| 1 |
On 2017-12-06 13:28, Peter Humphrey wrote: |
| 2 |
> On Sunday, 3 December 2017 15:12:21 GMT Mick wrote: |
| 3 |
>> On 03-12-2017 ,10:57:33, Peter Humphrey wrote: |
| 4 |
>> > On Saturday, 2 December 2017 12:30:57 GMT Mick wrote: |
| 5 |
>> > > I'm getting this error after I changed my profile as per |
| 6 |
>> > > '2017-11-30-new-17- |
| 7 |
>> > > |
| 8 |
>> > > profiles' news item: |
| 9 |
>> > > >>> Compiling source in |
| 10 |
>> > > >>> /data/tmp_var/portage/sys-boot/grub-0.97-r16/work/ |
| 11 |
>> > |
| 12 |
>> > [...] |
| 13 |
>> > |
| 14 |
>> > > However, sys-boot/grub-0.97-r17 installed fine once keyworded on this |
| 15 |
>> > > (mostly) stable system. This may save time for others who come across |
| 16 |
>> > > the same problem. |
| 17 |
>> > sys-boot/grub-0.97-r17 |
| 18 |
>> > It has. Thanks Mick. |
| 19 |
>> |
| 20 |
>> Unfortunately, an older system with only 50MB /boot partition did not |
| 21 |
>> have enough space to allow sys-boot/grub-0.97-r17 to install all its |
| 22 |
>> files and fs drivers. I ended up restoring /boot from a back up. |
| 23 |
>> YMMV. |
| 24 |
> |
| 25 |
> I spoke too soon, too. Sys-boot/grub-0.97-r17 compiled and installed |
| 26 |
> all |
| 27 |
> right, as a package, but when I went to install it to the MBR I got an |
| 28 |
> error |
| 29 |
> complaining of a mismatch or corruption in stage X. The wording was |
| 30 |
> something like that, and I forget the value of X. There was no mention |
| 31 |
> of |
| 32 |
> disk space, and the boot partition is 2GB, so I think it's something |
| 33 |
> else. |
| 34 |
> |
| 35 |
> Installing sys-boot/grub-static-0.97-r12 instead went smoothly, so I've |
| 36 |
> left |
| 37 |
> it like that for the moment. |
| 38 |
> |
| 39 |
> Does the team think I should go back to grub-0.97-r17, take proper |
| 40 |
> records |
| 41 |
> and file a bug report? |
| 42 |
|
| 43 |
I question if this makes sense for a masked ebuild. |
| 44 |
|
| 45 |
I'm curious about what was discussed until now. The issue seems to be |
| 46 |
quite simple to solve. |
| 47 |
|
| 48 |
The build fails but portage/gcc does give clear info in this case: the |
| 49 |
option "-nopie" have to be changed to "-no-pie". This option is set in |
| 50 |
860_all_grub-0.97-pie.patch. Here is a diff: |
| 51 |
|
| 52 |
--- a/860_all_grub-0.97-pie.patch 2012-05-31 01:00:13.000000000 |
| 53 |
+0200 |
| 54 |
+++ b/860_all_grub-0.97-pie.patch 2017-12-07 11:28:57.536089642 |
| 55 |
+0100 |
| 56 |
@@ -17,8 +17,8 @@ |
| 57 |
+ grub_cv_cc_fpie=no) |
| 58 |
+ ]) |
| 59 |
+ if test "x$grub_cv_cc_fpie" = xyes; then |
| 60 |
-+ STAGE1_CFLAGS="$STAGE1_CFLAGS -nopie" |
| 61 |
-+ STAGE2_CFLAGS="$STAGE2_CFLAGS -nopie" |
| 62 |
++ STAGE1_CFLAGS="$STAGE1_CFLAGS -no-pie" |
| 63 |
++ STAGE2_CFLAGS="$STAGE2_CFLAGS -no-pie" |
| 64 |
+ fi |
| 65 |
fi |
| 66 |
fi |
| 67 |
|
| 68 |
Maybe the easiest way is to create a new grub-patches package, but there |
| 69 |
are other ways to change this too. I'm expected the upstream will change |
| 70 |
this soon - within the remaining 5 weeks ;-). |
| 71 |
|
| 72 |
Another thing is I question that grub-legacy have to be rebuild at all. |
| 73 |
I'm pretty sure it is save to remove it from the world file or comment |
| 74 |
it out. |
| 75 |
|
| 76 |
Anyhow, upgrading to grub2 is IMHO the right way. There are some |
| 77 |
examples given in parallel threads how to write a grub.cfg by hand - and |
| 78 |
keep it simple :-). Then nothing else then the grub2 binary and |
| 79 |
grub2-install is required usually. |
| 80 |
|
| 81 |
Kai |
| 82 |
-- |
| 83 |
Sent with eQmail-1.10 |
Archives