I've updated the description and such for:
Bug 213937 - stable/11 -r334246 on BPi-M3 (cortex-a7), amd64, and more, also
head: ktrace misreports bad svc (SIGSYS) number
The primary description is really Comment 2 from Oleksandr Tymoshenko :
> This is not ARM specific, the same bug can be reproduced in amd64 using
> syscall(2) function.
>
> The reason for this that in sys/sys/ktrace.h syscall code is defined as short:
> struct ktr_syscall {
> short ktr_code; /* syscall number */
> short ktr_narg; /* number of arguments */
> /*
> * followed by ktr_narg register_t
> */
> register_t ktr_args[1];
> };
>
> This definitely does not match syscall(2) API where code is defined as
> integer and all architectures define it as u_int in struct syscall_args.
It is that way in every stable/11 and head version of the file
that I have looked at.
===
Mark Millard
marklmi at yahoo.com
( dsl-only.net went
away in early 2018-Mar)
_______________________________________________
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"
