OpenCores
URL https://opencores.org/ocsvn/openrisc/openrisc/trunk

Subversion Repositories openrisc

[/] [openrisc/] [trunk/] [rtos/] [ecos-2.0/] [doc/] [html/] [ref/] [hal-smp-support.html] - Blame information for rev 615

Go to most recent revision | Details | Compare with Previous | View Log

Line No. Rev Author Line
1 28 unneback 
<!-- Copyright (C) 2003 Red Hat, Inc.                                -->
2 
<!-- This material may be distributed only subject to the terms      -->
3 
<!-- and conditions set forth in the Open Publication License, v1.0  -->
4 
<!-- or later (the latest version is presently available at          -->
5 
<!-- http://www.opencontent.org/openpub/).                           -->
6 
<!-- Distribution of the work or derivative of the work in any       -->
7 
<!-- standard (paper) book form is prohibited unless prior           -->
8 
<!-- permission is obtained from the copyright holder.               -->
9 
<HTML
10 
><HEAD
11 
><TITLE
12 
>SMP Support</TITLE
13 
><meta name="MSSmartTagsPreventParsing" content="TRUE">
14 
<META
15 
NAME="GENERATOR"
16 
CONTENT="Modular DocBook HTML Stylesheet Version 1.76b+
17 
"><LINK
18 
REL="HOME"
19 
TITLE="eCos Reference Manual"
20 
HREF="ecos-ref.html"><LINK
21 
REL="UP"
22 
TITLE="HAL Interfaces"
23 
HREF="hal-interfaces.html"><LINK
24 
REL="PREVIOUS"
25 
TITLE="Diagnostic Support"
26 
HREF="hal-diagnostic-support.html"><LINK
27 
REL="NEXT"
28 
TITLE="Exception Handling"
29 
HREF="hal-exception-handling.html"></HEAD
30 
><BODY
31 
CLASS="SECTION"
32 
BGCOLOR="#FFFFFF"
33 
TEXT="#000000"
34 
LINK="#0000FF"
35 
VLINK="#840084"
36 
ALINK="#0000FF"
37 
><DIV
38 
CLASS="NAVHEADER"
39 
><TABLE
40 
SUMMARY="Header navigation table"
41 
WIDTH="100%"
42 
BORDER="0"
43 
CELLPADDING="0"
44 
CELLSPACING="0"
45 
><TR
46 
><TH
47 
COLSPAN="3"
48 
ALIGN="center"
49 
>eCos Reference Manual</TH
50 
></TR
51 
><TR
52 
><TD
53 
WIDTH="10%"
54 
ALIGN="left"
55 
VALIGN="bottom"
56 
><A
57 
HREF="hal-diagnostic-support.html"
58 
ACCESSKEY="P"
59 
>Prev</A
60 
></TD
61 
><TD
62 
WIDTH="80%"
63 
ALIGN="center"
64 
VALIGN="bottom"
65 
>Chapter 9. HAL Interfaces</TD
66 
><TD
67 
WIDTH="10%"
68 
ALIGN="right"
69 
VALIGN="bottom"
70 
><A
71 
HREF="hal-exception-handling.html"
72 
ACCESSKEY="N"
73 
>Next</A
74 
></TD
75 
></TR
76 
></TABLE
77 
><HR
78 
ALIGN="LEFT"
79 
WIDTH="100%"></DIV
80 
><DIV
81 
CLASS="SECTION"
82 
><H1
83 
CLASS="SECTION"
84 
><A
85 
NAME="HAL-SMP-SUPPORT">SMP Support</H1
86 
><P
87 
>eCos contains support for limited Symmetric Multi-Processing
88 
(SMP). This is only available on selected architectures and platforms.</P
89 
><DIV
90 
CLASS="SECTION"
91 
><H2
92 
CLASS="SECTION"
93 
><A
94 
NAME="AEN8275">Target Hardware Limitations</H2
95 
><P
96 
>To allow a reasonable implementation of SMP, and to reduce the
97 
disruption to the existing source base, a number of assumptions have
98 
been made about the features of the target hardware.</P
99 
><P
100 
></P
101 
><UL
102 
><LI
103 
><P
104 
>Modest multiprocessing. The typical number of CPUs supported is two
105 
to four, with an upper limit around eight. While there are no
106 
inherent limits in the code, hardware and algorithmic limitations
107 
will probably become significant beyond this point.</P
108 
></LI
109 
><LI
110 
><P
111 
>SMP synchronization support. The hardware must supply a mechanism to
112 
allow software on two CPUs to synchronize. This is normally provided
113 
as part of the instruction set in the form of test-and-set,
114 
compare-and-swap or load-link/store-conditional instructions. An
115 
alternative approach is the provision of hardware semaphore
116 
registers which can be used to serialize implementations of these
117 
operations. Whatever hardware facilities are available, they are
118 
used in eCos to implement spinlocks.</P
119 
></LI
120 
><LI
121 
><P
122 
>Coherent caches. It is assumed that no extra effort will be required
123 
to access shared memory from any processor. This means that either
124 
there are no caches, they are shared by all processors, or are
125 
maintained in a coherent state by the hardware. It would be too
126 
disruptive to the eCos sources if every memory access had to be
127 
bracketed by cache load/flush operations. Any hardware that requires
128 
this is not supported.</P
129 
></LI
130 
><LI
131 
><P
132 
>Uniform addressing. It is assumed that all memory that is
133 
shared between CPUs is addressed at the same location from all
134 
CPUs. Like non-coherent caches, dealing with CPU-specific address
135 
translation is considered too disruptive to the eCos source
136 
base. This does not, however, preclude systems with non-uniform
137 
access costs for different CPUs.</P
138 
></LI
139 
><LI
140 
><P
141 
>Uniform device addressing. As with access to memory, it is assumed
142 
that all devices are equally accessible to all CPUs. Since device
143 
access is often made from thread contexts, it is not possible to
144 
restrict access to device control registers to certain CPUs, since
145 
there is currently no support for binding or migrating threads to CPUs.</P
146 
></LI
147 
><LI
148 
><P
149 
>Interrupt routing. The target hardware must have an interrupt
150 
controller that can route interrupts to specific CPUs. It is
151 
acceptable for all interrupts to be delivered to just one CPU, or
152 
for some interrupts to be bound to specific CPUs, or for some
153 
interrupts to be local to each CPU. At present dynamic routing,
154 
where a different CPU may be chosen each time an interrupt is
155 
delivered, is not supported. ECos cannot support hardware where all
156 
interrupts are delivered to all CPUs simultaneously with the
157 
expectation that software will resolve any conflicts.</P
158 
></LI
159 
><LI
160 
><P
161 
>Inter-CPU interrupts. A mechanism to allow one CPU to interrupt
162 
another is needed. This is necessary so that events on one CPU can
163 
cause rescheduling on other CPUs.</P
164 
></LI
165 
><LI
166 
><P
167 
>CPU Identifiers. Code running on a CPU must be able to determine
168 
which CPU it is running on. The CPU Id is usually provided either in
169 
a CPU status register, or in a register associated with the
170 
inter-CPU interrupt delivery subsystem. ECos expects CPU Ids to be
171 
small positive integers, although alternative representations, such
172 
as bitmaps, can be converted relatively easily. Complex mechanisms
173 
for getting the CPU Id cannot be supported. Getting the CPU Id must
174 
be a cheap operation, since it is done often, and in performance
175 
critical places such as interrupt handlers and the scheduler.</P
176 
></LI
177 
></UL
178 
></DIV
179 
><DIV
180 
CLASS="SECTION"
181 
><H2
182 
CLASS="SECTION"
183 
><A
184 
NAME="AEN8295">HAL Support</H2
185 
><P
186 
>SMP support in any platform depends on the HAL supplying the
187 
appropriate operations. All HAL SMP support is defined in the
188 
<TT
189 
CLASS="FILENAME"
190 
>cyg/hal/hal_smp.h</TT
191 
> header. Variant and platform
192 
specific definitions will be in <TT
193 
CLASS="FILENAME"
194 
>cyg/hal/var_smp.h</TT
195 
>
196 
and <TT
197 
CLASS="FILENAME"
198 
>cyg/hal/plf_smp.h</TT
199 
> respectively. These files
200 
are include automatically by this header, so need not be included
201 
explicitly.</P
202 
><P
203 
>SMP support falls into a number of functional groups.</P
204 
><DIV
205 
CLASS="SECTION"
206 
><H3
207 
CLASS="SECTION"
208 
><A
209 
NAME="AEN8302">CPU Control</H3
210 
><P
211 
>This group consists of descriptive and control macros for managing the
212 
CPUs in an SMP system.</P
213 
><P
214 
></P
215 
><DIV
216 
CLASS="VARIABLELIST"
217 
><DL
218 
><DT
219 
><TT
220 
CLASS="LITERAL"
221 
>HAL_SMP_CPU_TYPE</TT
222 
></DT
223 
><DD
224 
><P
225 
>A type that can contain a CPU id. A CPU id is
226 
usually a small integer that is used to index
227 
arrays of variables that are managed on an
228 
per-CPU basis.</P
229 
></DD
230 
><DT
231 
><TT
232 
CLASS="LITERAL"
233 
>HAL_SMP_CPU_MAX</TT
234 
></DT
235 
><DD
236 
><P
237 
>The maximum number of CPUs that can be
238 
supported. This is used to provide the size of
239 
any arrays that have an element per CPU.</P
240 
></DD
241 
><DT
242 
><TT
243 
CLASS="LITERAL"
244 
>HAL_SMP_CPU_COUNT()</TT
245 
></DT
246 
><DD
247 
><P
248 
>Returns the number of CPUs currently
249 
operational. This may differ from
250 
HAL_SMP_CPU_MAX depending on the runtime
251 
environment.</P
252 
></DD
253 
><DT
254 
><TT
255 
CLASS="LITERAL"
256 
>HAL_SMP_CPU_THIS()</TT
257 
></DT
258 
><DD
259 
><P
260 
>Returns the CPU id of the current CPU.</P
261 
></DD
262 
><DT
263 
><TT
264 
CLASS="LITERAL"
265 
>HAL_SMP_CPU_NONE</TT
266 
></DT
267 
><DD
268 
><P
269 
>A value that does not match any real CPU
270 
id. This is uses where a CPU type variable
271 
must be set to a null value.</P
272 
></DD
273 
><DT
274 
><TT
275 
CLASS="LITERAL"
276 
>HAL_SMP_CPU_START( cpu )</TT
277 
></DT
278 
><DD
279 
><P
280 
>Starts the given CPU executing at a defined
281 
HAL entry point. After performing any HAL
282 
level initialization, the CPU calls up into
283 
the kernel at <TT
284 
CLASS="FUNCTION"
285 
>cyg_kernel_cpu_startup()</TT
286 
>.</P
287 
></DD
288 
><DT
289 
><TT
290 
CLASS="LITERAL"
291 
>HAL_SMP_CPU_RESCHEDULE_INTERRUPT( cpu, wait )</TT
292 
></DT
293 
><DD
294 
><P
295 
>Sends the CPU a reschedule interrupt, and if
296 
<TT
297 
CLASS="PARAMETER"
298 
><I
299 
>wait</I
300 
></TT
301 
> is non-zero, waits for an
302 
acknowledgment. The interrupted CPU should call
303 
<TT
304 
CLASS="FUNCTION"
305 
>cyg_scheduler_set_need_reschedule()</TT
306 
> in its DSR to
307 
cause the reschedule to occur.</P
308 
></DD
309 
><DT
310 
><TT
311 
CLASS="LITERAL"
312 
>HAL_SMP_CPU_TIMESLICE_INTERRUPT( cpu, wait )</TT
313 
></DT
314 
><DD
315 
><P
316 
>Sends the CPU a timeslice interrupt, and if
317 
<TT
318 
CLASS="PARAMETER"
319 
><I
320 
>wait</I
321 
></TT
322 
> is non-zero, waits for an
323 
acknowledgment. The interrupted CPU should call
324 
<TT
325 
CLASS="FUNCTION"
326 
>cyg_scheduler_timeslice_cpu()</TT
327 
> to cause the
328 
timeslice event to be processed.</P
329 
></DD
330 
></DL
331 
></DIV
332 
></DIV
333 
><DIV
334 
CLASS="SECTION"
335 
><H3
336 
CLASS="SECTION"
337 
><A
338 
NAME="AEN8351">Test-and-set Support</H3
339 
><P
340 
>Test-and-set is the foundation of the SMP synchronization
341 
mechanisms.</P
342 
><P
343 
></P
344 
><DIV
345 
CLASS="VARIABLELIST"
346 
><DL
347 
><DT
348 
><TT
349 
CLASS="LITERAL"
350 
>HAL_TAS_TYPE</TT
351 
></DT
352 
><DD
353 
><P
354 
>The type for all test-and-set variables. The
355 
test-and-set macros only support operations on
356 
a single bit (usually the least significant
357 
bit) of this location. This allows for maximum
358 
flexibility in the implementation.</P
359 
></DD
360 
><DT
361 
><TT
362 
CLASS="LITERAL"
363 
>HAL_TAS_SET( tas, oldb )</TT
364 
></DT
365 
><DD
366 
><P
367 
>Performs a test and set operation on the
368 
location <TT
369 
CLASS="PARAMETER"
370 
><I
371 
>tas</I
372 
></TT
373 
>. <TT
374 
CLASS="PARAMETER"
375 
><I
376 
>oldb</I
377 
></TT
378 
> will contain <TT
379 
CLASS="LITERAL"
380 
>true</TT
381 
> if
382 
the location was already set, and <TT
383 
CLASS="LITERAL"
384 
>false</TT
385 
> if
386 
it was clear.</P
387 
></DD
388 
><DT
389 
><TT
390 
CLASS="LITERAL"
391 
>HAL_TAS_CLEAR( tas, oldb )</TT
392 
></DT
393 
><DD
394 
><P
395 
>Performs a test and clear operation on the
396 
location <TT
397 
CLASS="PARAMETER"
398 
><I
399 
>tas</I
400 
></TT
401 
>. <TT
402 
CLASS="PARAMETER"
403 
><I
404 
>oldb</I
405 
></TT
406 
> will contain <TT
407 
CLASS="LITERAL"
408 
>true</TT
409 
> if
410 
the location was already set, and <TT
411 
CLASS="LITERAL"
412 
>false</TT
413 
> if
414 
it was clear.</P
415 
></DD
416 
></DL
417 
></DIV
418 
></DIV
419 
><DIV
420 
CLASS="SECTION"
421 
><H3
422 
CLASS="SECTION"
423 
><A
424 
NAME="AEN8378">Spinlocks</H3
425 
><P
426 
>Spinlocks provide inter-CPU locking. Normally they will be implemented
427 
on top of the test-and-set mechanism above, but may also be
428 
implemented by other means if, for example, the hardware has more
429 
direct support for spinlocks.</P
430 
><P
431 
></P
432 
><DIV
433 
CLASS="VARIABLELIST"
434 
><DL
435 
><DT
436 
><TT
437 
CLASS="LITERAL"
438 
>HAL_SPINLOCK_TYPE</TT
439 
></DT
440 
><DD
441 
><P
442 
>The type for all spinlock variables.</P
443 
></DD
444 
><DT
445 
><TT
446 
CLASS="LITERAL"
447 
>HAL_SPINLOCK_INIT_CLEAR</TT
448 
></DT
449 
><DD
450 
><P
451 
>A value that may be assigned to a spinlock
452 
variable to initialize it to clear.</P
453 
></DD
454 
><DT
455 
><TT
456 
CLASS="LITERAL"
457 
>HAL_SPINLOCK_INIT_SET</TT
458 
></DT
459 
><DD
460 
><P
461 
>A value that may be assigned to a spinlock
462 
variable to initialize it to set.</P
463 
></DD
464 
><DT
465 
><TT
466 
CLASS="LITERAL"
467 
>HAL_SPINLOCK_SPIN( lock )</TT
468 
></DT
469 
><DD
470 
><P
471 
>The caller spins in a busy loop waiting for
472 
the lock to become clear. It then sets it and
473 
continues. This is all handled atomically, so
474 
that there are no race conditions between CPUs.</P
475 
></DD
476 
><DT
477 
><TT
478 
CLASS="LITERAL"
479 
>HAL_SPINLOCK_CLEAR( lock )</TT
480 
></DT
481 
><DD
482 
><P
483 
>The caller clears the lock. One of any waiting
484 
spinners will then be able to proceed.</P
485 
></DD
486 
><DT
487 
><TT
488 
CLASS="LITERAL"
489 
>HAL_SPINLOCK_TRY( lock, val )</TT
490 
></DT
491 
><DD
492 
><P
493 
>Attempts to set the lock. The value put in
494 
<TT
495 
CLASS="PARAMETER"
496 
><I
497 
>val</I
498 
></TT
499 
> will be <TT
500 
CLASS="LITERAL"
501 
>true</TT
502 
> if the lock was
503 
claimed successfully, and <TT
504 
CLASS="LITERAL"
505 
>false</TT
506 
> if it was
507 
not.</P
508 
></DD
509 
><DT
510 
><TT
511 
CLASS="LITERAL"
512 
>HAL_SPINLOCK_TEST( lock, val )</TT
513 
></DT
514 
><DD
515 
><P
516 
>Tests the current value of the lock. The value
517 
put in <TT
518 
CLASS="PARAMETER"
519 
><I
520 
>val</I
521 
></TT
522 
> will be <TT
523 
CLASS="LITERAL"
524 
>true</TT
525 
> if the lock is
526 
claimed and <TT
527 
CLASS="LITERAL"
528 
>false</TT
529 
> of it is clear.</P
530 
></DD
531 
></DL
532 
></DIV
533 
></DIV
534 
><DIV
535 
CLASS="SECTION"
536 
><H3
537 
CLASS="SECTION"
538 
><A
539 
NAME="AEN8423">Scheduler Lock</H3
540 
><P
541 
>The scheduler lock is the main protection for all kernel data
542 
structures. By default the kernel implements the scheduler lock itself
543 
using a spinlock. However, if spinlocks cannot be supported by the
544 
hardware, or there is a more efficient implementation available, the
545 
HAL may provide macros to implement the scheduler lock.</P
546 
><P
547 
></P
548 
><DIV
549 
CLASS="VARIABLELIST"
550 
><DL
551 
><DT
552 
><TT
553 
CLASS="LITERAL"
554 
>HAL_SMP_SCHEDLOCK_DATA_TYPE</TT
555 
></DT
556 
><DD
557 
><P
558 
>A data type, possibly a structure, that
559 
contains any data items needed by the
560 
scheduler lock implementation. A variable of
561 
this type will be instantiated as a static
562 
member of the Cyg_Scheduler_SchedLock class
563 
and passed to all the following macros.</P
564 
></DD
565 
><DT
566 
><TT
567 
CLASS="LITERAL"
568 
>HAL_SMP_SCHEDLOCK_INIT( lock, data )</TT
569 
></DT
570 
><DD
571 
><P
572 
>Initialize the scheduler lock. The <TT
573 
CLASS="PARAMETER"
574 
><I
575 
>lock</I
576 
></TT
577 
>
578 
argument is the scheduler lock counter and the
579 
<TT
580 
CLASS="PARAMETER"
581 
><I
582 
>data</I
583 
></TT
584 
> argument is a variable of
585 
HAL_SMP_SCHEDLOCK_DATA_TYPE type.</P
586 
></DD
587 
><DT
588 
><TT
589 
CLASS="LITERAL"
590 
>HAL_SMP_SCHEDLOCK_INC( lock, data )</TT
591 
></DT
592 
><DD
593 
><P
594 
>Increment the scheduler lock. The first
595 
increment of the lock from zero to one for any
596 
CPU may cause it to wait until the lock is
597 
zeroed by another CPU. Subsequent increments
598 
should be less expensive since this CPU
599 
already holds the lock.</P
600 
></DD
601 
><DT
602 
><TT
603 
CLASS="LITERAL"
604 
>HAL_SMP_SCHEDLOCK_ZERO( lock, data )</TT
605 
></DT
606 
><DD
607 
><P
608 
>Zero the scheduler lock. This operation will
609 
also clear the lock so that other CPUs may
610 
claim it.</P
611 
></DD
612 
><DT
613 
><TT
614 
CLASS="LITERAL"
615 
>HAL_SMP_SCHEDLOCK_SET( lock, data, new )</TT
616 
></DT
617 
><DD
618 
><P
619 
>Set the lock to a different value, in
620 
<TT
621 
CLASS="PARAMETER"
622 
><I
623 
>new</I
624 
></TT
625 
>. This is only called when the lock is
626 
already known to be owned by the current CPU. It is never called to
627 
zero the lock, or to increment it from zero.</P
628 
></DD
629 
></DL
630 
></DIV
631 
></DIV
632 
><DIV
633 
CLASS="SECTION"
634 
><H3
635 
CLASS="SECTION"
636 
><A
637 
NAME="AEN8455">Interrupt Routing</H3
638 
><P
639 
>The routing of interrupts to different CPUs is supported by two new
640 
interfaces in hal_intr.h.</P
641 
><P
642 
>Once an interrupt has been routed to a new CPU, the existing vector
643 
masking and configuration operations should take account of the CPU
644 
routing. For example, if the operation is not invoked on the
645 
destination CPU itself, then the HAL may need to arrange to transfer
646 
the operation to the destination CPU for correct application.</P
647 
><P
648 
></P
649 
><DIV
650 
CLASS="VARIABLELIST"
651 
><DL
652 
><DT
653 
><TT
654 
CLASS="LITERAL"
655 
>HAL_INTERRUPT_SET_CPU( vector, cpu )</TT
656 
></DT
657 
><DD
658 
><P
659 
>Route the interrupt for the given <TT
660 
CLASS="PARAMETER"
661 
><I
662 
>vector</I
663 
></TT
664 
> to
665 
the given <TT
666 
CLASS="PARAMETER"
667 
><I
668 
>cpu</I
669 
></TT
670 
>. </P
671 
></DD
672 
><DT
673 
><TT
674 
CLASS="LITERAL"
675 
>HAL_INTERRUPT_GET_CPU( vector, cpu )</TT
676 
></DT
677 
><DD
678 
><P
679 
>Set <TT
680 
CLASS="PARAMETER"
681 
><I
682 
>cpu</I
683 
></TT
684 
> to the id of the CPU to which this
685 
vector is routed.</P
686 
></DD
687 
></DL
688 
></DIV
689 
></DIV
690 
></DIV
691 
></DIV
692 
><DIV
693 
CLASS="NAVFOOTER"
694 
><HR
695 
ALIGN="LEFT"
696 
WIDTH="100%"><TABLE
697 
SUMMARY="Footer navigation table"
698 
WIDTH="100%"
699 
BORDER="0"
700 
CELLPADDING="0"
701 
CELLSPACING="0"
702 
><TR
703 
><TD
704 
WIDTH="33%"
705 
ALIGN="left"
706 
VALIGN="top"
707 
><A
708 
HREF="hal-diagnostic-support.html"
709 
ACCESSKEY="P"
710 
>Prev</A
711 
></TD
712 
><TD
713 
WIDTH="34%"
714 
ALIGN="center"
715 
VALIGN="top"
716 
><A
717 
HREF="ecos-ref.html"
718 
ACCESSKEY="H"
719 
>Home</A
720 
></TD
721 
><TD
722 
WIDTH="33%"
723 
ALIGN="right"
724 
VALIGN="top"
725 
><A
726 
HREF="hal-exception-handling.html"
727 
ACCESSKEY="N"
728 
>Next</A
729 
></TD
730 
></TR
731 
><TR
732 
><TD
733 
WIDTH="33%"
734 
ALIGN="left"
735 
VALIGN="top"
736 
>Diagnostic Support</TD
737 
><TD
738 
WIDTH="34%"
739 
ALIGN="center"
740 
VALIGN="top"
741 
><A
742 
HREF="hal-interfaces.html"
743 
ACCESSKEY="U"
744 
>Up</A
745 
></TD
746 
><TD
747 
WIDTH="33%"
748 
ALIGN="right"
749 
VALIGN="top"
750 
>Exception Handling</TD
751 
></TR
752 
></TABLE
753 
></DIV
754 
></BODY
755 
></HTML
756 
>

powered by: WebSVN 2.1.0

© copyright 1999-2024 OpenCores.org, equivalent to Oliscience, all rights reserved. OpenCores®, registered trademark.