Date: Tue, 7 Jun 2011 21:48:36 +0000 (UTC) From: John Baldwin <jhb@FreeBSD.org> To: cvs-src-old@freebsd.org Subject: cvs commit: src/sys/compat/linprocfs linprocfs.c src/sys/fs/nfsclient nfs_clport.c src/sys/kern kern_descrip.c kern_fork.c kern_proc.c kern_resource.c src/sys/vm vm_meter.c vm_pageout.c Message-ID: <201106072149.p57Ln1tL051537@repoman.freebsd.org>
next in thread | raw e-mail | index | archive | help
jhb 2011-06-07 21:48:36 UTC
FreeBSD src repository
Modified files: (Branch: RELENG_8)
sys/compat/linprocfs linprocfs.c
sys/fs/nfsclient nfs_clport.c
sys/kern kern_descrip.c kern_fork.c kern_proc.c
kern_resource.c
sys/vm vm_meter.c vm_pageout.c
Log:
SVN rev 222839 on 2011-06-07 21:48:36Z by jhb
MFC 219968:
Fix some locking nits with the p_state field of struct proc:
- Hold the proc lock while changing the state from PRS_NEW to PRS_NORMAL
in fork to honor the locking requirements. While here, expand the scope
of the PROC_LOCK() on the new process (p2) to avoid some LORs. Previously
the code was locking the new child process (p2) after it had locked the
parent process (p1). However, when locking two processes, the safe order
is to lock the child first, then the parent.
- Fix various places that were checking p_state against PRS_NEW without
having the process locked to use PROC_LOCK(). Every place was already
locking the process, just after the PRS_NEW check.
- Remove or reduce the use of PROC_SLOCK() for places that were checking
p_state against PRS_NEW. The PROC_LOCK() alone is sufficient for reading
the current state.
Revision Changes Path
1.142.2.8 +0 -2 src/sys/compat/linprocfs/linprocfs.c
1.5.2.14 +2 -2 src/sys/fs/nfsclient/nfs_clport.c
1.357.2.8 +4 -2 src/sys/kern/kern_descrip.c
1.306.2.10 +5 -10 src/sys/kern/kern_fork.c
1.285.2.17 +5 -7 src/sys/kern/kern_proc.c
1.194.2.6 +2 -4 src/sys/kern/kern_resource.c
1.101.2.2 +0 -3 src/sys/vm/vm_meter.c
1.308.2.8 +4 -2 src/sys/vm/vm_pageout.c
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201106072149.p57Ln1tL051537>