getrusage/getrusage04.c: Fix factor_nr = 1 -> factor_nr = 5
PATCH 1/1 getrusage/getrusage04.c: Fix factor_nr = 1 -> factor_nr = 5
To make udelta or sdelta's values isnot great than expected, we change the factor_nr to 5.
--- testcases/kernel/syscalls/getrusage/getrusage04.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
(quoted diff) +0 -0
b/testcases/kernel/syscalls/getrusage/getrusage04.c index 96e24a4..41d0052 100644
testcases/kernel/syscalls/getrusage/getrusage04.c +1 -1
@@ -67,7 +67,7 @@ static long BIAS_MAX;static int opt_factor;static char *factor_str; -static long factor_nr =1 ; +static long factor_nr =5 ;option_t child_options[] = { {"m:" , &opt_factor, &factor_str},
Signature · 15 lines
1.7.1 ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk _______________________________________________ Ltp-list mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ltp-list
----- Original Message -----
Quoted 7 lines
From: "曾令钢" <[email protected]>
To: [email protected]
Sent: Thursday, 10 October, 2013 7:56:05 AM
Subject: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->factor_nr= 5To make udelta or sdelta's values isnot great than expected, we change the
factor_nrto 5.
Note, that there is already a parameter -m to set factor_nr.
Can you share more details why this should be the default? On which system/architecture do you see it failing?
Regards,
Jan
Quoted 26 lines
Signed-off-by Zeng Linggang <[email protected]>--- testcases/kernel/syscalls/getrusage/getrusage04.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)testcases/kernel/syscalls/getrusage/getrusage04.c +1
-2@@ -67,7 +67,7 @@ static long BIAS_MAX;static int opt_factor;static char *factor_str; -static long factor_nr =1 ; +static long factor_nr =5 ;option_t child_options[] = { {"m:" , &opt_factor, &factor_str}, --1.7.1
Quoted 11 lines
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
Quoted 11 lines
----- Original Message -----
Quoted 4 lines
From: "曾令钢" <[email protected]>
To: [email protected]
Sent: Thursday, 10 October, 2013 7:56:05 AM
Subject: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->
factor_nr= 5Quoted 1 line
Quoted 7 lines
Quoted 2 lines
To make udelta or sdelta's values isnot great than expected, we change the
factor_nrto 5.Note, that there is already a parameter -m to set
factor_nr.Can you share more details why this should be the default? On which system/architecture do you see it failing?
Yes. We have some problems in RHEL5.10GA.
It could not work as we expected. And I will be more careful. Thanks.
Quoted 30 lines
Regards,
JanQuoted 26 lines
Signed-off-by Zeng Linggang <[email protected]>--- testcases/kernel/syscalls/getrusage/getrusage04.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)testcases/kernel/syscalls/getrusage/getrusage04.c +1
-2@@ -67,7 +67,7 @@ static long BIAS_MAX;static int opt_factor;static char *factor_str; -static long factor_nr =1 ; +static long factor_nr =5 ;option_t child_options[] = { {"m:" , &opt_factor, &factor_str}, --1.7.1
Quoted 13 lines
Quoted 7 lines
----------------------------------------------------------------------
-------- October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.c
Quoted 5 lines
lktrk _______________________________________________ Ltp-list mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
Quoted 24 lines
-----Original Message-----
From: 曾令钢 [mailto:[email protected]]
Sent: Thursday, October 10, 2013 4:16 PM
To: 'Jan Stancek'
Cc: '[email protected]'
Subject: 答复: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->factor_nr= 5Quoted 12 lines
----- Original Message -----
Quoted 4 lines
From: "曾令钢" <[email protected]>
To: [email protected]
Sent: Thursday, 10 October, 2013 7:56:05 AM
Subject: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->
factor_nr= 5Quoted 3 lines
To make udelta or sdelta's values isnot great than expected, we change the
factor_nrto 5.
Quoted 9 lines
Quoted 4 lines
Note, that there is already a parameter -m to set
factor_nr.Can you share more details why this should be the default? On which system/architecture do you see it failing?
Yes. We have some problems in RHEL5.10GA.
It could not work as we expected. And I will be more careful. Thanks.
Maybe the case is not suit for xen kernel.
When kernel is vmlinuz-2.6.18-371.el5xen, the case will fail. But with vmlinuz-2.6.18-371.el5, it works well.
Regards,
Zeng
Quoted 39 lines
Quoted 30 lines
Regards,
JanQuoted 26 lines
Signed-off-by Zeng Linggang <[email protected]>--- testcases/kernel/syscalls/getrusage/getrusage04.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)testcases/kernel/syscalls/getrusage/getrusage04.c +1
-2@@ -67,7 +67,7 @@ static long BIAS_MAX;static int opt_factor;static char *factor_str; -static long factor_nr =1 ; +static long factor_nr =5 ;option_t child_options[] = { {"m:" , &opt_factor, &factor_str}, --1.7.1
Quoted 9 lines
Quoted 8 lines
--------------------------------------------------------------------
--
-------- October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >
Quoted 6 lines
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.c
Quoted 5 lines
Quoted 5 lines
lktrk _______________________________________________ Ltp-list mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
----- Original Message -----
Quoted 35 lines
From: "Zeng Linggang" <[email protected]>
To: "Jan Stancek" <[email protected]>
Cc: [email protected]
Sent: Friday, 11 October, 2013 12:19:19 PM
Subject: RE: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->factor_nr= 5Quoted 27 lines
-----Original Message-----
From: 曾令钢 [mailto:[email protected]]
Sent: Thursday, October 10, 2013 4:16 PM
To: 'Jan Stancek'
Cc: '[email protected]'
Subject: 答复: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->factor_nr= 5Quoted 16 lines
----- Original Message -----
Quoted 4 lines
From: "曾令钢" <[email protected]>
To: [email protected]
Sent: Thursday, 10 October, 2013 7:56:05 AM
Subject: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->
factor_nr= 5Quoted 3 lines
To make udelta or sdelta's values isnot great than expected, we change the
factor_nrto 5.Note, that there is already a parameter -m to set
factor_nr.Can you share more details why this should be the default? On which system/architecture do you see it failing?
Quoted 8 lines
Quoted 4 lines
Yes. We have some problems in RHEL5.10GA.
It could not work as we expected. And I will be more careful. Thanks.Maybe the case is not suit for xen kernel.
When kernel is vmlinuz-2.6.18-371.el5xen, the case will fail. But with vmlinuz-2.6.18-371.el5, it works well.
Can you send the output from failed run?
Quoted 52 lines
Regards,
ZengQuoted 45 lines
Quoted 30 lines
Regards,
JanQuoted 26 lines
Signed-off-by Zeng Linggang <[email protected]>--- testcases/kernel/syscalls/getrusage/getrusage04.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)testcases/kernel/syscalls/getrusage/getrusage04.c +1
-2@@ -67,7 +67,7 @@ static long BIAS_MAX;static int opt_factor;static char *factor_str; -static long factor_nr =1 ; +static long factor_nr =5 ;option_t child_options[] = { {"m:" , &opt_factor, &factor_str}, --1.7.1
Quoted 9 lines
Quoted 8 lines
--------------------------------------------------------------------
--
-------- October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.c
Quoted 5 lines
Quoted 5 lines
lktrk _______________________________________________ Ltp-list mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
Quoted 57 lines
-----Original Message-----
From: Jan Stancek [mailto:[email protected]]
Sent: Friday, October 11, 2013 6:35 PM
To: Zeng Linggang
Cc: [email protected]
Subject: Re: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->factor_nr= 5----- Original Message -----
Quoted 5 lines
From: "Zeng Linggang" <[email protected]>
To: "Jan Stancek" <[email protected]>
Cc: [email protected]
Sent: Friday, 11 October, 2013 12:19:19 PM
Subject: RE: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->
factor_nr= 5Quoted 36 lines
Quoted 35 lines
-----Original Message-----
From: 曾令钢 [mailto:[email protected]]
Sent: Thursday, October 10, 2013 4:16 PM
To: 'Jan Stancek'
Cc: '[email protected]'
Subject: 答复: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->factor_nr= 5Quoted 10 lines
----- Original Message -----
Quoted 5 lines
From: "曾令钢" <[email protected]>
To: [email protected]
Sent: Thursday, 10 October, 2013 7:56:05 AM
Subject: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->
factor_nr= 5Quoted 1 line
Quoted 7 lines
Quoted 2 lines
To make udelta or sdelta's values isnot great than expected, we change the
factor_nrto 5.Note, that there is already a parameter -m to set
factor_nr.Can you share more details why this should be the default? On which system/architecture do you see it failing?
Yes. We have some problems in RHEL5.10GA.
It could not work as we expected. And I will be more careful. Thanks.
Quoted 5 lines
Quoted 3 lines
Maybe the case is not suit for xen kernel.
When kernel is vmlinuz-2.6.18-371.el5xen, the case will fail. But with vmlinuz-2.6.18-371.el5, it works well.Can you send the output from failed run?
<<<test_start>>>
tag=getrusage04 stime=1381285379
cmdline="getrusage04"
contacts=""
analysis=exit
<<<test_output>>>
getrusage04 0 TINFO : CLOCK_REALTIME_COARSE not supported, using 4000 us getrusage04 0 TINFO : Using 1 as multiply factor for max [us]time increment (1000+4000us)! getrusage04 0 TINFO : utime: 0us; stime: 4000us getrusage04 0 TINFO : utime: 0us; stime: 16001us getrusage04 1 TFAIL : stime increased > 5000us:
<<<execution_status>>>
initiation_status="ok"
duration=0 termination_type=exited termination_id=1 corefile=no cutime=0 cstime=2
<<<test_end>>>
Regards,
Zeng
Quoted 47 lines
Quoted 47 lines
Regards,
ZengQuoted 41 lines
Quoted 40 lines
Regards,
JanQuoted 35 lines
Signed-off-by Zeng Linggang <[email protected]>--- testcases/kernel/syscalls/getrusage/getrusage04.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)testcases/kernel/syscalls/getrusage/getrusage04.c +1
-2@@ -67,7 +67,7 @@ static long BIAS_MAX;static int opt_factor;static char *factor_str; -static long factor_nr =1 ; +static long factor_nr =5 ;option_t child_options[] = { {"m:" , &opt_factor, &factor_str}, --1.7.1
----------------------------------------------------------------
----
--
-------- October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >
Quoted 10 lines
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg
Quoted 9 lines
Quoted 6 lines
.c
Quoted 5 lines
Quoted 5 lines
lktrk _______________________________________________ Ltp-list mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
----- Original Message -----
Quoted 11 lines
From: "Zeng Linggang" <[email protected]>
To: "Jan Stancek" <[email protected]>
Cc: [email protected]
Sent: Friday, 11 October, 2013 12:19:19 PM
Subject: RE: [LTP] [PATCH]getrusage/getrusage04.c: Fixfactor_nr= 1 ->factor_nr= 5Quoted 6 lines
Quoted 5 lines
Note, that there is already a parameter -m to set
factor_nr.Can you share more details why this should be the default? On which system/architecture do you see it failing?
Quoted 7 lines
Quoted 4 lines
Yes. We have some problems in RHEL5.10GA.
It could not work as we expected. And I will be more careful. Thanks.When kernel is vmlinuz-2.6.18-371.el5xen, the case will fail. But with vmlinuz-2.6.18-371.el5, it works well.
I'm seeing the same as you also with previous minor RHEL5 releases on dom0:
5.3 2.6.18-128.el5xen
5.6 2.6.18-238.el5xen
5.9 2.6.18-348.el5xen
5.10 2.6.18-371.el5xen
Quoted 1 line
Maybe the case is not suit for xen kernel.
It's possible, it doesn't look like regression to me so far.
The systems are definitely not "slow" ones, and test can pass on bare-metal kernel.
Regards,
Jan
Quoted 8 lines
Regards,
ZengQuoted 4 lines
Quoted 4 lines
Regards,
Jan
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
Hi!
Quoted 9 lines
Quoted 2 lines
When kernel is vmlinuz-2.6.18-371.el5xen, the case will fail. But with vmlinuz-2.6.18-371.el5, it works well.
I'm seeing the same as you also with previous minor RHEL5 releases on dom0:
5.3 2.6.18-128.el5xen
5.6 2.6.18-238.el5xen
5.9 2.6.18-348.el5xen
5.10 2.6.18-371.el5xen
Quoted 4 lines
Quoted 1 line
Maybe the case is not suit for xen kernel.
It's possible, it doesn't look like regression to me so far.
The systems are definitely not "slow" ones, and test can pass on bare-metal kernel.
The problem with the test is that it tries to assert that granurality of gerusage CPU time accounting is as good as possible. The granularity itself depends on kernel config options, mainly on CONFIG_HZ. But as the kernel compilation options are not exported (by running kernel) in any reliable way the code I've added tries to guess them by side channel (the CLOCK_REALTIME_COARSE resolution). If CLOCK_REALTIME_COARSE is not supported it goes with 4ms which should correspond with CONFIG_HZ=250.
Now the accuracy on xen may be off for several reasons. It can have CONFIG_HZ=100. Or xen may not be able to account virtual CPU time with the same accuracy as on the real CPU...
Signature · 13 lines
Cyril Hrubis [email protected] ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk _______________________________________________ Ltp-list mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ltp-list
A message that is not in the archive (sent off-list or not yet received): [email protected]
Hi!
Quoted 8 lines
Quoted 8 lines
The problem with the test is that it tries to assert that granurality of gerusage CPU time accounting is as good as possible. The granularity itself depends on kernel config options, mainly on
CONFIG_HZ. But as the kernel compilation options are not exported (by running kernel) in any reliable way the code I've added tries to guess them by side channel (theCLOCK_REALTIME_COARSEresolution). IfCLOCK_REALTIME_COARSEis not supported it goes with 4ms which should correspond withCONFIG_HZ=250.
Quoted 8 lines
Quoted 3 lines
Now the accuracy on xen may be off for several reasons. It can have
CONFIG_HZ=100. Or xen may not be able to account virtual CPU time with the same accuracy as on the real CPU...Drew (from xen team) mentioned something along these lines as well.
Do we have some simple way to detect that test is running on xen (dom0/domU), so we could end test with TCONF in that case?
I'm not aware of any. But feel free to add a function into LTP lib/.
Signature · 13 lines
Cyril Hrubis [email protected] ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk _______________________________________________ Ltp-list mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ltp-list