"SfR Fresh" - the SfR Freeware/Shareware Archive

Member "gmemusage-0.2/TODO" of archive gmemusage-0.2.tar.gz:


As a special service "SfR Fresh" has tried to format the requested source page into HTML format using source code syntax highlighting with prefixed line numbers. Alternatively you can here view or download the uninterpreted source code file. That can be also achieved for any archive member file by clicking within an archive contents listing on the first character of the file(path) respectively on the according byte size field.
    1 Here's the wish list:
    2 
    3 - Make the stack clickable so that clicing on a process will display
    4   some details about the process.
    5 
    6 - Allow user to switch between core and (core + swap) views. Shouldn't
    7   be too difficult -- the infrastructure is already in place as far as
    8   I can make out. Make a command-line option for the same.
    9 
   10 - Make a help screen (key "h").
   11 
   12 - Fix unbalanced drawables (currently the graphics and text seem to
   13   prefer to hang around at the bottom of the window).
   14 
   15 - Make a better icon/pixmap :)
   16 
   17   Update: I've been sent a pixmap but have not incorporated it yet
   18   since it'll probably necessitate my learning how to use libXpm. Some 
   19   time soon...
   20 
   21 - The speed delay due to opening and reading files in /proc seems to
   22   be minimal on my 486/66. However the process does take up a fair
   23   amount of CPU when it's scanning /proc. Maybe I can find out some
   24   way of getting these values from the kernel directly.
   25 
   26 - sum(RSS) for all procs works out to more than physical memory when
   27   memory is nearly or completely full. Leave this one for Linus to fix
   28   :-)
   29 
   30   Update: Apparently Linus is not to blame, it was my understanding of 
   31   the memory values which was faulty. I am not compensating for shared 
   32   memory (especially libraries) and hence the values which you see are 
   33   pretty approximate. Is there someplace on the web which discusses
   34   process memory usage vis-a-vis the values found in /proc/process-id?
   35 
   36 - Get a decent color combination for the bars, labels and background.
   37   R G B is easy but not particularly beautiful. There's also a new
   38   pastel color scheme (define USE_PASTELS in the Makefile) though the
   39   contrast in the colors in that could be better.
   40 
   41 - I have reports of gmemusage behaving erratically on systems with
   42   128Mb of RAM. I forgot to stock up on those this quarter, so any
   43   inputs will help. The problem is currently under investigation.
   44 
   45 - SPARCLinux gmemusage doesn't seem to read the installed system RAM
   46   correctly. Does stat(/proc/kcore) on SPARCS return something
   47   different on SPARCS?