I don't know either, it just continually failed at that point until I 'massaged' the origonal patch so it was set to about the correct line numbers, when it then 'just' continually worked. Which is why I submitted this thread and patch update. I origonally disabled the patch, but on a read of it and rethink I decided it was wanted.

And looking at your patch update (message below) it didn't mention the three sets of line number changes I did. Maybe they had moved too far for the patch command to work with, or has patch itself been updated since (no), who knows, all I do know is that i'm using my modified patch and it works every time for me. I'll leave it at that for now and re-test when I start with the latest revision download.

I'd rather not play with it while i'm working with a 'mixed' revision at the moment as i've found a few changes as I go through against your packages that are working, you'll get some diff's when i've finished and tested them against a fresh download of the latest revision (at that point as you don't want them now). I'm just building 20-minimal. Nothing major, mainly just build orders and stages so far.

And that glibc/glibc32 problem of v2.30 builds everything while v2.31 'kills' the build at [2] gcc v9.2.0 every time is a pest, so i've gone back to using v2.30 for now. Do you have time to see what's up, or am I missing an already created 'fix' revision.

regards and thanks
scsijon

On 01/03/20 05:21, René Rebe wrote:
Hi,

prior to writing this email I had checked that the mentioned revision includes 
updating the patch:

        svn di -c r49190 https://svn.exactcode.de/t2/trunk

I also just test build binutils again in a clean tree, and it just built 
without any issue for me.

I really don’t know what happens on your side there, copying patches in 
manually?

In any case, as we did not had “versioned” releases for a decade, sigh, I will 
start tagging versions after massive test and fix cables, like last year, and 
before major updates and cleanup, like now. So that we have more stable 
versions to mention ike t2 19.12 20.1 etc.

        René

On 28. Feb 2020, at 22:20, scsijon <[email protected]> wrote:

I downloaded a totally new t2 9.0 tree and that's what I got. ?Did you by 
chance only update the package, as the ld-glob.patch is one of yours in the 
package/develop/binutils directory.

regards

On 28/02/20 20:31, René Rebe wrote:
Hi,
I remember updating the patch live during some youtube livestream, and indeed 
it looks like that in the SVN history:
r49190 | rene | 2020-02-01 22:11:49 +0000 (Sat, 01 Feb 2020) | 2 lines
         * updated binutils (2.33.1 -> 2.34)
Any change you have random old files in your tree? How did you update it?
        René
On 27. Feb 2020, at 16:21, scsijon <[email protected]> wrote:

Downloading a fresh T2 at r49451 to build at 20-minimum.in, the config part for 
0-binutils failed at first go at this patch, after investigation it seems the 
binutils-2.34/ld/ldelf.c file has changed internally, moving the area marked 
for this patch up by a few hundred lines, thus of course failing the patch.

I've attempted to ammend the patch, but it still needs checking before 
releasing by Rene in case i've forgotten something.


<ld-glob.patch>-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
[email protected] with a subject of: unsubscribe t2




----------------------------------------------------------- If you wish to unsubscribe from this mailing, send mail to
[email protected] with a subject of: unsubscribe t2

Reply via email to