00000040000b0002 0000000000000000 0000000000040001 0000000000000400 000000fdfd300400 0000000000000100
4 : 0000005700040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000000000 0000000000000000 000000fdfd300000 0000000000000100
5 : 0000000000000000 ffffffff81043ea8 00000000000a0001 0000000000000000 0000000000000000 0000000000000514 000000fdfd300514 0000000000000100
6 : 0000004200040000 ffffffff813d8c74 00000000000b0006 0000000000000000 0000000000000000 0000000000000000 000000fdfd300000 0000000000000100
7 : 0000000000000000 ffffffff81043ea8 00000000000a0000 0000000000000000 0000000000000000 0000000000000514 000000fdfd300514 0000000000000100
8 : 0000002300040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000000000 0000000000000000 000000fdfd300000 0000000000000100
9 : 0000004d00040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000040002 0000000000000400 000000fdfd300400 0000000000000100
10 : 00001fe500000000 ffffffff813d8c6d 00000058000b0002 0000000000000000 0000000000000000 0000000000000000 000000fdfd300000 0000000000000100
11 : 0000000000000000 ffffffff81043ea8 00000000000a0001 0000000000000000 0000000000040001 0000000000000514 000000fdfd300514 0000000000000100
12 : 0000008100040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000000000 0000000000000000 000000fdfd300000 0000000000000100
13 : 0000006900040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000040002 0000000000000400 000000fdfd300400 0000000000000100
14 : 0000004900040000 ffffffff813d8c6d 00000040000b0002 0000000000000000 0000000000040001 0000000000000000 000000fdfd300000 0000000000000100
15 : 0000002300040000 ffffffff813d8c6d 00000040000b0002 0000000000000000 0000000000040001 0000000000000400 000000fdfd300400 0000000000000100
16 : 0000000f00040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000000000 0000000000000000 000000fdfd300000 0000000000000100
17 : 0000004b00040000 ffffffff813d8c6d 00000058000b0002 0000000000000000 0000000000000000 0000000000000400 000000fdfd300400 0000000000000100
18 : 0000003d00040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000000000 00000000000002b8 000000fdfd3002b8 0000000000000100
19 : 0000004400040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000000000 0000000000000400 000000fdfd300400 0000000000000100
20 : 0000001800040000 ffffffff813d8d27 0000000000060001 0000000000000000 0000000000000000 0000000000000000 000000fdfd300000 0000000000000100
21 : 0000002900040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000000000 0000000000000400 000000fdfd300400 0000000000000100
22 : 0000005900040000 ffffffff813d8c6d 00000040000b0002 0000000000000000 0000000000040001 0000000000000000 000000fdfd300000 0000000000000100
23 : 0000001500040000 ffffffff813d8c6d 00000058000b0002 0000000000000000 0000000000000000 0000000000000400 000000fdfd300400 0000000000000100
24 : 0000006100040000 ffffffff8133e844 00000028001e000f 0000000000000000 0000000000000000 0000000000000000 000000fdfd300000 0000000000000100
25 : 0000005400040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000000000 0000000000000400 000000fdfd300400 0000000000000100
26 : 0000002900040000 ffffffff813d8c6d 00000040000b0002 0000000000000000 0000000000040001 0000000000000000 000000fdfd300000 0000000000000100
27 : 0000000e00040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000000000 0000000000000400 000000fdfd300400 0000000000000100
28 : 0000007e00040000 ffffffff813d8c6d 00000040000b0006 0000000000000000 0000000000040042 0000000000000000 000000fdfd300000 0000000000000100
29 : 0000000000000000 ffffffff81043ea8 00000000000a0001 0000000000000000 0000000000040001 0000000000000514 000000fdfd300514 0000000000000100
30 : 00003f4a00000000 ffffffff813d8c6d 00000040000b0006 0000000000000000 000000000004000a 0000000000000000 000000fdfd300000 0000000000000100
31 : 0001147800000000 ffffffff810b9400 00000000003c0001 0000000000000000 0000000000040009 00000000000005dc 000000fdfd3005dc 0000000000000100
------------------------------------------------------------
IBS MSR VALUES FROM CASE 3:
core : 0xc0011033 0xc0011034 0xc0011035 0xc0011036 0xc0011037 0xc0011038 0xc0011039 0xc001103a
0 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
1 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
2 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
3 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
4 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
5 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
6 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
7 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
8 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
9 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
10 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
11 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
12 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
13 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
14 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
15 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
16 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
17 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
18 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
19 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
20 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
21 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
22 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
23 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
24 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
25 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
26 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
27 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
28 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
29 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
30 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000100
31 : 00034d8900000000 ffffffff811370cc 0000000000120000 0000000000000000 0000000000000008 ffff88082592bc58 000000082592bc58 0000000000000100
Suravee
On Mon, 2012-12-17 at 10:44 +0100, Robert Richter wrote:
> On 16.12.12 10:04:10, Ingo Molnar wrote:
> >
> > * suravee.suthikulpanit@....com <suravee.suthikulpanit@....com> wrote:
> >
> > > From: Suravee Suthikulpanit <suravee.suthikulpanit@....com>
> > >
> > > Currently, the AMD IBS PMU initialize pmu.task_ctx_nr to
> > > perf_invalid_context which only allows IBS to be running only
> > > in system-wide mode (e.g. perf record -a). IBS hardware is
> > > available in each core and should be per-context. This patch
> > > modifies the task_ctx_nr to use the perf_hw_context (default)
> > > instead.
> >
> > I'm wondering how extensively was it tested/verified that it's
> > safe to enable IBS in per context mode as well, and that the
> > profiling results are precise and accurate?
>
> From the implementation's point of view this is very similar to hw
> perf counters. I wouldn't expect any issues here. Since IBS can be
> immediatly started/stopped and there is no caching, there won't be any
> incomming sample that is not related to that context.
>
> The only potential problem I see could be a security risk in a way
> that an IBS sample might expose data related to other contexts such as
> cache information. This is similar to uncore/northbridge events so I
> don't think this is an issue, but we might want to evaluate this.
>
> > We never used the IBS hardware in this fashion before, so some
> > extra care is prudent - and traces of that extra care should be
> > visible in the changelog as well.
>
> Yeah, a comparison of numbers for IBS and hw counter (-e r076:p,r076
> and -e r0C1:p,r0C1) in per-context mode would be useful here.
>
> -Robert
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists