I'm switching to non-partial repositories for most of my GHC trees now. This
way I avoid the whole class of get_extra failures, which seem to be cropping up
quite a lot these days. To start, you can grab:
http://darcs.haskell.org/ghc-HEAD-20061109.tar.bz2
And then pull into that tree. Doing 'darcs get' without --partial probably
won't work on Windows because some of the patches interact badly with
case-insensitive file systems.
Non-partial trees aren't so bad: you only need to download the whole thing once,
then you can just copy it to make new trees. Don't 'darcs get' on Windows,
again due to the case-insensitivity problem, but definitely *do* 'darcs get' on
Unix, because it'll share the patch files between the two trees.
Cheers,
Simon
Simon Peyton-Jones wrote:
Sorry. I mis-typed. Then I tried to un-record the patch I had recorded.
Darcs crashed (log below).
Since I cannot un-record it, all I could do was to push it.
Sigh
Anyway, it's a small change fixing a bug in TcSimplify that mixed up
implication constraints and implicit parameters; it fixes one of the two bugs
reported in Trac #1033
Simon
| -----Original Message-----
| From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Lemmih
| Sent: 01 December 2006 15:28
| To: Simon Peyton-Jones
| Cc: [EMAIL PROTECTED]
| Subject: Re: patch applied (ghc): q
|
| On 12/1/06, Simon Peyton Jones <[EMAIL PROTECTED]> wrote:
| > Thu Nov 30 19:42:07 PST 2006 [EMAIL PROTECTED]
| > * q
| >
| > M ./compiler/typecheck/Inst.lhs -1 +3
| > M ./compiler/typecheck/TcSimplify.lhs -4 +1
|
| q?
sh-2.04$ darcs unrecord
Fri Dec 1 03:42:07 GMT Standard Time 2006 [EMAIL PROTECTED]
* q
Shall I unrecord this patch? (1/1121) [ynWvpxqadjk], or ? for help: y
Wed Nov 29 22:52:43 GMT Standard Time 2006 [EMAIL PROTECTED]
* Remove trace
Shall I unrecord this patch? (2/1121) [ynWvpxqadjk], or ? for help: d
darcs.exe: failed to read patch in get_extra:
Fri Apr 7 11:18:39 GMT Standard Time 2006 Simon Marlow <[EMAIL PROTECTED]>
* Fix a bug related to threads blocked on blackholes
We weren't making them live early enough, with the result that
finalizable objects referred to only by a thread blocked on a black
hole could be finalized too early (see conc057 test).
Perhaps this is a 'partial' repository?
_______________________________________________
Cvs-ghc mailing list
[EMAIL PROTECTED]
http://www.haskell.org/mailman/listinfo/cvs-ghc
