Date: Tue, 31 Aug 2010 05:30:06 GMT From: David Xu <davidxu@freebsd.org> To: freebsd-bugs@FreeBSD.org Subject: Re: kern/150138: [patch] signal sent to stopped, traced process not immediately handled on continue Message-ID: <201008310530.o7V5U65M007254@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
The following reply was made to PR kern/150138; it has been noted by GNATS.
From: David Xu <davidxu@freebsd.org>
To: Dan McNulty <dkmcnulty@gmail.com>
Cc: FreeBSD-gnats-submit@freebsd.org
Subject: Re: kern/150138: [patch] signal sent to stopped, traced process not
immediately handled on continue
Date: Tue, 31 Aug 2010 13:27:01 +0000
Dan McNulty wrote:
> On Tue, Aug 31, 2010 at 6:09 AM, David Xu <davidxu@freebsd.org> wrote:
>> User & wrote:
>>
>>>> Fix:
>>> The attached patch ( to CURRENT ) takes a stab at fixing the problem. The
>>> patch
>>> adds code so that even if the process is being traced, a signal sent to a
>>> interruptibly sleeping thread wakes up the thread so it will handle the
>>> signal
>>> when the process leaves the stopped state. The added code was copied from
>>> the
>>> code that handled stopped, non-traced processes later in the same
>>> function.
>> No patch ?
>>
>
> Sorry, in trying to figure out send-pr, didn't actually attach the
> patch and other files.
>
> They should be attached now.
>
I also propose a new patch, it seems if the process is being traced,
we should hand off job control to debugger, all job control caused
by STOP/CONT signals should be skipped, but normal signal should
remove thread from sleep queue.
Index: kern_sig.c
===================================================================
--- kern_sig.c (版本 211859)
+++ kern_sig.c (工作副本)
@@ -2139,20 +2139,9 @@
* We try do the per-process part here.
*/
if (P_SHOULDSTOP(p)) {
- /*
- * The process is in stopped mode. All the threads should be
- * either winding down or already on the suspended queue.
- */
- if (p->p_flag & P_TRACED) {
- /*
- * The traced process is already stopped,
- * so no further action is necessary.
- * No signal can restart us.
- */
- goto out;
- }
-
if (sig == SIGKILL) {
+ if (p->p_flag & P_TRACED)
+ goto out;
/*
* SIGKILL sets process running.
* It will die elsewhere.
@@ -2163,6 +2152,8 @@
}
if (prop & SA_CONT) {
+ if (p->p_flag & P_TRACED)
+ goto out;
/*
* If SIGCONT is default (or ignored), we continue the
* process but don't leave the signal in sigqueue as
@@ -2207,6 +2198,8 @@
}
if (prop & SA_STOP) {
+ if (p->p_flag & P_TRACED)
+ goto out;
/*
* Already stopped, don't need to stop again
* (If we did the shell could get confused).
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201008310530.o7V5U65M007254>