3 * Bill Paul <wpaul@windriver.com>. All rights reserved.
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
8 * 1. Redistributions of source code must retain the above copyright
9 * notice, this list of conditions and the following disclaimer.
10 * 2. Redistributions in binary form must reproduce the above copyright
11 * notice, this list of conditions and the following disclaimer in the
12 * documentation and/or other materials provided with the distribution.
13 * 3. All advertising materials mentioning features or use of this software
14 * must display the following acknowledgement:
15 * This product includes software developed by Bill Paul.
16 * 4. Neither the name of the author nor the names of any co-contributors
17 * may be used to endorse or promote products derived from this software
18 * without specific prior written permission.
20 * THIS SOFTWARE IS PROVIDED BY Bill Paul AND CONTRIBUTORS ``AS IS'' AND
21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 * ARE DISCLAIMED. IN NO EVENT SHALL Bill Paul OR THE VOICES IN HIS HEAD
24 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
25 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
26 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
27 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
28 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
29 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
30 * THE POSSIBILITY OF SUCH DAMAGE.
33 #include <sys/cdefs.h>
34 __FBSDID("$FreeBSD$");
36 #include <sys/param.h>
37 #include <sys/types.h>
38 #include <sys/errno.h>
40 #include <sys/callout.h>
41 #include <sys/kernel.h>
43 #include <sys/mutex.h>
45 #include <sys/sched.h>
46 #include <sys/module.h>
48 #include <sys/systm.h>
49 #include <machine/bus.h>
54 #include <compat/ndis/pe_var.h>
55 #include <compat/ndis/resource_var.h>
56 #include <compat/ndis/cfg_var.h>
57 #include <compat/ndis/ntoskrnl_var.h>
58 #include <compat/ndis/hal_var.h>
60 static void KeStallExecutionProcessor(uint32_t);
61 static void WRITE_PORT_BUFFER_ULONG(uint32_t *,
62 uint32_t *, uint32_t);
63 static void WRITE_PORT_BUFFER_USHORT(uint16_t *,
64 uint16_t *, uint32_t);
65 static void WRITE_PORT_BUFFER_UCHAR(uint8_t *,
67 static void WRITE_PORT_ULONG(uint32_t *, uint32_t);
68 static void WRITE_PORT_USHORT(uint16_t *, uint16_t);
69 static void WRITE_PORT_UCHAR(uint8_t *, uint8_t);
70 static uint32_t READ_PORT_ULONG(uint32_t *);
71 static uint16_t READ_PORT_USHORT(uint16_t *);
72 static uint8_t READ_PORT_UCHAR(uint8_t *);
73 static void READ_PORT_BUFFER_ULONG(uint32_t *,
74 uint32_t *, uint32_t);
75 static void READ_PORT_BUFFER_USHORT(uint16_t *,
76 uint16_t *, uint32_t);
77 static void READ_PORT_BUFFER_UCHAR(uint8_t *,
79 static uint64_t KeQueryPerformanceCounter(uint64_t *);
80 static void _KeLowerIrql(uint8_t);
81 static uint8_t KeRaiseIrqlToDpcLevel(void);
82 static void dummy (void);
84 #define NDIS_MAXCPUS 64
85 static struct mtx disp_lock[NDIS_MAXCPUS];
90 image_patch_table *patch;
93 for (i = 0; i < NDIS_MAXCPUS; i++)
94 mtx_init(&disp_lock[i], "HAL preemption lock",
95 "HAL lock", MTX_RECURSE|MTX_DEF);
98 while (patch->ipt_func != NULL) {
99 windrv_wrap((funcptr)patch->ipt_func,
100 (funcptr *)&patch->ipt_wrap,
101 patch->ipt_argcnt, patch->ipt_ftype);
112 image_patch_table *patch;
115 for (i = 0; i < NDIS_MAXCPUS; i++)
116 mtx_destroy(&disp_lock[i]);
119 while (patch->ipt_func != NULL) {
120 windrv_unwrap(patch->ipt_wrap);
128 KeStallExecutionProcessor(usecs)
136 WRITE_PORT_ULONG(port, val)
140 bus_space_write_4(NDIS_BUS_SPACE_IO, 0x0, (bus_size_t)port, val);
145 WRITE_PORT_USHORT(port, val)
149 bus_space_write_2(NDIS_BUS_SPACE_IO, 0x0, (bus_size_t)port, val);
154 WRITE_PORT_UCHAR(port, val)
158 bus_space_write_1(NDIS_BUS_SPACE_IO, 0x0, (bus_size_t)port, val);
163 WRITE_PORT_BUFFER_ULONG(port, val, cnt)
168 bus_space_write_multi_4(NDIS_BUS_SPACE_IO, 0x0,
169 (bus_size_t)port, val, cnt);
174 WRITE_PORT_BUFFER_USHORT(port, val, cnt)
179 bus_space_write_multi_2(NDIS_BUS_SPACE_IO, 0x0,
180 (bus_size_t)port, val, cnt);
185 WRITE_PORT_BUFFER_UCHAR(port, val, cnt)
190 bus_space_write_multi_1(NDIS_BUS_SPACE_IO, 0x0,
191 (bus_size_t)port, val, cnt);
196 READ_PORT_USHORT(port)
199 return(bus_space_read_2(NDIS_BUS_SPACE_IO, 0x0, (bus_size_t)port));
203 READ_PORT_ULONG(port)
206 return(bus_space_read_4(NDIS_BUS_SPACE_IO, 0x0, (bus_size_t)port));
210 READ_PORT_UCHAR(port)
213 return(bus_space_read_1(NDIS_BUS_SPACE_IO, 0x0, (bus_size_t)port));
217 READ_PORT_BUFFER_ULONG(port, val, cnt)
222 bus_space_read_multi_4(NDIS_BUS_SPACE_IO, 0x0,
223 (bus_size_t)port, val, cnt);
228 READ_PORT_BUFFER_USHORT(port, val, cnt)
233 bus_space_read_multi_2(NDIS_BUS_SPACE_IO, 0x0,
234 (bus_size_t)port, val, cnt);
239 READ_PORT_BUFFER_UCHAR(port, val, cnt)
244 bus_space_read_multi_1(NDIS_BUS_SPACE_IO, 0x0,
245 (bus_size_t)port, val, cnt);
250 * The spinlock implementation in Windows differs from that of FreeBSD.
251 * The basic operation of spinlocks involves two steps: 1) spin in a
252 * tight loop while trying to acquire a lock, 2) after obtaining the
253 * lock, disable preemption. (Note that on uniprocessor systems, you're
254 * allowed to skip the first step and just lock out pre-emption, since
255 * it's not possible for you to be in contention with another running
256 * thread.) Later, you release the lock then re-enable preemption.
257 * The difference between Windows and FreeBSD lies in how preemption
258 * is disabled. In FreeBSD, it's done using critical_enter(), which on
259 * the x86 arch translates to a cli instruction. This masks off all
260 * interrupts, and effectively stops the scheduler from ever running
261 * so _nothing_ can execute except the current thread. In Windows,
262 * preemption is disabled by raising the processor IRQL to DISPATCH_LEVEL.
263 * This stops other threads from running, but does _not_ block device
264 * interrupts. This means ISRs can still run, and they can make other
265 * threads runable, but those other threads won't be able to execute
266 * until the current thread lowers the IRQL to something less than
269 * There's another commonly used IRQL in Windows, which is APC_LEVEL.
270 * An APC is an Asynchronous Procedure Call, which differs from a DPC
271 * (Defered Procedure Call) in that a DPC is queued up to run in
272 * another thread, while an APC runs in the thread that scheduled
273 * it (similar to a signal handler in a UNIX process). We don't
274 * actually support the notion of APCs in FreeBSD, so for now, the
275 * only IRQLs we're interested in are DISPATCH_LEVEL and PASSIVE_LEVEL.
277 * To simulate DISPATCH_LEVEL, we raise the current thread's priority
278 * to PI_REALTIME, which is the highest we can give it. This should,
279 * if I understand things correctly, prevent anything except for an
280 * interrupt thread from preempting us. PASSIVE_LEVEL is basically
283 * Be aware that, at least on the x86 arch, the Windows spinlock
284 * functions are divided up in peculiar ways. The actual spinlock
285 * functions are KfAcquireSpinLock() and KfReleaseSpinLock(), and
286 * they live in HAL.dll. Meanwhile, KeInitializeSpinLock(),
287 * KefAcquireSpinLockAtDpcLevel() and KefReleaseSpinLockFromDpcLevel()
288 * live in ntoskrnl.exe. Most Windows source code will call
289 * KeAcquireSpinLock() and KeReleaseSpinLock(), but these are just
290 * macros that call KfAcquireSpinLock() and KfReleaseSpinLock().
291 * KefAcquireSpinLockAtDpcLevel() and KefReleaseSpinLockFromDpcLevel()
292 * perform the lock aquisition/release functions without doing the
293 * IRQL manipulation, and are used when one is already running at
294 * DISPATCH_LEVEL. Make sense? Good.
296 * According to the Microsoft documentation, any thread that calls
297 * KeAcquireSpinLock() must be running at IRQL <= DISPATCH_LEVEL. If
298 * we detect someone trying to acquire a spinlock from DEVICE_LEVEL
299 * or HIGH_LEVEL, we panic.
301 * Alternate sleep-lock-based spinlock implementation
302 * --------------------------------------------------
304 * The earlier spinlock implementation was arguably a bit of a hack
305 * and presented several problems. It was basically designed to provide
306 * the functionality of spinlocks without incurring the wrath of
307 * WITNESS. We could get away with using both our spinlock implementation
308 * and FreeBSD sleep locks at the same time, but if WITNESS knew what
309 * we were really up to, it would have spanked us rather severely.
311 * There's another method we can use based entirely on sleep locks.
312 * First, it's important to realize that everything we're locking
313 * resides inside Project Evil itself: any critical data being locked
314 * by drivers belongs to the drivers, and should not be referenced
315 * by any other OS code outside of the NDISulator. The priority-based
316 * locking scheme has system-wide effects, just like real spinlocks
317 * (blocking preemption affects the whole CPU), but since we keep all
318 * our critical data private, we can use a simpler mechanism that
319 * affects only code/threads directly related to Project Evil.
321 * The idea is to create a sleep lock mutex for each CPU in the system.
322 * When a CPU running in the NDISulator wants to acquire a spinlock, it
323 * does the following:
324 * - Pin ourselves to the current CPU
325 * - Acquire the mutex for the current CPU
326 * - Spin on the spinlock variable using atomic test and set, just like
328 * - Once we have the lock, we execute our critical code
330 * To give up the lock, we do:
331 * - Clear the spinlock variable with an atomic op
332 * - Release the per-CPU mutex
333 * - Unpin ourselves from the current CPU.
335 * On a uniprocessor system, this means all threads that access protected
336 * data are serialized through the per-CPU mutex. After one thread
337 * acquires the 'spinlock,' any other thread that uses a spinlock on the
338 * current CPU will block on the per-CPU mutex, which has the same general
339 * effect of blocking pre-emption, but _only_ for those threads that are
340 * running NDISulator code.
342 * On a multiprocessor system, threads on different CPUs all block on
343 * their respective per-CPU mutex, and the atomic test/set operation
344 * on the spinlock variable provides inter-CPU synchronization, though
345 * only for threads running NDISulator code.
347 * This method solves an important problem. In Windows, you're allowed
348 * to do an ExAllocatePoolWithTag() with a spinlock held, provided you
349 * allocate from NonPagedPool. This implies an atomic heap allocation
350 * that will not cause the current thread to sleep. (You can't sleep
351 * while holding real spinlock: clowns will eat you.) But in FreeBSD,
352 * malloc(9) _always_ triggers the acquisition of a sleep lock, even
353 * when you use M_NOWAIT. This is not a problem for FreeBSD native
354 * code: you're allowed to sleep in things like interrupt threads. But
355 * it is a problem with the old priority-based spinlock implementation:
356 * even though we get away with it most of the time, we really can't
357 * do a malloc(9) after doing a KeAcquireSpinLock() or KeRaiseIrql().
358 * With the new implementation, it's not a problem: you're allowed to
359 * acquire more than one sleep lock (as long as you avoid lock order
362 * The one drawback to this approach is that now we have a lot of
363 * contention on one per-CPU mutex within the NDISulator code. Whether
364 * or not this is preferable to the expected Windows spinlock behavior
365 * of blocking pre-emption is debatable.
369 KfAcquireSpinLock(lock)
374 KeRaiseIrql(DISPATCH_LEVEL, &oldirql);
375 KeAcquireSpinLockAtDpcLevel(lock);
381 KfReleaseSpinLock(lock, newirql)
385 KeReleaseSpinLockFromDpcLevel(lock);
386 KeLowerIrql(newirql);
394 if (mtx_owned(&disp_lock[curthread->td_oncpu]))
395 return(DISPATCH_LEVEL);
396 return(PASSIVE_LEVEL);
400 KeQueryPerformanceCounter(freq)
406 return((uint64_t)ticks);
415 oldirql = KeGetCurrentIrql();
417 /* I am so going to hell for this. */
419 panic("IRQL_NOT_LESS_THAN");
421 if (oldirql != DISPATCH_LEVEL) {
423 mtx_lock(&disp_lock[curthread->td_oncpu]);
425 /*printf("RAISE IRQL: %d %d\n", irql, oldirql);*/
434 if (oldirql == DISPATCH_LEVEL)
437 if (KeGetCurrentIrql() != DISPATCH_LEVEL)
438 panic("IRQL_NOT_GREATER_THAN");
440 mtx_unlock(&disp_lock[curthread->td_oncpu]);
447 KeRaiseIrqlToDpcLevel(void)
451 KeRaiseIrql(DISPATCH_LEVEL, &irql);
456 _KeLowerIrql(oldirql)
459 KeLowerIrql(oldirql);
465 printf ("hal dummy called...\n");
469 image_patch_table hal_functbl[] = {
470 IMPORT_SFUNC(KeStallExecutionProcessor, 1),
471 IMPORT_SFUNC(WRITE_PORT_ULONG, 2),
472 IMPORT_SFUNC(WRITE_PORT_USHORT, 2),
473 IMPORT_SFUNC(WRITE_PORT_UCHAR, 2),
474 IMPORT_SFUNC(WRITE_PORT_BUFFER_ULONG, 3),
475 IMPORT_SFUNC(WRITE_PORT_BUFFER_USHORT, 3),
476 IMPORT_SFUNC(WRITE_PORT_BUFFER_UCHAR, 3),
477 IMPORT_SFUNC(READ_PORT_ULONG, 1),
478 IMPORT_SFUNC(READ_PORT_USHORT, 1),
479 IMPORT_SFUNC(READ_PORT_UCHAR, 1),
480 IMPORT_SFUNC(READ_PORT_BUFFER_ULONG, 3),
481 IMPORT_SFUNC(READ_PORT_BUFFER_USHORT, 3),
482 IMPORT_SFUNC(READ_PORT_BUFFER_UCHAR, 3),
483 IMPORT_FFUNC(KfAcquireSpinLock, 1),
484 IMPORT_FFUNC(KfReleaseSpinLock, 1),
485 IMPORT_SFUNC(KeGetCurrentIrql, 0),
486 IMPORT_SFUNC(KeQueryPerformanceCounter, 1),
487 IMPORT_FFUNC(KfLowerIrql, 1),
488 IMPORT_FFUNC(KfRaiseIrql, 1),
489 IMPORT_SFUNC(KeRaiseIrqlToDpcLevel, 0),
491 IMPORT_SFUNC_MAP(KeLowerIrql, _KeLowerIrql, 1),
494 * This last entry is a catch-all for any function we haven't
495 * implemented yet. The PE import list patching routine will
496 * use it for any function that doesn't have an explicit match
500 { NULL, (FUNC)dummy, NULL, 0, WINDRV_WRAP_STDCALL },