pkgsrc-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: ERROR: [...] missing RELRO

On Fri, 1 Oct 2021 15:08:57 +0200, Hauke Fath wrote:
> one more of these things that just get dropped on pkgsrc by 
> well-meaning people: RELRO.

Both on -9 and current, the editors/xemacs build fails with

Dumping under the name xemacs
Testing for Lisp shadows ...
[1]   Segmentation fault (core dumped) ./xemacs -batch -vanilla -f 
*** [xemacs.dmp] Error code 139

Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x00000000962a51a2 in pdump_load_finish () at dumper.c:1112
1112          (* ptr.address) = ptr.value + delta;
Really redefine built-in command "ptype"? (y or n) [answered Y; input 
not from terminal]
(gdb) bt
#0  0x00000000962a51a2 in pdump_load_finish () at dumper.c:1112
#1  pdump_load (argv0=<optimized out>) at dumper.c:1447
#2  0x0000000096291ef9 in xemacs_21_4_24_x86_64__netbsd (argc=5, 
argv=0x7f7fff122ce8, envp=<optimized out>, restart=restart@entry=0) at 
#3  0x00000000963ba6b0 in main (argc=<optimized out>, argv=<optimized 
out>, envp=0x7f7fff122d18) at emacs.c:2829

Setting PKGSRC_USE_RELRO=no (in mk.conf) and/or RELRO_SUPPORTED=no (in 
the package Makefile) does not make a difference.

Would the use of RELRO in dependencies matter? Has anything else global 
landed, equally well communicated?


I would really appreciate if the people who intend to throw global 
switches would 

o run a bulk build beforehand, to get an idea of the fallout
o announce the change to package maintainers, explaining what it is 
o explain how to mitigate the breakage



Hauke Fath                        <hauke%Espresso.Rhein-Neckar.DE@localhost>
Linnéweg 7
64342 Seeheim-Jugenheim

Home | Main Index | Thread Index | Old Index