Bug in spline2

[mvandam]

Bug in spline2

[frigaut]

Hi Nieder,

Can you be more explicit on what type of machine produce the error?
What OS are you running? Are you building imutil from source? If not,
could it be a glibc version issue?
Thanks
Francois

On 27/02/14 10:07, Marcos van Dam wrote:

On some machines but not others, I get the following error when using
spline2

ERROR (spline2) Floating point interrupt (SIGFPE)

This error is totally repeateable (i.e., running the same commands
leads to the same errors). The funny thing is that after the bug is
produced, the same command can be called and it does not crash the
second time.

—
Reply to this email directly or view it on GitHub
#2.

[mvandam]

Hello!

The machine that produces this error is a 64-bit Ubuntu 12.04:

legolas:~$ uname -a

Linux legolas 3.2.0-26-generic #41-Ubuntu SMP Thu Jun 14 17:49:24 UTC
2012 x86_64 x86_64 x86_64 GNU/Linux

legolas:~$ cat /etc/lsb-release
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=12.04
DISTRIB_CODENAME=precise
DISTRIB_DESCRIPTION="Ubuntu 12.04.3 LTS"

Intel i7-3930K (6 cores), 24 GB ram

I built imutil from source (latest version on github).

Thanks,
Marcos

On 27/02/14 12:32, Francois Rigaut wrote:

Hi Nieder,

Can you be more explicit on what type of machine produce the error?
What OS are you running? Are you building imutil from source? If not,
could it be a glibc version issue?
Thanks
Francois

On 27/02/14 10:07, Marcos van Dam wrote:

On some machines but not others, I get the following error when using
spline2

ERROR (spline2) Floating point interrupt (SIGFPE)

This error is totally repeateable (i.e., running the same commands
leads to the same errors). The funny thing is that after the bug is
produced, the same command can be called and it does not crash the
second time.

—
Reply to this email directly or view it on GitHub
#2.

—
Reply to this email directly or view it on GitHub
#2 (comment).

Marcos van Dam
Adaptive Optics Scientist
Flat Wavefronts
Postal Address: PO BOX 1060, Christchurch 8140, New Zealand
Cell phone: [phone]
Skype: marcos.van.dam
Website: www.flatwavefronts.com

[frigaut]

Marcos !
I didn't realised it was you first...

several things:

1. Thanks for the ygsl commit on yao
2. Can you send me a precise example of the command that makes spline2 fail?
   If it triggers on a special array, could you also send that to me?
   Cheers
   Francois

PS: what is it about the "nieder" pseudo?

On 27/02/14 10:54, Marcos van Dam wrote:

Hello!

The machine that produces this error is a 64-bit Ubuntu 12.04:

legolas:~$ uname -a

Linux legolas 3.2.0-26-generic #41-Ubuntu SMP Thu Jun 14 17:49:24 UTC
2012 x86_64 x86_64 x86_64 GNU/Linux

legolas:~$ cat /etc/lsb-release
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=12.04
DISTRIB_CODENAME=precise
DISTRIB_DESCRIPTION="Ubuntu 12.04.3 LTS"

Intel i7-3930K (6 cores), 24 GB ram

I built imutil from source (latest version on github).

Thanks,
Marcos

On 27/02/14 12:32, Francois Rigaut wrote:

Hi Nieder,

Can you be more explicit on what type of machine produce the error?
What OS are you running? Are you building imutil from source? If not,
could it be a glibc version issue?
Thanks
Francois

On 27/02/14 10:07, Marcos van Dam wrote:

On some machines but not others, I get the following error when using
spline2

ERROR (spline2) Floating point interrupt (SIGFPE)

This error is totally repeateable (i.e., running the same commands
leads to the same errors). The funny thing is that after the bug is
produced, the same command can be called and it does not crash the
second time.

—
Reply to this email directly or view it on GitHub
#2.

—
Reply to this email directly or view it on GitHub

#2 (comment).

Marcos van Dam
Adaptive Optics Scientist
Flat Wavefronts
Postal Address: PO BOX 1060, Christchurch 8140, New Zealand
Cell phone: [phone]
Skype: marcos.van.dam
Website: www.flatwavefronts.com

—
Reply to this email directly or view it on GitHub
#2 (comment).

[mvandam]

I don't have a nieder pseudonym... I am not sure where this came from!!!

The funny thing is that I can't get it to happen on its own (i.e., the
spline2 fails only if another spline2 call happens beforehand).

I will send you an email on how to see this on the computer where it
fails in a minute.
Marcos

On 27/02/14 13:03, Francois Rigaut wrote:

Marcos !
I didn't realised it was you first...

several things:

1. Thanks for the ygsl commit on yao
2. Can you send me a precise example of the command that makes spline2
   fail?
   If it triggers on a special array, could you also send that to me?
   Cheers
   Francois

PS: what is it about the "nieder" pseudo?

On 27/02/14 10:54, Marcos van Dam wrote:

Hello!

The machine that produces this error is a 64-bit Ubuntu 12.04:

legolas:~$ uname -a

Linux legolas 3.2.0-26-generic #41-Ubuntu SMP Thu Jun 14 17:49:24 UTC
2012 x86_64 x86_64 x86_64 GNU/Linux

legolas:~$ cat /etc/lsb-release
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=12.04
DISTRIB_CODENAME=precise
DISTRIB_DESCRIPTION="Ubuntu 12.04.3 LTS"

Intel i7-3930K (6 cores), 24 GB ram

I built imutil from source (latest version on github).

Thanks,
Marcos

On 27/02/14 12:32, Francois Rigaut wrote:

Hi Nieder,

Can you be more explicit on what type of machine produce the error?
What OS are you running? Are you building imutil from source? If not,
could it be a glibc version issue?
Thanks
Francois

On 27/02/14 10:07, Marcos van Dam wrote:

On some machines but not others, I get the following error when
using
spline2

ERROR (spline2) Floating point interrupt (SIGFPE)

This error is totally repeateable (i.e., running the same commands
leads to the same errors). The funny thing is that after the bug is
produced, the same command can be called and it does not crash the
second time.

—
Reply to this email directly or view it on GitHub
#2.

—
Reply to this email directly or view it on GitHub

#2 (comment).

Marcos van Dam
Adaptive Optics Scientist
Flat Wavefronts
Postal Address: PO BOX 1060, Christchurch 8140, New Zealand
Cell phone: [phone]
Skype: marcos.van.dam
Website: www.flatwavefronts.com

—
Reply to this email directly or view it on GitHub

#2 (comment).

—
Reply to this email directly or view it on GitHub
#2 (comment).

Marcos van Dam
Adaptive Optics Scientist
Flat Wavefronts
Postal Address: PO BOX 1060, Christchurch 8140, New Zealand
Cell phone: [phone]
Skype: marcos.van.dam
Website: www.flatwavefronts.com

[frigaut]

hum...
that's gotta be a thunderbird mischief then. Look at what the GUI says:

On 27/02/14 11:15, Marcos van Dam wrote:

I don't have a nieder pseudonym... I am not sure where this came from!!!

The funny thing is that I can't get it to happen on its own (i.e., the
spline2 fails only if another spline2 call happens beforehand).

I will send you an email on how to see this on the computer where it
fails in a minute.
Marcos

On 27/02/14 13:03, Francois Rigaut wrote:

Marcos !
I didn't realised it was you first...

several things:

1. Thanks for the ygsl commit on yao
2. Can you send me a precise example of the command that makes spline2
   fail?
   If it triggers on a special array, could you also send that to me?
   Cheers
   Francois

PS: what is it about the "nieder" pseudo?

On 27/02/14 10:54, Marcos van Dam wrote:

Hello!

The machine that produces this error is a 64-bit Ubuntu 12.04:

legolas:~$ uname -a

Linux legolas 3.2.0-26-generic #41-Ubuntu SMP Thu Jun 14 17:49:24 UTC
2012 x86_64 x86_64 x86_64 GNU/Linux

legolas:~$ cat /etc/lsb-release
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=12.04
DISTRIB_CODENAME=precise
DISTRIB_DESCRIPTION="Ubuntu 12.04.3 LTS"

Intel i7-3930K (6 cores), 24 GB ram

I built imutil from source (latest version on github).

Thanks,
Marcos

On 27/02/14 12:32, Francois Rigaut wrote:

Hi Nieder,

Can you be more explicit on what type of machine produce the error?
What OS are you running? Are you building imutil from source? If
not,
could it be a glibc version issue?
Thanks
Francois

On 27/02/14 10:07, Marcos van Dam wrote:

On some machines but not others, I get the following error when
using
spline2

ERROR (spline2) Floating point interrupt (SIGFPE)

This error is totally repeateable (i.e., running the same commands
leads to the same errors). The funny thing is that after the
bug is
produced, the same command can be called and it does not crash the
second time.

—
Reply to this email directly or view it on GitHub
#2.

—
Reply to this email directly or view it on GitHub

#2 (comment).

Marcos van Dam
Adaptive Optics Scientist
Flat Wavefronts
Postal Address: PO BOX 1060, Christchurch 8140, New Zealand
Cell phone: [phone]
Skype: marcos.van.dam
Website: www.flatwavefronts.com

—
Reply to this email directly or view it on GitHub

#2 (comment).

—
Reply to this email directly or view it on GitHub

#2 (comment).

Marcos van Dam
Adaptive Optics Scientist
Flat Wavefronts
Postal Address: PO BOX 1060, Christchurch 8140, New Zealand
Cell phone: [phone]
Skype: marcos.van.dam
Website: www.flatwavefronts.com

—
Reply to this email directly or view it on GitHub
#2 (comment).