Gentoo Archives: gentoo-alt

From: Joey Dumont <joey.dumont@×××××.com>
To: gentoo-alt@l.g.o
Subject: Re: [gentoo-alt] Issues bootstrapping Gentoo Prefix on Arch Linux
Date: Thu, 17 Dec 2020 02:47:06
Message-Id: CAB6-_ier66pWiMG=Tvev+hd6m=GnY7m6FqfJTHKFGqPFEvyxAg@mail.gmail.com
In Reply to: Re: [gentoo-alt] Issues bootstrapping Gentoo Prefix on Arch Linux by Fabian Groffen
1 Done: https://bugs.gentoo.org/760411
2
3 Thanks!
4
5 Joey Dumont (Profile <http://blog.joey-dumont.ca/>)
6 The supreme elegance of Nature lies in its apparent simplicity.
7
8
9 On Wed, 16 Dec 2020 at 02:04, Fabian Groffen <grobian@g.o> wrote:
10
11 > I think RAP simply pulls in attr, and we need to see if we can fix it.
12 > Can you file a bug for it on bugs.gentoo.org if you haven't yet?
13 >
14 > Thanks,
15 > Fabian
16 >
17 > On 15-12-2020 21:44:01 -0500, Joey Dumont wrote:
18 > > At this exploratory stage, I do not need a libc in my prefix. Arch is a
19 > rolling
20 > > release, so I currently glibc-2.32, which I believe is the latest.
21 > >
22 > > Good to hear that it's working on both Ubuntu and CentOS: those are the
23 > likely
24 > > platforms for any production work. I was mostly curious to see how it
25 > would fare
26 > > on my home system. It's a nice way of getting acquainted with portage as
27 > well.
28 > >
29 > > The PREFIX_DISABLE_RAP=yes bootstrap worked! This is enough for me on
30 > Arch
31 > > Linux, although I'd be curious to know why RAP fails. I just tried
32 > emerging attr
33 > > on my non-RAP prefix, and it also fails. Do you know if it's a function
34 > of the
35 > > host glibc?
36 > >
37 > > In any case, I think this is fine. I'll set prefix up on a CentOS or
38 > Ubuntu
39 > > system with a prefixed libc.
40 > >
41 > > Thanks!
42 > >
43 > > Joey Dumont (Profile[1])
44 > > The supreme elegance of Nature lies in its apparent simplicity.
45 > >
46 > >
47 > > On Tue, 15 Dec 2020 at 02:55, Fabian Groffen <grobian@g.o[2]>
48 > wrote:
49 > > > I know this is not a small ask, but do you need a libc in your Prefix?
50 > > > E.g. do you anticipate one is neccesary because Arch's is too old?
51 > > >
52 > > > If not, could you try bootstrapping again from scratch with
53 > > > PREFIX_DISABLE_RAP=yes in your environment set.
54 > > >
55 > > > I see that yesterday's Ubuntu bootstrap (using RAP) succeeded, and the
56 > > > CentOS bootstrap succeeded not too long ago, so this may be a total
57 > > > pointless ask.
58 > > >
59 > > > Thanks,
60 > > > Fabian
61 > > >
62 > > >
63 > > > On 14-12-2020 20:59:52 -0500, Joey Dumont wrote:
64 > > > > I tried contacting the mailing list a couple times, but it seems my
65 > > > > messages weren't going through. Trying with plain text email, sorry
66 > if
67 > > > > I generated noise.
68 > > > >
69 > > > > I've been trying to bootstrap Gentoo Prefix on Arch Linux. However, I
70 > > > > am having issues during stage3. Specifically, I am having trouble
71 > > > > building sys-apps/attr-2.4.48-r4. Actually, it builds fine, but then
72 > > > > the symbol version sanity check fails:
73 > > > >
74 > > > > * ERROR: sys-apps/attr-2.4.48-r4::gentoo failed (install phase):
75 > > > > * symbol version sanity check failed; please comment on
76 > > > > https://bugs.gentoo.org/644048[3]
77 > > > > *
78 > > > > * Call stack:
79 > > > > * ebuild.sh, line 125: Called src_install
80 > > > > * environment, line 2163: Called multilib-minimal_src_install
81 > > > > * environment, line 1501: Called multilib_foreach_abi
82 > > > > 'multilib-minimal_abi_src_install'
83 > > > > * environment, line 1734: Called multibuild_foreach_variant
84 > > > > '_multilib_multibuild_wrapper' 'multilib-minimal_abi_src_install'
85 > > > > * environment, line 1388: Called _multibuild_run
86 > > > > '_multilib_multibuild_wrapper' 'multilib-minimal_abi_src_install'
87 > > > > * environment, line 1386: Called _multilib_multibuild_wrapper
88 > > > > 'multilib-minimal_abi_src_install'
89 > > > > * environment, line 474: Called multilib-minimal_abi_src_install
90 > > > > * environment, line 1491: Called multilib_src_install
91 > > > > * environment, line 1965: Called die
92 > > > > * The specific snippet of code:
93 > > > > * die "symbol version sanity check failed; please
94 > > > > comment on https://bugs.gentoo.org/644048[4]";
95 > > > >
96 > > > > I've checked the symbols, and it does seem that the issue fits the
97 > > > > parameters of bug 644048:
98 > > > >
99 > > > > ~/software/gentoo/2020.12/tmp/bin/x86_64-pc-linux-gnu-readelf -sW
100 > > > >
101 > /home/valandil/software/gentoo/2020.12/var/tmp/portage/sys-apps/attr-2.4.48-
102 > > > r4/image/home/valandil/software/gentoo/2020.12/usr/lib64/libattr.so.1
103 > > > > | grep getxattr
104 > > > > 13: 0000000000000000 0 FUNC GLOBAL DEFAULT UND
105 > > > > lgetxattr@GLIBC_2.3 (7)
106 > > > > 26: 0000000000000000 0 FUNC GLOBAL DEFAULT UND
107 > > > > fgetxattr@GLIBC_2.3 (7)
108 > > > > 31: 0000000000000000 0 FUNC GLOBAL DEFAULT UND
109 > > > > getxattr@GLIBC_2.3 (7)
110 > > > > 43: 00000000000040e0 0 FUNC GLOBAL DEFAULT 13
111 > fgetxattr@ATTR_1.0
112 > > > > 54: 00000000000040a0 0 FUNC GLOBAL DEFAULT 13
113 > getxattr@ATTR_1.0
114 > > > > 62: 00000000000040c0 0 FUNC GLOBAL DEFAULT 13
115 > lgetxattr@ATTR_1.0
116 > > > > 47: 00000000000040c0 24 FUNC LOCAL DEFAULT 13
117 > libattr_lgetxattr
118 > > > > 50: 00000000000040a0 24 FUNC LOCAL DEFAULT 13
119 > libattr_getxattr
120 > > > > 57: 00000000000040e0 23 FUNC LOCAL DEFAULT 13
121 > libattr_fgetxattr
122 > > > > 83: 0000000000000000 0 FUNC GLOBAL DEFAULT UND
123 > > > lgetxattr@@GLIBC_2.3
124 > > > > 89: 00000000000040a0 0 FUNC GLOBAL DEFAULT 13
125 > getxattr@ATTR_1.0
126 > > > > 97: 00000000000040c0 0 FUNC GLOBAL DEFAULT 13
127 > lgetxattr@ATTR_1.0
128 > > > > 113: 0000000000000000 0 FUNC GLOBAL DEFAULT UND
129 > > > fgetxattr@@GLIBC_2.3
130 > > > > 123: 0000000000000000 0 FUNC GLOBAL DEFAULT UND getxattr@
131 > @GLIBC_2.3
132 > > > > 127: 00000000000040e0 0 FUNC GLOBAL DEFAULT 13
133 > fgetxattr@ATTR_1.0
134 > > > >
135 > > > > I tried adding the ebuild/patch attached to the bug above by adding a
136 > > > > local repo, and masking earlier versions of sys-apps/attr in
137 > > > > tmp/etc/portage/package.mask, but since the patch modifies a file
138 > > > > alled Makemodules.am, portage triggers automake-1.15, which is not
139 > > > > available on the prefix at this point. I tried installing it by
140 > adding
141 > > > > sys-devel/automake to the list of pkgs installed at this point, but
142 > > > > this fails at the install_qa_check_prefix stage, as automake contains
143 > > > > non-prefixed shebangs.
144 > > > >
145 > > > > What's the way forward here? Should I write my own automake patch to
146 > > > > fix the non-prefixed shebang? Or is that a known issue in Prefix with
147 > > > > a better (known) solution?
148 > > > >
149 > > > > Thanks for any help!
150 > > > >
151 > > > > Joey Dumont (Profile)
152 > > > > The supreme elegance of Nature lies in its apparent simplicity.
153 > > > >
154 > > >
155 > > > --
156 > > > Fabian Groffen
157 > > > Gentoo on a different level
158 > >
159 > >
160 > > References
161 > > 1. http://blog.joey-dumont.ca/
162 > > 2. mailto:grobian@g.o
163 > > 3. https://bugs.gentoo.org/644048
164 > > 4. https://bugs.gentoo.org/644048
165 >
166 > --
167 > Fabian Groffen
168 > Gentoo on a different level
169 >