avr-gcc-list
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [avr-gcc-list] More results from the testsuite with avrtest


From: Paulo Marques
Subject: Re: [avr-gcc-list] More results from the testsuite with avrtest
Date: Wed, 23 Jan 2008 01:02:08 +0000
User-agent: Internet Messaging Program (IMP) H3 (4.1.2)

Quoting Andrew Hutchinson <address@hidden>:

Ok,

here are my test results - same GCC as Paulo - except I have install some patches in avr to fix some reported bugs (hence I'm testing patches)

I did NOT include Paulo's testcase patch for gcc.c-torture/execute/builtins/pr23484-chk.c: (Paulo please state effect on numbers for this)

      === gcc Summary ===

# of expected passes        11785
# of unexpected failures    59
# of unresolved testcases    30
# of unsupported tests        682

/cygdrive/e/awhconf/gcc/xgcc  version 4.3.0 20080121 (experimental) (GCC)

Ok, I tried to remove the patch to pr23484-chk.c only to find out that it isn't part of the execute.exp section :P

So, I still get the same result as before. This means that there must be something different between our setups. I just hope it isn't something that I changed and forgot :(

Could you please compare the "failing tests list" that I sent in my mail yesterday to see which specific tests are failing for you and not for me? That might help figure out where's the difference.

Unlike Paulo, it takes a while to run test on Cygwin as make spawns tasks.

I can see you're almost ready to come into the light ;)

On the otherhand, I can download full GCC in maybe 30 mins and a new rev in minutes!

I can download a new rev in minutes, too (or even much less, depending on what changed from my local revision to the new one).

The problem is that if you're going to bisect a lot of bugs, you're going to test probably hundreds of revisions, and a few users doing that around the world would put an enormous strain on the subversion server itself.

From http://gcc.gnu.org/bugs/reghunt.html:

Using rsync to get a local copy of the GCC CVS repository is highly recommended
for regression hunts. You'll be checking out the tree used for the regression
search over and over again and won't want to affect access times for other GCC
developers who are using the real repository, and it will also be faster for
you.

(this documentation seems older than the move from CVS to SVN, but I guess the same principles still apply)

In that case doing just a rsync of the repository and then bisecting from your local copy is much better. It is even better if you expect to do it frequently in the future, because rsync only sends the changes between your local copy and the master repository (and that shouldn't be much worse than an svn update).

I'm rsync'ing the repository right now.

I'll let you guys know how that turns out,

--
Paulo Marques


----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.





reply via email to

[Prev in Thread] Current Thread [Next in Thread]