[Bug runtime/30415] conflicting types for ‘kallsyms_on_each_symbol’

[mcermak at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=30415

--- Comment #7 from Martin Cermak <mcermak at redhat dot com> ---
Created attachment 14905
  --> https://sourceware.org/bugzilla/attachment.cgi?id=14905&action=edit
possible patch

I'm wondering if something as simple as the attached patch would work.  My
impression is that with this, either stapkp_symbol_callback() runs as a
callback for a module symbol, and then module is known.  Or it runs as a
callback for a kernel symbol, the module isn't known, but even in this case the
if-else branching in the function should proceed meaningfully (?).

However, when testing this I'm getting following warnings when trying to
compile alias_suffixes.stp, and I'm not sure if it's related and what to do
about it (if anything at all):

39 x86_64 # stap --poison-cache -p4
/usr/local/share/systemtap/examples/general/alias_suffixes.stp
/tmp/stapBnU41h/stap_12c1257a7321a3bb17fbe980edd13be4_84218.o: warning:
objtool: _stp_vsprint_memory+0x1e9: call to __get_user_nocheck_1() with UACCESS
enabled
/tmp/stapBnU41h/stap_12c1257a7321a3bb17fbe980edd13be4_84218.o: warning:
objtool: .altinstr_replacement+0x1d: recursive UACCESS enable
/tmp/stapBnU41h/stap_12c1257a7321a3bb17fbe980edd13be4_84218.o: warning:
objtool: probe_6413+0x3a7: call to __get_user_nocheck_1() with UACCESS enabled
/root/.systemtap/cache/12/stap_12c1257a7321a3bb17fbe980edd13be4_84218.ko
39 x86_64 #

-- 
You are receiving this mail because:
You are the assignee for the bug.