Loading ...
Sorry, an error occurred while loading the content.
 

Re: [Jal_developers] 18F mnemonics - testsuite checked.

Expand Messages
  • Craig Franklin
    ... If you are working in cvs you can also: cvs diff -u file.diff to retrieve all the changes you have made but not yet committed. cvs diff -r FIRST_TAG -r
    Message 1 of 45 , Sep 9, 2003
      On Tue, 2003-09-09 at 22:47, mark gross wrote:
      > Javier Martínez wrote:
      >
      > > Another mistake found:
      > >
      > > In simulator.c line 1876:
      > >
      > > IFM(0xFE00, 0x1C00, sim_comf)
      > >
      > > ... should be:
      > >
      > > IFM(0xFC00, 0x1C00, sim_comf)
      > >
      > > How can I make a diff file from a diff file of the sources?
      >
      >
      > diff -u old_version new_version > file.diff
      >

      If you are working in cvs you can also:

      cvs diff -u > file.diff

      to retrieve all the changes you have made but not yet committed.

      cvs diff -r FIRST_TAG -r SECOND_TAG > file.diff

      to see the differences between two versions.

      Check the cvs manual for more options.

      > >
      > >
      > >
      > >
      > > Testsuite:
      > >
      > > I've currently this errors compiling with f877_20 (bug reappearance
      > > again):
      > >
      > > t0031 -> if a > 1 then ... when a=1
      > > t0085 -> the same
      > > t0106 -> the same
      > > t0134 -> the same
      > > t0135 -> the same
      > > t0099 -> movlw const - 1
      > >
      > >
      > > I've currently this errors compiling with f452_20:
      > >
      > > all f877_20s plus:
      > >
      > > lcd1 ..lcd9 -> jpic library instead of jpic16
      > > t0087 -> the same
      > > schakel6 -> eeprom_get not foud
      > > t0090 -> file_put not found
      > > t0116 -> corrected with this email
      > >
      > >
      > >
      > > [PS] We need to split the testsuite for diferent cores.
      > >
      > Your change has been incorporated along with the others.
      >
      > Ok, doing the regression test for f877 against jal vers 0.4.56 we do
      > have the following new regressions given your changes:
      > t0135.jal
      > t0134.jal
      > t0114.jal
      > t0106.jal
      > t0085.jal
      > t0031.jal
      > all now fail where they passed before, looks like all of them in the
      > simulation.
      >
      > Note the asm files are the same, so I think there is a new bug for f877
      > simulation that your patch snuck in.
      >
      > Running the do_test.sh reg-last I get the following regressions:
      > t0031.jal,
      > t0085.jal,
      > t0106.jal,
      > t0116.jal,
      > t0134.jal
      > t0135.jal
      > all have some new issue WRT 0.4.56.
      > There are other failures, these are just the ones unique the this
      > updated code.
      >
      > I have commited these changes to the source forge CVS, so that you can
      > work from the same code I have. We'll hold off doing a relase until we
      > fix the regressions, but I think this work is cool stuff.
      >
      > I've also updated the do_test.sh script to include "reg-last" parrameter
      > to do a regression against the currently installed version of jal WRT
      > the one thats in the build tree.
      >
      > Yeah I think having PIC core specific test dirctories would be nice
      > too. Perhaps someday we'll get to it.
      >
      > --mgross
      >
      >
      >
      > To unsubscribe from this group, send an email to:
      > Jal_developers-unsubscribe@yahoogroups.com
      >
      >
      >
      > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
      >
      >
    • Javier Martínez
      Hi Mark, ... I know what are you saying, that s why I left you to make changes in cvs this time. Hi, hi :-)) Now, going ahead: - here s my proposal for 6 new
      Message 45 of 45 , Sep 11, 2003
        Hi Mark,


        > Hmm, I must have botch something on my end.
        >


        I know what are you saying, that's why I left you to make changes in
        cvs this time. Hi, hi :-))


        Now, going ahead:

        - here's my proposal for 6 new test file. Can be anhaced,
        modified,etc.

        - A new update of 0.4.57, corrects negative jumps in simulator.
        Hex code seems to be OK.



        Javi.
      Your message has been successfully submitted and would be delivered to recipients shortly.