I'm seeing a strange behavior on 3.7.0 RC4 on linux, with included sample
scenes.

First, the particulars:

Slackware 13.37, stock 2.6.37.6 kernel, Slackware default "everything" install
RC4 compiled with Slackware stock gcc 4.52
configure script run with no options except the usual "COMPILED_BY"
Dell R210 hardware, Pentium G620 2.6ghz

What's happening is povray looks to be hanging in the parsing stage.  I tried to
render the "box.pov" file that comes with the distribution.
(/usr/local/share/povray-3.7/scenes/objects/box.pov) and it hangs with the last
of the output like so:


Parser Options
  Input file: box.pov
  Remove bounds........On
  Split unions.........Off
  Library paths:
    /usr/local/share/povray-3.7
    /usr/local/share/povray-3.7/ini
    /usr/local/share/povray-3.7/include
  Clock value:    0.000  (Animation off)
Image Output Options
  Image resolution.....1600 by 1200 (rows 1 to 1200, columns 1 to 1600).
  Output file..........box.png, 24 bpp PNG
  Dithering............Off
  Graphic display......On  (gamma: sRGB)
  Mosaic preview.......Off
  Continued trace......Off
Information Output Options
  All Streams to console..........On
  Debug Stream to console.........On
  Fatal Stream to console.........On
  Render Stream to console........On
  Statistics Stream to console....On
  Warning Stream to console.......On
==== [Parsing...]=======================

Parsing never completes.  Povray seems to be using a single thread, and seems to
allocate memory for the parse and then stop so it doesn't look like some kind of
leak.  It consumes 100% of CPU on the one core as long as I let it run but
either never finishes the parse or never starts the actual render.  Povray
doesn't respond to an abort signal (ctl-c) and has to be killed with signal 9.

Does this sound like a user error or should I file a bug report?  I have done
the same install of the OS and RC4 on different hardware and had no problems but
I'm not sure if I need to pass some configure option for the G620 processor.

Thanks!
harry

Post a reply to this message

And, I'm retarded.  Apparently povray won't abort when it can't connect to a
local display, it just hangs now.  Not very intuitive but definitely not a bug.

Post a reply to this message

On 03/09/2012 12:52 PM, harry wrote:
> And, I'm retarded.  Apparently povray won't abort when it can't connect to a
> local display, it just hangs now.  Not very intuitive but definitely not a bug.
>

I'm not sure if this will make a difference ... but maybe it's worth 
mentioning: A bug (my bad) was spotted with with the RC4 linux source 
distribution, that was corrected in the pending RC5 release.

at the head of your build directory try:

find . -type d -name '.deps'

then remove or empty those returned directories

and do configure to rebuild those directory and files (to reflect what's 
on your system) ... then of course make

Jim

Post a reply to this message

James Holsenback <nom### [at] nonecom> wrote:
> On 03/09/2012 12:52 PM, harry wrote:
> > And, I'm retarded.  Apparently povray won't abort when it can't connect to a
> > local display, it just hangs now.  Not very intuitive but definitely not a bug.
> >
>
> I'm not sure if this will make a difference ... but maybe it's worth
> mentioning: A bug (my bad) was spotted with with the RC4 linux source
> distribution, that was corrected in the pending RC5 release.
>
> at the head of your build directory try:
>
> find . -type d -name '.deps'
>
> then remove or empty those returned directories
>
> and do configure to rebuild those directory and files (to reflect what's
> on your system) ... then of course make
>
> Jim

Thanks Jim, I'll try that.  With no render display enabled the binary I compiled
works fine.  It's really nothing more than an inconvenience but I'll give that a
shot.

thanks!

Post a reply to this message

James Holsenback <nom### [at] nonecom> wrote:
> I'm not sure if this will make a difference ... but maybe it's worth 
> mentioning: A bug (my bad) was spotted with with the RC4 linux source 
> distribution, that was corrected in the pending RC5 release.

> at the head of your build directory try:

> find . -type d -name '.deps'

> then remove or empty those returned directories

  Isn't "make clean" the standard way of doing that?

-- 
                                                          - Warp

Post a reply to this message

James Holsenback <nom### [at] nonecom> wrote:
> On 03/09/2012 12:52 PM, harry wrote:
> > And, I'm retarded.  Apparently povray won't abort when it can't connect to a
> > local display, it just hangs now.  Not very intuitive but definitely not a bug.
> >
>
> I'm not sure if this will make a difference ... but maybe it's worth
> mentioning: A bug (my bad) was spotted with with the RC4 linux source
> distribution, that was corrected in the pending RC5 release.
>
> at the head of your build directory try:
>
> find . -type d -name '.deps'
>
> then remove or empty those returned directories
>
> and do configure to rebuild those directory and files (to reflect what's
> on your system) ... then of course make
>
> Jim

No change after doing that and a recompile.  Don't spend any more time on this,
it's a beta and it's easy enough to just specify -D on the command line to work
around it.

thanks!
harry

Post a reply to this message

On 03/09/2012 02:22 PM, Warp wrote:
> James Holsenback<nom### [at] nonecom>  wrote:
>> I'm not sure if this will make a difference ... but maybe it's worth
>> mentioning: A bug (my bad) was spotted with with the RC4 linux source
>> distribution, that was corrected in the pending RC5 release.
>
>> at the head of your build directory try:
>
>> find . -type d -name '.deps'
>
>> then remove or empty those returned directories
>
>    Isn't "make clean" the standard way of doing that?
>
LOL didn't occur until you mentioned ... I'm just coming up to speed on 
this ;-) Since the OP has already done a build I'd imagine it wouldn't 
apply in this case (my advice). IIRC wouldn't someone trying to do a 
build for the first time (just unpacked the tarball) get a "no such 
rule" (or something to that affect) when doing a make clean?

Post a reply to this message

James Holsenback <nom### [at] nonecom> wrote:
> IIRC wouldn't someone trying to do a 
> build for the first time (just unpacked the tarball) get a "no such 
> rule" (or something to that affect) when doing a make clean?

  IIRC there's no makefile directly in the package. It's created by
running 'configure' first.

-- 
                                                          - Warp

Post a reply to this message

"harry" <har### [at] gmailcom> wrote:
> Slackware 13.37, stock 2.6.37.6 kernel, Slackware default "everything" install
> RC4 compiled with Slackware stock gcc 4.52
> configure script run with no options except the usual "COMPILED_BY"
> Dell R210 hardware, Pentium G620 2.6ghz
>
> What's happening is povray looks to be hanging in the parsing stage.  I tried to
> render the "box.pov" file that comes with the distribution.
> (/usr/local/share/povray-3.7/scenes/objects/box.pov) and it hangs with the last
On my Slackware 13.37 - but the x86_64 version - running on Intel Core2Duo I
build thus:
Unpack povray-3.7.0.RC4.tar.gz and then simply do:
../configure  --with-boost-thread=boost_thread COMPILED_BY="RC4"

The example box.pov renders OK.
Just a thought - does "make check" render coffee and biscuits OK?

Cheers,
Peter

Post a reply to this message