Re: PR/52117 CVS commit: src/tests/kernel

["Kamil Rytarowski" <n54%gmx.com@localhost>]

The following reply was made to PR kern/52117; it has been noted by GNATS.

From: "Kamil Rytarowski" <n54%gmx.com@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: PR/52117 CVS commit: src/tests/kernel
Date: Tue, 28 Mar 2017 15:57:57 +0200

 > Sent: Tuesday, March 28, 2017 at 3:40 PM
 > From: "Martin Husemann" <martin%duskware.de@localhost>
 > To: kern-bug-people%netbsd.org@localhost, gnats-admin%netbsd.org@localhost, netbsd-bugs%netbsd.org@localhost, n54%gmx.com@localhost
 > Subject: Re: PR/52117 CVS commit: src/tests/kernel
 >
 > The following reply was made to PR kern/52117; it has been noted by GNATS.
 > 
 > From: Martin Husemann <martin%duskware.de@localhost>
 > To: gnats-bugs%NetBSD.org@localhost
 > Cc: 
 > Subject: Re: PR/52117 CVS commit: src/tests/kernel
 > Date: Tue, 28 Mar 2017 15:37:35 +0200
 > 
 >  On Tue, Mar 28, 2017 at 01:20:01PM +0000, Kamil Rytarowski wrote:
 >  >  Add message similar to:
 >  >      We expect two SIGTRAP events, for child 22199 (TRAP_CHLD,
 >  >      pe_report_event=PTRACE_FORK, state.pe_other_pid=child2) and for child2
 >  >      (TRAP_CHLD, pe_report_event=PTRACE_FORK, state.pe_other_pid=child)
 >  
 >  Where is the kernel code hiding that causes this for the forkee?
 >  
 >  I see the code for the process calling fork() at the end of
 >  kern_fork.c, but the "child2" process does not run that code, and I
 >  don't understand how it works on amd64. nxr does not find other kernel
 >  referencs for TRAP_CHLD (and this one works, the forker properly gets
 >  signaled).
 >  
 >  Martin
 >  
 > 
 
 It's a bug in the kernel. It sits hardcoded in x86 specific code:
 
 void
 child_return(void *arg)
 {
 	struct lwp *l = arg;
 	struct trapframe *tf = l->l_md.md_regs;
 	struct proc *p = l->l_proc;
 	if (p->p_slflag & PSL_TRACED) {
 		ksiginfo_t ksi;
 		mutex_enter(proc_lock);
                 KSI_INIT_EMPTY(&ksi);
                 ksi.ksi_signo = SIGTRAP;
                 ksi.ksi_lid = l->l_lid;
                 kpsignal(p, &ksi, NULL);
 		mutex_exit(proc_lock);
 	}
 	X86_TF_RAX(tf) = 0;
 	X86_TF_RFLAGS(tf) &= ~PSL_C;
 	userret(l);
 	ktrsysret(SYS_fork, 0, 0);
 }
 
 I've verified that this fires SIGTRAP for fork(2).