Just this week I found a problem with the 'here' accounting in
showdetail. The new base library isn't available yet, but the change
was small: replace the line
her=. 2 >: bgn usage end
with
her =. 1 = +/\ bgn - end
If that doesn't help, see if you can create a reproducible testcase,
perhaps using 6!:3 to represent lengthy computation, and send it to me.
Henry Rich
On 2/28/2020 1:24 AM, Devon McCormick wrote:
I am confused by the output of jpm because I just ran something that took
48 seconds but the detail report of the top-level function shows a total
time of about 70 seconds. How can this be? Am I mis-interpreting the
numbers?
For example:
6!:2 '''FLNMS FLDTS FLSZS FLPARENT DIRNMS DIRDEP''=. PllDirInfoEG
''E:\'''
48.7943
0 1 showtotal_jpm_ ''
Time (seconds)
+------------------+---------+---------+-----+----+-----+
|name (all locales)|all |here |here%|cum%|rep |
+------------------+---------+---------+-----+----+-----+
|wait |47.982074|47.982074| 67.6| 68 | 48|
|cd |21.512789|21.512789| 30.3| 98 |14519|
|[rest] | | 1.454809| 2.1|100 | |
|[total] | |70.949672|100.0|100 | |
+------------------+---------+---------+-----+----+-----+
0 1 showdetail_jpm_ 'PllDirInfoEG'
Time (seconds)
+---------+--------+---+-----------------------------------------------------...
|all |here |rep|PllDirInfoEG
...
+---------+--------+---+-----------------------------------------------------...
| 0.000002|0.000002|1 |monad
...
| 0.000003|0.000003|1 |[1] exeID=.'J8Pll'
...
|68.921268|0.000019|1 |[2] 'tmpd srcd flnms fldts flszs
exeIDs'=.pllParseDir...
| 0.000036|0.000036|1 |[3] exeID=.'0123456789'(]{.~[:>:0
i:~]e.[)'.'(]{.~]i:...
| 0.000102|0.000015|1 |[4] flszs=.;flszs[fldts=.cvtTS21Num&>fldts
...
| 0.000002|0.000002|1 |[5] flparent=.0$~#flszs
...
| 0.000003|0.000003|1 |[6] dirdep=._1,0$~#srcd[dirnms=.y;srcd
...
| 0.822916|0.007631|1 |[7] 'flnms fldts flszs flparent dirnms
dirdep'=.cumul...
| 0.735149|0.000015|1 |[8] cleanupTempDirs'PllDTmp'
...
| 0.380437|0.008926|1 |[9] 'dirnms flparent'=.adjustDirInfo
dirnms;<flparent...
| 0.089712|0.016307|1 |[10] dirdep=.dirDependencies dirnms
...
| 0.000014|0.000014|1 |[11] flnms;fldts;flszs;flparent;dirnms;<dirdep
...
|70.949643|0.032973|1 |total monad
...
+---------+--------+---+-----------------------------------------------------...
I don't understand how the 70.9 in the last row jibes with the 48.7943
seconds the entire run took.
Can anyone explain?
Thanks,
Devon
On Thu, Feb 27, 2020 at 9:56 PM bill lam <bbill....@gmail.com> wrote:
I checked the viewtotal required the grid addon which had been
decommitted.
Thu, 27 Feb 2020, Devon McCormick написал(а):
Hi - I'm following the Performance Monitor documentation but I get a
value
error when I try to invoke "viewtotal_jpm_". Does anyone know why?
The example output shows what look like graphical windows so I'm
wondering
if I have to run this under Qt.
Thanks,
Devon
--
Devon McCormick, CFA
Quantitative Consultant
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
--
regards,
====================================================
GPG key 1024D/4434BAB3 2008-08-24
gpg --keyserver subkeys.pgp.net --armor --export 4434BAB3
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
