* [Bug runtime/27465] ERROR: Couldn't insert module&Invalid module format
2021-02-25 2:13 [Bug runtime/27465] New: ERROR: Couldn't insert module&Invalid module format lyeeer at outlook dot com
@ 2021-03-02 20:14 ` fche at redhat dot com
2022-02-04 18:21 ` v.mayatskih at gmail dot com
` (7 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: fche at redhat dot com @ 2021-03-02 20:14 UTC (permalink / raw)
To: systemtap
https://sourceware.org/bugzilla/show_bug.cgi?id=27465
Frank Ch. Eigler <fche at redhat dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Last reconfirmed| |2021-03-02
Status|UNCONFIRMED |WAITING
Ever confirmed|0 |1
CC| |fche at redhat dot com
--- Comment #1 from Frank Ch. Eigler <fche at redhat dot com> ---
dmesg might give further information about why the kernel rejected loading the
systemtap module. Can you try it again, and report what the kernel says?
Also, increasing the pass-5 verbosity to a higher level may give us more tips.
# stap --vp 00004 ....
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug runtime/27465] ERROR: Couldn't insert module&Invalid module format
2021-02-25 2:13 [Bug runtime/27465] New: ERROR: Couldn't insert module&Invalid module format lyeeer at outlook dot com
2021-03-02 20:14 ` [Bug runtime/27465] " fche at redhat dot com
@ 2022-02-04 18:21 ` v.mayatskih at gmail dot com
2022-02-04 18:23 ` fche at redhat dot com
` (6 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: v.mayatskih at gmail dot com @ 2022-02-04 18:21 UTC (permalink / raw)
To: systemtap
https://sourceware.org/bugzilla/show_bug.cgi?id=27465
Vitaly Mayatskikh <v.mayatskih at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |v.mayatskih at gmail dot com
--- Comment #2 from Vitaly Mayatskikh <v.mayatskih at gmail dot com> ---
Created attachment 13957
--> https://sourceware.org/bugzilla/attachment.cgi?id=13957&action=edit
Module init trace
Looks like it fails to resolve a symbol:
74 insmod(34216): ->find_symbol fsa=0xffffa1dc610c7d10
76 insmod(34216): ->find_exported_symbol_in_section
syms=0xffffffffb2c288c0 owner=0x0 data=0xffffa1dc610c7d10
...
103 insmod(34216): ->cmp_name name=0xffffffffb2fcd7ee
sym=0xffffffffb310b754
104 insmod(34216): <-cmp_name return=0x1
105 insmod(34216): ->cmp_name name=0xffffffffb2fcd7ee
sym=0xffffffffb310b760
106 insmod(34216): <-cmp_name return=0x0
107 insmod(34216): <-find_exported_symbol_in_section return=0x1
108 insmod(34216): <-find_symbol return=0x1
118 insmod(34216): <-check_version return=0x0
122 insmod(34216): <-load_module return=0xfffffffffffffff8
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug runtime/27465] ERROR: Couldn't insert module&Invalid module format
2021-02-25 2:13 [Bug runtime/27465] New: ERROR: Couldn't insert module&Invalid module format lyeeer at outlook dot com
2021-03-02 20:14 ` [Bug runtime/27465] " fche at redhat dot com
2022-02-04 18:21 ` v.mayatskih at gmail dot com
@ 2022-02-04 18:23 ` fche at redhat dot com
2022-02-04 18:53 ` v.mayatskih at gmail dot com
` (5 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: fche at redhat dot com @ 2022-02-04 18:23 UTC (permalink / raw)
To: systemtap
https://sourceware.org/bugzilla/show_bug.cgi?id=27465
--- Comment #3 from Frank Ch. Eigler <fche at redhat dot com> ---
It'd be helpful to see the last relevant lines of the output of dmesg. Module
loading related diagnostics are often printed there, to the kernel log.
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug runtime/27465] ERROR: Couldn't insert module&Invalid module format
2021-02-25 2:13 [Bug runtime/27465] New: ERROR: Couldn't insert module&Invalid module format lyeeer at outlook dot com
` (2 preceding siblings ...)
2022-02-04 18:23 ` fche at redhat dot com
@ 2022-02-04 18:53 ` v.mayatskih at gmail dot com
2022-02-04 19:22 ` v.mayatskih at gmail dot com
` (4 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: v.mayatskih at gmail dot com @ 2022-02-04 18:53 UTC (permalink / raw)
To: systemtap
https://sourceware.org/bugzilla/show_bug.cgi?id=27465
--- Comment #4 from Vitaly Mayatskikh <v.mayatskih at gmail dot com> ---
Nothing in dmesg. After enabling pr_debug for kernel/modules.c I see only that:
[ 4686.786467] finit_module: fd=3, uargs=0000000066cecbb4, flags=0
Something is broken w.r.t. how Ubuntu's kernel is configured:
$ uname -v
#31~20.04.1-Ubuntu SMP Wed Jan 19 14:08:10 UTC 2022
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug runtime/27465] ERROR: Couldn't insert module&Invalid module format
2021-02-25 2:13 [Bug runtime/27465] New: ERROR: Couldn't insert module&Invalid module format lyeeer at outlook dot com
` (3 preceding siblings ...)
2022-02-04 18:53 ` v.mayatskih at gmail dot com
@ 2022-02-04 19:22 ` v.mayatskih at gmail dot com
2022-02-04 19:30 ` v.mayatskih at gmail dot com
` (3 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: v.mayatskih at gmail dot com @ 2022-02-04 19:22 UTC (permalink / raw)
To: systemtap
https://sourceware.org/bugzilla/show_bug.cgi?id=27465
--- Comment #5 from Vitaly Mayatskikh <v.mayatskih at gmail dot com> ---
check_version fails for module_layout:
204 insmod(43808): ->check_version info={.name="stap_30878",
.mod=0xffffa1dc617d1740, .hdr=0xffffa1dc617c1000, .len=142104,
.sechdrs=0xffffa1dc617e3098, .secstrings="", .strtab="", .symoffs=0,
.stroffs=0, .init_typeoffs=0, .core_typeoffs=0, .debug=0x0, .num_debug=0,
.sig_ok=0, .mod_kallsyms_init_off=0, .index={.sym=39, .str=40, .mod=34,
.vers=0, .info=21, .pcpu=32}} symname="module_layout" crc=-1290631832 mod=ERROR
207 insmod(43808): <-check_version return=0x0
This is for non-guru case (does not fail):
311 insmod(43893): ->check_version info={.name="stap_11738",
.mod=0xffffa1dc617d1c00, .hdr=0xffffa1dc617c1000, .len=139264,
.sechdrs=0xffffa1dc617e2540, .secstrings="", .strtab="", .symoffs=0,
.stroffs=0, .init_typeoffs=0, .core_typeoffs=0, .debug=0x0, .num_debug=0,
.sig_ok=0, .mod_kallsyms_init_off=0, .index={.sym=40, .str=41, .mod=35,
.vers=24, .info=21, .pcpu=33}} symname="module_layout" crc=-1290631832
mod=ERROR
319 insmod(43893): <-check_version return=0x1
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug runtime/27465] ERROR: Couldn't insert module&Invalid module format
2021-02-25 2:13 [Bug runtime/27465] New: ERROR: Couldn't insert module&Invalid module format lyeeer at outlook dot com
` (4 preceding siblings ...)
2022-02-04 19:22 ` v.mayatskih at gmail dot com
@ 2022-02-04 19:30 ` v.mayatskih at gmail dot com
2022-02-04 19:40 ` v.mayatskih at gmail dot com
` (2 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: v.mayatskih at gmail dot com @ 2022-02-04 19:30 UTC (permalink / raw)
To: systemtap
https://sourceware.org/bugzilla/show_bug.cgi?id=27465
--- Comment #6 from Vitaly Mayatskikh <v.mayatskih at gmail dot com> ---
check_version fails for module_layout:
204 insmod(43808): ->check_version info={.name="stap_30878",
.mod=0xffffa1dc617d1740, .hdr=0xffffa1dc617c1000, .len=142104,
.sechdrs=0xffffa1dc617e3098, .secstrings="", .strtab="", .symoffs=0,
.stroffs=0, .init_typeoffs=0, .core_typeoffs=0, .debug=0x0, .num_debug=0,
.sig_ok=0, .mod_kallsyms_init_off=0, .index={.sym=39, .str=40, .mod=34,
.vers=0, .info=21, .pcpu=32}} symname="module_layout" crc=-1290631832 mod=ERROR
207 insmod(43808): <-check_version return=0x0
This is for non-guru case (does not fail):
311 insmod(43893): ->check_version info={.name="stap_11738",
.mod=0xffffa1dc617d1c00, .hdr=0xffffa1dc617c1000, .len=139264,
.sechdrs=0xffffa1dc617e2540, .secstrings="", .strtab="", .symoffs=0,
.stroffs=0, .init_typeoffs=0, .core_typeoffs=0, .debug=0x0, .num_debug=0,
.sig_ok=0, .mod_kallsyms_init_off=0, .index={.sym=40, .str=41, .mod=35,
.vers=24, .info=21, .pcpu=33}} symname="module_layout" crc=-1290631832
mod=ERROR
319 insmod(43893): <-check_version return=0x1
____versions symbol is missing in the "broken" module:
$ sudo readelf -a /tmp/stapzTtc4M/stap_11738.ko | grep version
[24] __versions PROGBITS 0000000000000000 0000cc80
27: 0000000000000076 35 OBJECT LOCAL DEFAULT 21
__UNIQUE_ID_srcversion127
29: 0000000000000000 8832 OBJECT LOCAL DEFAULT 24 ____versions
$ sudo readelf -a /tmp/stapykqAs1/stap_30878.ko | grep version
26: 00000000000000a3 35 OBJECT LOCAL DEFAULT 21
__UNIQUE_ID_srcversion127
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug runtime/27465] ERROR: Couldn't insert module&Invalid module format
2021-02-25 2:13 [Bug runtime/27465] New: ERROR: Couldn't insert module&Invalid module format lyeeer at outlook dot com
` (5 preceding siblings ...)
2022-02-04 19:30 ` v.mayatskih at gmail dot com
@ 2022-02-04 19:40 ` v.mayatskih at gmail dot com
2022-02-09 23:57 ` fche at redhat dot com
2024-02-21 13:07 ` fche at redhat dot com
8 siblings, 0 replies; 10+ messages in thread
From: v.mayatskih at gmail dot com @ 2022-02-04 19:40 UTC (permalink / raw)
To: systemtap
https://sourceware.org/bugzilla/show_bug.cgi?id=27465
--- Comment #7 from Vitaly Mayatskikh <v.mayatskih at gmail dot com> ---
check_version fails for module_layout:
204 insmod(43808): ->check_version info={.name="stap_30878",
.mod=0xffffa1dc617d1740, .hdr=0xffffa1dc617c1000, .len=142104,
.sechdrs=0xffffa1dc617e3098, .secstrings="", .strtab="", .symoffs=0,
.stroffs=0, .init_typeoffs=0, .core_typeoffs=0, .debug=0x0, .num_debug=0,
.sig_ok=0, .mod_kallsyms_init_off=0, .index={.sym=39, .str=40, .mod=34,
.vers=0, .info=21, .pcpu=32}} symname="module_layout" crc=-1290631832 mod=ERROR
207 insmod(43808): <-check_version return=0x0
This is for non-guru case (does not fail):
311 insmod(43893): ->check_version info={.name="stap_11738",
.mod=0xffffa1dc617d1c00, .hdr=0xffffa1dc617c1000, .len=139264,
.sechdrs=0xffffa1dc617e2540, .secstrings="", .strtab="", .symoffs=0,
.stroffs=0, .init_typeoffs=0, .core_typeoffs=0, .debug=0x0, .num_debug=0,
.sig_ok=0, .mod_kallsyms_init_off=0, .index={.sym=40, .str=41, .mod=35,
.vers=24, .info=21, .pcpu=33}} symname="module_layout" crc=-1290631832
mod=ERROR
319 insmod(43893): <-check_version return=0x1
____versions symbol is missing in the "broken" module:
$ sudo readelf -a /tmp/stapzTtc4M/stap_11738.ko | grep version
[24] __versions PROGBITS 0000000000000000 0000cc80
27: 0000000000000076 35 OBJECT LOCAL DEFAULT 21
__UNIQUE_ID_srcversion127
29: 0000000000000000 8832 OBJECT LOCAL DEFAULT 24 ____versions
$ sudo readelf -a /tmp/stapykqAs1/stap_30878.ko | grep version
26: 00000000000000a3 35 OBJECT LOCAL DEFAULT 21
__UNIQUE_ID_srcversion127
Adding -B CONFIG_MODVERSIONS=y solved it for me. And this is where it gets
messed up:
buildrun.cxx:
// PR10280: suppress symbol versioning to restrict to exact kernel version
//
// XXX (PR24720): this was reported to cause problems on a PPC machine,
// and I've spotted kernel commits (43e24e82f35) which vary the vermagic
// string for PPC. Still, a standard PPC configuration tested fine, so
// I'm cautiously re-enabling this.
if (s.guru_mode)
make_cmd.push_back("CONFIG_MODVERSIONS=");
// Note: can re-enable from command line with "-B CONFIG_MODVERSIONS=y".
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug runtime/27465] ERROR: Couldn't insert module&Invalid module format
2021-02-25 2:13 [Bug runtime/27465] New: ERROR: Couldn't insert module&Invalid module format lyeeer at outlook dot com
` (6 preceding siblings ...)
2022-02-04 19:40 ` v.mayatskih at gmail dot com
@ 2022-02-09 23:57 ` fche at redhat dot com
2024-02-21 13:07 ` fche at redhat dot com
8 siblings, 0 replies; 10+ messages in thread
From: fche at redhat dot com @ 2022-02-09 23:57 UTC (permalink / raw)
To: systemtap
https://sourceware.org/bugzilla/show_bug.cgi?id=27465
--- Comment #8 from Frank Ch. Eigler <fche at redhat dot com> ---
modversion / module_layout checks tend to fail mainly if the system compiler
used to build the kernel is different enough from the one used to build
systemtap's kernel modules. There are some gcc version-level features that
vary the struct-module and other data types in linux, so modversions machinery
can catch that.
I suggest checking compiler versions too.
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug runtime/27465] ERROR: Couldn't insert module&Invalid module format
2021-02-25 2:13 [Bug runtime/27465] New: ERROR: Couldn't insert module&Invalid module format lyeeer at outlook dot com
` (7 preceding siblings ...)
2022-02-09 23:57 ` fche at redhat dot com
@ 2024-02-21 13:07 ` fche at redhat dot com
8 siblings, 0 replies; 10+ messages in thread
From: fche at redhat dot com @ 2024-02-21 13:07 UTC (permalink / raw)
To: systemtap
https://sourceware.org/bugzilla/show_bug.cgi?id=27465
Frank Ch. Eigler <fche at redhat dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |OBSOLETE
Status|WAITING |RESOLVED
--- Comment #9 from Frank Ch. Eigler <fche at redhat dot com> ---
no recent investigation progress
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 10+ messages in thread