[mesa-users] pgstar SIGSEGV using 7184 and latest SDK

[Bill Paxton]

Hi Marcin,

Long time since we've heard from you!  Thanks to your detailed bug report, I was able to quickly fix the problem.  If you'd like to patch your current version, replace "function abundance_line" in star/private/pgstar_abundance by the following, then cd star/test;./mkx and you should be good to go.

         integer function abundance_line(cnt, j, str)
            integer, intent(in) :: cnt, j
            character (len=*), intent(in) :: str
            integer :: i, ii
            real :: x, frac, y
            include 'formats'
            call set_line_style(cnt)
            call pgslw(lw)
            call pgline(npts, xvec(grid_min:grid_max), yvec(grid_min:grid_max))
            call pgslw(lw_sav)
            call pgsch(txt_scale*s% Abundance_line_txt_scale_factor)
            abundance_line = cnt + 1
            do i=1,num_labels
               ii = iloc_abundance_label(i)
               if (ii > grid_max .or. ii < grid_min) cycle
               x = xloc_abundance_label(i)
               if (ii > 1) then
                  frac = (x-xvec(ii-1))/(xvec(ii)-xvec(ii-1))
                  y = frac*yvec(ii) + (1.0-frac)*yvec(ii-1) + dylbl
               else
                  y = yvec(ii) + dylbl
               end if
               if (y < ymin .or. y > ymax) cycle
               call pgptxt(x, y, 0.0, 0.5, str)
            end do
         end function abundance_line

The bug was that I forgot to check for ii == 1 inside the loop.

Cheers,
Bill






On Nov 28, 2014, at 5:49 AM, Marcin Mackiewicz wrote:

> Hello MESA Users,
> 
> Recently I was testing 1M_pre_ms_to_wd case from the test suite.
> The files I modified were history_colums.list, profile_colums.list and inlist _1.0. The rest I left intact.
> The mases tested were between 0.15 and 1.30 Msun with 0.05 step.
> All test cases ran fine with one exeption for 1.05 Msun.
> When I commented out pgstar_flag option situation did not occur.
> When I left pgstar_flag option set to .true. but commented out *_win_flag options problem occured again.
> 
> Test environment was Ubuntu 14.10 x64 with 8 GB of RAM.
> MESA v. 7184 with latest SDK.
> All modified MESA files and a terminal output with error message are attached.
> Terminal output suggests that SIGSEGV happened when pgplot was trying to save .png file.
> 
> Can you try to reproduce this error on Ubuntu x64 or any other platforms?
> 
> Regards
> Marcin Mackiewicz
> 
> 
> <history_columns.list><inlist_1.0><pgplot_fault.txt><profile_columns.list>------------------------------------------------------------------------------
> Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> with Interactivity, Sharing, Native Excel Exports, App Integration & more
> Get technology previously reserved for billion-dollar corporations, FREE
> http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk_______________________________________________
> mesa-users mailing list
> mesa-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/mesa-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20141128/c452a19c/attachment.html>