Possible bug, Rprof() and scan(pipe()) (PR#1140)

2 messages · Brian Ripley, Luke Tierney

Sun, Nov 4, 2001 5:01 AM #

This does not happen on Solaris or Linux, so must be related to the use of
signals on darwin.

It probably means Rprof should be avoided on that platform, but the
porter (Jan de Leeuw) needs to investigate.

On Tue, 23 Oct 2001 macq@llnl.gov wrote:

This looks like a bug?
Unable to use scan(pipe()) while profiling.
I have no idea whether this version of R violates the
    "do not use `Rprof' in an executable built for profiling"
warning in ?Rprof.

Thanks
-Don

 version

          _
platform powerpc-apple-darwin1.4
arch     powerpc
os       darwin1.4
system   powerpc, darwin1.4
status   Patched
major    1
minor    3.1
year     2001
month    10
day      15
language R

 foo <- scan(pipe('ls -1 | head -3'),what='')

Read 3 items

foo

[1] "README"      "RMfiles.dat" "RMfiles.tmp"

 Rprof()
 foo <- scan(pipe('ls -1 | head -3'),what='')

Read 0 items

foo

character(0)

 Rprof(NULL)
 foo <- scan(pipe('ls -1 | head -3'),what='')

Read 3 items

note, profiling does appear to be functional:

 Rprof() ; hist(rnorm(10000)) ; Rprof(NULL)
 system('R CMD Rprof Rprof.out')

Each sample represents 0.02 seconds.
Total run time: 0.1 seconds.

Total seconds: time spent in function and callees.
Self seconds: time spent in function alone.

    %       total       %       self
  total    seconds     self    seconds    name
100.00      0.10      0.00      0.00     "hist"
  60.00      0.06      0.00      0.00     "hist.default"
  40.00      0.04     40.00      0.04     "rnorm"
  40.00      0.04      0.00      0.00     "plot"
  40.00      0.04      0.00      0.00     "plot.histogram"
  20.00      0.02      0.00      0.00     "storage.mode<-"
  20.00      0.02     20.00      0.02     "X11"
  20.00      0.02      0.00      0.00     "paste"
  20.00      0.02      0.00      0.00     "rect"
  20.00      0.02      0.00      0.00     "plot.new"
  20.00      0.02     20.00      0.02     "seq.default"
  20.00      0.02      0.00      0.00     "seq"
  20.00      0.02     20.00      0.02     "par"

    %       self        %       total
  self     seconds    total    seconds    name
  40.00      0.04     40.00      0.04     "rnorm"
  20.00      0.02     20.00      0.02     "X11"
  20.00      0.02     20.00      0.02     "seq.default"
  20.00      0.02     20.00      0.02     "par"

--
--------------------------------------
Don MacQueen
Environmental Protection Department
Lawrence Livermore National Laboratory
Livermore, CA, USA
--------------------------------------

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Luke Tierney

Sun, Nov 4, 2001 5:18 AM #

This feels like a system call restarting issue.  Some OS's restart
system calls after signals, others do not. For those that don't, low
level reads need to check for EINTR and such.  Connections use high
level reads, which should be insulated by the stdio library.  Possibly
dawrin's gets this wrong.  If strace is available you might be able to
tell that way.

luke

On Sun, Nov 04, 2001 at 02:01:10PM +0100, ripley@stats.ox.ac.uk wrote:

This does not happen on Solaris or Linux, so must be related to the use of
signals on darwin.

It probably means Rprof should be avoided on that platform, but the
porter (Jan de Leeuw) needs to investigate.


On Tue, 23 Oct 2001 macq@llnl.gov wrote:

This looks like a bug?
Unable to use scan(pipe()) while profiling.
I have no idea whether this version of R violates the
    "do not use `Rprof' in an executable built for profiling"
warning in ?Rprof.

Thanks
-Don

 version

          _
platform powerpc-apple-darwin1.4
arch     powerpc
os       darwin1.4
system   powerpc, darwin1.4
status   Patched
major    1
minor    3.1
year     2001
month    10
day      15
language R

 foo <- scan(pipe('ls -1 | head -3'),what='')

Read 3 items

foo

[1] "README"      "RMfiles.dat" "RMfiles.tmp"

 Rprof()
 foo <- scan(pipe('ls -1 | head -3'),what='')

Read 0 items

foo

character(0)

 Rprof(NULL)
 foo <- scan(pipe('ls -1 | head -3'),what='')

Read 3 items

note, profiling does appear to be functional:

 Rprof() ; hist(rnorm(10000)) ; Rprof(NULL)
 system('R CMD Rprof Rprof.out')

Each sample represents 0.02 seconds.
Total run time: 0.1 seconds.

Total seconds: time spent in function and callees.
Self seconds: time spent in function alone.

    %       total       %       self
  total    seconds     self    seconds    name
100.00      0.10      0.00      0.00     "hist"
  60.00      0.06      0.00      0.00     "hist.default"
  40.00      0.04     40.00      0.04     "rnorm"
  40.00      0.04      0.00      0.00     "plot"
  40.00      0.04      0.00      0.00     "plot.histogram"
  20.00      0.02      0.00      0.00     "storage.mode<-"
  20.00      0.02     20.00      0.02     "X11"
  20.00      0.02      0.00      0.00     "paste"
  20.00      0.02      0.00      0.00     "rect"
  20.00      0.02      0.00      0.00     "plot.new"
  20.00      0.02     20.00      0.02     "seq.default"
  20.00      0.02      0.00      0.00     "seq"
  20.00      0.02     20.00      0.02     "par"

    %       self        %       total
  self     seconds    total    seconds    name
  40.00      0.04     40.00      0.04     "rnorm"
  20.00      0.02     20.00      0.02     "X11"
  20.00      0.02     20.00      0.02     "seq.default"
  20.00      0.02     20.00      0.02     "par"

--
--------------------------------------
Don MacQueen
Environmental Protection Department
Lawrence Livermore National Laboratory
Livermore, CA, USA
--------------------------------------

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Luke Tierney
University of Minnesota                      Phone:           612-625-7843
School of Statistics                         Fax:             612-624-8868
313 Ford Hall, 224 Church St. S.E.           email:      luke@stat.umn.edu
Minneapolis, MN 55455 USA                    WWW:  http://www.stat.umn.edu
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._