Thread (53 messages) 53 messages, 9 authors, 2012-02-29

Re: [PATCH v11 08/12] signal, x86: add SIGSYS info and make it synchronous.

From: Will Drewry <wad@chromium.org>
Date: 2012-02-27 20:24:14
Also in: linux-arch, lkml

On Mon, Feb 27, 2012 at 11:22 AM, Oleg Nesterov [off-list ref] wrote:
On 02/24, Will Drewry wrote:
quoted
To ensure that SIGSYS delivery occurs on return from the triggering
system call, SIGSYS is added to the SYNCHRONOUS_MASK macro.
Hmm. Can't understand... please help.
quoted
 #define SYNCHRONOUS_MASK \
      (sigmask(SIGSEGV) | sigmask(SIGBUS) | sigmask(SIGILL) | \
-      sigmask(SIGTRAP) | sigmask(SIGFPE))
+      sigmask(SIGTRAP) | sigmask(SIGFPE) | sigmask(SIGSYS))
Why?

SYNCHRONOUS_MASK just tells dequeue_signal() "pick them first".
This is needed to make sure that the handler for, say SIGSEGV,
can use ucontext->ip as a faulting addr.
I think that Roland covered this.  (Since the syscall_rollback was
called it's nice to let our handler get first go.)
But seccomp adds info->si_call_addr, this looks unneeded.
True enough.  I can drop it.  It'd only be useful if the SIGSYS wasn't
being forced and the signal was being handled without ucontext_t
access.

thanks!
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help