Alfred Perlstein 426da3bcfb SMP Lock struct file, filedesc and the global file list.
Seigo Tanimura (tanimura) posted the initial delta.

I've polished it quite a bit reducing the need for locking and
adapting it for KSE.

Locks:

1 mutex in each filedesc
   protects all the fields.
   protects "struct file" initialization, while a struct file
     is being changed from &badfileops -> &pipeops or something
     the filedesc should be locked.

1 mutex in each struct file
   protects the refcount fields.
   doesn't protect anything else.
   the flags used for garbage collection have been moved to
     f_gcflag which was the FILLER short, this doesn't need
     locking because the garbage collection is a single threaded
     container.
  could likely be made to use a pool mutex.

1 sx lock for the global filelist.

struct file *	fhold(struct file *fp);
        /* increments reference count on a file */

struct file *	fhold_locked(struct file *fp);
        /* like fhold but expects file to locked */

struct file *	ffind_hold(struct thread *, int fd);
        /* finds the struct file in thread, adds one reference and
                returns it unlocked */

struct file *	ffind_lock(struct thread *, int fd);
        /* ffind_hold, but returns file locked */

I still have to smp-safe the fget cruft, I'll get to that asap.
2002-01-13 11:58:06 +00:00
..
2001-09-20 05:15:04 +00:00

$NetBSD: README.mach-traps,v 1.2 1999/03/23 09:19:25 itohy Exp $
$FreeBSD$

Some Alpha AXP OSF/1 binaries directly use the facilities provided by
the Mach kernel that is the basis for OSF/1.  These include (but are
surely not limited to) 'dd', 'ps', and 'w'.

Invariably, the symptom that these binaries display is that they crash
with an "unimplemented system call" trap (SIGSYS signal) for a syscall
that has a negative number.  In general, binaries that use the Mach
syscalls appear to invoke task_self() as their first syscall.

The name, number, and number of arguments for each Mach syscall is
given below; this information was gleaned by looking through the OSF/1
libmach.a's object files with dbx, then double-checked against the
contents of OSF/1's <mach/syscall_sw.h>.

These calls would be very difficult to implement properly in the
OSF/1 emulation code; by its very nature, NetBSD is not Mach, and we
don't and can't provide the underlying facilities that it does.

-- cgd

trap name			number	nargs	notes
---- ----			------	-----	-----
task_self			-10	0
thread_reply			-11	0
task_notify			-12	0
thread_self			-13	0
msg_send_old			-14	3
msg_receive_old			-15	3
msg_rpc_old			-16	5
msg_send_trap			-20	4
msg_receive_trap		-21	5
msg_rpc_trap			-22	6
lw_wire				-30	3
lw_unwire			-31	1
nxm_task_init			-33	2
nxm_sched_thread		-34	1
nxm_idle			-35	1
nxm_wakeup_idle			-36	1
nxm_set_pthid			-37	2
nxm_thread_kill			-38	2
nxm_thread_block		-39	1
nxm_thread_wakeup		-40	1
inode_swap_preference		-40	3	old call?
init_process			-41	0
map_fd				-43	5
nxm_resched			-44	2
htg_unix_syscall		-52	3
host_self			-55	1
host_priv_self			-56	1
swtch_pri			-59	1
swtch				-60	0
thread_switch			-61	3
semop_fast			-62	4
mach_sctimes_0			-70	0	only if MACH_SCTIMES defined
mach_sctimes_1			-71	1	only if MACH_SCTIMES defined
mach_sctimes_2			-72	2	only if MACH_SCTIMES defined
mach_sctimes_3			-73	3	only if MACH_SCTIMES defined
mach_sctimes_4			-74	4	only if MACH_SCTIMES defined
mach_sctimes_5			-75	5	only if MACH_SCTIMES defined
mach_sctimes_6			-76	6	only if MACH_SCTIMES defined
mach_sctimes_7			-77	0	only if MACH_SCTIMES defined
mach_sctimes_8			-78	6	only if MACH_SCTIMES defined
mach_sctimes_9			-79	1	only if MACH_SCTIMES defined
mach_sctimes_10			-80	2	only if MACH_SCTIMES defined
mach_sctimes_11			-81	2	only if MACH_SCTIMES defined
mach_sctimes_port_alloc_dealloc	-82	1	only if MACH_SCTIMES defined