Building cdds failed (Maven & Java version) #6
Comments
|
My apologies for having misinterpreted some of the stack trace when you reported this against the old atolab/cdds repository. I am not so well-versed in Java/Maven issues, but I do know that I have no issues building it using the Oracle JDK, and that the README points there as well. Is it possible that this is an OpenJDK-specific issue? |
|
Same issue : Java : Java check : |
|
seem to be a issue in Debian Maven... Connected to https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=901881 |
Theosakamg
changed the title
|
And after upgrade to java >= 1.9 |
Merged
|
Thank you for figuring all this out! |
eboasson
added a commit
to eboasson/cyclonedds
that referenced
this issue
Dec 7, 2023
This fixes a global buffer overflow (underflow?) reported by ASAN as
follows:
==93535==ERROR: AddressSanitizer: global-buffer-overflow on address \
0x000100144e3f at pc 0x000100084424 bp 0x00016fdfd380 sp 0x00016fdfd378
READ of size 1 at 0x000100144e3f thread T0
#0 0x100084420 in put_a_line main.c:975
#1 0x100082974 in putout main.c:895
eclipse-cyclonedds#2 0x100080ab4 in mcpp_main main.c:753
eclipse-cyclonedds#3 0x10007d3fc in mcpp_lib_main main.c:430
eclipse-cyclonedds#4 0x10000a230 in idlc_parse idlc.c:375
eclipse-cyclonedds#5 0x100006f1c in main idlc.c:772
eclipse-cyclonedds#6 0x181d51054 (<unknown module>)
0x000100144e3f is located 1 bytes before global variable 'output' \
defined in '/Users/erik/C/cdds/src/tools/idlpp/src/main.c' (0x100144e40) of size 262144
0x000100144e3f is located 23 bytes after global variable 'src_col' \
defined in '/Users/erik/C/cdds/src/tools/idlpp/src/main.c' (0x100144e20) of size 8
Triggered by the following input:
hexdump -C ~/Downloads/asan-global-bof.idl
00000000 23 fc ff 0a 23 64 65 66 69 6e 65 20 69 0a 23 64 |#...#define i.#d|
00000010 65 66 69 6e 65 20 4a 00 0a 20 4a 00 0a 23 64 69 |efine J.. J..#di|
00000020 44 66 7f f8 ff ff 66 00 32 44 0a 23 64 69 69 6e |Df....f.2D.#diin|
00000030 65 20 4a 32 32 32 44 12 64 27 ff 7f ff 23 3e 69 |e J222D.d'...#>i|
00000040 0a |.|
00000041
(I gave up trying to reduce it further.)
The problem is caused by scanning for the last non-whitespace character
just before outputting a line without accounting for the case where the
line consists solely of whitespace.
Credits for finding the bug:
- Carlos Andres Ramirez (https://carlos.engineer)
- Goktug Serez (https://github.com/g0ku704)
- Xin Huang (https://github.com/xinhuang)
Signed-off-by: Erik Boasson <eb@ilities.com>
eboasson
added a commit
to eboasson/cyclonedds
that referenced
this issue
Dec 8, 2023
This fixes a global buffer overflow (underflow?) reported by ASAN as
follows:
```
ERROR: AddressSanitizer: global-buffer-overflow on address \
0x000100144e3f at pc 0x000100084424 bp 0x00016fdfd380 sp 0x00016fdfd378
READ of size 1 at 0x000100144e3f thread T0
#0 0x100084420 in put_a_line main.c:975
#1 0x100082974 in putout main.c:895
eclipse-cyclonedds#2 0x100080ab4 in mcpp_main main.c:753
eclipse-cyclonedds#3 0x10007d3fc in mcpp_lib_main main.c:430
eclipse-cyclonedds#4 0x10000a230 in idlc_parse idlc.c:375
eclipse-cyclonedds#5 0x100006f1c in main idlc.c:772
eclipse-cyclonedds#6 0x181d51054 (<unknown module>)
0x000100144e3f is located 1 bytes before global variable 'output' \
defined in 'src/tools/idlpp/src/main.c' (0x100144e40) of size 262144
0x000100144e3f is located 23 bytes after global variable 'src_col' \
defined in 'src/tools/idlpp/src/main.c' (0x100144e20) of size 8
```
Triggered by the following input (`hexdump -C` output):
```
00000000 23 fc ff 0a 23 64 65 66 69 6e 65 20 69 0a 23 64 |#...#define i.#d|
00000010 65 66 69 6e 65 20 4a 00 0a 20 4a 00 0a 23 64 69 |efine J.. J..#di|
00000020 44 66 7f f8 ff ff 66 00 32 44 0a 23 64 69 69 6e |Df....f.2D.#diin|
00000030 65 20 4a 32 32 32 44 12 64 27 ff 7f ff 23 3e 69 |e J222D.d'...#>i|
00000040 0a |.|
00000041
```
(I tried to reduce it further, but gave up.)
The problem is caused by scanning for the last non-whitespace character
just before outputting a line without accounting for the case where the
line consists solely of whitespace.
Credits for finding the bug:
- Carlos Andres Ramirez (https://carlos.engineer)
- Goktug Serez (https://github.com/g0ku704)
- Xin Huang (https://github.com/xinhuang)
Signed-off-by: Erik Boasson <eb@ilities.com>
eboasson
added a commit
to eboasson/cyclonedds
that referenced
this issue
Dec 8, 2023
This fixes a global buffer overflow (underflow?) reported by ASAN as
follows:
ERROR: AddressSanitizer: global-buffer-overflow on address \
0x000100144e3f at pc 0x000100084424 bp 0x00016fdfd380 sp 0x00016fdfd378
READ of size 1 at 0x000100144e3f thread T0
#0 0x100084420 in put_a_line main.c:975
#1 0x100082974 in putout main.c:895
eclipse-cyclonedds#2 0x100080ab4 in mcpp_main main.c:753
eclipse-cyclonedds#3 0x10007d3fc in mcpp_lib_main main.c:430
eclipse-cyclonedds#4 0x10000a230 in idlc_parse idlc.c:375
eclipse-cyclonedds#5 0x100006f1c in main idlc.c:772
eclipse-cyclonedds#6 0x181d51054 (<unknown module>)
0x000100144e3f is located 1 bytes before global variable 'output' \
defined in 'src/tools/idlpp/src/main.c' (0x100144e40) of size 262144
0x000100144e3f is located 23 bytes after global variable 'src_col' \
defined in 'src/tools/idlpp/src/main.c' (0x100144e20) of size 8
Triggered by the following input ("hexdump -C" output):
00000000 23 fc ff 0a 23 64 65 66 69 6e 65 20 69 0a 23 64 |#...#define i.#d|
00000010 65 66 69 6e 65 20 4a 00 0a 20 4a 00 0a 23 64 69 |efine J.. J..#di|
00000020 44 66 7f f8 ff ff 66 00 32 44 0a 23 64 69 69 6e |Df....f.2D.#diin|
00000030 65 20 4a 32 32 32 44 12 64 27 ff 7f ff 23 3e 69 |e J222D.d'...#>i|
00000040 0a |.|
00000041
(I tried to reduce it further, but gave up.)
The problem is caused by scanning for the last non-whitespace character
just before outputting a line without accounting for the case where the
line consists solely of whitespace.
Credits for finding the bug:
- Carlos Andres Ramirez (https://carlos.engineer)
- Goktug Serez (https://github.com/g0ku704)
- Xin Huang (https://github.com/xinhuang)
Signed-off-by: Erik Boasson <eb@ilities.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi,
I tried compiling the cdds.
Context :
Debian Buster GNU/linux
GCC : Debian 7.3.0-21
cmake : version 3.11.2
JAVA_HOME="/usr/lib/jvm/java-8-openjdk-amd64"
Java :
maven :
but on make
The text was updated successfully, but these errors were encountered: