| commit | bef653d92e4e1c47c60f5c020ff6de69dba83478 | |
| author | Peter Maydell <peter.maydell@linaro.org> | |
| Fri, 27 May 2016 14:51:57 +0000 (27 15:51 +0100) | ||
| committer | Riku Voipio <riku.voipio@linaro.org> | |
| Tue, 7 Jun 2016 13:39:08 +0000 (7 16:39 +0300) | ||
| tree | cbdcf2af2e17339f44c303f7cb91541c152b8ca5 | treesnapshot (tar.gz zip) |
| parent | a0995886e23323258d0612c71f8b0416a02806f2 | commitdiff |
linux-user: Use safe_syscall for kill, tkill and tgkill syscalls
Use the safe_syscall wrapper for the kill, tkill and tgkill syscalls.
Without this, if a thread sent a SIGKILL to itself it could kill the
thread before we had a chance to process a signal that arrived just
before the SIGKILL, and that signal would get lost.
We drop all the ifdeffery for tkill and tgkill, because every guest
architecture we support implements them, and they've been in Linux
since 2003 so we can assume the host headers define the __NR_tkill
and __NR_tgkill constants.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Use the safe_syscall wrapper for the kill, tkill and tgkill syscalls.
Without this, if a thread sent a SIGKILL to itself it could kill the
thread before we had a chance to process a signal that arrived just
before the SIGKILL, and that signal would get lost.
We drop all the ifdeffery for tkill and tgkill, because every guest
architecture we support implements them, and they've been in Linux
since 2003 so we can assume the host headers define the __NR_tkill
and __NR_tgkill constants.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
| linux-user/syscall.c | diffblobblamehistory |