Bug 695736 – sys-libs/ncurses-6.1_p20190609 - x86_64-pc-linux-gnu-gcc ... -masm=intel ... ../objects/lib_longname.o: Assembler messages: {standard input}:113: Error: invalid use of register

• Status: UNCONFIRMED
• Product: Gentoo Linux

[Description 7_9j~HTz 2019-09-27 15:22:41 UTC]

Created attachment 591168 [details]
build.log

log enclosed

[Comment 1 Ben Kohler 2019-09-27 15:38:23 UTC]

Please also attach emerge --info output, and config.log

[Comment 2 7_9j~HTz 2019-09-27 15:46:08 UTC]

Created attachment 591170 [details]
emerge info

[Comment 3 Ben Kohler 2019-09-27 15:47:45 UTC]

Can you attach make.conf as well? The fact that it's configuring in a "cross" dir rather than "ncurses-6.1-abi_x86_64.amd64" makes me suspect some environment variable pollution.

[Comment 4 7_9j~HTz 2019-09-27 15:50:38 UTC]

Created attachment 591172 [details]
make.conf

[Comment 5 7_9j~HTz 2019-09-27 15:56:43 UTC]

Created attachment 591174 [details]
config.log

[Comment 6 Ben Kohler 2019-09-27 15:57:36 UTC]

x86_64-pc-linux-gnu-gcc: error: unrecognized command line option '--malign-data=abi'; did you mean '-malign-data=abi'?

[Comment 7 7_9j~HTz 2019-09-27 17:07:57 UTC]

I was aware of it and rectified it and ran again emerge --ask --verbose --update --deep --newuse @world

Which should have sorted it but it did not. Just cleaned all drives and restart fresh and the issue is still present however.

[Comment 8 7_9j~HTz 2019-09-27 17:13:13 UTC]

Created attachment 591176 [details]
build.log

[Comment 9 7_9j~HTz 2019-09-27 17:18:28 UTC]

Created attachment 591178 [details]
build environment

[Comment 10 7_9j~HTz 2019-09-27 19:12:57 UTC]

emerge -vauDN @world appears to be rebuilding packages only when the user changes the USE_FLAG but not the COMMON_FLAGS...

That learned then used emerge -ve for rebuilding...
____

...and selectively removed COMMON_FLAGS. It would that -masm=intel is causing the issue, which is curious though since other packages compiled just fine with the flag in place. Having removed -masm=intel the package then compiled.


Perhaps because the CPU being virtualized 

Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
Address sizes:       40 bits physical, 48 bits virtual
CPU(s):              10
On-line CPU(s) list: 0-9
Thread(s) per core:  1
Core(s) per socket:  1
Socket(s):           10
Vendor ID:           GenuineIntel
CPU family:          6
Model:               13
Model name:          QEMU Virtual CPU version 1.5.3
Stepping:            3
CPU MHz:             2933.408
BogoMIPS:            5866.81
Hypervisor vendor:   KVM
Virtualization type: full
L1d cache:           32K
L1i cache:           32K
L2 cache:            4096K
Flags:               fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pse36 clflush mmx fxsr sse sse2 syscall nx lm rep_good nopl cpuid tsc_known_freq pni cx16 hypervisor lahf_lm pti

[Comment 11 Jeroen Roovers (RETIRED) 2019-09-28 06:12:49 UTC]

Comment on attachment 591174 [details]
config.log

So this config.log is obsolete because it's about the incorrect `--malign-data=abi`.