4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jamal Hadi Salim jhs@mojatatu.com

commit 051d442098421c28c7951625652f61b1e15c4bd5 upstream.

While this amazing qdisc has served us well over the years it has not been getting any tender love and care and has bitrotted over time. It has become mostly a shooting target for syzkaller lately. For this reason, we are retiring it. Goodbye CBQ - we loved you.

Signed-off-by: Jamal Hadi Salim jhs@mojatatu.com Acked-by: Jiri Pirko jiri@nvidia.com Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/sched/Kconfig | 17 net/sched/Makefile | 1 net/sched/sch_cbq.c | 1823 ---------------------------------------------------- 3 files changed, 1841 deletions(-) delete mode 100644 net/sched/sch_cbq.c delete mode 100644 tools/testing/selftests/tc-testing/tc-tests/qdiscs/cbq.json

--- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -44,23 +44,6 @@ if NET_SCHED

comment "Queueing/Scheduling"

-config NET_SCH_CBQ - tristate "Class Based Queueing (CBQ)" - ---help--- - Say Y here if you want to use the Class-Based Queueing (CBQ) packet - scheduling algorithm. This algorithm classifies the waiting packets - into a tree-like hierarchy of classes; the leaves of this tree are - in turn scheduled by separate algorithms. - - See the top of file:net/sched/sch_cbq.c for more details. - - CBQ is a commonly used scheduler, so if you're unsure, you should - say Y here. Then say Y to all the queueing algorithms below that you - want to use as leaf disciplines. - - To compile this code as a module, choose M here: the - module will be called sch_cbq. - config NET_SCH_HTB tristate "Hierarchical Token Bucket (HTB)" ---help--- --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -28,7 +28,6 @@ obj-$(CONFIG_NET_IFE_SKBPRIO) += act_met obj-$(CONFIG_NET_IFE_SKBTCINDEX) += act_meta_skbtcindex.o obj-$(CONFIG_NET_ACT_TUNNEL_KEY)+= act_tunnel_key.o obj-$(CONFIG_NET_SCH_FIFO) += sch_fifo.o -obj-$(CONFIG_NET_SCH_CBQ) += sch_cbq.o obj-$(CONFIG_NET_SCH_HTB) += sch_htb.o obj-$(CONFIG_NET_SCH_HFSC) += sch_hfsc.o obj-$(CONFIG_NET_SCH_RED) += sch_red.o --- a/net/sched/sch_cbq.c +++ /dev/null @@ -1,1823 +0,0 @@ -/* - * net/sched/sch_cbq.c Class-Based Queueing discipline. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * Authors: Alexey Kuznetsov, kuznet@ms2.inr.ac.ru - * - */ - -#include <linux/module.h> -#include <linux/slab.h> -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/string.h> -#include <linux/errno.h> -#include <linux/skbuff.h> -#include <net/netlink.h> -#include <net/pkt_sched.h> -#include <net/pkt_cls.h> - - -/* Class-Based Queueing (CBQ) algorithm. - ======================================= - - Sources: [1] Sally Floyd and Van Jacobson, "Link-sharing and Resource - Management Models for Packet Networks", - IEEE/ACM Transactions on Networking, Vol.3, No.4, 1995 - - [2] Sally Floyd, "Notes on CBQ and Guaranteed Service", 1995 - - [3] Sally Floyd, "Notes on Class-Based Queueing: Setting - Parameters", 1996 - - [4] Sally Floyd and Michael Speer, "Experimental Results - for Class-Based Queueing", 1998, not published. - - ----------------------------------------------------------------------- - - Algorithm skeleton was taken from NS simulator cbq.cc. - If someone wants to check this code against the LBL version, - he should take into account that ONLY the skeleton was borrowed, - the implementation is different. Particularly: - - --- The WRR algorithm is different. Our version looks more - reasonable (I hope) and works when quanta are allowed to be - less than MTU, which is always the case when real time classes - have small rates. Note, that the statement of [3] is - incomplete, delay may actually be estimated even if class - per-round allotment is less than MTU. Namely, if per-round - allotment is W*r_i, and r_1+...+r_k = r < 1 - - delay_i <= ([MTU/(W*r_i)]*W*r + W*r + k*MTU)/B - - In the worst case we have IntServ estimate with D = W*r+k*MTU - and C = MTU*r. The proof (if correct at all) is trivial. - - - --- It seems that cbq-2.0 is not very accurate. At least, I cannot - interpret some places, which look like wrong translations - from NS. Anyone is advised to find these differences - and explain to me, why I am wrong 8). - - --- Linux has no EOI event, so that we cannot estimate true class - idle time. Workaround is to consider the next dequeue event - as sign that previous packet is finished. This is wrong because of - internal device queueing, but on a permanently loaded link it is true. - Moreover, combined with clock integrator, this scheme looks - very close to an ideal solution. */ - -struct cbq_sched_data; - - -struct cbq_class { - struct Qdisc_class_common common; - struct cbq_class *next_alive; /* next class with backlog in this priority band */ - -/* Parameters */ - unsigned char priority; /* class priority */ - unsigned char priority2; /* priority to be used after overlimit */ - unsigned char ewma_log; /* time constant for idle time calculation */ - - u32 defmap; - - /* Link-sharing scheduler parameters */ - long maxidle; /* Class parameters: see below. */ - long offtime; - long minidle; - u32 avpkt; - struct qdisc_rate_table *R_tab; - - /* General scheduler (WRR) parameters */ - long allot; - long quantum; /* Allotment per WRR round */ - long weight; /* Relative allotment: see below */ - - struct Qdisc *qdisc; /* Ptr to CBQ discipline */ - struct cbq_class *split; /* Ptr to split node */ - struct cbq_class *share; /* Ptr to LS parent in the class tree */ - struct cbq_class *tparent; /* Ptr to tree parent in the class tree */ - struct cbq_class *borrow; /* NULL if class is bandwidth limited; - parent otherwise */ - struct cbq_class *sibling; /* Sibling chain */ - struct cbq_class *children; /* Pointer to children chain */ - - struct Qdisc *q; /* Elementary queueing discipline */ - - -/* Variables */ - unsigned char cpriority; /* Effective priority */ - unsigned char delayed; - unsigned char level; /* level of the class in hierarchy: - 0 for leaf classes, and maximal - level of children + 1 for nodes. - */ - - psched_time_t last; /* Last end of service */ - psched_time_t undertime; - long avgidle; - long deficit; /* Saved deficit for WRR */ - psched_time_t penalized; - struct gnet_stats_basic_packed bstats; - struct gnet_stats_queue qstats; - struct net_rate_estimator __rcu *rate_est; - struct tc_cbq_xstats xstats; - - struct tcf_proto __rcu *filter_list; - struct tcf_block *block; - - int filters; - - struct cbq_class *defaults[TC_PRIO_MAX + 1]; -}; - -struct cbq_sched_data { - struct Qdisc_class_hash clhash; /* Hash table of all classes */ - int nclasses[TC_CBQ_MAXPRIO + 1]; - unsigned int quanta[TC_CBQ_MAXPRIO + 1]; - - struct cbq_class link; - - unsigned int activemask; - struct cbq_class *active[TC_CBQ_MAXPRIO + 1]; /* List of all classes - with backlog */ - -#ifdef CONFIG_NET_CLS_ACT - struct cbq_class *rx_class; -#endif - struct cbq_class *tx_class; - struct cbq_class *tx_borrowed; - int tx_len; - psched_time_t now; /* Cached timestamp */ - unsigned int pmask; - - struct hrtimer delay_timer; - struct qdisc_watchdog watchdog; /* Watchdog timer, - started when CBQ has - backlog, but cannot - transmit just now */ - psched_tdiff_t wd_expires; - int toplevel; - u32 hgenerator; -}; - - -#define L2T(cl, len) qdisc_l2t((cl)->R_tab, len) - -static inline struct cbq_class * -cbq_class_lookup(struct cbq_sched_data *q, u32 classid) -{ - struct Qdisc_class_common *clc; - - clc = qdisc_class_find(&q->clhash, classid); - if (clc == NULL) - return NULL; - return container_of(clc, struct cbq_class, common); -} - -#ifdef CONFIG_NET_CLS_ACT - -static struct cbq_class * -cbq_reclassify(struct sk_buff *skb, struct cbq_class *this) -{ - struct cbq_class *cl; - - for (cl = this->tparent; cl; cl = cl->tparent) { - struct cbq_class *new = cl->defaults[TC_PRIO_BESTEFFORT]; - - if (new != NULL && new != this) - return new; - } - return NULL; -} - -#endif - -/* Classify packet. The procedure is pretty complicated, but - * it allows us to combine link sharing and priority scheduling - * transparently. - * - * Namely, you can put link sharing rules (f.e. route based) at root of CBQ, - * so that it resolves to split nodes. Then packets are classified - * by logical priority, or a more specific classifier may be attached - * to the split node. - */ - -static struct cbq_class * -cbq_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *head = &q->link; - struct cbq_class **defmap; - struct cbq_class *cl = NULL; - u32 prio = skb->priority; - struct tcf_proto *fl; - struct tcf_result res; - - /* - * Step 1. If skb->priority points to one of our classes, use it. - */ - if (TC_H_MAJ(prio ^ sch->handle) == 0 && - (cl = cbq_class_lookup(q, prio)) != NULL) - return cl; - - *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; - for (;;) { - int result = 0; - defmap = head->defaults; - - fl = rcu_dereference_bh(head->filter_list); - /* - * Step 2+n. Apply classifier. - */ - result = tcf_classify(skb, fl, &res, true); - if (!fl || result < 0) - goto fallback; - if (result == TC_ACT_SHOT) - return NULL; - - cl = (void *)res.class; - if (!cl) { - if (TC_H_MAJ(res.classid)) - cl = cbq_class_lookup(q, res.classid); - else if ((cl = defmap[res.classid & TC_PRIO_MAX]) == NULL) - cl = defmap[TC_PRIO_BESTEFFORT]; - - if (cl == NULL) - goto fallback; - } - if (cl->level >= head->level) - goto fallback; -#ifdef CONFIG_NET_CLS_ACT - switch (result) { - case TC_ACT_QUEUED: - case TC_ACT_STOLEN: - case TC_ACT_TRAP: - *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; - /* fall through */ - case TC_ACT_RECLASSIFY: - return cbq_reclassify(skb, cl); - } -#endif - if (cl->level == 0) - return cl; - - /* - * Step 3+n. If classifier selected a link sharing class, - * apply agency specific classifier. - * Repeat this procdure until we hit a leaf node. - */ - head = cl; - } - -fallback: - cl = head; - - /* - * Step 4. No success... - */ - if (TC_H_MAJ(prio) == 0 && - !(cl = head->defaults[prio & TC_PRIO_MAX]) && - !(cl = head->defaults[TC_PRIO_BESTEFFORT])) - return head; - - return cl; -} - -/* - * A packet has just been enqueued on the empty class. - * cbq_activate_class adds it to the tail of active class list - * of its priority band. - */ - -static inline void cbq_activate_class(struct cbq_class *cl) -{ - struct cbq_sched_data *q = qdisc_priv(cl->qdisc); - int prio = cl->cpriority; - struct cbq_class *cl_tail; - - cl_tail = q->active[prio]; - q->active[prio] = cl; - - if (cl_tail != NULL) { - cl->next_alive = cl_tail->next_alive; - cl_tail->next_alive = cl; - } else { - cl->next_alive = cl; - q->activemask |= (1<<prio); - } -} - -/* - * Unlink class from active chain. - * Note that this same procedure is done directly in cbq_dequeue* - * during round-robin procedure. - */ - -static void cbq_deactivate_class(struct cbq_class *this) -{ - struct cbq_sched_data *q = qdisc_priv(this->qdisc); - int prio = this->cpriority; - struct cbq_class *cl; - struct cbq_class *cl_prev = q->active[prio]; - - do { - cl = cl_prev->next_alive; - if (cl == this) { - cl_prev->next_alive = cl->next_alive; - cl->next_alive = NULL; - - if (cl == q->active[prio]) { - q->active[prio] = cl_prev; - if (cl == q->active[prio]) { - q->active[prio] = NULL; - q->activemask &= ~(1<<prio); - return; - } - } - return; - } - } while ((cl_prev = cl) != q->active[prio]); -} - -static void -cbq_mark_toplevel(struct cbq_sched_data *q, struct cbq_class *cl) -{ - int toplevel = q->toplevel; - - if (toplevel > cl->level) { - psched_time_t now = psched_get_time(); - - do { - if (cl->undertime < now) { - q->toplevel = cl->level; - return; - } - } while ((cl = cl->borrow) != NULL && toplevel > cl->level); - } -} - -static int -cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch, - struct sk_buff **to_free) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - int ret; - struct cbq_class *cl = cbq_classify(skb, sch, &ret); - -#ifdef CONFIG_NET_CLS_ACT - q->rx_class = cl; -#endif - if (cl == NULL) { - if (ret & __NET_XMIT_BYPASS) - qdisc_qstats_drop(sch); - __qdisc_drop(skb, to_free); - return ret; - } - - ret = qdisc_enqueue(skb, cl->q, to_free); - if (ret == NET_XMIT_SUCCESS) { - sch->q.qlen++; - cbq_mark_toplevel(q, cl); - if (!cl->next_alive) - cbq_activate_class(cl); - return ret; - } - - if (net_xmit_drop_count(ret)) { - qdisc_qstats_drop(sch); - cbq_mark_toplevel(q, cl); - cl->qstats.drops++; - } - return ret; -} - -/* Overlimit action: penalize leaf class by adding offtime */ -static void cbq_overlimit(struct cbq_class *cl) -{ - struct cbq_sched_data *q = qdisc_priv(cl->qdisc); - psched_tdiff_t delay = cl->undertime - q->now; - - if (!cl->delayed) { - delay += cl->offtime; - - /* - * Class goes to sleep, so that it will have no - * chance to work avgidle. Let's forgive it 8) - * - * BTW cbq-2.0 has a crap in this - * place, apparently they forgot to shift it by cl->ewma_log. - */ - if (cl->avgidle < 0) - delay -= (-cl->avgidle) - ((-cl->avgidle) >> cl->ewma_log); - if (cl->avgidle < cl->minidle) - cl->avgidle = cl->minidle; - if (delay <= 0) - delay = 1; - cl->undertime = q->now + delay; - - cl->xstats.overactions++; - cl->delayed = 1; - } - if (q->wd_expires == 0 || q->wd_expires > delay) - q->wd_expires = delay; - - /* Dirty work! We must schedule wakeups based on - * real available rate, rather than leaf rate, - * which may be tiny (even zero). - */ - if (q->toplevel == TC_CBQ_MAXLEVEL) { - struct cbq_class *b; - psched_tdiff_t base_delay = q->wd_expires; - - for (b = cl->borrow; b; b = b->borrow) { - delay = b->undertime - q->now; - if (delay < base_delay) { - if (delay <= 0) - delay = 1; - base_delay = delay; - } - } - - q->wd_expires = base_delay; - } -} - -static psched_tdiff_t cbq_undelay_prio(struct cbq_sched_data *q, int prio, - psched_time_t now) -{ - struct cbq_class *cl; - struct cbq_class *cl_prev = q->active[prio]; - psched_time_t sched = now; - - if (cl_prev == NULL) - return 0; - - do { - cl = cl_prev->next_alive; - if (now - cl->penalized > 0) { - cl_prev->next_alive = cl->next_alive; - cl->next_alive = NULL; - cl->cpriority = cl->priority; - cl->delayed = 0; - cbq_activate_class(cl); - - if (cl == q->active[prio]) { - q->active[prio] = cl_prev; - if (cl == q->active[prio]) { - q->active[prio] = NULL; - return 0; - } - } - - cl = cl_prev->next_alive; - } else if (sched - cl->penalized > 0) - sched = cl->penalized; - } while ((cl_prev = cl) != q->active[prio]); - - return sched - now; -} - -static enum hrtimer_restart cbq_undelay(struct hrtimer *timer) -{ - struct cbq_sched_data *q = container_of(timer, struct cbq_sched_data, - delay_timer); - struct Qdisc *sch = q->watchdog.qdisc; - psched_time_t now; - psched_tdiff_t delay = 0; - unsigned int pmask; - - now = psched_get_time(); - - pmask = q->pmask; - q->pmask = 0; - - while (pmask) { - int prio = ffz(~pmask); - psched_tdiff_t tmp; - - pmask &= ~(1<<prio); - - tmp = cbq_undelay_prio(q, prio, now); - if (tmp > 0) { - q->pmask |= 1<<prio; - if (tmp < delay || delay == 0) - delay = tmp; - } - } - - if (delay) { - ktime_t time; - - time = 0; - time = ktime_add_ns(time, PSCHED_TICKS2NS(now + delay)); - hrtimer_start(&q->delay_timer, time, HRTIMER_MODE_ABS_PINNED); - } - - __netif_schedule(qdisc_root(sch)); - return HRTIMER_NORESTART; -} - -/* - * It is mission critical procedure. - * - * We "regenerate" toplevel cutoff, if transmitting class - * has backlog and it is not regulated. It is not part of - * original CBQ description, but looks more reasonable. - * Probably, it is wrong. This question needs further investigation. - */ - -static inline void -cbq_update_toplevel(struct cbq_sched_data *q, struct cbq_class *cl, - struct cbq_class *borrowed) -{ - if (cl && q->toplevel >= borrowed->level) { - if (cl->q->q.qlen > 1) { - do { - if (borrowed->undertime == PSCHED_PASTPERFECT) { - q->toplevel = borrowed->level; - return; - } - } while ((borrowed = borrowed->borrow) != NULL); - } -#if 0 - /* It is not necessary now. Uncommenting it - will save CPU cycles, but decrease fairness. - */ - q->toplevel = TC_CBQ_MAXLEVEL; -#endif - } -} - -static void -cbq_update(struct cbq_sched_data *q) -{ - struct cbq_class *this = q->tx_class; - struct cbq_class *cl = this; - int len = q->tx_len; - psched_time_t now; - - q->tx_class = NULL; - /* Time integrator. We calculate EOS time - * by adding expected packet transmission time. - */ - now = q->now + L2T(&q->link, len); - - for ( ; cl; cl = cl->share) { - long avgidle = cl->avgidle; - long idle; - - cl->bstats.packets++; - cl->bstats.bytes += len; - - /* - * (now - last) is total time between packet right edges. - * (last_pktlen/rate) is "virtual" busy time, so that - * - * idle = (now - last) - last_pktlen/rate - */ - - idle = now - cl->last; - if ((unsigned long)idle > 128*1024*1024) { - avgidle = cl->maxidle; - } else { - idle -= L2T(cl, len); - - /* true_avgidle := (1-W)*true_avgidle + W*idle, - * where W=2^{-ewma_log}. But cl->avgidle is scaled: - * cl->avgidle == true_avgidle/W, - * hence: - */ - avgidle += idle - (avgidle>>cl->ewma_log); - } - - if (avgidle <= 0) { - /* Overlimit or at-limit */ - - if (avgidle < cl->minidle) - avgidle = cl->minidle; - - cl->avgidle = avgidle; - - /* Calculate expected time, when this class - * will be allowed to send. - * It will occur, when: - * (1-W)*true_avgidle + W*delay = 0, i.e. - * idle = (1/W - 1)*(-true_avgidle) - * or - * idle = (1 - W)*(-cl->avgidle); - */ - idle = (-avgidle) - ((-avgidle) >> cl->ewma_log); - - /* - * That is not all. - * To maintain the rate allocated to the class, - * we add to undertime virtual clock, - * necessary to complete transmitted packet. - * (len/phys_bandwidth has been already passed - * to the moment of cbq_update) - */ - - idle -= L2T(&q->link, len); - idle += L2T(cl, len); - - cl->undertime = now + idle; - } else { - /* Underlimit */ - - cl->undertime = PSCHED_PASTPERFECT; - if (avgidle > cl->maxidle) - cl->avgidle = cl->maxidle; - else - cl->avgidle = avgidle; - } - if ((s64)(now - cl->last) > 0) - cl->last = now; - } - - cbq_update_toplevel(q, this, q->tx_borrowed); -} - -static inline struct cbq_class * -cbq_under_limit(struct cbq_class *cl) -{ - struct cbq_sched_data *q = qdisc_priv(cl->qdisc); - struct cbq_class *this_cl = cl; - - if (cl->tparent == NULL) - return cl; - - if (cl->undertime == PSCHED_PASTPERFECT || q->now >= cl->undertime) { - cl->delayed = 0; - return cl; - } - - do { - /* It is very suspicious place. Now overlimit - * action is generated for not bounded classes - * only if link is completely congested. - * Though it is in agree with ancestor-only paradigm, - * it looks very stupid. Particularly, - * it means that this chunk of code will either - * never be called or result in strong amplification - * of burstiness. Dangerous, silly, and, however, - * no another solution exists. - */ - cl = cl->borrow; - if (!cl) { - this_cl->qstats.overlimits++; - cbq_overlimit(this_cl); - return NULL; - } - if (cl->level > q->toplevel) - return NULL; - } while (cl->undertime != PSCHED_PASTPERFECT && q->now < cl->undertime); - - cl->delayed = 0; - return cl; -} - -static inline struct sk_buff * -cbq_dequeue_prio(struct Qdisc *sch, int prio) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *cl_tail, *cl_prev, *cl; - struct sk_buff *skb; - int deficit; - - cl_tail = cl_prev = q->active[prio]; - cl = cl_prev->next_alive; - - do { - deficit = 0; - - /* Start round */ - do { - struct cbq_class *borrow = cl; - - if (cl->q->q.qlen && - (borrow = cbq_under_limit(cl)) == NULL) - goto skip_class; - - if (cl->deficit <= 0) { - /* Class exhausted its allotment per - * this round. Switch to the next one. - */ - deficit = 1; - cl->deficit += cl->quantum; - goto next_class; - } - - skb = cl->q->dequeue(cl->q); - - /* Class did not give us any skb :-( - * It could occur even if cl->q->q.qlen != 0 - * f.e. if cl->q == "tbf" - */ - if (skb == NULL) - goto skip_class; - - cl->deficit -= qdisc_pkt_len(skb); - q->tx_class = cl; - q->tx_borrowed = borrow; - if (borrow != cl) { -#ifndef CBQ_XSTATS_BORROWS_BYTES - borrow->xstats.borrows++; - cl->xstats.borrows++; -#else - borrow->xstats.borrows += qdisc_pkt_len(skb); - cl->xstats.borrows += qdisc_pkt_len(skb); -#endif - } - q->tx_len = qdisc_pkt_len(skb); - - if (cl->deficit <= 0) { - q->active[prio] = cl; - cl = cl->next_alive; - cl->deficit += cl->quantum; - } - return skb; - -skip_class: - if (cl->q->q.qlen == 0 || prio != cl->cpriority) { - /* Class is empty or penalized. - * Unlink it from active chain. - */ - cl_prev->next_alive = cl->next_alive; - cl->next_alive = NULL; - - /* Did cl_tail point to it? */ - if (cl == cl_tail) { - /* Repair it! */ - cl_tail = cl_prev; - - /* Was it the last class in this band? */ - if (cl == cl_tail) { - /* Kill the band! */ - q->active[prio] = NULL; - q->activemask &= ~(1<<prio); - if (cl->q->q.qlen) - cbq_activate_class(cl); - return NULL; - } - - q->active[prio] = cl_tail; - } - if (cl->q->q.qlen) - cbq_activate_class(cl); - - cl = cl_prev; - } - -next_class: - cl_prev = cl; - cl = cl->next_alive; - } while (cl_prev != cl_tail); - } while (deficit); - - q->active[prio] = cl_prev; - - return NULL; -} - -static inline struct sk_buff * -cbq_dequeue_1(struct Qdisc *sch) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct sk_buff *skb; - unsigned int activemask; - - activemask = q->activemask & 0xFF; - while (activemask) { - int prio = ffz(~activemask); - activemask &= ~(1<<prio); - skb = cbq_dequeue_prio(sch, prio); - if (skb) - return skb; - } - return NULL; -} - -static struct sk_buff * -cbq_dequeue(struct Qdisc *sch) -{ - struct sk_buff *skb; - struct cbq_sched_data *q = qdisc_priv(sch); - psched_time_t now; - - now = psched_get_time(); - - if (q->tx_class) - cbq_update(q); - - q->now = now; - - for (;;) { - q->wd_expires = 0; - - skb = cbq_dequeue_1(sch); - if (skb) { - qdisc_bstats_update(sch, skb); - sch->q.qlen--; - return skb; - } - - /* All the classes are overlimit. - * - * It is possible, if: - * - * 1. Scheduler is empty. - * 2. Toplevel cutoff inhibited borrowing. - * 3. Root class is overlimit. - * - * Reset 2d and 3d conditions and retry. - * - * Note, that NS and cbq-2.0 are buggy, peeking - * an arbitrary class is appropriate for ancestor-only - * sharing, but not for toplevel algorithm. - * - * Our version is better, but slower, because it requires - * two passes, but it is unavoidable with top-level sharing. - */ - - if (q->toplevel == TC_CBQ_MAXLEVEL && - q->link.undertime == PSCHED_PASTPERFECT) - break; - - q->toplevel = TC_CBQ_MAXLEVEL; - q->link.undertime = PSCHED_PASTPERFECT; - } - - /* No packets in scheduler or nobody wants to give them to us :-( - * Sigh... start watchdog timer in the last case. - */ - - if (sch->q.qlen) { - qdisc_qstats_overlimit(sch); - if (q->wd_expires) - qdisc_watchdog_schedule(&q->watchdog, - now + q->wd_expires); - } - return NULL; -} - -/* CBQ class maintanance routines */ - -static void cbq_adjust_levels(struct cbq_class *this) -{ - if (this == NULL) - return; - - do { - int level = 0; - struct cbq_class *cl; - - cl = this->children; - if (cl) { - do { - if (cl->level > level) - level = cl->level; - } while ((cl = cl->sibling) != this->children); - } - this->level = level + 1; - } while ((this = this->tparent) != NULL); -} - -static void cbq_normalize_quanta(struct cbq_sched_data *q, int prio) -{ - struct cbq_class *cl; - unsigned int h; - - if (q->quanta[prio] == 0) - return; - - for (h = 0; h < q->clhash.hashsize; h++) { - hlist_for_each_entry(cl, &q->clhash.hash[h], common.hnode) { - /* BUGGGG... Beware! This expression suffer of - * arithmetic overflows! - */ - if (cl->priority == prio) { - cl->quantum = (cl->weight*cl->allot*q->nclasses[prio])/ - q->quanta[prio]; - } - if (cl->quantum <= 0 || - cl->quantum > 32*qdisc_dev(cl->qdisc)->mtu) { - pr_warn("CBQ: class %08x has bad quantum==%ld, repaired.\n", - cl->common.classid, cl->quantum); - cl->quantum = qdisc_dev(cl->qdisc)->mtu/2 + 1; - } - } - } -} - -static void cbq_sync_defmap(struct cbq_class *cl) -{ - struct cbq_sched_data *q = qdisc_priv(cl->qdisc); - struct cbq_class *split = cl->split; - unsigned int h; - int i; - - if (split == NULL) - return; - - for (i = 0; i <= TC_PRIO_MAX; i++) { - if (split->defaults[i] == cl && !(cl->defmap & (1<<i))) - split->defaults[i] = NULL; - } - - for (i = 0; i <= TC_PRIO_MAX; i++) { - int level = split->level; - - if (split->defaults[i]) - continue; - - for (h = 0; h < q->clhash.hashsize; h++) { - struct cbq_class *c; - - hlist_for_each_entry(c, &q->clhash.hash[h], - common.hnode) { - if (c->split == split && c->level < level && - c->defmap & (1<<i)) { - split->defaults[i] = c; - level = c->level; - } - } - } - } -} - -static void cbq_change_defmap(struct cbq_class *cl, u32 splitid, u32 def, u32 mask) -{ - struct cbq_class *split = NULL; - - if (splitid == 0) { - split = cl->split; - if (!split) - return; - splitid = split->common.classid; - } - - if (split == NULL || split->common.classid != splitid) { - for (split = cl->tparent; split; split = split->tparent) - if (split->common.classid == splitid) - break; - } - - if (split == NULL) - return; - - if (cl->split != split) { - cl->defmap = 0; - cbq_sync_defmap(cl); - cl->split = split; - cl->defmap = def & mask; - } else - cl->defmap = (cl->defmap & ~mask) | (def & mask); - - cbq_sync_defmap(cl); -} - -static void cbq_unlink_class(struct cbq_class *this) -{ - struct cbq_class *cl, **clp; - struct cbq_sched_data *q = qdisc_priv(this->qdisc); - - qdisc_class_hash_remove(&q->clhash, &this->common); - - if (this->tparent) { - clp = &this->sibling; - cl = *clp; - do { - if (cl == this) { - *clp = cl->sibling; - break; - } - clp = &cl->sibling; - } while ((cl = *clp) != this->sibling); - - if (this->tparent->children == this) { - this->tparent->children = this->sibling; - if (this->sibling == this) - this->tparent->children = NULL; - } - } else { - WARN_ON(this->sibling != this); - } -} - -static void cbq_link_class(struct cbq_class *this) -{ - struct cbq_sched_data *q = qdisc_priv(this->qdisc); - struct cbq_class *parent = this->tparent; - - this->sibling = this; - qdisc_class_hash_insert(&q->clhash, &this->common); - - if (parent == NULL) - return; - - if (parent->children == NULL) { - parent->children = this; - } else { - this->sibling = parent->children->sibling; - parent->children->sibling = this; - } -} - -static void -cbq_reset(struct Qdisc *sch) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *cl; - int prio; - unsigned int h; - - q->activemask = 0; - q->pmask = 0; - q->tx_class = NULL; - q->tx_borrowed = NULL; - qdisc_watchdog_cancel(&q->watchdog); - hrtimer_cancel(&q->delay_timer); - q->toplevel = TC_CBQ_MAXLEVEL; - q->now = psched_get_time(); - - for (prio = 0; prio <= TC_CBQ_MAXPRIO; prio++) - q->active[prio] = NULL; - - for (h = 0; h < q->clhash.hashsize; h++) { - hlist_for_each_entry(cl, &q->clhash.hash[h], common.hnode) { - qdisc_reset(cl->q); - - cl->next_alive = NULL; - cl->undertime = PSCHED_PASTPERFECT; - cl->avgidle = cl->maxidle; - cl->deficit = cl->quantum; - cl->cpriority = cl->priority; - } - } - sch->q.qlen = 0; -} - - -static int cbq_set_lss(struct cbq_class *cl, struct tc_cbq_lssopt *lss) -{ - if (lss->change & TCF_CBQ_LSS_FLAGS) { - cl->share = (lss->flags & TCF_CBQ_LSS_ISOLATED) ? NULL : cl->tparent; - cl->borrow = (lss->flags & TCF_CBQ_LSS_BOUNDED) ? NULL : cl->tparent; - } - if (lss->change & TCF_CBQ_LSS_EWMA) - cl->ewma_log = lss->ewma_log; - if (lss->change & TCF_CBQ_LSS_AVPKT) - cl->avpkt = lss->avpkt; - if (lss->change & TCF_CBQ_LSS_MINIDLE) - cl->minidle = -(long)lss->minidle; - if (lss->change & TCF_CBQ_LSS_MAXIDLE) { - cl->maxidle = lss->maxidle; - cl->avgidle = lss->maxidle; - } - if (lss->change & TCF_CBQ_LSS_OFFTIME) - cl->offtime = lss->offtime; - return 0; -} - -static void cbq_rmprio(struct cbq_sched_data *q, struct cbq_class *cl) -{ - q->nclasses[cl->priority]--; - q->quanta[cl->priority] -= cl->weight; - cbq_normalize_quanta(q, cl->priority); -} - -static void cbq_addprio(struct cbq_sched_data *q, struct cbq_class *cl) -{ - q->nclasses[cl->priority]++; - q->quanta[cl->priority] += cl->weight; - cbq_normalize_quanta(q, cl->priority); -} - -static int cbq_set_wrr(struct cbq_class *cl, struct tc_cbq_wrropt *wrr) -{ - struct cbq_sched_data *q = qdisc_priv(cl->qdisc); - - if (wrr->allot) - cl->allot = wrr->allot; - if (wrr->weight) - cl->weight = wrr->weight; - if (wrr->priority) { - cl->priority = wrr->priority - 1; - cl->cpriority = cl->priority; - if (cl->priority >= cl->priority2) - cl->priority2 = TC_CBQ_MAXPRIO - 1; - } - - cbq_addprio(q, cl); - return 0; -} - -static int cbq_set_fopt(struct cbq_class *cl, struct tc_cbq_fopt *fopt) -{ - cbq_change_defmap(cl, fopt->split, fopt->defmap, fopt->defchange); - return 0; -} - -static const struct nla_policy cbq_policy[TCA_CBQ_MAX + 1] = { - [TCA_CBQ_LSSOPT] = { .len = sizeof(struct tc_cbq_lssopt) }, - [TCA_CBQ_WRROPT] = { .len = sizeof(struct tc_cbq_wrropt) }, - [TCA_CBQ_FOPT] = { .len = sizeof(struct tc_cbq_fopt) }, - [TCA_CBQ_OVL_STRATEGY] = { .len = sizeof(struct tc_cbq_ovl) }, - [TCA_CBQ_RATE] = { .len = sizeof(struct tc_ratespec) }, - [TCA_CBQ_RTAB] = { .type = NLA_BINARY, .len = TC_RTAB_SIZE }, - [TCA_CBQ_POLICE] = { .len = sizeof(struct tc_cbq_police) }, -}; - -static int cbq_opt_parse(struct nlattr *tb[TCA_CBQ_MAX + 1], - struct nlattr *opt, - struct netlink_ext_ack *extack) -{ - int err; - - if (!opt) { - NL_SET_ERR_MSG(extack, "CBQ options are required for this operation"); - return -EINVAL; - } - - err = nla_parse_nested(tb, TCA_CBQ_MAX, opt, cbq_policy, extack); - if (err < 0) - return err; - - if (tb[TCA_CBQ_WRROPT]) { - const struct tc_cbq_wrropt *wrr = nla_data(tb[TCA_CBQ_WRROPT]); - - if (wrr->priority > TC_CBQ_MAXPRIO) { - NL_SET_ERR_MSG(extack, "priority is bigger than TC_CBQ_MAXPRIO"); - err = -EINVAL; - } - } - return err; -} - -static int cbq_init(struct Qdisc *sch, struct nlattr *opt, - struct netlink_ext_ack *extack) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct nlattr *tb[TCA_CBQ_MAX + 1]; - struct tc_ratespec *r; - int err; - - qdisc_watchdog_init(&q->watchdog, sch); - hrtimer_init(&q->delay_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); - q->delay_timer.function = cbq_undelay; - - err = cbq_opt_parse(tb, opt, extack); - if (err < 0) - return err; - - if (!tb[TCA_CBQ_RTAB] || !tb[TCA_CBQ_RATE]) { - NL_SET_ERR_MSG(extack, "Rate specification missing or incomplete"); - return -EINVAL; - } - - r = nla_data(tb[TCA_CBQ_RATE]); - - q->link.R_tab = qdisc_get_rtab(r, tb[TCA_CBQ_RTAB], extack); - if (!q->link.R_tab) - return -EINVAL; - - err = tcf_block_get(&q->link.block, &q->link.filter_list, sch, extack); - if (err) - goto put_rtab; - - err = qdisc_class_hash_init(&q->clhash); - if (err < 0) - goto put_block; - - q->link.sibling = &q->link; - q->link.common.classid = sch->handle; - q->link.qdisc = sch; - q->link.q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, - sch->handle, NULL); - if (!q->link.q) - q->link.q = &noop_qdisc; - else - qdisc_hash_add(q->link.q, true); - - q->link.priority = TC_CBQ_MAXPRIO - 1; - q->link.priority2 = TC_CBQ_MAXPRIO - 1; - q->link.cpriority = TC_CBQ_MAXPRIO - 1; - q->link.allot = psched_mtu(qdisc_dev(sch)); - q->link.quantum = q->link.allot; - q->link.weight = q->link.R_tab->rate.rate; - - q->link.ewma_log = TC_CBQ_DEF_EWMA; - q->link.avpkt = q->link.allot/2; - q->link.minidle = -0x7FFFFFFF; - - q->toplevel = TC_CBQ_MAXLEVEL; - q->now = psched_get_time(); - - cbq_link_class(&q->link); - - if (tb[TCA_CBQ_LSSOPT]) - cbq_set_lss(&q->link, nla_data(tb[TCA_CBQ_LSSOPT])); - - cbq_addprio(q, &q->link); - return 0; - -put_block: - tcf_block_put(q->link.block); - -put_rtab: - qdisc_put_rtab(q->link.R_tab); - return err; -} - -static int cbq_dump_rate(struct sk_buff *skb, struct cbq_class *cl) -{ - unsigned char *b = skb_tail_pointer(skb); - - if (nla_put(skb, TCA_CBQ_RATE, sizeof(cl->R_tab->rate), &cl->R_tab->rate)) - goto nla_put_failure; - return skb->len; - -nla_put_failure: - nlmsg_trim(skb, b); - return -1; -} - -static int cbq_dump_lss(struct sk_buff *skb, struct cbq_class *cl) -{ - unsigned char *b = skb_tail_pointer(skb); - struct tc_cbq_lssopt opt; - - opt.flags = 0; - if (cl->borrow == NULL) - opt.flags |= TCF_CBQ_LSS_BOUNDED; - if (cl->share == NULL) - opt.flags |= TCF_CBQ_LSS_ISOLATED; - opt.ewma_log = cl->ewma_log; - opt.level = cl->level; - opt.avpkt = cl->avpkt; - opt.maxidle = cl->maxidle; - opt.minidle = (u32)(-cl->minidle); - opt.offtime = cl->offtime; - opt.change = ~0; - if (nla_put(skb, TCA_CBQ_LSSOPT, sizeof(opt), &opt)) - goto nla_put_failure; - return skb->len; - -nla_put_failure: - nlmsg_trim(skb, b); - return -1; -} - -static int cbq_dump_wrr(struct sk_buff *skb, struct cbq_class *cl) -{ - unsigned char *b = skb_tail_pointer(skb); - struct tc_cbq_wrropt opt; - - memset(&opt, 0, sizeof(opt)); - opt.flags = 0; - opt.allot = cl->allot; - opt.priority = cl->priority + 1; - opt.cpriority = cl->cpriority + 1; - opt.weight = cl->weight; - if (nla_put(skb, TCA_CBQ_WRROPT, sizeof(opt), &opt)) - goto nla_put_failure; - return skb->len; - -nla_put_failure: - nlmsg_trim(skb, b); - return -1; -} - -static int cbq_dump_fopt(struct sk_buff *skb, struct cbq_class *cl) -{ - unsigned char *b = skb_tail_pointer(skb); - struct tc_cbq_fopt opt; - - if (cl->split || cl->defmap) { - opt.split = cl->split ? cl->split->common.classid : 0; - opt.defmap = cl->defmap; - opt.defchange = ~0; - if (nla_put(skb, TCA_CBQ_FOPT, sizeof(opt), &opt)) - goto nla_put_failure; - } - return skb->len; - -nla_put_failure: - nlmsg_trim(skb, b); - return -1; -} - -static int cbq_dump_attr(struct sk_buff *skb, struct cbq_class *cl) -{ - if (cbq_dump_lss(skb, cl) < 0 || - cbq_dump_rate(skb, cl) < 0 || - cbq_dump_wrr(skb, cl) < 0 || - cbq_dump_fopt(skb, cl) < 0) - return -1; - return 0; -} - -static int cbq_dump(struct Qdisc *sch, struct sk_buff *skb) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct nlattr *nest; - - nest = nla_nest_start(skb, TCA_OPTIONS); - if (nest == NULL) - goto nla_put_failure; - if (cbq_dump_attr(skb, &q->link) < 0) - goto nla_put_failure; - return nla_nest_end(skb, nest); - -nla_put_failure: - nla_nest_cancel(skb, nest); - return -1; -} - -static int -cbq_dump_stats(struct Qdisc *sch, struct gnet_dump *d) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - - q->link.xstats.avgidle = q->link.avgidle; - return gnet_stats_copy_app(d, &q->link.xstats, sizeof(q->link.xstats)); -} - -static int -cbq_dump_class(struct Qdisc *sch, unsigned long arg, - struct sk_buff *skb, struct tcmsg *tcm) -{ - struct cbq_class *cl = (struct cbq_class *)arg; - struct nlattr *nest; - - if (cl->tparent) - tcm->tcm_parent = cl->tparent->common.classid; - else - tcm->tcm_parent = TC_H_ROOT; - tcm->tcm_handle = cl->common.classid; - tcm->tcm_info = cl->q->handle; - - nest = nla_nest_start(skb, TCA_OPTIONS); - if (nest == NULL) - goto nla_put_failure; - if (cbq_dump_attr(skb, cl) < 0) - goto nla_put_failure; - return nla_nest_end(skb, nest); - -nla_put_failure: - nla_nest_cancel(skb, nest); - return -1; -} - -static int -cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg, - struct gnet_dump *d) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *cl = (struct cbq_class *)arg; - - cl->xstats.avgidle = cl->avgidle; - cl->xstats.undertime = 0; - - if (cl->undertime != PSCHED_PASTPERFECT) - cl->xstats.undertime = cl->undertime - q->now; - - if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), - d, NULL, &cl->bstats) < 0 || - gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 || - gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->q->q.qlen) < 0) - return -1; - - return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats)); -} - -static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, - struct Qdisc **old, struct netlink_ext_ack *extack) -{ - struct cbq_class *cl = (struct cbq_class *)arg; - - if (new == NULL) { - new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, - cl->common.classid, extack); - if (new == NULL) - return -ENOBUFS; - } - - *old = qdisc_replace(sch, new, &cl->q); - return 0; -} - -static struct Qdisc *cbq_leaf(struct Qdisc *sch, unsigned long arg) -{ - struct cbq_class *cl = (struct cbq_class *)arg; - - return cl->q; -} - -static void cbq_qlen_notify(struct Qdisc *sch, unsigned long arg) -{ - struct cbq_class *cl = (struct cbq_class *)arg; - - cbq_deactivate_class(cl); -} - -static unsigned long cbq_find(struct Qdisc *sch, u32 classid) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - - return (unsigned long)cbq_class_lookup(q, classid); -} - -static void cbq_destroy_class(struct Qdisc *sch, struct cbq_class *cl) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - - WARN_ON(cl->filters); - - tcf_block_put(cl->block); - qdisc_put(cl->q); - qdisc_put_rtab(cl->R_tab); - gen_kill_estimator(&cl->rate_est); - if (cl != &q->link) - kfree(cl); -} - -static void cbq_destroy(struct Qdisc *sch) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct hlist_node *next; - struct cbq_class *cl; - unsigned int h; - -#ifdef CONFIG_NET_CLS_ACT - q->rx_class = NULL; -#endif - /* - * Filters must be destroyed first because we don't destroy the - * classes from root to leafs which means that filters can still - * be bound to classes which have been destroyed already. --TGR '04 - */ - for (h = 0; h < q->clhash.hashsize; h++) { - hlist_for_each_entry(cl, &q->clhash.hash[h], common.hnode) { - tcf_block_put(cl->block); - cl->block = NULL; - } - } - for (h = 0; h < q->clhash.hashsize; h++) { - hlist_for_each_entry_safe(cl, next, &q->clhash.hash[h], - common.hnode) - cbq_destroy_class(sch, cl); - } - qdisc_class_hash_destroy(&q->clhash); -} - -static int -cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **tca, - unsigned long *arg, struct netlink_ext_ack *extack) -{ - int err; - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *cl = (struct cbq_class *)*arg; - struct nlattr *opt = tca[TCA_OPTIONS]; - struct nlattr *tb[TCA_CBQ_MAX + 1]; - struct cbq_class *parent; - struct qdisc_rate_table *rtab = NULL; - - err = cbq_opt_parse(tb, opt, extack); - if (err < 0) - return err; - - if (tb[TCA_CBQ_OVL_STRATEGY] || tb[TCA_CBQ_POLICE]) { - NL_SET_ERR_MSG(extack, "Neither overlimit strategy nor policing attributes can be used for changing class params"); - return -EOPNOTSUPP; - } - - if (cl) { - /* Check parent */ - if (parentid) { - if (cl->tparent && - cl->tparent->common.classid != parentid) { - NL_SET_ERR_MSG(extack, "Invalid parent id"); - return -EINVAL; - } - if (!cl->tparent && parentid != TC_H_ROOT) { - NL_SET_ERR_MSG(extack, "Parent must be root"); - return -EINVAL; - } - } - - if (tb[TCA_CBQ_RATE]) { - rtab = qdisc_get_rtab(nla_data(tb[TCA_CBQ_RATE]), - tb[TCA_CBQ_RTAB], extack); - if (rtab == NULL) - return -EINVAL; - } - - if (tca[TCA_RATE]) { - err = gen_replace_estimator(&cl->bstats, NULL, - &cl->rate_est, - NULL, - qdisc_root_sleeping_running(sch), - tca[TCA_RATE]); - if (err) { - NL_SET_ERR_MSG(extack, "Failed to replace specified rate estimator"); - qdisc_put_rtab(rtab); - return err; - } - } - - /* Change class parameters */ - sch_tree_lock(sch); - - if (cl->next_alive != NULL) - cbq_deactivate_class(cl); - - if (rtab) { - qdisc_put_rtab(cl->R_tab); - cl->R_tab = rtab; - } - - if (tb[TCA_CBQ_LSSOPT]) - cbq_set_lss(cl, nla_data(tb[TCA_CBQ_LSSOPT])); - - if (tb[TCA_CBQ_WRROPT]) { - cbq_rmprio(q, cl); - cbq_set_wrr(cl, nla_data(tb[TCA_CBQ_WRROPT])); - } - - if (tb[TCA_CBQ_FOPT]) - cbq_set_fopt(cl, nla_data(tb[TCA_CBQ_FOPT])); - - if (cl->q->q.qlen) - cbq_activate_class(cl); - - sch_tree_unlock(sch); - - return 0; - } - - if (parentid == TC_H_ROOT) - return -EINVAL; - - if (!tb[TCA_CBQ_WRROPT] || !tb[TCA_CBQ_RATE] || !tb[TCA_CBQ_LSSOPT]) { - NL_SET_ERR_MSG(extack, "One of the following attributes MUST be specified: WRR, rate or link sharing"); - return -EINVAL; - } - - rtab = qdisc_get_rtab(nla_data(tb[TCA_CBQ_RATE]), tb[TCA_CBQ_RTAB], - extack); - if (rtab == NULL) - return -EINVAL; - - if (classid) { - err = -EINVAL; - if (TC_H_MAJ(classid ^ sch->handle) || - cbq_class_lookup(q, classid)) { - NL_SET_ERR_MSG(extack, "Specified class not found"); - goto failure; - } - } else { - int i; - classid = TC_H_MAKE(sch->handle, 0x8000); - - for (i = 0; i < 0x8000; i++) { - if (++q->hgenerator >= 0x8000) - q->hgenerator = 1; - if (cbq_class_lookup(q, classid|q->hgenerator) == NULL) - break; - } - err = -ENOSR; - if (i >= 0x8000) { - NL_SET_ERR_MSG(extack, "Unable to generate classid"); - goto failure; - } - classid = classid|q->hgenerator; - } - - parent = &q->link; - if (parentid) { - parent = cbq_class_lookup(q, parentid); - err = -EINVAL; - if (!parent) { - NL_SET_ERR_MSG(extack, "Failed to find parentid"); - goto failure; - } - } - - err = -ENOBUFS; - cl = kzalloc(sizeof(*cl), GFP_KERNEL); - if (cl == NULL) - goto failure; - - err = tcf_block_get(&cl->block, &cl->filter_list, sch, extack); - if (err) { - kfree(cl); - goto failure; - } - - if (tca[TCA_RATE]) { - err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est, - NULL, - qdisc_root_sleeping_running(sch), - tca[TCA_RATE]); - if (err) { - NL_SET_ERR_MSG(extack, "Couldn't create new estimator"); - tcf_block_put(cl->block); - kfree(cl); - goto failure; - } - } - - cl->R_tab = rtab; - rtab = NULL; - cl->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid, - NULL); - if (!cl->q) - cl->q = &noop_qdisc; - else - qdisc_hash_add(cl->q, true); - - cl->common.classid = classid; - cl->tparent = parent; - cl->qdisc = sch; - cl->allot = parent->allot; - cl->quantum = cl->allot; - cl->weight = cl->R_tab->rate.rate; - - sch_tree_lock(sch); - cbq_link_class(cl); - cl->borrow = cl->tparent; - if (cl->tparent != &q->link) - cl->share = cl->tparent; - cbq_adjust_levels(parent); - cl->minidle = -0x7FFFFFFF; - cbq_set_lss(cl, nla_data(tb[TCA_CBQ_LSSOPT])); - cbq_set_wrr(cl, nla_data(tb[TCA_CBQ_WRROPT])); - if (cl->ewma_log == 0) - cl->ewma_log = q->link.ewma_log; - if (cl->maxidle == 0) - cl->maxidle = q->link.maxidle; - if (cl->avpkt == 0) - cl->avpkt = q->link.avpkt; - if (tb[TCA_CBQ_FOPT]) - cbq_set_fopt(cl, nla_data(tb[TCA_CBQ_FOPT])); - sch_tree_unlock(sch); - - qdisc_class_hash_grow(sch, &q->clhash); - - *arg = (unsigned long)cl; - return 0; - -failure: - qdisc_put_rtab(rtab); - return err; -} - -static int cbq_delete(struct Qdisc *sch, unsigned long arg) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *cl = (struct cbq_class *)arg; - unsigned int qlen, backlog; - - if (cl->filters || cl->children || cl == &q->link) - return -EBUSY; - - sch_tree_lock(sch); - - qlen = cl->q->q.qlen; - backlog = cl->q->qstats.backlog; - qdisc_reset(cl->q); - qdisc_tree_reduce_backlog(cl->q, qlen, backlog); - - if (cl->next_alive) - cbq_deactivate_class(cl); - - if (q->tx_borrowed == cl) - q->tx_borrowed = q->tx_class; - if (q->tx_class == cl) { - q->tx_class = NULL; - q->tx_borrowed = NULL; - } -#ifdef CONFIG_NET_CLS_ACT - if (q->rx_class == cl) - q->rx_class = NULL; -#endif - - cbq_unlink_class(cl); - cbq_adjust_levels(cl->tparent); - cl->defmap = 0; - cbq_sync_defmap(cl); - - cbq_rmprio(q, cl); - sch_tree_unlock(sch); - - cbq_destroy_class(sch, cl); - return 0; -} - -static struct tcf_block *cbq_tcf_block(struct Qdisc *sch, unsigned long arg, - struct netlink_ext_ack *extack) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *cl = (struct cbq_class *)arg; - - if (cl == NULL) - cl = &q->link; - - return cl->block; -} - -static unsigned long cbq_bind_filter(struct Qdisc *sch, unsigned long parent, - u32 classid) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *p = (struct cbq_class *)parent; - struct cbq_class *cl = cbq_class_lookup(q, classid); - - if (cl) { - if (p && p->level <= cl->level) - return 0; - cl->filters++; - return (unsigned long)cl; - } - return 0; -} - -static void cbq_unbind_filter(struct Qdisc *sch, unsigned long arg) -{ - struct cbq_class *cl = (struct cbq_class *)arg; - - cl->filters--; -} - -static void cbq_walk(struct Qdisc *sch, struct qdisc_walker *arg) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *cl; - unsigned int h; - - if (arg->stop) - return; - - for (h = 0; h < q->clhash.hashsize; h++) { - hlist_for_each_entry(cl, &q->clhash.hash[h], common.hnode) { - if (arg->count < arg->skip) { - arg->count++; - continue; - } - if (arg->fn(sch, (unsigned long)cl, arg) < 0) { - arg->stop = 1; - return; - } - arg->count++; - } - } -} - -static const struct Qdisc_class_ops cbq_class_ops = { - .graft = cbq_graft, - .leaf = cbq_leaf, - .qlen_notify = cbq_qlen_notify, - .find = cbq_find, - .change = cbq_change_class, - .delete = cbq_delete, - .walk = cbq_walk, - .tcf_block = cbq_tcf_block, - .bind_tcf = cbq_bind_filter, - .unbind_tcf = cbq_unbind_filter, - .dump = cbq_dump_class, - .dump_stats = cbq_dump_class_stats, -}; - -static struct Qdisc_ops cbq_qdisc_ops __read_mostly = { - .next = NULL, - .cl_ops = &cbq_class_ops, - .id = "cbq", - .priv_size = sizeof(struct cbq_sched_data), - .enqueue = cbq_enqueue, - .dequeue = cbq_dequeue, - .peek = qdisc_peek_dequeued, - .init = cbq_init, - .reset = cbq_reset, - .destroy = cbq_destroy, - .change = NULL, - .dump = cbq_dump, - .dump_stats = cbq_dump_stats, - .owner = THIS_MODULE, -}; - -static int __init cbq_module_init(void) -{ - return register_qdisc(&cbq_qdisc_ops); -} -static void __exit cbq_module_exit(void) -{ - unregister_qdisc(&cbq_qdisc_ops); -} -module_init(cbq_module_init) -module_exit(cbq_module_exit) -MODULE_LICENSE("GPL");

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jamal Hadi Salim jhs@mojatatu.com

commit bbe77c14ee6185a61ba6d5e435c1cbb489d2a9ed upstream.

The dsmark qdisc has served us well over the years for diffserv but has not been getting much attention due to other more popular approaches to do diffserv services. Most recently it has become a shooting target for syzkaller. For this reason, we are retiring it.

Signed-off-by: Jamal Hadi Salim jhs@mojatatu.com Acked-by: Jiri Pirko jiri@nvidia.com Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/sched/Kconfig | 11 - net/sched/Makefile | 1 net/sched/sch_dsmark.c | 519 ------------------------------------------------- 3 files changed, 531 deletions(-) delete mode 100644 net/sched/sch_dsmark.c delete mode 100644 tools/testing/selftests/tc-testing/tc-tests/qdiscs/dsmark.json

--- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -174,17 +174,6 @@ config NET_SCH_GRED To compile this code as a module, choose M here: the module will be called sch_gred.

-config NET_SCH_DSMARK - tristate "Differentiated Services marker (DSMARK)" - ---help--- - Say Y if you want to schedule packets according to the - Differentiated Services architecture proposed in RFC 2475. - Technical information on this method, with pointers to associated - RFCs, is available at http://www.gta.ufrj.br/diffserv/. - - To compile this code as a module, choose M here: the - module will be called sch_dsmark. - config NET_SCH_NETEM tristate "Network emulator (NETEM)" ---help--- --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -33,7 +33,6 @@ obj-$(CONFIG_NET_SCH_HFSC) += sch_hfsc.o obj-$(CONFIG_NET_SCH_RED) += sch_red.o obj-$(CONFIG_NET_SCH_GRED) += sch_gred.o obj-$(CONFIG_NET_SCH_INGRESS) += sch_ingress.o -obj-$(CONFIG_NET_SCH_DSMARK) += sch_dsmark.o obj-$(CONFIG_NET_SCH_SFB) += sch_sfb.o obj-$(CONFIG_NET_SCH_SFQ) += sch_sfq.o obj-$(CONFIG_NET_SCH_TBF) += sch_tbf.o --- a/net/sched/sch_dsmark.c +++ /dev/null @@ -1,519 +0,0 @@ -/* net/sched/sch_dsmark.c - Differentiated Services field marker */ - -/* Written 1998-2000 by Werner Almesberger, EPFL ICA */ - - -#include <linux/module.h> -#include <linux/init.h> -#include <linux/slab.h> -#include <linux/types.h> -#include <linux/string.h> -#include <linux/errno.h> -#include <linux/skbuff.h> -#include <linux/rtnetlink.h> -#include <linux/bitops.h> -#include <net/pkt_sched.h> -#include <net/pkt_cls.h> -#include <net/dsfield.h> -#include <net/inet_ecn.h> -#include <asm/byteorder.h> - -/* - * classid class marking - * ------- ----- ------- - * n/a 0 n/a - * x:0 1 use entry [0] - * ... ... ... - * x:y y>0 y+1 use entry [y] - * ... ... ... - * x:indices-1 indices use entry [indices-1] - * ... ... ... - * x:y y+1 use entry [y & (indices-1)] - * ... ... ... - * 0xffff 0x10000 use entry [indices-1] - */ - - -#define NO_DEFAULT_INDEX (1 << 16) - -struct mask_value { - u8 mask; - u8 value; -}; - -struct dsmark_qdisc_data { - struct Qdisc *q; - struct tcf_proto __rcu *filter_list; - struct tcf_block *block; - struct mask_value *mv; - u16 indices; - u8 set_tc_index; - u32 default_index; /* index range is 0...0xffff */ -#define DSMARK_EMBEDDED_SZ 16 - struct mask_value embedded[DSMARK_EMBEDDED_SZ]; -}; - -static inline int dsmark_valid_index(struct dsmark_qdisc_data *p, u16 index) -{ - return index <= p->indices && index > 0; -} - -/* ------------------------- Class/flow operations ------------------------- */ - -static int dsmark_graft(struct Qdisc *sch, unsigned long arg, - struct Qdisc *new, struct Qdisc **old, - struct netlink_ext_ack *extack) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - - pr_debug("%s(sch %p,[qdisc %p],new %p,old %p)\n", - __func__, sch, p, new, old); - - if (new == NULL) { - new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, - sch->handle, NULL); - if (new == NULL) - new = &noop_qdisc; - } - - *old = qdisc_replace(sch, new, &p->q); - return 0; -} - -static struct Qdisc *dsmark_leaf(struct Qdisc *sch, unsigned long arg) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - return p->q; -} - -static unsigned long dsmark_find(struct Qdisc *sch, u32 classid) -{ - return TC_H_MIN(classid) + 1; -} - -static unsigned long dsmark_bind_filter(struct Qdisc *sch, - unsigned long parent, u32 classid) -{ - pr_debug("%s(sch %p,[qdisc %p],classid %x)\n", - __func__, sch, qdisc_priv(sch), classid); - - return dsmark_find(sch, classid); -} - -static void dsmark_unbind_filter(struct Qdisc *sch, unsigned long cl) -{ -} - -static const struct nla_policy dsmark_policy[TCA_DSMARK_MAX + 1] = { - [TCA_DSMARK_INDICES] = { .type = NLA_U16 }, - [TCA_DSMARK_DEFAULT_INDEX] = { .type = NLA_U16 }, - [TCA_DSMARK_SET_TC_INDEX] = { .type = NLA_FLAG }, - [TCA_DSMARK_MASK] = { .type = NLA_U8 }, - [TCA_DSMARK_VALUE] = { .type = NLA_U8 }, -}; - -static int dsmark_change(struct Qdisc *sch, u32 classid, u32 parent, - struct nlattr **tca, unsigned long *arg, - struct netlink_ext_ack *extack) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - struct nlattr *opt = tca[TCA_OPTIONS]; - struct nlattr *tb[TCA_DSMARK_MAX + 1]; - int err = -EINVAL; - - pr_debug("%s(sch %p,[qdisc %p],classid %x,parent %x), arg 0x%lx\n", - __func__, sch, p, classid, parent, *arg); - - if (!dsmark_valid_index(p, *arg)) { - err = -ENOENT; - goto errout; - } - - if (!opt) - goto errout; - - err = nla_parse_nested(tb, TCA_DSMARK_MAX, opt, dsmark_policy, NULL); - if (err < 0) - goto errout; - - if (tb[TCA_DSMARK_VALUE]) - p->mv[*arg - 1].value = nla_get_u8(tb[TCA_DSMARK_VALUE]); - - if (tb[TCA_DSMARK_MASK]) - p->mv[*arg - 1].mask = nla_get_u8(tb[TCA_DSMARK_MASK]); - - err = 0; - -errout: - return err; -} - -static int dsmark_delete(struct Qdisc *sch, unsigned long arg) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - - if (!dsmark_valid_index(p, arg)) - return -EINVAL; - - p->mv[arg - 1].mask = 0xff; - p->mv[arg - 1].value = 0; - - return 0; -} - -static void dsmark_walk(struct Qdisc *sch, struct qdisc_walker *walker) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - int i; - - pr_debug("%s(sch %p,[qdisc %p],walker %p)\n", - __func__, sch, p, walker); - - if (walker->stop) - return; - - for (i = 0; i < p->indices; i++) { - if (p->mv[i].mask == 0xff && !p->mv[i].value) - goto ignore; - if (walker->count >= walker->skip) { - if (walker->fn(sch, i + 1, walker) < 0) { - walker->stop = 1; - break; - } - } -ignore: - walker->count++; - } -} - -static struct tcf_block *dsmark_tcf_block(struct Qdisc *sch, unsigned long cl, - struct netlink_ext_ack *extack) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - - return p->block; -} - -/* --------------------------- Qdisc operations ---------------------------- */ - -static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch, - struct sk_buff **to_free) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - int err; - - pr_debug("%s(skb %p,sch %p,[qdisc %p])\n", __func__, skb, sch, p); - - if (p->set_tc_index) { - int wlen = skb_network_offset(skb); - - switch (skb_protocol(skb, true)) { - case htons(ETH_P_IP): - wlen += sizeof(struct iphdr); - if (!pskb_may_pull(skb, wlen) || - skb_try_make_writable(skb, wlen)) - goto drop; - - skb->tc_index = ipv4_get_dsfield(ip_hdr(skb)) - & ~INET_ECN_MASK; - break; - - case htons(ETH_P_IPV6): - wlen += sizeof(struct ipv6hdr); - if (!pskb_may_pull(skb, wlen) || - skb_try_make_writable(skb, wlen)) - goto drop; - - skb->tc_index = ipv6_get_dsfield(ipv6_hdr(skb)) - & ~INET_ECN_MASK; - break; - default: - skb->tc_index = 0; - break; - } - } - - if (TC_H_MAJ(skb->priority) == sch->handle) - skb->tc_index = TC_H_MIN(skb->priority); - else { - struct tcf_result res; - struct tcf_proto *fl = rcu_dereference_bh(p->filter_list); - int result = tcf_classify(skb, fl, &res, false); - - pr_debug("result %d class 0x%04x\n", result, res.classid); - - switch (result) { -#ifdef CONFIG_NET_CLS_ACT - case TC_ACT_QUEUED: - case TC_ACT_STOLEN: - case TC_ACT_TRAP: - __qdisc_drop(skb, to_free); - return NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; - - case TC_ACT_SHOT: - goto drop; -#endif - case TC_ACT_OK: - skb->tc_index = TC_H_MIN(res.classid); - break; - - default: - if (p->default_index != NO_DEFAULT_INDEX) - skb->tc_index = p->default_index; - break; - } - } - - err = qdisc_enqueue(skb, p->q, to_free); - if (err != NET_XMIT_SUCCESS) { - if (net_xmit_drop_count(err)) - qdisc_qstats_drop(sch); - return err; - } - - qdisc_qstats_backlog_inc(sch, skb); - sch->q.qlen++; - - return NET_XMIT_SUCCESS; - -drop: - qdisc_drop(skb, sch, to_free); - return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; -} - -static struct sk_buff *dsmark_dequeue(struct Qdisc *sch) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - struct sk_buff *skb; - u32 index; - - pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p); - - skb = qdisc_dequeue_peeked(p->q); - if (skb == NULL) - return NULL; - - qdisc_bstats_update(sch, skb); - qdisc_qstats_backlog_dec(sch, skb); - sch->q.qlen--; - - index = skb->tc_index & (p->indices - 1); - pr_debug("index %d->%d\n", skb->tc_index, index); - - switch (skb_protocol(skb, true)) { - case htons(ETH_P_IP): - ipv4_change_dsfield(ip_hdr(skb), p->mv[index].mask, - p->mv[index].value); - break; - case htons(ETH_P_IPV6): - ipv6_change_dsfield(ipv6_hdr(skb), p->mv[index].mask, - p->mv[index].value); - break; - default: - /* - * Only complain if a change was actually attempted. - * This way, we can send non-IP traffic through dsmark - * and don't need yet another qdisc as a bypass. - */ - if (p->mv[index].mask != 0xff || p->mv[index].value) - pr_warn("%s: unsupported protocol %d\n", - __func__, ntohs(skb_protocol(skb, true))); - break; - } - - return skb; -} - -static struct sk_buff *dsmark_peek(struct Qdisc *sch) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - - pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p); - - return p->q->ops->peek(p->q); -} - -static int dsmark_init(struct Qdisc *sch, struct nlattr *opt, - struct netlink_ext_ack *extack) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - struct nlattr *tb[TCA_DSMARK_MAX + 1]; - int err = -EINVAL; - u32 default_index = NO_DEFAULT_INDEX; - u16 indices; - int i; - - pr_debug("%s(sch %p,[qdisc %p],opt %p)\n", __func__, sch, p, opt); - - if (!opt) - goto errout; - - err = tcf_block_get(&p->block, &p->filter_list, sch, extack); - if (err) - return err; - - err = nla_parse_nested(tb, TCA_DSMARK_MAX, opt, dsmark_policy, NULL); - if (err < 0) - goto errout; - - err = -EINVAL; - if (!tb[TCA_DSMARK_INDICES]) - goto errout; - indices = nla_get_u16(tb[TCA_DSMARK_INDICES]); - - if (hweight32(indices) != 1) - goto errout; - - if (tb[TCA_DSMARK_DEFAULT_INDEX]) - default_index = nla_get_u16(tb[TCA_DSMARK_DEFAULT_INDEX]); - - if (indices <= DSMARK_EMBEDDED_SZ) - p->mv = p->embedded; - else - p->mv = kmalloc_array(indices, sizeof(*p->mv), GFP_KERNEL); - if (!p->mv) { - err = -ENOMEM; - goto errout; - } - for (i = 0; i < indices; i++) { - p->mv[i].mask = 0xff; - p->mv[i].value = 0; - } - p->indices = indices; - p->default_index = default_index; - p->set_tc_index = nla_get_flag(tb[TCA_DSMARK_SET_TC_INDEX]); - - p->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, sch->handle, - NULL); - if (p->q == NULL) - p->q = &noop_qdisc; - else - qdisc_hash_add(p->q, true); - - pr_debug("%s: qdisc %p\n", __func__, p->q); - - err = 0; -errout: - return err; -} - -static void dsmark_reset(struct Qdisc *sch) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - - pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p); - if (p->q) - qdisc_reset(p->q); - sch->qstats.backlog = 0; - sch->q.qlen = 0; -} - -static void dsmark_destroy(struct Qdisc *sch) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - - pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p); - - tcf_block_put(p->block); - qdisc_put(p->q); - if (p->mv != p->embedded) - kfree(p->mv); -} - -static int dsmark_dump_class(struct Qdisc *sch, unsigned long cl, - struct sk_buff *skb, struct tcmsg *tcm) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - struct nlattr *opts = NULL; - - pr_debug("%s(sch %p,[qdisc %p],class %ld\n", __func__, sch, p, cl); - - if (!dsmark_valid_index(p, cl)) - return -EINVAL; - - tcm->tcm_handle = TC_H_MAKE(TC_H_MAJ(sch->handle), cl - 1); - tcm->tcm_info = p->q->handle; - - opts = nla_nest_start(skb, TCA_OPTIONS); - if (opts == NULL) - goto nla_put_failure; - if (nla_put_u8(skb, TCA_DSMARK_MASK, p->mv[cl - 1].mask) || - nla_put_u8(skb, TCA_DSMARK_VALUE, p->mv[cl - 1].value)) - goto nla_put_failure; - - return nla_nest_end(skb, opts); - -nla_put_failure: - nla_nest_cancel(skb, opts); - return -EMSGSIZE; -} - -static int dsmark_dump(struct Qdisc *sch, struct sk_buff *skb) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - struct nlattr *opts = NULL; - - opts = nla_nest_start(skb, TCA_OPTIONS); - if (opts == NULL) - goto nla_put_failure; - if (nla_put_u16(skb, TCA_DSMARK_INDICES, p->indices)) - goto nla_put_failure; - - if (p->default_index != NO_DEFAULT_INDEX && - nla_put_u16(skb, TCA_DSMARK_DEFAULT_INDEX, p->default_index)) - goto nla_put_failure; - - if (p->set_tc_index && - nla_put_flag(skb, TCA_DSMARK_SET_TC_INDEX)) - goto nla_put_failure; - - return nla_nest_end(skb, opts); - -nla_put_failure: - nla_nest_cancel(skb, opts); - return -EMSGSIZE; -} - -static const struct Qdisc_class_ops dsmark_class_ops = { - .graft = dsmark_graft, - .leaf = dsmark_leaf, - .find = dsmark_find, - .change = dsmark_change, - .delete = dsmark_delete, - .walk = dsmark_walk, - .tcf_block = dsmark_tcf_block, - .bind_tcf = dsmark_bind_filter, - .unbind_tcf = dsmark_unbind_filter, - .dump = dsmark_dump_class, -}; - -static struct Qdisc_ops dsmark_qdisc_ops __read_mostly = { - .next = NULL, - .cl_ops = &dsmark_class_ops, - .id = "dsmark", - .priv_size = sizeof(struct dsmark_qdisc_data), - .enqueue = dsmark_enqueue, - .dequeue = dsmark_dequeue, - .peek = dsmark_peek, - .init = dsmark_init, - .reset = dsmark_reset, - .destroy = dsmark_destroy, - .change = NULL, - .dump = dsmark_dump, - .owner = THIS_MODULE, -}; - -static int __init dsmark_module_init(void) -{ - return register_qdisc(&dsmark_qdisc_ops); -} - -static void __exit dsmark_module_exit(void) -{ - unregister_qdisc(&dsmark_qdisc_ops); -} - -module_init(dsmark_module_init) -module_exit(dsmark_module_exit) - -MODULE_LICENSE("GPL");

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Aaro Koskinen aaro.koskinen@nokia.com

commit 474a31e13a4e9749fb3ee55794d69d0f17ee0998 upstream.

We cannot register the same netdev notifier multiple times when probing stmmac devices. Register the notifier only once in module init, and also make debugfs creation/deletion safe against simultaneous notifier call.

Fixes: 481a7d154cbb ("stmmac: debugfs entry name is not be changed when udev rename device name.") Signed-off-by: Aaro Koskinen aaro.koskinen@nokia.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Hugo SIMELIERE hsimeliere.opensource@witekio.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-)

--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -4067,6 +4067,8 @@ static void stmmac_init_fs(struct net_de { struct stmmac_priv *priv = netdev_priv(dev);

+ rtnl_lock(); + /* Create per netdev entries */ priv->dbgfs_dir = debugfs_create_dir(dev->name, stmmac_fs_dir);

@@ -4078,14 +4080,13 @@ static void stmmac_init_fs(struct net_de debugfs_create_file("dma_cap", 0444, priv->dbgfs_dir, dev, &stmmac_dma_cap_fops);

- register_netdevice_notifier(&stmmac_notifier); + rtnl_unlock(); }

static void stmmac_exit_fs(struct net_device *dev) { struct stmmac_priv *priv = netdev_priv(dev);

- unregister_netdevice_notifier(&stmmac_notifier); debugfs_remove_recursive(priv->dbgfs_dir); } #endif /* CONFIG_DEBUG_FS */ @@ -4455,14 +4456,14 @@ int stmmac_dvr_remove(struct device *dev

netdev_info(priv->dev, "%s: removing driver", __func__);

-#ifdef CONFIG_DEBUG_FS - stmmac_exit_fs(ndev); -#endif stmmac_stop_all_dma(priv);

stmmac_mac_set(priv, priv->ioaddr, false); netif_carrier_off(ndev); unregister_netdev(ndev); +#ifdef CONFIG_DEBUG_FS + stmmac_exit_fs(ndev); +#endif if (priv->plat->stmmac_rst) reset_control_assert(priv->plat->stmmac_rst); clk_disable_unprepare(priv->plat->pclk); @@ -4679,6 +4680,7 @@ static int __init stmmac_init(void) /* Create debugfs main directory if it doesn't exist yet */ if (!stmmac_fs_dir) stmmac_fs_dir = debugfs_create_dir(STMMAC_RESOURCE_NAME, NULL); + register_netdevice_notifier(&stmmac_notifier); #endif

return 0; @@ -4687,6 +4689,7 @@ static int __init stmmac_init(void) static void __exit stmmac_exit(void) { #ifdef CONFIG_DEBUG_FS + unregister_netdevice_notifier(&stmmac_notifier); debugfs_remove_recursive(stmmac_fs_dir); #endif }

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ryusuke Konishi konishi.ryusuke@gmail.com

commit 5124a0a549857c4b87173280e192eea24dea72ad upstream.

If DAT metadata file block access fails due to corruption of the DAT file or abnormal virtual block numbers held by b-trees or inodes, a kernel warning is generated.

This replaces the WARN_ONs by error output, so that a kernel, booted with panic_on_warn, does not panic. This patch also replaces the detected return code -ENOENT with another internal code -EINVAL to notify the bmap layer of metadata corruption. When the bmap layer sees -EINVAL, it handles the abnormal situation with nilfs_bmap_convert_error() and finally returns code -EIO as it should.

Link: https://lkml.kernel.org/r/0000000000005cc3d205ea23ddcf@google.com Link: https://lkml.kernel.org/r/20230126164114.6911-1-konishi.ryusuke@gmail.com Signed-off-by: Ryusuke Konishi konishi.ryusuke@gmail.com Reported-by: syzbot+5d5d25f90f195a3cfcb4@syzkaller.appspotmail.com Tested-by: Ryusuke Konishi konishi.ryusuke@gmail.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- fs/nilfs2/dat.c | 27 +++++++++++++++++---------- 1 file changed, 17 insertions(+), 10 deletions(-)

--- a/fs/nilfs2/dat.c +++ b/fs/nilfs2/dat.c @@ -40,8 +40,21 @@ static inline struct nilfs_dat_info *NIL static int nilfs_dat_prepare_entry(struct inode *dat, struct nilfs_palloc_req *req, int create) { - return nilfs_palloc_get_entry_block(dat, req->pr_entry_nr, - create, &req->pr_entry_bh); + int ret; + + ret = nilfs_palloc_get_entry_block(dat, req->pr_entry_nr, + create, &req->pr_entry_bh); + if (unlikely(ret == -ENOENT)) { + nilfs_msg(dat->i_sb, KERN_ERR, + "DAT doesn't have a block to manage vblocknr = %llu", + (unsigned long long)req->pr_entry_nr); + /* + * Return internal code -EINVAL to notify bmap layer of + * metadata corruption. + */ + ret = -EINVAL; + } + return ret; }

static void nilfs_dat_commit_entry(struct inode *dat, @@ -123,11 +136,7 @@ static void nilfs_dat_commit_free(struct

int nilfs_dat_prepare_start(struct inode *dat, struct nilfs_palloc_req *req) { - int ret; - - ret = nilfs_dat_prepare_entry(dat, req, 0); - WARN_ON(ret == -ENOENT); - return ret; + return nilfs_dat_prepare_entry(dat, req, 0); }

void nilfs_dat_commit_start(struct inode *dat, struct nilfs_palloc_req *req, @@ -154,10 +163,8 @@ int nilfs_dat_prepare_end(struct inode * int ret;

ret = nilfs_dat_prepare_entry(dat, req, 0); - if (ret < 0) { - WARN_ON(ret == -ENOENT); + if (ret < 0) return ret; - }

kaddr = kmap_atomic(req->pr_entry_bh->b_page); entry = nilfs_palloc_block_get_entry(dat, req->pr_entry_nr,

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cyril Hrubis chrubis@suse.cz

[ Upstream commit c7fcb99877f9f542c918509b2801065adcaf46fa ]

There is a 10% rounding error in the intial value of the sysctl_sched_rr_timeslice with CONFIG_HZ_300=y.

This was found with LTP test sched_rr_get_interval01:

sched_rr_get_interval01.c:57: TPASS: sched_rr_get_interval() passed sched_rr_get_interval01.c:64: TPASS: Time quantum 0s 99999990ns sched_rr_get_interval01.c:72: TFAIL: /proc/sys/kernel/sched_rr_timeslice_ms != 100 got 90 sched_rr_get_interval01.c:57: TPASS: sched_rr_get_interval() passed sched_rr_get_interval01.c:64: TPASS: Time quantum 0s 99999990ns sched_rr_get_interval01.c:72: TFAIL: /proc/sys/kernel/sched_rr_timeslice_ms != 100 got 90

What this test does is to compare the return value from the sched_rr_get_interval() and the sched_rr_timeslice_ms sysctl file and fails if they do not match.

The problem it found is the intial sysctl file value which was computed as:

static int sysctl_sched_rr_timeslice = (MSEC_PER_SEC / HZ) * RR_TIMESLICE;

which works fine as long as MSEC_PER_SEC is multiple of HZ, however it introduces 10% rounding error for CONFIG_HZ_300:

(MSEC_PER_SEC / HZ) * (100 * HZ / 1000)

(1000 / 300) * (100 * 300 / 1000)

3 * 30 = 90

This can be easily fixed by reversing the order of the multiplication and division. After this fix we get:

(MSEC_PER_SEC * (100 * HZ / 1000)) / HZ

(1000 * (100 * 300 / 1000)) / 300

(1000 * 30) / 300 = 100

Fixes: 975e155ed873 ("sched/rt: Show the 'sched_rr_timeslice' SCHED_RR timeslice tuning knob in milliseconds") Signed-off-by: Cyril Hrubis chrubis@suse.cz Signed-off-by: Peter Zijlstra (Intel) peterz@infradead.org Reviewed-by: Petr Vorel pvorel@suse.cz Acked-by: Mel Gorman mgorman@suse.de Tested-by: Petr Vorel pvorel@suse.cz Link: https://lore.kernel.org/r/20230802151906.25258-2-chrubis@suse.cz [ pvorel: rebased for 4.19 ] Signed-off-by: Petr Vorel pvorel@suse.cz Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- kernel/sched/rt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -8,7 +8,7 @@ #include "pelt.h"

int sched_rr_timeslice = RR_TIMESLICE; -int sysctl_sched_rr_timeslice = (MSEC_PER_SEC / HZ) * RR_TIMESLICE; +int sysctl_sched_rr_timeslice = (MSEC_PER_SEC * RR_TIMESLICE) / HZ;

static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun);

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cyril Hrubis chrubis@suse.cz

[ Upstream commit 079be8fc630943d9fc70a97807feb73d169ee3fc ]

The validation of the value written to sched_rt_period_us was broken because:

- the sysclt_sched_rt_period is declared as unsigned int - parsed by proc_do_intvec() - the range is asserted after the value parsed by proc_do_intvec()

Because of this negative values written to the file were written into a unsigned integer that were later on interpreted as large positive integers which did passed the check:

if (sysclt_sched_rt_period <= 0) return EINVAL;

This commit fixes the parsing by setting explicit range for both perid_us and runtime_us into the sched_rt_sysctls table and processes the values with proc_dointvec_minmax() instead.

Alternatively if we wanted to use full range of unsigned int for the period value we would have to split the proc_handler and use proc_douintvec() for it however even the Documentation/scheduller/sched-rt-group.rst describes the range as 1 to INT_MAX.

As far as I can tell the only problem this causes is that the sysctl file allows writing negative values which when read back may confuse userspace.

There is also a LTP test being submitted for these sysctl files at:

http://patchwork.ozlabs.org/project/ltp/patch/20230901144433.2526-1-chrubis@...

Signed-off-by: Cyril Hrubis chrubis@suse.cz Signed-off-by: Ingo Molnar mingo@kernel.org Link: https://lore.kernel.org/r/20231002115553.3007-2-chrubis@suse.cz [ pvorel: rebased for 4.19 ] Reviewed-by: Petr Vorel pvorel@suse.cz Signed-off-by: Petr Vorel pvorel@suse.cz Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- kernel/sched/rt.c | 5 +---- kernel/sysctl.c | 5 +++++ 2 files changed, 6 insertions(+), 4 deletions(-)

--- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -2658,9 +2658,6 @@ static int sched_rt_global_constraints(v

static int sched_rt_global_validate(void) { - if (sysctl_sched_rt_period <= 0) - return -EINVAL; - if ((sysctl_sched_rt_runtime != RUNTIME_INF) && (sysctl_sched_rt_runtime > sysctl_sched_rt_period)) return -EINVAL; @@ -2690,7 +2687,7 @@ int sched_rt_handler(struct ctl_table *t old_period = sysctl_sched_rt_period; old_runtime = sysctl_sched_rt_runtime;

- ret = proc_dointvec(table, write, buffer, lenp, ppos); + ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);

if (!ret && write) { ret = sched_rt_global_validate(); --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -127,6 +127,7 @@ static int zero; static int __maybe_unused one = 1; static int __maybe_unused two = 2; static int __maybe_unused four = 4; +static int int_max = INT_MAX; static unsigned long zero_ul; static unsigned long one_ul = 1; static unsigned long long_max = LONG_MAX; @@ -464,6 +465,8 @@ static struct ctl_table kern_table[] = { .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = sched_rt_handler, + .extra1 = &one, + .extra2 = &int_max, }, { .procname = "sched_rt_runtime_us", @@ -471,6 +474,8 @@ static struct ctl_table kern_table[] = { .maxlen = sizeof(int), .mode = 0644, .proc_handler = sched_rt_handler, + .extra1 = &neg_one, + .extra2 = &int_max, }, { .procname = "sched_rr_timeslice_ms",

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vinod Koul vkoul@kernel.org

[ Upstream commit 404290240827c3bb5c4e195174a8854eef2f89ac ]

We seem to have hit warnings of 'output may be truncated' which is fixed by increasing the size of 'dev_id'

drivers/dma/sh/shdmac.c: In function ‘sh_dmae_probe’: drivers/dma/sh/shdmac.c:541:34: error: ‘%d’ directive output may be truncated writing between 1 and 10 bytes into a region of size 9 [-Werror=format-truncation=] 541 | "sh-dmae%d.%d", pdev->id, id); | ^~ In function ‘sh_dmae_chan_probe’, inlined from ‘sh_dmae_probe’ at drivers/dma/sh/shdmac.c:845:9: drivers/dma/sh/shdmac.c:541:26: note: directive argument in the range [0, 2147483647] 541 | "sh-dmae%d.%d", pdev->id, id); | ^~~~~~~~~~~~~~ drivers/dma/sh/shdmac.c:541:26: note: directive argument in the range [0, 19] drivers/dma/sh/shdmac.c:540:17: note: ‘snprintf’ output between 11 and 21 bytes into a destination of size 16 540 | snprintf(sh_chan->dev_id, sizeof(sh_chan->dev_id), | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 541 | "sh-dmae%d.%d", pdev->id, id); | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Vinod Koul vkoul@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/dma/sh/shdma.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/dma/sh/shdma.h b/drivers/dma/sh/shdma.h index 2c0a969adc9fb..35987a35a8480 100644 --- a/drivers/dma/sh/shdma.h +++ b/drivers/dma/sh/shdma.h @@ -29,7 +29,7 @@ struct sh_dmae_chan { const struct sh_dmae_slave_config *config; /* Slave DMA configuration */ int xmit_shift; /* log_2(bytes_per_xfer) */ void __iomem *base; - char dev_id[16]; /* unique name per DMAC of channel */ + char dev_id[32]; /* unique name per DMAC of channel */ int pm_error; dma_addr_t slave_addr; };

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Felix Fietkau nbd@nbd.name

[ Upstream commit bcbc84af1183c8cf3d1ca9b78540c2185cd85e7f ]

fast-xmit must only be enabled after the sta has been uploaded to the driver, otherwise it could end up passing the not-yet-uploaded sta via drv_tx calls to the driver, leading to potential crashes because of uninitialized drv_priv data. Add a missing sta->uploaded check and re-check fast xmit after inserting a sta.

Signed-off-by: Felix Fietkau nbd@nbd.name Link: https://msgid.link/20240104181059.84032-1-nbd@nbd.name Signed-off-by: Johannes Berg johannes.berg@intel.com Signed-off-by: Sasha Levin sashal@kernel.org --- net/mac80211/sta_info.c | 2 ++ net/mac80211/tx.c | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 5c209f72de701..714d0b01ea629 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -629,6 +629,8 @@ static int sta_info_insert_finish(struct sta_info *sta) __acquires(RCU) if (ieee80211_vif_is_mesh(&sdata->vif)) mesh_accept_plinks_update(sdata);

+ ieee80211_check_fast_xmit(sta); + return 0; out_remove: sta_info_hash_del(local, sta); diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 3a0aadf881fc9..89500b1fe3016 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -2868,7 +2868,7 @@ void ieee80211_check_fast_xmit(struct sta_info *sta) sdata->vif.type == NL80211_IFTYPE_STATION) goto out;

- if (!test_sta_flag(sta, WLAN_STA_AUTHORIZED)) + if (!test_sta_flag(sta, WLAN_STA_AUTHORIZED) || !sta->uploaded) goto out;

if (test_sta_flag(sta, WLAN_STA_PS_STA) ||

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Fullway Wang fullwaywang@outlook.com

[ Upstream commit e421946be7d9bf545147bea8419ef8239cb7ca52 ]

The userspace program could pass any values to the driver through ioctl() interface. If the driver doesn't check the value of pixclock, it may cause divide-by-zero error.

In sisfb_check_var(), var->pixclock is used as a divisor to caculate drate before it is checked against zero. Fix this by checking it at the beginning.

This is similar to CVE-2022-3061 in i740fb which was fixed by commit 15cf0b8.

Signed-off-by: Fullway Wang fullwaywang@outlook.com Signed-off-by: Helge Deller deller@gmx.de Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/video/fbdev/sis/sis_main.c | 2 ++ 1 file changed, 2 insertions(+)

diff --git a/drivers/video/fbdev/sis/sis_main.c b/drivers/video/fbdev/sis/sis_main.c index 20aff90059781..b7f9da690db27 100644 --- a/drivers/video/fbdev/sis/sis_main.c +++ b/drivers/video/fbdev/sis/sis_main.c @@ -1488,6 +1488,8 @@ sisfb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)

vtotal = var->upper_margin + var->lower_margin + var->vsync_len;

+ if (!var->pixclock) + return -EINVAL; pixclock = var->pixclock;

if((var->vmode & FB_VMODE_MASK) == FB_VMODE_NONINTERLACED) {

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Baokun Li libaokun1@huawei.com

[ Upstream commit 4530b3660d396a646aad91a787b6ab37cf604b53 ]

Determine if the group block bitmap is corrupted before using ac_b_ex in ext4_mb_try_best_found() to avoid allocating blocks from a group with a corrupted block bitmap in the following concurrency and making the situation worse.

ext4_mb_regular_allocator ext4_lock_group(sb, group) ext4_mb_good_group // check if the group bbitmap is corrupted ext4_mb_complex_scan_group // Scan group gets ac_b_ex but doesn't use it ext4_unlock_group(sb, group) ext4_mark_group_bitmap_corrupted(group) // The block bitmap was corrupted during // the group unlock gap. ext4_mb_try_best_found ext4_lock_group(ac->ac_sb, group) ext4_mb_use_best_found mb_mark_used // Allocating blocks in block bitmap corrupted group

Signed-off-by: Baokun Li libaokun1@huawei.com Reviewed-by: Jan Kara jack@suse.cz Link: https://lore.kernel.org/r/20240104142040.2835097-7-libaokun1@huawei.com Signed-off-by: Theodore Ts'o tytso@mit.edu Signed-off-by: Sasha Levin sashal@kernel.org --- fs/ext4/mballoc.c | 4 ++++ 1 file changed, 4 insertions(+)

diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index df390979058fd..e0dd01cb1a0e7 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -1802,6 +1802,9 @@ int ext4_mb_try_best_found(struct ext4_allocation_context *ac, return err;

ext4_lock_group(ac->ac_sb, group); + if (unlikely(EXT4_MB_GRP_BBITMAP_CORRUPT(e4b->bd_info))) + goto out; + max = mb_find_extent(e4b, ex.fe_start, ex.fe_len, &ex);

if (max > 0) { @@ -1809,6 +1812,7 @@ int ext4_mb_try_best_found(struct ext4_allocation_context *ac, ext4_mb_use_best_found(ac, e4b); }

+out: ext4_unlock_group(ac->ac_sb, group); ext4_mb_unload_buddy(e4b);

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Martin Blumenstingl martin.blumenstingl@googlemail.com

[ Upstream commit c92688cac239794e4a1d976afa5203a4d3a2ac0e ]

Continuous regulators can be configured to operate only in a certain duty cycle range (for example from 0..91%). Add a check to error out if the duty cycle translates to an unsupported (or out of range) voltage.

Suggested-by: Uwe Kleine-König u.kleine-koenig@pengutronix.de Signed-off-by: Martin Blumenstingl martin.blumenstingl@googlemail.com Link: https://msgid.link/r/20240113224628.377993-2-martin.blumenstingl@googlemail.... Signed-off-by: Mark Brown broonie@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/regulator/pwm-regulator.c | 3 +++ 1 file changed, 3 insertions(+)

diff --git a/drivers/regulator/pwm-regulator.c b/drivers/regulator/pwm-regulator.c index 34f3b9778ffa1..215581dfd3330 100644 --- a/drivers/regulator/pwm-regulator.c +++ b/drivers/regulator/pwm-regulator.c @@ -164,6 +164,9 @@ static int pwm_regulator_get_voltage(struct regulator_dev *rdev) pwm_get_state(drvdata->pwm, &pstate);

voltage = pwm_get_relative_duty_cycle(&pstate, duty_unit); + if (voltage < min(max_uV_duty, min_uV_duty) || + voltage > max(max_uV_duty, min_uV_duty)) + return -ENOTRECOVERABLE;

/* * The dutycycle for min_uV might be greater than the one for max_uV.

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Takashi Sakamoto o-takashi@sakamocchi.jp

[ Upstream commit 7ed4380009e96d9e9c605e12822e987b35b05648 ]

If we are bus manager and the bus has inconsistent gap counts, send a bus reset immediately instead of trying to read the root node's config ROM first. Otherwise, we could spend a lot of time trying to read the config ROM but never succeeding.

This eliminates a 50+ second delay before the FireWire bus is usable after a newly connected device is powered on in certain circumstances.

The delay occurs if a gap count inconsistency occurs, we are not the root node, and we become bus manager. One scenario that causes this is with a TI XIO2213B OHCI, the first time a Sony DSR-25 is powered on after being connected to the FireWire cable. In this configuration, the Linux box will not receive the initial PHY configuration packet sent by the DSR-25 as IRM, resulting in the DSR-25 having a gap count of 44 while the Linux box has a gap count of 63.

FireWire devices have a gap count parameter, which is set to 63 on power-up and can be changed with a PHY configuration packet. This determines the duration of the subaction and arbitration gaps. For reliable communication, all nodes on a FireWire bus must have the same gap count.

A node may have zero or more of the following roles: root node, bus manager (BM), isochronous resource manager (IRM), and cycle master. Unless a root node was forced with a PHY configuration packet, any node might become root node after a bus reset. Only the root node can become cycle master. If the root node is not cycle master capable, the BM or IRM should force a change of root node.

After a bus reset, each node sends a self-ID packet, which contains its current gap count. A single bus reset does not change the gap count, but two bus resets in a row will set the gap count to 63. Because a consistent gap count is required for reliable communication, IEEE 1394a-2000 requires that the bus manager generate a bus reset if it detects that the gap count is inconsistent.

When the gap count is inconsistent, build_tree() will notice this after the self identification process. It will set card->gap_count to the invalid value 0. If we become bus master, this will force bm_work() to send a bus reset when it performs gap count optimization.

After a bus reset, there is no bus manager. We will almost always try to become bus manager. Once we become bus manager, we will first determine whether the root node is cycle master capable. Then, we will determine if the gap count should be changed. If either the root node or the gap count should be changed, we will generate a bus reset.

To determine if the root node is cycle master capable, we read its configuration ROM. bm_work() will wait until we have finished trying to read the configuration ROM.

However, an inconsistent gap count can make this take a long time. read_config_rom() will read the first few quadlets from the config ROM. Due to the gap count inconsistency, eventually one of the reads will time out. When read_config_rom() fails, fw_device_init() calls it again until MAX_RETRIES is reached. This takes 50+ seconds.

Once we give up trying to read the configuration ROM, bm_work() will wake up, assume that the root node is not cycle master capable, and do a bus reset. Hopefully, this will resolve the gap count inconsistency.

This change makes bm_work() check for an inconsistent gap count before waiting for the root node's configuration ROM. If the gap count is inconsistent, bm_work() will immediately do a bus reset. This eliminates the 50+ second delay and rapidly brings the bus to a working state.

I considered that if the gap count is inconsistent, a PHY configuration packet might not be successful, so it could be desirable to skip the PHY configuration packet before the bus reset in this case. However, IEEE 1394a-2000 and IEEE 1394-2008 say that the bus manager may transmit a PHY configuration packet before a bus reset when correcting a gap count error. Since the standard endorses this, I decided it's safe to retain the PHY configuration packet transmission.

Normally, after a topology change, we will reset the bus a maximum of 5 times to change the root node and perform gap count optimization. However, if there is a gap count inconsistency, we must always generate a bus reset. Otherwise the gap count inconsistency will persist and communication will be unreliable. For that reason, if there is a gap count inconstency, we generate a bus reset even if we already reached the 5 reset limit.

Signed-off-by: Adam Goldman adamg@pobox.com Reference: https://sourceforge.net/p/linux1394/mailman/message/58727806/ Signed-off-by: Takashi Sakamoto o-takashi@sakamocchi.jp Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/firewire/core-card.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/firewire/core-card.c b/drivers/firewire/core-card.c index 11c634125c7df..0e6f96c0e3957 100644 --- a/drivers/firewire/core-card.c +++ b/drivers/firewire/core-card.c @@ -442,7 +442,23 @@ static void bm_work(struct work_struct *work) */ card->bm_generation = generation;

- if (root_device == NULL) { + if (card->gap_count == 0) { + /* + * If self IDs have inconsistent gap counts, do a + * bus reset ASAP. The config rom read might never + * complete, so don't wait for it. However, still + * send a PHY configuration packet prior to the + * bus reset. The PHY configuration packet might + * fail, but 1394-2008 8.4.5.2 explicitly permits + * it in this case, so it should be safe to try. + */ + new_root_id = local_id; + /* + * We must always send a bus reset if the gap count + * is inconsistent, so bypass the 5-reset limit. + */ + card->bm_retries = 0; + } else if (root_device == NULL) { /* * Either link_on is false, or we failed to read the * config rom. In either case, pick another root.

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alexandra Winter wintera@linux.ibm.com

[ Upstream commit 2fe8a236436fe40d8d26a1af8d150fc80f04ee1a ]

Symptom: In case of a bad cable connection (e.g. dirty optics) a fast sequence of network DOWN-UP-DOWN-UP could happen. UP triggers recovery of the qeth interface. In case of a second DOWN while recovery is still ongoing, it can happen that the IP@ of a Layer3 qeth interface is lost and will not be recovered by the second UP.

Problem: When registration of IP addresses with Layer 3 qeth devices fails, (e.g. because of bad address format) the respective IP address is deleted from its hash-table in the driver. If registration fails because of a ENETDOWN condition, the address should stay in the hashtable, so a subsequent recovery can restore it.

3caa4af834df ("qeth: keep ip-address after LAN_OFFLINE failure") fixes this for registration failures during normal operation, but not during recovery.

Solution: Keep L3-IP address in case of ENETDOWN in qeth_l3_recover_ip(). For consistency with qeth_l3_add_ip() we also keep it in case of EADDRINUSE, i.e. for some reason the card already/still has this address registered.

Fixes: 4a71df50047f ("qeth: new qeth device driver") Cc: stable@vger.kernel.org Signed-off-by: Alexandra Winter wintera@linux.ibm.com Link: https://lore.kernel.org/r/20240206085849.2902775-1-wintera@linux.ibm.com Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/s390/net/qeth_l3_main.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index 52e0ae4dc7241..e9102359e3134 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -285,9 +285,10 @@ static void qeth_l3_clear_ip_htable(struct qeth_card *card, int recover) if (!recover) { hash_del(&addr->hnode); kfree(addr); - continue; + } else { + /* prepare for recovery */ + addr->disp_flag = QETH_DISP_ADDR_ADD; } - addr->disp_flag = QETH_DISP_ADDR_ADD; }

spin_unlock_bh(&card->ip_lock); @@ -325,11 +326,13 @@ static void qeth_l3_recover_ip(struct qeth_card *card) } else rc = qeth_l3_register_addr_entry(card, addr);

- if (!rc) { + if (!rc || rc == -EADDRINUSE || rc == -ENETDOWN) { + /* keep it in the records */ addr->disp_flag = QETH_DISP_ADDR_DO_NOTHING; if (addr->ref_counter < 1) qeth_l3_delete_ip(card, addr); } else { + /* bad address */ hash_del(&addr->hnode); kfree(addr); }

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Daniel Vacek neelx@redhat.com

commit e6f57c6881916df39db7d95981a8ad2b9c3458d6 upstream.

Unfortunately the commit `fd8958efe877` introduced another error causing the `descs` array to overflow. This reults in further crashes easily reproducible by `sendmsg` system call.

[ 1080.836473] general protection fault, probably for non-canonical address 0x400300015528b00a: 0000 [#1] PREEMPT SMP PTI [ 1080.869326] RIP: 0010:hfi1_ipoib_build_ib_tx_headers.constprop.0+0xe1/0x2b0 [hfi1] -- [ 1080.974535] Call Trace: [ 1080.976990] <TASK> [ 1081.021929] hfi1_ipoib_send_dma_common+0x7a/0x2e0 [hfi1] [ 1081.027364] hfi1_ipoib_send_dma_list+0x62/0x270 [hfi1] [ 1081.032633] hfi1_ipoib_send+0x112/0x300 [hfi1] [ 1081.042001] ipoib_start_xmit+0x2a9/0x2d0 [ib_ipoib] [ 1081.046978] dev_hard_start_xmit+0xc4/0x210 -- [ 1081.148347] __sys_sendmsg+0x59/0xa0

crash> ipoib_txreq 0xffff9cfeba229f00 struct ipoib_txreq { txreq = { list = { next = 0xffff9cfeba229f00, prev = 0xffff9cfeba229f00 }, descp = 0xffff9cfeba229f40, coalesce_buf = 0x0, wait = 0xffff9cfea4e69a48, complete = 0xffffffffc0fe0760 <hfi1_ipoib_sdma_complete>, packet_len = 0x46d, tlen = 0x0, num_desc = 0x0, desc_limit = 0x6, next_descq_idx = 0x45c, coalesce_idx = 0x0, flags = 0x0, descs = {{ qw = {0x8024000120dffb00, 0x4} # SDMA_DESC0_FIRST_DESC_FLAG (bit 63) }, { qw = { 0x3800014231b108, 0x4} }, { qw = { 0x310000e4ee0fcf0, 0x8} }, { qw = { 0x3000012e9f8000, 0x8} }, { qw = { 0x59000dfb9d0000, 0x8} }, { qw = { 0x78000e02e40000, 0x8} }} }, sdma_hdr = 0x400300015528b000, <<< invalid pointer in the tx request structure sdma_status = 0x0, SDMA_DESC0_LAST_DESC_FLAG (bit 62) complete = 0x0, priv = 0x0, txq = 0xffff9cfea4e69880, skb = 0xffff9d099809f400 }

If an SDMA send consists of exactly 6 descriptors and requires dword padding (in the 7th descriptor), the sdma_txreq descriptor array is not properly expanded and the packet will overflow into the container structure. This results in a panic when the send completion runs. The exact panic varies depending on what elements of the container structure get corrupted. The fix is to use the correct expression in _pad_sdma_tx_descs() to test the need to expand the descriptor array.

With this patch the crashes are no longer reproducible and the machine is stable.

Fixes: fd8958efe877 ("IB/hfi1: Fix sdma.h tx->num_descs off-by-one errors") Cc: stable@vger.kernel.org Reported-by: Mats Kronberg kronberg@nsc.liu.se Tested-by: Mats Kronberg kronberg@nsc.liu.se Signed-off-by: Daniel Vacek neelx@redhat.com Link: https://lore.kernel.org/r/20240201081009.1109442-1-neelx@redhat.com Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/infiniband/hw/hfi1/sdma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/drivers/infiniband/hw/hfi1/sdma.c +++ b/drivers/infiniband/hw/hfi1/sdma.c @@ -3202,7 +3202,7 @@ int _pad_sdma_tx_descs(struct hfi1_devda { int rval = 0;

- if ((unlikely(tx->num_desc + 1 == tx->desc_limit))) { + if ((unlikely(tx->num_desc == tx->desc_limit))) { rval = _extend_sdma_tx_descs(dd, tx); if (rval) { __sdma_txclean(dd, tx);

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mikulas Patocka mpatocka@redhat.com

commit 50c70240097ce41fe6bce6478b80478281e4d0f7 upstream.

It was said that authenticated encryption could produce invalid tag when the data that is being encrypted is modified [1]. So, fix this problem by copying the data into the clone bio first and then encrypt them inside the clone bio.

This may reduce performance, but it is needed to prevent the user from corrupting the device by writing data with O_DIRECT and modifying them at the same time.

[1] https://lore.kernel.org/all/20240207004723.GA35324@sol.localdomain/T/

Signed-off-by: Mikulas Patocka mpatocka@redhat.com Cc: stable@vger.kernel.org Signed-off-by: Mike Snitzer snitzer@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/md/dm-crypt.c | 6 ++++++ 1 file changed, 6 insertions(+)

--- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -1734,6 +1734,12 @@ static void kcryptd_crypt_write_convert( io->ctx.bio_out = clone; io->ctx.iter_out = clone->bi_iter;

+ if (crypt_integrity_aead(cc)) { + bio_copy_data(clone, io->base_bio); + io->ctx.bio_in = clone; + io->ctx.iter_in = clone->bi_iter; + } + sector += bio_sectors(clone);

crypt_inc_pending(io);

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tom Parkin tparkin@katalix.com

commit 359e54a93ab43d32ee1bff3c2f9f10cb9f6b6e79 upstream.

l2tp_ip6_sendmsg needs to avoid accounting for the transport header twice when splicing more data into an already partially-occupied skbuff.

To manage this, we check whether the skbuff contains data using skb_queue_empty when deciding how much data to append using ip6_append_data.

However, the code which performed the calculation was incorrect:

ulen = len + skb_queue_empty(&sk->sk_write_queue) ? transhdrlen : 0;

...due to C operator precedence, this ends up setting ulen to transhdrlen for messages with a non-zero length, which results in corrupted packets on the wire.

Add parentheses to correct the calculation in line with the original intent.

Fixes: 9d4c75800f61 ("ipv4, ipv6: Fix handling of transhdrlen in __ip{,6}_append_data()") Cc: David Howells dhowells@redhat.com Cc: stable@vger.kernel.org Signed-off-by: Tom Parkin tparkin@katalix.com Reviewed-by: Simon Horman horms@kernel.org Link: https://lore.kernel.org/r/20240220122156.43131-1-tparkin@katalix.com Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/l2tp/l2tp_ip6.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/net/l2tp/l2tp_ip6.c +++ b/net/l2tp/l2tp_ip6.c @@ -648,7 +648,7 @@ static int l2tp_ip6_sendmsg(struct sock

back_from_confirm: lock_sock(sk); - ulen = len + skb_queue_empty(&sk->sk_write_queue) ? transhdrlen : 0; + ulen = len + (skb_queue_empty(&sk->sk_write_queue) ? transhdrlen : 0); err = ip6_append_data(sk, ip_generic_getfrag, msg, ulen, transhdrlen, &ipc6, &fl6, (struct rt6_info *)dst,

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Krishna Kurapati quic_kriskura@quicinc.com

commit 76c51146820c5dac629f21deafab0a7039bc3ccd upstream.

It is observed sometimes when tethering is used over NCM with Windows 11 as host, at some instances, the gadget_giveback has one byte appended at the end of a proper NTB. When the NTB is parsed, unwrap call looks for any leftover bytes in SKB provided by u_ether and if there are any pending bytes, it treats them as a separate NTB and parses it. But in case the second NTB (as per unwrap call) is faulty/corrupt, all the datagrams that were parsed properly in the first NTB and saved in rx_list are dropped.

Adding a few custom traces showed the following: [002] d..1 7828.532866: dwc3_gadget_giveback: ep1out: req 000000003868811a length 1025/16384 zsI ==> 0 [002] d..1 7828.532867: ncm_unwrap_ntb: K: ncm_unwrap_ntb toprocess: 1025 [002] d..1 7828.532867: ncm_unwrap_ntb: K: ncm_unwrap_ntb nth: 1751999342 [002] d..1 7828.532868: ncm_unwrap_ntb: K: ncm_unwrap_ntb seq: 0xce67 [002] d..1 7828.532868: ncm_unwrap_ntb: K: ncm_unwrap_ntb blk_len: 0x400 [002] d..1 7828.532868: ncm_unwrap_ntb: K: ncm_unwrap_ntb ndp_len: 0x10 [002] d..1 7828.532869: ncm_unwrap_ntb: K: Parsed NTB with 1 frames

In this case, the giveback is of 1025 bytes and block length is 1024. The rest 1 byte (which is 0x00) won't be parsed resulting in drop of all datagrams in rx_list.

Same is case with packets of size 2048: [002] d..1 7828.557948: dwc3_gadget_giveback: ep1out: req 0000000011dfd96e length 2049/16384 zsI ==> 0 [002] d..1 7828.557949: ncm_unwrap_ntb: K: ncm_unwrap_ntb nth: 1751999342 [002] d..1 7828.557950: ncm_unwrap_ntb: K: ncm_unwrap_ntb blk_len: 0x800

Lecroy shows one byte coming in extra confirming that the byte is coming in from PC:

Transfer 2959 - Bytes Transferred(1025) Timestamp((18.524 843 590) - Transaction 8391 - Data(1025 bytes) Timestamp(18.524 843 590) --- Packet 4063861 Data(1024 bytes) Duration(2.117us) Idle(14.700ns) Timestamp(18.524 843 590) --- Packet 4063863 Data(1 byte) Duration(66.160ns) Time(282.000ns) Timestamp(18.524 845 722)

According to Windows driver, no ZLP is needed if wBlockLength is non-zero, because the non-zero wBlockLength has already told the function side the size of transfer to be expected. However, there are in-market NCM devices that rely on ZLP as long as the wBlockLength is multiple of wMaxPacketSize. To deal with such devices, it pads an extra 0 at end so the transfer is no longer multiple of wMaxPacketSize.

Cc: stable@vger.kernel.org Fixes: 9f6ce4240a2b ("usb: gadget: f_ncm.c added") Signed-off-by: Krishna Kurapati quic_kriskura@quicinc.com Reviewed-by: Maciej Żenczykowski maze@google.com Link: https://lore.kernel.org/r/20240205074650.200304-1-quic_kriskura@quicinc.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/usb/gadget/function/f_ncm.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-)

--- a/drivers/usb/gadget/function/f_ncm.c +++ b/drivers/usb/gadget/function/f_ncm.c @@ -1349,7 +1349,15 @@ parse_ntb: "Parsed NTB with %d frames\n", dgram_counter);

to_process -= block_len; - if (to_process != 0) { + + /* + * Windows NCM driver avoids USB ZLPs by adding a 1-byte + * zero pad as needed. + */ + if (to_process == 1 && + (*(unsigned char *)(ntb_ptr + block_len) == 0x00)) { + to_process--; + } else if (to_process > 0) { ntb_ptr = (unsigned char *)(ntb_ptr + block_len); goto parse_ntb; }

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhipeng Lu alexious@zju.edu.cn

[ Upstream commit 809aa64ebff51eb170ee31a95f83b2d21efa32e2 ]

When dma_alloc_coherent fails to allocate dd->cr_base[i].va, init_credit_return should deallocate dd->cr_base and dd->cr_base[i] that allocated before. Or those resources would be never freed and a memleak is triggered.

Fixes: 7724105686e7 ("IB/hfi1: add driver files") Signed-off-by: Zhipeng Lu alexious@zju.edu.cn Link: https://lore.kernel.org/r/20240112085523.3731720-1-alexious@zju.edu.cn Acked-by: Dennis Dalessandro dennis.dalessandro@cornelisnetworks.com Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/infiniband/hw/hfi1/pio.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/hfi1/pio.c b/drivers/infiniband/hw/hfi1/pio.c index 3fcbf56f8be23..c2a129ed17743 100644 --- a/drivers/infiniband/hw/hfi1/pio.c +++ b/drivers/infiniband/hw/hfi1/pio.c @@ -2118,7 +2118,7 @@ int init_credit_return(struct hfi1_devdata *dd) "Unable to allocate credit return DMA range for NUMA %d\n", i); ret = -ENOMEM; - goto done; + goto free_cr_base; } } set_dev_node(&dd->pcidev->dev, dd->node); @@ -2126,6 +2126,10 @@ int init_credit_return(struct hfi1_devdata *dd) ret = 0; done: return ret; + +free_cr_base: + free_credit_return(dd); + goto done; }

void free_credit_return(struct hfi1_devdata *dd)

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bart Van Assche bvanassche@acm.org

[ Upstream commit fdfa083549de5d50ebf7f6811f33757781e838c0 ]

Make loading ib_srpt with this parameter set work. The current behavior is that setting that parameter while loading the ib_srpt kernel module triggers the following kernel crash:

BUG: kernel NULL pointer dereference, address: 0000000000000000 Call Trace: <TASK> parse_one+0x18c/0x1d0 parse_args+0xe1/0x230 load_module+0x8de/0xa60 init_module_from_file+0x8b/0xd0 idempotent_init_module+0x181/0x240 __x64_sys_finit_module+0x5a/0xb0 do_syscall_64+0x5f/0xe0 entry_SYSCALL_64_after_hwframe+0x6e/0x76

Cc: LiHonggang honggangli@163.com Reported-by: LiHonggang honggangli@163.com Fixes: a42d985bd5b2 ("ib_srpt: Initial SRP Target merge for v3.3-rc1") Signed-off-by: Bart Van Assche bvanassche@acm.org Link: https://lore.kernel.org/r/20240205004207.17031-1-bvanassche@acm.org Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/infiniband/ulp/srpt/ib_srpt.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c index 6090f1ce0c56f..f3d83a05aa4fa 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.c +++ b/drivers/infiniband/ulp/srpt/ib_srpt.c @@ -82,12 +82,16 @@ module_param(srpt_srq_size, int, 0444); MODULE_PARM_DESC(srpt_srq_size, "Shared receive queue (SRQ) size.");

+static int srpt_set_u64_x(const char *buffer, const struct kernel_param *kp) +{ + return kstrtou64(buffer, 16, (u64 *)kp->arg); +} static int srpt_get_u64_x(char *buffer, const struct kernel_param *kp) { return sprintf(buffer, "0x%016llx", *(u64 *)kp->arg); } -module_param_call(srpt_service_guid, NULL, srpt_get_u64_x, &srpt_service_guid, - 0444); +module_param_call(srpt_service_guid, srpt_set_u64_x, srpt_get_u64_x, + &srpt_service_guid, 0444); MODULE_PARM_DESC(srpt_service_guid, "Using this value for ioc_guid, id_ext, and cm_listen_id" " instead of using the node_guid of the first HCA.");

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bart Van Assche bvanassche@acm.org

[ Upstream commit d4ee7f3a4445ec1b0b88af216f4032c4d30abf5a ]

Since the session name by itself is not sufficient to uniquely identify a queue pair, include the queue pair number. Show the ASCII channel state name instead of the numeric value. This change makes the ib_srpt debug output more consistent.

Link: https://lore.kernel.org/r/20200525172212.14413-3-bvanassche@acm.org Signed-off-by: Bart Van Assche bvanassche@acm.org Signed-off-by: Jason Gunthorpe jgg@mellanox.com Stable-dep-of: eb5c7465c324 ("RDMA/srpt: fix function pointer cast warnings") Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/infiniband/ulp/srpt/ib_srpt.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c index fd3d8da6a9db8..4a696bf280ef9 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.c +++ b/drivers/infiniband/ulp/srpt/ib_srpt.c @@ -225,8 +225,9 @@ static const char *get_ch_state_name(enum rdma_ch_state s) */ static void srpt_qp_event(struct ib_event *event, struct srpt_rdma_ch *ch) { - pr_debug("QP event %d on ch=%p sess_name=%s state=%d\n", - event->event, ch, ch->sess_name, ch->state); + pr_debug("QP event %d on ch=%p sess_name=%s-%d state=%s\n", + event->event, ch, ch->sess_name, ch->qp->qp_num, + get_ch_state_name(ch->state));

switch (event->event) { case IB_EVENT_COMM_EST: @@ -1967,8 +1968,8 @@ static void __srpt_close_all_ch(struct srpt_port *sport) list_for_each_entry(nexus, &sport->nexus_list, entry) { list_for_each_entry(ch, &nexus->ch_list, list) { if (srpt_disconnect_ch(ch) >= 0) - pr_info("Closing channel %s because target %s_%d has been disabled\n", - ch->sess_name, + pr_info("Closing channel %s-%d because target %s_%d has been disabled\n", + ch->sess_name, ch->qp->qp_num, dev_name(&sport->sdev->device->dev), sport->port); srpt_close_ch(ch);

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andrii Nakryiko andriin@fb.com

[ Upstream commit 7a387bed47f7e80e257d966cd64a3e92a63e26a1 ]

Enhance scripts/bpf_helpers_doc.py to emit C header with BPF helper definitions (to be included from libbpf's bpf_helpers.h).

Signed-off-by: Andrii Nakryiko andriin@fb.com Signed-off-by: Alexei Starovoitov ast@kernel.org Stable-dep-of: e37243b65d52 ("bpf, scripts: Correct GPL license name") Signed-off-by: Sasha Levin sashal@kernel.org --- scripts/bpf_helpers_doc.py | 155 ++++++++++++++++++++++++++++++++++++- 1 file changed, 154 insertions(+), 1 deletion(-)

diff --git a/scripts/bpf_helpers_doc.py b/scripts/bpf_helpers_doc.py index 5010a4d5bfbab..dcddacc25dff3 100755 --- a/scripts/bpf_helpers_doc.py +++ b/scripts/bpf_helpers_doc.py @@ -391,6 +391,154 @@ SEE ALSO

print('')

+class PrinterHelpers(Printer): + """ + A printer for dumping collected information about helpers as C header to + be included from BPF program. + @helpers: array of Helper objects to print to standard output + """ + + type_fwds = [ + 'struct bpf_fib_lookup', + 'struct bpf_perf_event_data', + 'struct bpf_perf_event_value', + 'struct bpf_sock', + 'struct bpf_sock_addr', + 'struct bpf_sock_ops', + 'struct bpf_sock_tuple', + 'struct bpf_spin_lock', + 'struct bpf_sysctl', + 'struct bpf_tcp_sock', + 'struct bpf_tunnel_key', + 'struct bpf_xfrm_state', + 'struct pt_regs', + 'struct sk_reuseport_md', + 'struct sockaddr', + 'struct tcphdr', + + 'struct __sk_buff', + 'struct sk_msg_md', + 'struct xpd_md', + ] + known_types = { + '...', + 'void', + 'const void', + 'char', + 'const char', + 'int', + 'long', + 'unsigned long', + + '__be16', + '__be32', + '__wsum', + + 'struct bpf_fib_lookup', + 'struct bpf_perf_event_data', + 'struct bpf_perf_event_value', + 'struct bpf_sock', + 'struct bpf_sock_addr', + 'struct bpf_sock_ops', + 'struct bpf_sock_tuple', + 'struct bpf_spin_lock', + 'struct bpf_sysctl', + 'struct bpf_tcp_sock', + 'struct bpf_tunnel_key', + 'struct bpf_xfrm_state', + 'struct pt_regs', + 'struct sk_reuseport_md', + 'struct sockaddr', + 'struct tcphdr', + } + mapped_types = { + 'u8': '__u8', + 'u16': '__u16', + 'u32': '__u32', + 'u64': '__u64', + 's8': '__s8', + 's16': '__s16', + 's32': '__s32', + 's64': '__s64', + 'size_t': 'unsigned long', + 'struct bpf_map': 'void', + 'struct sk_buff': 'struct __sk_buff', + 'const struct sk_buff': 'const struct __sk_buff', + 'struct sk_msg_buff': 'struct sk_msg_md', + 'struct xdp_buff': 'struct xdp_md', + } + + def print_header(self): + header = '''\ +/* This is auto-generated file. See bpf_helpers_doc.py for details. */ + +/* Forward declarations of BPF structs */''' + + print(header) + for fwd in self.type_fwds: + print('%s;' % fwd) + print('') + + def print_footer(self): + footer = '' + print(footer) + + def map_type(self, t): + if t in self.known_types: + return t + if t in self.mapped_types: + return self.mapped_types[t] + print("") + print("Unrecognized type '%s', please add it to known types!" % t) + sys.exit(1) + + seen_helpers = set() + + def print_one(self, helper): + proto = helper.proto_break_down() + + if proto['name'] in self.seen_helpers: + return + self.seen_helpers.add(proto['name']) + + print('/*') + print(" * %s" % proto['name']) + print(" *") + if (helper.desc): + # Do not strip all newline characters: formatted code at the end of + # a section must be followed by a blank line. + for line in re.sub('\n$', '', helper.desc, count=1).split('\n'): + print(' *{}{}'.format(' \t' if line else '', line)) + + if (helper.ret): + print(' *') + print(' * Returns') + for line in helper.ret.rstrip().split('\n'): + print(' *{}{}'.format(' \t' if line else '', line)) + + print(' */') + print('static %s %s(*%s)(' % (self.map_type(proto['ret_type']), + proto['ret_star'], proto['name']), end='') + comma = '' + for i, a in enumerate(proto['args']): + t = a['type'] + n = a['name'] + if proto['name'] == 'bpf_get_socket_cookie' and i == 0: + t = 'void' + n = 'ctx' + one_arg = '{}{}'.format(comma, self.map_type(t)) + if n: + if a['star']: + one_arg += ' {}'.format(a['star']) + else: + one_arg += ' ' + one_arg += '{}'.format(n) + comma = ', ' + print(one_arg, end='') + + print(') = (void *) %d;' % len(self.seen_helpers)) + print('') + ###############################################################################

# If script is launched from scripts/ from kernel tree and can access @@ -405,6 +553,8 @@ Parse eBPF header file and generate documentation for eBPF helper functions. The RST-formatted output produced can be turned into a manual page with the rst2man utility. """) +argParser.add_argument('--header', action='store_true', + help='generate C header file') if (os.path.isfile(bpfh)): argParser.add_argument('--filename', help='path to include/uapi/linux/bpf.h', default=bpfh) @@ -417,5 +567,8 @@ headerParser = HeaderParser(args.filename) headerParser.run()

# Print formatted output to standard output. -printer = PrinterRST(headerParser.helpers) +if args.header: + printer = PrinterHelpers(headerParser.helpers) +else: + printer = PrinterRST(headerParser.helpers) printer.print_all()

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Randy Dunlap rdunlap@infradead.org

[ Upstream commit 9ddf190a7df77b77817f955fdb9c2ae9d1c9c9a3 ]

JAZZ_ESP is a bool kconfig symbol that selects SCSI_SPI_ATTRS. When CONFIG_SCSI=m, this results in SCSI_SPI_ATTRS=m while JAZZ_ESP=y, which causes many undefined symbol linker errors.

Fix this by only offering to build this driver when CONFIG_SCSI=y.

[mkp: JAZZ_ESP is unique in that it does not support being compiled as a module unlike the remaining SPI SCSI HBA drivers]

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Randy Dunlap rdunlap@infradead.org Link: https://lore.kernel.org/r/20240214055953.9612-1-rdunlap@infradead.org Cc: Thomas Bogendoerfer tsbogend@alpha.franken.de Cc: linux-mips@vger.kernel.org Cc: Arnd Bergmann arnd@arndb.de Cc: Masahiro Yamada masahiroy@kernel.org Cc: Nicolas Schier nicolas@fjasle.eu Cc: James E.J. Bottomley jejb@linux.ibm.com Cc: Martin K. Petersen martin.petersen@oracle.com Cc: linux-scsi@vger.kernel.org Cc: Geert Uytterhoeven geert@linux-m68k.org Reported-by: kernel test robot lkp@intel.com Closes: https://lore.kernel.org/oe-kbuild-all/202402112222.Gl0udKyU-lkp@intel.com/ Signed-off-by: Martin K. Petersen martin.petersen@oracle.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/scsi/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index 6047f0284f73a..5c3635378c321 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -1252,7 +1252,7 @@ source "drivers/scsi/arm/Kconfig"

config JAZZ_ESP bool "MIPS JAZZ FAS216 SCSI support" - depends on MACH_JAZZ && SCSI + depends on MACH_JAZZ && SCSI=y select SCSI_SPI_ATTRS help This is the driver for the onboard SCSI host adapter of MIPS Magnum

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vasiliy Kovalev kovalev@altlinux.org

[ Upstream commit 5559cea2d5aa3018a5f00dd2aca3427ba09b386b ]

The pernet operations structure for the subsystem must be registered before registering the generic netlink family.

Fixes: 915d7e5e5930 ("ipv6: sr: add code base for control plane support of SR-IPv6") Signed-off-by: Vasiliy Kovalev kovalev@altlinux.org Link: https://lore.kernel.org/r/20240215202717.29815-1-kovalev@altlinux.org Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Sasha Levin sashal@kernel.org --- net/ipv6/seg6.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-)

diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c index 89d55770ac74b..4bd601c964343 100644 --- a/net/ipv6/seg6.c +++ b/net/ipv6/seg6.c @@ -445,22 +445,24 @@ int __init seg6_init(void) { int err = -ENOMEM;

- err = genl_register_family(&seg6_genl_family); + err = register_pernet_subsys(&ip6_segments_ops); if (err) goto out;

- err = register_pernet_subsys(&ip6_segments_ops); + err = genl_register_family(&seg6_genl_family); if (err) - goto out_unregister_genl; + goto out_unregister_pernet;

#ifdef CONFIG_IPV6_SEG6_LWTUNNEL err = seg6_iptunnel_init(); if (err) - goto out_unregister_pernet; + goto out_unregister_genl;

err = seg6_local_init(); - if (err) - goto out_unregister_pernet; + if (err) { + seg6_iptunnel_exit(); + goto out_unregister_genl; + } #endif

#ifdef CONFIG_IPV6_SEG6_HMAC @@ -481,11 +483,11 @@ int __init seg6_init(void) #endif #endif #ifdef CONFIG_IPV6_SEG6_LWTUNNEL -out_unregister_pernet: - unregister_pernet_subsys(&ip6_segments_ops); -#endif out_unregister_genl: genl_unregister_family(&seg6_genl_family); +#endif +out_unregister_pernet: + unregister_pernet_subsys(&ip6_segments_ops); goto out; }

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jason Gunthorpe jgg@nvidia.com

[ Upstream commit 723a2cc8d69d4342b47dfddbfe6c19f1b135f09b ]

The signature for __iowrite64_copy() requires the number of 64 bit quantities, not bytes. Multiple by 8 to get to a byte length before invoking zpci_memcpy_toio()

Fixes: 87bc359b9822 ("s390/pci: speed up __iowrite64_copy by using pci store block insn") Acked-by: Niklas Schnelle schnelle@linux.ibm.com Signed-off-by: Jason Gunthorpe jgg@nvidia.com Link: https://lore.kernel.org/r/0-v1-9223d11a7662+1d7785-s390_iowrite64_jgg@nvidia... Signed-off-by: Heiko Carstens hca@linux.ibm.com Signed-off-by: Sasha Levin sashal@kernel.org --- arch/s390/pci/pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/arch/s390/pci/pci.c +++ b/arch/s390/pci/pci.c @@ -273,7 +273,7 @@ resource_size_t pcibios_align_resource(v /* combine single writes by using store-block insn */ void __iowrite64_copy(void __iomem *to, const void *from, size_t count) { - zpci_memcpy_toio(to, from, count); + zpci_memcpy_toio(to, from, count * 8); }

/* Create a virtual mapping cookie for a PCI BAR */

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Oliver Upton oliver.upton@linux.dev

commit 8d3a7dfb801d157ac423261d7cd62c33e95375f8 upstream.

vgic_get_irq() may not return a valid descriptor if there is no ITS that holds a valid translation for the specified INTID. If that is the case, it is safe to silently ignore it and continue processing the LPI pending table.

Cc: stable@vger.kernel.org Fixes: 33d3bc9556a7 ("KVM: arm64: vgic-its: Read initial LPI pending table") Signed-off-by: Oliver Upton oliver.upton@linux.dev Link: https://lore.kernel.org/r/20240221092732.4126848-2-oliver.upton@linux.dev Signed-off-by: Marc Zyngier maz@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- virt/kvm/arm/vgic/vgic-its.c | 3 +++ 1 file changed, 3 insertions(+)

--- a/virt/kvm/arm/vgic/vgic-its.c +++ b/virt/kvm/arm/vgic/vgic-its.c @@ -469,6 +469,9 @@ static int its_sync_lpi_pending_table(st }

irq = vgic_get_irq(vcpu->kvm, NULL, intids[i]); + if (!irq) + continue; + spin_lock_irqsave(&irq->irq_lock, flags); irq->pending_latch = pendmask & (1U << bit_nr); vgic_queue_irq_unlock(vcpu->kvm, irq, flags);

4.19-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bart Van Assche bvanassche@acm.org

commit b820de741ae48ccf50dd95e297889c286ff4f760 upstream.

If kiocb_set_cancel_fn() is called for I/O submitted via io_uring, the following kernel warning appears:

WARNING: CPU: 3 PID: 368 at fs/aio.c:598 kiocb_set_cancel_fn+0x9c/0xa8 Call trace: kiocb_set_cancel_fn+0x9c/0xa8 ffs_epfile_read_iter+0x144/0x1d0 io_read+0x19c/0x498 io_issue_sqe+0x118/0x27c io_submit_sqes+0x25c/0x5fc __arm64_sys_io_uring_enter+0x104/0xab0 invoke_syscall+0x58/0x11c el0_svc_common+0xb4/0xf4 do_el0_svc+0x2c/0xb0 el0_svc+0x2c/0xa4 el0t_64_sync_handler+0x68/0xb4 el0t_64_sync+0x1a4/0x1a8

Fix this by setting the IOCB_AIO_RW flag for read and write I/O that is submitted by libaio.

Suggested-by: Jens Axboe axboe@kernel.dk Cc: Christoph Hellwig hch@lst.de Cc: Avi Kivity avi@scylladb.com Cc: Sandeep Dhavale dhavale@google.com Cc: Jens Axboe axboe@kernel.dk Cc: Greg Kroah-Hartman gregkh@linuxfoundation.org Cc: Kent Overstreet kent.overstreet@linux.dev Cc: stable@vger.kernel.org Signed-off-by: Bart Van Assche bvanassche@acm.org Link: https://lore.kernel.org/r/20240215204739.2677806-2-bvanassche@acm.org Signed-off-by: Christian Brauner brauner@kernel.org Signed-off-by: Bart Van Assche bvanassche@acm.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- fs/aio.c | 9 ++++++++- include/linux/fs.h | 2 ++ 2 files changed, 10 insertions(+), 1 deletion(-)

--- a/fs/aio.c +++ b/fs/aio.c @@ -566,6 +566,13 @@ void kiocb_set_cancel_fn(struct kiocb *i struct kioctx *ctx = req->ki_ctx; unsigned long flags;

+ /* + * kiocb didn't come from aio or is neither a read nor a write, hence + * ignore it. + */ + if (!(iocb->ki_flags & IOCB_AIO_RW)) + return; + if (WARN_ON_ONCE(!list_empty(&req->ki_list))) return;

@@ -1446,7 +1453,7 @@ static int aio_prep_rw(struct kiocb *req req->ki_complete = aio_complete_rw; req->private = NULL; req->ki_pos = iocb->aio_offset; - req->ki_flags = iocb_flags(req->ki_filp); + req->ki_flags = iocb_flags(req->ki_filp) | IOCB_AIO_RW; if (iocb->aio_flags & IOCB_FLAG_RESFD) req->ki_flags |= IOCB_EVENTFD; req->ki_hint = ki_hint_validate(file_write_hint(req->ki_filp)); --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -304,6 +304,8 @@ enum rw_hint { #define IOCB_SYNC (1 << 5) #define IOCB_WRITE (1 << 6) #define IOCB_NOWAIT (1 << 7) +/* kiocb is a read or write operation submitted by fs/aio.c. */ +#define IOCB_AIO_RW (1 << 23)

struct kiocb { struct file *ki_filp;