[lfs-dev] Coreutils 8.25 i18n patch breaks cut

Bruce Dubbs bruce.dubbs at gmail.com
Mon Feb 8 20:26:59 PST 2016


DJ Lucas wrote:
>
>
> On 2/7/2016 10:27 PM, Chris Staub wrote:
>
>> The problem is that the "cut" program isn't giving the expected output.
>> After some further experimentation, and checking with William Harrington
>> in IRC who got the same results, this problem only occurs when using
>> Coreutils 8.25 with the i18n patch that's in LFS.
>
> SUSE has a considerably different set of patches for 8.25. This particular
> issue is resolved by substituting the SUSE MB patch for the MB patch
> provided by RH, but I'm not sure what kind of coverage we'll have from
> this one. It's considerably different than the direction that RedHat is
> taking (mentioned in my previous post), but less invasive that the one
> we've used previously (no additional library - probably won't have to do
> UNSAFE_CONFIGURE make check, but didn't test that way). Anyway, test
> results look good (from a partially completed system with additional deps):
>
> ============================================================================
> Testsuite summary for GNU coreutils 8.25
> ============================================================================
> # TOTAL: 30
> # PASS:  23
> # SKIP:  7
> # XFAIL: 0
> # FAIL:  0
> # XPASS: 0
> # ERROR: 0
> --
> ============================================================================
> Testsuite summary for GNU coreutils 8.25
> ============================================================================
> # TOTAL: 583
> # PASS:  469
> # SKIP:  114
> # XFAIL: 0
> # FAIL:  0
> # XPASS: 0
> # ERROR: 0
> --
> ============================================================================
> Testsuite summary for GNU coreutils 8.25
> ============================================================================
> # TOTAL: 308
> # PASS:  296
> # SKIP:  12
> # XFAIL: 0
> # FAIL:  0
> # XPASS: 0
> # ERROR: 0
>
> Options (as I see it):
>
> 1. Use the SUSE patch as is and keep an eye on upstream
> 2. Use the main RedHat patch (since we have a history with it) and grab
> patches for cut, expand, and unexpand (the suspect ones from RH) from the
> SUSE patch (SUSE patch is not dependent on additional libs).
> 3. Kill cut patch (and possibly expand/unexpand).

I do not like 2.  Too invasive.  I'm open to the others.

> I'm going to give the SUSE patch a shot. I'll run it through jhalfs
> tonight. Wanna get in a test run with new grep and texinfo anyway.

I've tested the new texinfo and the only thing that changes in the book is 
version, size, and md5sum.

> Will
> upload the rediffed SUSE patch in a sec (can always kill it or increment
> if one of the other options is selected).

I'll go with your recommendation, but we need to settle on it soon.  I 
plan on releasing LFS 7.9-rc1 on Sunday.  After that we will be in a 
package freeze until we can release a stable LFS/BLFS stable 7.9.

   -- Bruce




More information about the lfs-dev mailing list