[PATCH v5 0/8] Always use 32-bit time_t for certain syscalls

[Adhemerval Zanella]

On 06/04/2020 11:03, Andreas Schwab wrote:
> On Apr 06 2020, Adhemerval Zanella wrote:
> 
>> On 04/04/2020 10:58, Andreas Schwab wrote:
>>> One of the patches in the series has broken the ABI for all 32-bit
>>> architectures.
>>>
>>> 0bcd0c5100 sysv/alpha: Use generic __timeval32 and helpers
>>> 600f00b747 linux: Use long time_t for wait4/getrusage
>>> 5d24ba82c4 resource: Add a __rusage64 struct
>>> a51e035889 linux: Use long time_t __getitimer/__setitimer
>>> 1c634e677f sysv: Define __KERNEL_OLD_TIMEVAL_MATCHES_TIMEVAL64
>>>
>>
>> Which failure are you seeing?
> 
> [   24s] [   10.164945] sysrq: Changing Loglevel
> [   24s] [   10.164997] sysrq: Loglevel set to 4
> [   24s] Setting up swapspace version 1, size = 1024 MiB (1073676288 bytes)
> [   24s] no label, UUID=c7fc94c6-28a5-4bf9-a465-52f5926b0054
> [   24s] swapon: /dev/vdb: found signature [pagesize=65536, signature=swap]
> [   24s] swapon: /dev/vdb: pagesize=65536, swapsize=1073741824, devsize=1073741824
> [   24s] swapon /dev/vdb
> [   24s] WARNING: udev not running, creating extra device nodes
> [   24s] logging output to //.build.log...
> [   24s] processing recipe /.build-srcdir/glibc.spec ...
> [   24s] /.build/build-vm: line 501:   382 Segmentation fault      $0 "$@"
> [   24s] 
> [   24s]  failed "build glibc.spec" at Sat Apr  4 02:17:45 UTC 2020.
> [   24s] 
> [   24s] [   10.303401] Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b
> [   24s] [   10.303503] CPU: 3 PID: 1 Comm: build Not tainted 5.6.0-1-default #1 openSUSE Tumbleweed (unreleased)
> [   24s] [   10.303611] Call Trace:
> [   24s] [   10.303654] [c000000003753940] [c000000000d79ec0] .dump_stack+0xec/0x15c (unreliable)
> [   24s] [   10.303754] [c0000000037539e0] [c00000000016ff9c] .panic+0x16c/0x420
> [   24s] [   10.303837] [c000000003753a90] [c00000000017825c] .do_exit+0xbcc/0xc60
> [   24s] [   10.303921] [c000000003753b60] [c0000000001783c0] .do_group_exit+0x60/0xd0
> [   24s] [   10.304007] [c000000003753bf0] [c00000000018c584] .get_signal+0x1e4/0xbf0
> [   24s] [   10.304093] [c000000003753d20] [c000000000023e0c] .do_notify_resume+0x1cc/0x460
> [   24s] [   10.304191] [c000000003753e20] [c00000000000e3c4] .ret_from_except_lite+0x70/0x74

I am not familiar with build-vm command, which would be a easy
way to try reproduce it? Does it issue {get,set}itimer by
chance?