2 * Copyright (c) 2010 Fabio Checconi, Luigi Rizzo, Paolo Valente
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.
14 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
15 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
16 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
17 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
18 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
19 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
20 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
21 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
22 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
23 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 #include <sys/malloc.h>
33 #include <sys/socket.h>
34 #include <sys/socketvar.h>
35 #include <sys/kernel.h>
38 #include <sys/module.h>
39 #include <sys/rwlock.h>
40 #include <net/if.h> /* IFNAMSIZ */
41 #include <netinet/in.h>
42 #include <netinet/ip_var.h> /* ipfw_rule_ref */
43 #include <netinet/ip_fw.h> /* flow_id */
44 #include <netinet/ip_dummynet.h>
45 #include <netpfil/ipfw/ip_fw_private.h>
46 #include <netpfil/ipfw/dn_heap.h>
47 #include <netpfil/ipfw/ip_dn_private.h>
49 #include <netpfil/ipfw/dn_aqm.h>
51 #include <netpfil/ipfw/dn_sched.h>
57 #define _P64 unsigned long long /* cast for printing uint64_t */
59 static void dump_sched(struct qfq_sched *q, const char *msg);
64 #define DN_SCHED_QFQ 4 // XXX Where?
65 typedef unsigned long bitmap;
68 * bitmaps ops are critical. Some linux versions have __fls
69 * and the bitmap ops. Some machines have ffs
70 * NOTE: fls() returns 1 for the least significant bit,
71 * __fls() returns 0 for the same case.
72 * We use the base-0 version __fls() to match the description in
75 #if defined(_WIN32) || (defined(__MIPSEL__) && defined(LINUX_24))
76 int fls(unsigned int n)
79 for (i = 0; n > 0; n >>= 1, i++)
85 #if !defined(_KERNEL) || defined( __FreeBSD__ ) || defined(_WIN32) || (defined(__MIPSEL__) && defined(LINUX_24))
86 static inline unsigned long __fls(unsigned long word)
92 #if !defined(_KERNEL) || !defined(__linux__)
94 static int test_bit(int ix, bitmap *p)
96 if (ix < 0 || ix > 31)
97 D("bad index %d", ix);
100 static void __set_bit(int ix, bitmap *p)
102 if (ix < 0 || ix > 31)
103 D("bad index %d", ix);
106 static void __clear_bit(int ix, bitmap *p)
108 if (ix < 0 || ix > 31)
109 D("bad index %d", ix);
112 #else /* !QFQ_DEBUG */
113 /* XXX do we have fast version, or leave it to the compiler ? */
114 #define test_bit(ix, pData) ((*pData) & (1<<(ix)))
115 #define __set_bit(ix, pData) (*pData) |= (1<<(ix))
116 #define __clear_bit(ix, pData) (*pData) &= ~(1<<(ix))
117 #endif /* !QFQ_DEBUG */
118 #endif /* !__linux__ */
121 #define __clear_bit(ix, pData) (*pData) &= ~(1<<(ix))
124 /*-------------------------------------------*/
127 Virtual time computations.
129 S, F and V are all computed in fixed point arithmetic with
130 FRAC_BITS decimal bits.
132 QFQ_MAX_INDEX is the maximum index allowed for a group. We need
134 QFQ_MAX_WSHIFT is the maximum power of two supported as a weight.
135 The layout of the bits is as below:
137 [ MTU_SHIFT ][ FRAC_BITS ]
138 [ MAX_INDEX ][ MIN_SLOT_SHIFT ]
142 where MIN_SLOT_SHIFT is derived by difference from the others.
144 The max group index corresponds to Lmax/w_min, where
145 Lmax=1<<MTU_SHIFT, w_min = 1 .
146 From this, and knowing how many groups (MAX_INDEX) we want,
147 we can derive the shift corresponding to each group.
149 Because we often need to compute
150 F = S + len/w_i and V = V + len/wsum
151 instead of storing w_i store the value
152 inv_w = (1<<FRAC_BITS)/w_i
153 so we can do F = S + len * inv_w * wsum.
154 We use W_TOT in the formulas so we can easily move between
155 static and adaptive weight sum.
157 The per-scheduler-instance data contain all the data structures
158 for the scheduler: bitmaps and bucket lists.
162 * Maximum number of consecutive slots occupied by backlogged classes
163 * inside a group. This is approx lmax/lmin + 5.
164 * XXX check because it poses constraints on MAX_INDEX
166 #define QFQ_MAX_SLOTS 32
168 * Shifts used for class<->group mapping. Class weights are
169 * in the range [1, QFQ_MAX_WEIGHT], we to map each class i to the
170 * group with the smallest index that can support the L_i / r_i
171 * configured for the class.
173 * grp->index is the index of the group; and grp->slot_shift
174 * is the shift for the corresponding (scaled) sigma_i.
176 * When computing the group index, we do (len<<FP_SHIFT)/weight,
177 * then compute an FLS (which is like a log2()), and if the result
178 * is below the MAX_INDEX region we use 0 (which is the same as
179 * using a larger len).
181 #define QFQ_MAX_INDEX 19
182 #define QFQ_MAX_WSHIFT 16 /* log2(max_weight) */
184 #define QFQ_MAX_WEIGHT (1<<QFQ_MAX_WSHIFT)
185 #define QFQ_MAX_WSUM (2*QFQ_MAX_WEIGHT)
187 #define FRAC_BITS 30 /* fixed point arithmetic */
188 #define ONE_FP (1UL << FRAC_BITS)
190 #define QFQ_MTU_SHIFT 11 /* log2(max_len) */
191 #define QFQ_MIN_SLOT_SHIFT (FRAC_BITS + QFQ_MTU_SHIFT - QFQ_MAX_INDEX)
194 * Possible group states, also indexes for the bitmaps array in
195 * struct qfq_queue. We rely on ER, IR, EB, IB being numbered 0..3
197 enum qfq_state { ER, IR, EB, IB, QFQ_MAX_STATE };
201 * additional queue info. Some of this info should come from
202 * the flowset, we copy them here for faster processing.
203 * This is an overlay of the struct dn_queue
207 uint64_t S, F; /* flow timestamps (exact) */
208 struct qfq_class *next; /* Link for the slot list. */
210 /* group we belong to. In principle we would need the index,
211 * which is log_2(lmax/weight), but we never reference it
212 * directly, only the group.
214 struct qfq_group *grp;
216 /* these are copied from the flowset. */
217 uint32_t inv_w; /* ONE_FP/weight */
218 uint32_t lmax; /* Max packet size for this flow. */
221 /* Group descriptor, see the paper for details.
222 * Basically this contains the bucket lists
225 uint64_t S, F; /* group timestamps (approx). */
226 unsigned int slot_shift; /* Slot shift. */
227 unsigned int index; /* Group index. */
228 unsigned int front; /* Index of the front slot. */
229 bitmap full_slots; /* non-empty slots */
231 /* Array of lists of active classes. */
232 struct qfq_class *slots[QFQ_MAX_SLOTS];
235 /* scheduler instance descriptor. */
237 uint64_t V; /* Precise virtual time. */
238 uint32_t wsum; /* weight sum */
239 uint32_t iwsum; /* inverse weight sum */
240 NO(uint32_t i_wsum;) /* ONE_FP/w_sum */
241 NO(uint32_t queued;) /* debugging */
242 NO(uint32_t loops;) /* debugging */
243 bitmap bitmaps[QFQ_MAX_STATE]; /* Group bitmaps. */
244 struct qfq_group groups[QFQ_MAX_INDEX + 1]; /* The groups. */
247 /*---- support functions ----------------------------*/
249 /* Generic comparison function, handling wraparound. */
250 static inline int qfq_gt(uint64_t a, uint64_t b)
252 return (int64_t)(a - b) > 0;
255 /* Round a precise timestamp to its slotted value. */
256 static inline uint64_t qfq_round_down(uint64_t ts, unsigned int shift)
258 return ts & ~((1ULL << shift) - 1);
261 /* return the pointer to the group with lowest index in the bitmap */
262 static inline struct qfq_group *qfq_ffs(struct qfq_sched *q,
263 unsigned long bitmap)
265 int index = ffs(bitmap) - 1; // zero-based
266 return &q->groups[index];
270 * Calculate a flow index, given its weight and maximum packet length.
271 * index = log_2(maxlen/weight) but we need to apply the scaling.
272 * This is used only once at flow creation.
274 static int qfq_calc_index(uint32_t inv_w, unsigned int maxlen)
276 uint64_t slot_size = (uint64_t)maxlen *inv_w;
277 unsigned long size_map;
280 size_map = (unsigned long)(slot_size >> QFQ_MIN_SLOT_SHIFT);
284 index = __fls(size_map) + 1; // basically a log_2()
285 index -= !(slot_size - (1ULL << (index + QFQ_MIN_SLOT_SHIFT - 1)));
291 ND("W = %d, L = %d, I = %d\n", ONE_FP/inv_w, maxlen, index);
294 /*---- end support functions ----*/
296 /*-------- API calls --------------------------------*/
298 * Validate and copy parameters from flowset.
301 qfq_new_queue(struct dn_queue *_q)
303 struct qfq_sched *q = (struct qfq_sched *)(_q->_si + 1);
304 struct qfq_class *cl = (struct qfq_class *)_q;
306 uint32_t w; /* approximated weight */
308 /* import parameters from the flowset. They should be correct
311 w = _q->fs->fs.par[0];
312 cl->lmax = _q->fs->fs.par[1];
313 if (!w || w > QFQ_MAX_WEIGHT) {
315 D("rounding weight to 1");
317 cl->inv_w = ONE_FP/w;
318 w = ONE_FP/cl->inv_w;
319 if (q->wsum + w > QFQ_MAX_WSUM)
322 i = qfq_calc_index(cl->inv_w, cl->lmax);
323 cl->grp = &q->groups[i];
325 q->iwsum = ONE_FP / q->wsum; /* XXX note theory */
326 // XXX cl->S = q->V; ?
330 /* remove an empty queue */
332 qfq_free_queue(struct dn_queue *_q)
334 struct qfq_sched *q = (struct qfq_sched *)(_q->_si + 1);
335 struct qfq_class *cl = (struct qfq_class *)_q;
337 q->wsum -= ONE_FP/cl->inv_w;
339 q->iwsum = ONE_FP / q->wsum;
340 cl->inv_w = 0; /* reset weight to avoid run twice */
345 /* Calculate a mask to mimic what would be ffs_from(). */
346 static inline unsigned long
347 mask_from(unsigned long bitmap, int from)
349 return bitmap & ~((1UL << from) - 1);
353 * The state computation relies on ER=0, IR=1, EB=2, IB=3
354 * First compute eligibility comparing grp->S, q->V,
355 * then check if someone is blocking us and possibly add EB
357 static inline unsigned int
358 qfq_calc_state(struct qfq_sched *q, struct qfq_group *grp)
360 /* if S > V we are not eligible */
361 unsigned int state = qfq_gt(grp->S, q->V);
362 unsigned long mask = mask_from(q->bitmaps[ER], grp->index);
363 struct qfq_group *next;
366 next = qfq_ffs(q, mask);
367 if (qfq_gt(grp->F, next->F))
376 * q->bitmaps[dst] |= q->bitmaps[src] & mask;
377 * q->bitmaps[src] &= ~mask;
378 * but we should make sure that src != dst
381 qfq_move_groups(struct qfq_sched *q, unsigned long mask, int src, int dst)
383 q->bitmaps[dst] |= q->bitmaps[src] & mask;
384 q->bitmaps[src] &= ~mask;
388 qfq_unblock_groups(struct qfq_sched *q, int index, uint64_t old_finish)
390 unsigned long mask = mask_from(q->bitmaps[ER], index + 1);
391 struct qfq_group *next;
394 next = qfq_ffs(q, mask);
395 if (!qfq_gt(next->F, old_finish))
399 mask = (1UL << index) - 1;
400 qfq_move_groups(q, mask, EB, ER);
401 qfq_move_groups(q, mask, IB, IR);
408 old_V >>= QFQ_MIN_SLOT_SHIFT;
415 qfq_make_eligible(struct qfq_sched *q, uint64_t old_V)
417 unsigned long mask, vslot, old_vslot;
419 vslot = q->V >> QFQ_MIN_SLOT_SHIFT;
420 old_vslot = old_V >> QFQ_MIN_SLOT_SHIFT;
422 if (vslot != old_vslot) {
423 /* must be 2ULL, see ToN QFQ article fig.5, we use base-0 fls */
424 mask = (2ULL << (__fls(vslot ^ old_vslot))) - 1;
425 qfq_move_groups(q, mask, IR, ER);
426 qfq_move_groups(q, mask, IB, EB);
431 * XXX we should make sure that slot becomes less than 32.
432 * This is guaranteed by the input values.
433 * roundedS is always cl->S rounded on grp->slot_shift bits.
436 qfq_slot_insert(struct qfq_group *grp, struct qfq_class *cl, uint64_t roundedS)
438 uint64_t slot = (roundedS - grp->S) >> grp->slot_shift;
439 unsigned int i = (grp->front + slot) % QFQ_MAX_SLOTS;
441 cl->next = grp->slots[i];
443 __set_bit(slot, &grp->full_slots);
447 * remove the entry from the slot
450 qfq_front_slot_remove(struct qfq_group *grp)
452 struct qfq_class **h = &grp->slots[grp->front];
456 __clear_bit(0, &grp->full_slots);
460 * Returns the first full queue in a group. As a side effect,
461 * adjust the bucket list so the first non-empty bucket is at
462 * position 0 in full_slots.
464 static inline struct qfq_class *
465 qfq_slot_scan(struct qfq_group *grp)
469 ND("grp %d full %x", grp->index, grp->full_slots);
470 if (!grp->full_slots)
473 i = ffs(grp->full_slots) - 1; // zero-based
475 grp->front = (grp->front + i) % QFQ_MAX_SLOTS;
476 grp->full_slots >>= i;
479 return grp->slots[grp->front];
483 * adjust the bucket list. When the start time of a group decreases,
484 * we move the index down (modulo QFQ_MAX_SLOTS) so we don't need to
485 * move the objects. The mask of occupied slots must be shifted
486 * because we use ffs() to find the first non-empty slot.
487 * This covers decreases in the group's start time, but what about
488 * increases of the start time ?
489 * Here too we should make sure that i is less than 32
492 qfq_slot_rotate(struct qfq_sched *q, struct qfq_group *grp, uint64_t roundedS)
494 unsigned int i = (grp->S - roundedS) >> grp->slot_shift;
497 grp->full_slots <<= i;
498 grp->front = (grp->front - i) % QFQ_MAX_SLOTS;
503 qfq_update_eligible(struct qfq_sched *q, uint64_t old_V)
507 ineligible = q->bitmaps[IR] | q->bitmaps[IB];
509 if (!q->bitmaps[ER]) {
510 struct qfq_group *grp;
511 grp = qfq_ffs(q, ineligible);
512 if (qfq_gt(grp->S, q->V))
515 qfq_make_eligible(q, old_V);
520 * Updates the class, returns true if also the group needs to be updated.
523 qfq_update_class(struct qfq_sched *q, struct qfq_group *grp,
524 struct qfq_class *cl)
529 if (cl->_q.mq.head == NULL) {
530 qfq_front_slot_remove(grp);
535 len = cl->_q.mq.head->m_pkthdr.len;
536 cl->F = cl->S + (uint64_t)len * cl->inv_w;
537 roundedS = qfq_round_down(cl->S, grp->slot_shift);
538 if (roundedS == grp->S)
541 qfq_front_slot_remove(grp);
542 qfq_slot_insert(grp, cl, roundedS);
548 qfq_dequeue(struct dn_sch_inst *si)
550 struct qfq_sched *q = (struct qfq_sched *)(si + 1);
551 struct qfq_group *grp;
552 struct qfq_class *cl;
557 if (!q->bitmaps[ER]) {
559 dump_sched(q, "start dequeue");)
563 grp = qfq_ffs(q, q->bitmaps[ER]);
565 cl = grp->slots[grp->front];
566 /* extract from the first bucket in the bucket list */
567 m = dn_dequeue(&cl->_q);
570 D("BUG/* non-workconserving leaf */");
575 q->V += (uint64_t)m->m_pkthdr.len * q->iwsum;
576 ND("m is %p F 0x%llx V now 0x%llx", m, cl->F, q->V);
578 if (qfq_update_class(q, grp, cl)) {
579 uint64_t old_F = grp->F;
580 cl = qfq_slot_scan(grp);
581 if (!cl) { /* group gone, remove from ER */
582 __clear_bit(grp->index, &q->bitmaps[ER]);
583 // grp->S = grp->F + 1; // XXX debugging only
585 uint64_t roundedS = qfq_round_down(cl->S, grp->slot_shift);
588 if (grp->S == roundedS)
591 grp->F = roundedS + (2ULL << grp->slot_shift);
592 /* remove from ER and put in the new set */
593 __clear_bit(grp->index, &q->bitmaps[ER]);
594 s = qfq_calc_state(q, grp);
595 __set_bit(grp->index, &q->bitmaps[s]);
597 /* we need to unblock even if the group has gone away */
598 qfq_unblock_groups(q, grp->index, old_F);
602 qfq_update_eligible(q, old_V);
603 NO(if (!q->bitmaps[ER] && q->queued)
604 dump_sched(q, "end dequeue");)
610 * Assign a reasonable start time for a new flow k in group i.
611 * Admissible values for \hat(F) are multiples of \sigma_i
612 * no greater than V+\sigma_i . Larger values mean that
613 * we had a wraparound so we consider the timestamp to be stale.
615 * If F is not stale and F >= V then we set S = F.
616 * Otherwise we should assign S = V, but this may violate
617 * the ordering in ER. So, if we have groups in ER, set S to
618 * the F_j of the first group j which would be blocking us.
619 * We are guaranteed not to move S backward because
620 * otherwise our group i would still be blocked.
623 qfq_update_start(struct qfq_sched *q, struct qfq_class *cl)
626 uint64_t limit, roundedF;
627 int slot_shift = cl->grp->slot_shift;
629 roundedF = qfq_round_down(cl->F, slot_shift);
630 limit = qfq_round_down(q->V, slot_shift) + (1ULL << slot_shift);
632 if (!qfq_gt(cl->F, q->V) || qfq_gt(roundedF, limit)) {
633 /* timestamp was stale */
634 mask = mask_from(q->bitmaps[ER], cl->grp->index);
636 struct qfq_group *next = qfq_ffs(q, mask);
637 if (qfq_gt(roundedF, next->F)) {
638 /* from pv 71261956973ba9e0637848a5adb4a5819b4bae83 */
639 if (qfq_gt(limit, next->F))
641 else /* preserve timestamp correctness */
647 } else { /* timestamp is not stale */
653 qfq_enqueue(struct dn_sch_inst *si, struct dn_queue *_q, struct mbuf *m)
655 struct qfq_sched *q = (struct qfq_sched *)(si + 1);
656 struct qfq_group *grp;
657 struct qfq_class *cl = (struct qfq_class *)_q;
662 DX(4, "len %d flow %p inv_w 0x%x grp %d", m->m_pkthdr.len,
663 _q, cl->inv_w, cl->grp->index);
664 /* XXX verify that the packet obeys the parameters */
665 if (m != _q->mq.head) {
666 if (dn_enqueue(_q, m, 0)) /* packet was dropped */
669 if (m != _q->mq.head)
672 /* If reach this point, queue q was idle */
674 qfq_update_start(q, cl); /* adjust start time */
675 /* compute new finish time and rounded start. */
676 cl->F = cl->S + (uint64_t)(m->m_pkthdr.len) * cl->inv_w;
677 roundedS = qfq_round_down(cl->S, grp->slot_shift);
680 * insert cl in the correct bucket.
681 * If cl->S >= grp->S we don't need to adjust the
682 * bucket list and simply go to the insertion phase.
683 * Otherwise grp->S is decreasing, we must make room
684 * in the bucket list, and also recompute the group state.
685 * Finally, if there were no flows in this group and nobody
686 * was in ER make sure to adjust V.
688 if (grp->full_slots) {
689 if (!qfq_gt(grp->S, cl->S))
691 /* create a slot for this cl->S */
692 qfq_slot_rotate(q, grp, roundedS);
693 /* group was surely ineligible, remove */
694 __clear_bit(grp->index, &q->bitmaps[IR]);
695 __clear_bit(grp->index, &q->bitmaps[IB]);
696 } else if (!q->bitmaps[ER] && qfq_gt(roundedS, q->V))
700 grp->F = roundedS + (2ULL << grp->slot_shift); // i.e. 2\sigma_i
701 s = qfq_calc_state(q, grp);
702 __set_bit(grp->index, &q->bitmaps[s]);
703 ND("new state %d 0x%x", s, q->bitmaps[s]);
704 ND("S %llx F %llx V %llx", cl->S, cl->F, q->V);
706 qfq_slot_insert(grp, cl, roundedS);
714 qfq_slot_remove(struct qfq_sched *q, struct qfq_group *grp,
715 struct qfq_class *cl, struct qfq_class **pprev)
717 unsigned int i, offset;
720 roundedS = qfq_round_down(cl->S, grp->slot_shift);
721 offset = (roundedS - grp->S) >> grp->slot_shift;
722 i = (grp->front + offset) % QFQ_MAX_SLOTS;
726 pprev = &grp->slots[i];
727 while (*pprev && *pprev != cl)
728 pprev = &(*pprev)->next;
734 __clear_bit(offset, &grp->full_slots);
738 * called to forcibly destroy a queue.
739 * If the queue is not in the front bucket, or if it has
740 * other queues in the front bucket, we can simply remove
741 * the queue with no other side effects.
742 * Otherwise we must propagate the event up.
743 * XXX description to be completed.
746 qfq_deactivate_class(struct qfq_sched *q, struct qfq_class *cl,
747 struct qfq_class **pprev)
749 struct qfq_group *grp = &q->groups[cl->index];
754 cl->F = cl->S; // not needed if the class goes away.
755 qfq_slot_remove(q, grp, cl, pprev);
757 if (!grp->full_slots) {
758 /* nothing left in the group, remove from all sets.
759 * Do ER last because if we were blocking other groups
760 * we must unblock them.
762 __clear_bit(grp->index, &q->bitmaps[IR]);
763 __clear_bit(grp->index, &q->bitmaps[EB]);
764 __clear_bit(grp->index, &q->bitmaps[IB]);
766 if (test_bit(grp->index, &q->bitmaps[ER]) &&
767 !(q->bitmaps[ER] & ~((1UL << grp->index) - 1))) {
768 mask = q->bitmaps[ER] & ((1UL << grp->index) - 1);
770 mask = ~((1UL << __fls(mask)) - 1);
773 qfq_move_groups(q, mask, EB, ER);
774 qfq_move_groups(q, mask, IB, IR);
776 __clear_bit(grp->index, &q->bitmaps[ER]);
777 } else if (!grp->slots[grp->front]) {
778 cl = qfq_slot_scan(grp);
779 roundedS = qfq_round_down(cl->S, grp->slot_shift);
780 if (grp->S != roundedS) {
781 __clear_bit(grp->index, &q->bitmaps[ER]);
782 __clear_bit(grp->index, &q->bitmaps[IR]);
783 __clear_bit(grp->index, &q->bitmaps[EB]);
784 __clear_bit(grp->index, &q->bitmaps[IB]);
786 grp->F = roundedS + (2ULL << grp->slot_shift);
787 s = qfq_calc_state(q, grp);
788 __set_bit(grp->index, &q->bitmaps[s]);
791 qfq_update_eligible(q, q->V);
796 qfq_new_fsk(struct dn_fsk *f)
798 ipdn_bound_var(&f->fs.par[0], 1, 1, QFQ_MAX_WEIGHT, "qfq weight");
799 ipdn_bound_var(&f->fs.par[1], 1500, 1, 2000, "qfq maxlen");
800 ND("weight %d len %d\n", f->fs.par[0], f->fs.par[1]);
805 * initialize a new scheduler instance
808 qfq_new_sched(struct dn_sch_inst *si)
810 struct qfq_sched *q = (struct qfq_sched *)(si + 1);
811 struct qfq_group *grp;
814 for (i = 0; i <= QFQ_MAX_INDEX; i++) {
817 grp->slot_shift = QFQ_MTU_SHIFT + FRAC_BITS -
824 * QFQ scheduler descriptor
826 static struct dn_alg qfq_desc = {
827 _SI( .type = ) DN_SCHED_QFQ,
828 _SI( .name = ) "QFQ",
829 _SI( .flags = ) DN_MULTIQUEUE,
831 _SI( .schk_datalen = ) 0,
832 _SI( .si_datalen = ) sizeof(struct qfq_sched),
833 _SI( .q_datalen = ) sizeof(struct qfq_class) - sizeof(struct dn_queue),
835 _SI( .enqueue = ) qfq_enqueue,
836 _SI( .dequeue = ) qfq_dequeue,
838 _SI( .config = ) NULL,
839 _SI( .destroy = ) NULL,
840 _SI( .new_sched = ) qfq_new_sched,
841 _SI( .free_sched = ) NULL,
842 _SI( .new_fsk = ) qfq_new_fsk,
843 _SI( .free_fsk = ) NULL,
844 _SI( .new_queue = ) qfq_new_queue,
845 _SI( .free_queue = ) qfq_free_queue,
847 _SI( .getconfig = ) NULL,
851 DECLARE_DNSCHED_MODULE(dn_qfq, &qfq_desc);
855 dump_groups(struct qfq_sched *q, uint32_t mask)
859 for (i = 0; i < QFQ_MAX_INDEX + 1; i++) {
860 struct qfq_group *g = &q->groups[i];
862 if (0 == (mask & (1<<i)))
864 for (j = 0; j < QFQ_MAX_SLOTS; j++) {
866 D(" bucket %d %p", j, g->slots[j]);
868 D("full_slots 0x%llx", (_P64)g->full_slots);
869 D(" %2d S 0x%20llx F 0x%llx %c", i,
870 (_P64)g->S, (_P64)g->F,
871 mask & (1<<i) ? '1' : '0');
876 dump_sched(struct qfq_sched *q, const char *msg)
878 D("--- in %s: ---", msg);
879 D("loops %d queued %d V 0x%llx", q->loops, q->queued, (_P64)q->V);
880 D(" ER 0x%08x", (unsigned)q->bitmaps[ER]);
881 D(" EB 0x%08x", (unsigned)q->bitmaps[EB]);
882 D(" IR 0x%08x", (unsigned)q->bitmaps[IR]);
883 D(" IB 0x%08x", (unsigned)q->bitmaps[IB]);
884 dump_groups(q, 0xffffffff);
886 #endif /* QFQ_DEBUG */