| 1 |
On 3/23/2020 04:21, Jaco Kroon wrote: |
| 2 |
> Hi, |
| 3 |
> |
| 4 |
> https://bugs.gentoo.org/713668 relates. |
| 5 |
> |
| 6 |
> * Searching for /usr/include/execinfo.h ... |
| 7 |
> sys-libs/glibc-2.29-r7 (/usr/include/execinfo.h) |
| 8 |
> |
| 9 |
> As I see I can either add an explicit depend on glibc which I'd prefer |
| 10 |
> not to. Or someone from the musl team could possibly assist on how to |
| 11 |
> get the backtrace() set of calls on musl please? |
| 12 |
> |
| 13 |
> Alternatively I need to add a test and simply path debug.c to only |
| 14 |
> provide stub function for print_backtrace(FILE *fp) that just does |
| 15 |
> fprintf(fp, "No backtrace() available to print a backtrace.\n"); |
| 16 |
> |
| 17 |
> Suggestions? |
| 18 |
> |
| 19 |
> Kind Regards, |
| 20 |
> Jaco |
| 21 |
|
| 22 |
Some quick searching on google, it looks like the cleanest fix for that bug |
| 23 |
is dahdi-tools needs to be patched to only include execinfo.h or only use |
| 24 |
backtrace() on glibc-based systems, and that patch then sent to the |
| 25 |
dahdi-tools upstream developers for inclusion in a future release. That |
| 26 |
way, we're not dragging that patch around forever in the tree or in the musl |
| 27 |
overlay. |
| 28 |
|
| 29 |
It also doesn't look like musl itself will ever implement execinfo.h or |
| 30 |
backtrace(), per this message in 2015 from the lead musl developer: |
| 31 |
https://www.openwall.com/lists/musl/2015/04/09/3 |
| 32 |
|
| 33 |
-- |
| 34 |
Joshua Kinard |
| 35 |
Gentoo/MIPS |
| 36 |
kumba@g.o |
| 37 |
rsa6144/5C63F4E3F5C6C943 2015-04-27 |
| 38 |
177C 1972 1FB8 F254 BAD0 3E72 5C63 F4E3 F5C6 C943 |
| 39 |
|
| 40 |
"The past tempts us, the present confuses us, the future frightens us. And |
| 41 |
our lives slip away, moment by moment, lost in that vast, terrible in-between." |
| 42 |
|
| 43 |
--Emperor Turhan, Centauri Republic |
Archives