Hi, I get failures with elf x86 test tst-cpu-features-cpuinfo. My configuration is the following: - glibc sources are from commit 21181d1c7b181c4bb71e587c7944e100d923b393 - I use an x86_64 machine - I build and run tests for glibc in a --privileged docker container (not sure this has any influence). - My kernel is from RHEL 7: Linux 9f1b028ef771 3.10.0-1127.13.1.el7.x86_64 #1 SMP Fri Jun 12 14:34:17 EDT 2020 x86_64 GNU/Linux - This machine is a VM. The failure I get is this one: Checking HAS_CPU_FEATURE (ERMS): HAS_CPU_FEATURE (ERMS): 1 cpuinfo (erms): 0 *** failure *** I am pasting /proc/cpuinfo for the first processor: processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 63 model name : Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz stepping : 0 microcode : 0x2000065 cpu MHz : 2400.000 cache size : 28160 KB physical id : 0 siblings : 1 core id : 0 cpu cores : 1 apicid : 0 initial apicid : 0 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon nopl xt opology tsc_reliable nonstop_tsc eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm inv pcid_single ssbd ibrs ibpb stibp fsgsbase tsc_adjust bmi1 avx2 smep bmi2 invpcid xsaveopt arat md_clear spec_ctrl intel_stibp flush_l1d arch_capabilities bogomips : 4800.00 clflush size : 64 cache_alignment : 64 address sizes : 42 bits physical, 48 bits virtual power management: Is this something expected ? Cheers, Romain
I am still seeing this test failing with latest gcc trunk and glibc trunk.
Hi, I have now tried with the released glibc 2.33 and still see this issue. H.J., it looks like similar issues were posted on the list here https://sourceware.org/pipermail/libc-alpha/2021-February/122227.html, so is there anything which should be done to fix the ERMS case as well ? Cheers, Romain