[petsc-users] EXTERNAL: Re: EXTERNAL: Re: Upgrading from petsc 3.2-p7 to 3.3-p5

[Wadud.Miah at awe.co.uk]

Hi Satish,

When should the subroutine:

MatSetOption( mat,MAT_NEW_NONZERO_ALLOCATION_ERR,PETSC_FALSE )

be called? I currently have the following calling sequence:

1. MatCreate
2. MatSetSizes
3. MatSetType
4. MatMPIAIJSetPreallocation
5. MatSetValues
6. MatAssemblyBegin
7. MatAssemblyEnd

Regards,
Wadud.

-----Original Message-----
From: Satish Balay [mailto:balay at mcs.anl.gov] 
Sent: 23 January 2014 16:45
To: Miah Wadud AWE
Cc: knepley at gmail.com; petsc-users at mcs.anl.gov
Subject: EXTERNAL: Re: [petsc-users] EXTERNAL: Re: Upgrading from petsc 3.2-p7 to 3.3-p5

> I get a non-zero error code from the MatAssemblyBegin/End but nothing gets printed.


1. It appears you are not using CHKERRQ() on your calls to PETSc routines.

[yeah its combersome from fortran - so the petsc examples don't have
it - but that will print the complete error trace].

2. The latest petsc version is 3.4 - suggest upgrading your code to it.

3. from http://www.mcs.anl.gov/petsc/documentation/changes/33.html

>>>>
Preallocation routines now automatically set
MAT_NEW_NONZERO_ALLOCATION_ERR, if you intentionally preallocate less
than necessary then use
MatSetOption(mat,MAT_NEW_NONZERO_ALLOCATION_ERR,PETSC_FALSE) to
disable the error generation.
<<<<

Satish


On Thu, 23 Jan 2014, Wadud.Miah at awe.co.uk wrote:

> Hi Matthew,
> 
> I get a non-zero error code from the MatAssemblyBegin/End but nothing gets printed. Am I missing something in 3.3-p5? Please reply to wadud.miah at awe.co.uk with a reply as my email header gets re-written for some strange reason!
> 
> Regards,
> Wadud.
> 
> ________________________________
> From: Matthew Knepley [mailto:knepley at gmail.com]
> Sent: 23 January 2014 16:21
> To: Miah Wadud AWE
> Cc: petsc-users at mcs.anl.gov
> Subject: EXTERNAL: Re: [petsc-users] Upgrading from petsc 3.2-p7 to 3.3-p5
> 
> On Thu, Jan 23, 2014 at 10:05 AM, <Wadud.Miah at awe.co.uk<mailto:Wadud.Miah at awe.co.uk>> wrote:
> Hello petsc users,
> 
> I have upgraded to PETSc 3.3-p5 from 3.2-p7 and my matrix assembly is failing (MatAssemblyBegin and MatAssemblyEnd). I am using MatMPIAIJSetPreallocation prior to calling MatSetValues. I have looked at the 3.3-p5 changelog and cannot find anything. Any help will be greatly appreciated. Please reply to wadud.miah at awe.co.uk<mailto:wadud.miah at awe.co.uk> with a reply as my email header gets re-written for some strange reason!
> 
> Always send the full error message.
> 
>    Matt
> 
> 
> Regards,
> Wadud.
> 
> ___________________________________________________ ____________________________ The information in this email and in any attachment(s) is commercial in confidence. If you are not the named addressee(s) or if you receive this email in error then any distribution, copying or use of this communication or the information in it is strictly prohibited. Please notify us immediately by email at admin.internet(at)awe.co.uk<http://awe.co.uk>, and then delete this message from your computer. While attachments are virus checked, AWE plc does not accept any liability in respect of any virus which is not detected. AWE Plc Registered in England and Wales Registration No 02763902 AWE, Aldermaston, Reading, RG7 4PR
> 
> 
> 
> --
> What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.
> -- Norbert Wiener
> 
> 
> ___________________________________________________
> ____________________________
> 
> The information in this email and in any attachment(s) is
> commercial in confidence. If you are not the named addressee(s)
> or
> if you receive this email in error then any distribution, copying or
> use of this communication or the information in it is strictly
> prohibited. Please notify us immediately by email at
> admin.internet(at)awe.co.uk, and then delete this message from
> your computer. While attachments are virus checked, AWE plc
> does not accept any liability in respect of any virus which is not
> detected.
> 
> AWE Plc
> Registered in England and Wales
> Registration No 02763902
> AWE, Aldermaston, Reading, RG7 4PR
> 


___________________________________________________
____________________________

The information in this email and in any attachment(s) is 
commercial in confidence. If you are not the named addressee(s) 
or 
if you receive this email in error then any distribution, copying or 
use of this communication or the information in it is strictly 
prohibited. Please notify us immediately by email at 
admin.internet(at)awe.co.uk, and then delete this message from 
your computer. While attachments are virus checked, AWE plc 
does not accept any liability in respect of any virus which is not 
detected.

AWE Plc
Registered in England and Wales
Registration No 02763902
AWE, Aldermaston, Reading, RG7 4PR