FTBFS with -O3 on ppc64el
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
w1retap (Ubuntu) |
New
|
Undecided
|
Unassigned |
Bug Description
w1retap FTBFS with -O3 on ppc64el: https:/
I've uploaded a version that uses -O2 instead, but we should look into why there's a failure with -O3 so that we can avoid that workaround.
Relevant log snippet:
gcc -DHAVE_CONFIG_H -I. -I../.. -Wall -D_GNU_SOURCE=1 -I . -pthread -I/usr/
gcc -DHAVE_CONFIG_H -I. -I../.. -Wall -D_GNU_SOURCE=1 -I . -pthread -I/usr/
gcc -DHAVE_CONFIG_H -I. -I../.. -Wall -D_GNU_SOURCE=1 -I . -pthread -I/usr/
owprgm.c: In function ‘setJobData’:
owprgm.c:816:28: error: writing 1 byte into a region of size 0 [-Werror=
816 | job.Page[
| ~~~~~~~
In file included from owprgm.c:33:
owfile.h:115:17: note: at offset [29, 30] to object ‘data’ with size 29 declared here
115 | uchar data[29]; // data to write
| ^~~~
owprgm.c:816:28: error: writing 1 byte into a region of size 0 [-Werror=
816 | job.Page[
| ~~~~~~~
In file included from owprgm.c:33:
owfile.h:115:17: note: at offset 30 to object ‘data’ with size 29 declared here
115 | uchar data[29]; // data to write
| ^~~~
owfile.c: In function ‘owOpenFile’:
owfile.c:1271:16: note: the layout of aggregates containing vectors with 1-byte alignment has changed in GCC 5
1271 | Han[hn].Ext = flname->Ext & 0x7F;
| ~~~~~~~
gcc -DHAVE_CONFIG_H -I. -I../.. -Wall -D_GNU_SOURCE=1 -I . -pthread -I/usr/
cc1: all warnings being treated as errors
make[5]: *** [Makefile:926: owprgm.o] Error 1