diff options
53 files changed, 788 insertions, 1267 deletions
diff --git a/Documentation/scsi/BusLogic.rst b/Documentation/scsi/BusLogic.rst index b60169812358..d7d4d56981ca 100644 --- a/Documentation/scsi/BusLogic.rst +++ b/Documentation/scsi/BusLogic.rst @@ -251,8 +251,6 @@ BT-445C VLB Fast SCSI-2 BT-747C EISA Fast SCSI-2 BT-757C EISA Wide Fast SCSI-2 BT-757CD EISA Wide Differential Fast SCSI-2 -BT-545C ISA Fast SCSI-2 -BT-540CF ISA Fast SCSI-2 ======== ==== ============================== MultiMaster "S" Series Host Adapters: @@ -263,17 +261,13 @@ BT-747S EISA Fast SCSI-2 BT-747D EISA Differential Fast SCSI-2 BT-757S EISA Wide Fast SCSI-2 BT-757D EISA Wide Differential Fast SCSI-2 -BT-545S ISA Fast SCSI-2 -BT-542D ISA Differential Fast SCSI-2 BT-742A EISA SCSI-2 (742A revision H) -BT-542B ISA SCSI-2 (542B revision H) ======= ==== ============================== MultiMaster "A" Series Host Adapters: ======= ==== ============================== BT-742A EISA SCSI-2 (742A revisions A - G) -BT-542B ISA SCSI-2 (542B revisions A - G) ======= ==== ============================== AMI FastDisk Host Adapters that are true BusLogic MultiMaster clones are also @@ -400,26 +394,11 @@ selected host adapter. The BusLogic Driver Probing Options comprise the following: -IO:<integer> - - The "IO:" option specifies an ISA I/O Address to be probed for a non-PCI - MultiMaster Host Adapter. If neither "IO:" nor "NoProbeISA" options are - specified, then the standard list of BusLogic MultiMaster ISA I/O Addresses - will be probed (0x330, 0x334, 0x230, 0x234, 0x130, and 0x134). Multiple - "IO:" options may be specified to precisely determine the I/O Addresses to - be probed, but the probe order will always follow the standard list. - NoProbe The "NoProbe" option disables all probing and therefore no BusLogic Host Adapters will be detected. -NoProbeISA - - The "NoProbeISA" option disables probing of the standard BusLogic ISA I/O - Addresses and therefore only PCI MultiMaster and FlashPoint Host Adapters - will be detected. - NoProbePCI The "NoProbePCI" options disables the interrogation of PCI Configuration @@ -464,10 +443,7 @@ QueueDepth:<integer> Depth for devices that do not support Tagged Queuing. If no Queue Depth option is provided, the Queue Depth will be determined automatically based on the Host Adapter's Total Queue Depth and the number, type, speed, and - capabilities of the detected Target Devices. For Host Adapters that - require ISA Bounce Buffers, the Queue Depth is automatically set by default - to BusLogic_TaggedQueueDepthBB or BusLogic_UntaggedQueueDepthBB to avoid - excessive preallocation of DMA Bounce Buffer memory. Target Devices that + capabilities of the detected Target Devices. Target Devices that do not support Tagged Queuing always have their Queue Depth set to BusLogic_UntaggedQueueDepth or BusLogic_UntaggedQueueDepthBB, unless a lower Queue Depth option is provided. A Queue Depth of 1 automatically diff --git a/Documentation/scsi/scsi_mid_low_api.rst b/Documentation/scsi/scsi_mid_low_api.rst index 5bc17d012b25..096ffe9cae0e 100644 --- a/Documentation/scsi/scsi_mid_low_api.rst +++ b/Documentation/scsi/scsi_mid_low_api.rst @@ -1095,10 +1095,6 @@ of interest: - maximum number of commands that can be queued on devices controlled by the host. Overridden by LLD calls to scsi_change_queue_depth(). - unchecked_isa_dma - - 1=>only use bottom 16 MB of ram (ISA DMA addressing - restriction), 0=>can use full 32 bit (or better) DMA - address space no_async_abort - 1=>Asynchronous aborts are not supported - 0=>Timed-out commands will be aborted asynchronously diff --git a/block/bfq-cgroup.c b/block/bfq-cgroup.c index b791e2041e49..e2f14508f2d6 100644 --- a/block/bfq-cgroup.c +++ b/block/bfq-cgroup.c @@ -547,6 +547,8 @@ static void bfq_pd_init(struct blkg_policy_data *pd) entity->orig_weight = entity->weight = entity->new_weight = d->weight; entity->my_sched_data = &bfqg->sched_data; + entity->last_bfqq_created = NULL; + bfqg->my_entity = entity; /* * the root_group's will be set to NULL * in bfq_init_queue() diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index 95586137194e..0270cd7ca165 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -1012,7 +1012,7 @@ static void bfq_bfqq_resume_state(struct bfq_queue *bfqq, struct bfq_data *bfqd, struct bfq_io_cq *bic, bool bfq_already_existing) { - unsigned int old_wr_coeff = bfqq->wr_coeff; + unsigned int old_wr_coeff = 1; bool busy = bfq_already_existing && bfq_bfqq_busy(bfqq); if (bic->saved_has_short_ttime) @@ -1033,7 +1033,13 @@ bfq_bfqq_resume_state(struct bfq_queue *bfqq, struct bfq_data *bfqd, bfqq->ttime = bic->saved_ttime; bfqq->io_start_time = bic->saved_io_start_time; bfqq->tot_idle_time = bic->saved_tot_idle_time; - bfqq->wr_coeff = bic->saved_wr_coeff; + /* + * Restore weight coefficient only if low_latency is on + */ + if (bfqd->low_latency) { + old_wr_coeff = bfqq->wr_coeff; + bfqq->wr_coeff = bic->saved_wr_coeff; + } bfqq->service_from_wr = bic->saved_service_from_wr; bfqq->wr_start_at_switch_to_srt = bic->saved_wr_start_at_switch_to_srt; bfqq->last_wr_start_finish = bic->saved_last_wr_start_finish; @@ -1069,7 +1075,7 @@ bfq_bfqq_resume_state(struct bfq_queue *bfqq, struct bfq_data *bfqd, static int bfqq_process_refs(struct bfq_queue *bfqq) { return bfqq->ref - bfqq->allocated - bfqq->entity.on_st_or_in_serv - - (bfqq->weight_counter != NULL); + (bfqq->weight_counter != NULL) - bfqq->stable_ref; } /* Empty burst list and add just bfqq (see comments on bfq_handle_burst) */ @@ -2622,6 +2628,11 @@ static bool bfq_may_be_close_cooperator(struct bfq_queue *bfqq, return true; } +static bool idling_boosts_thr_without_issues(struct bfq_data *bfqd, + struct bfq_queue *bfqq); + +static void bfq_put_stable_ref(struct bfq_queue *bfqq); + /* * Attempt to schedule a merge of bfqq with the currently in-service * queue or with a close queue among the scheduled queues. Return @@ -2644,11 +2655,50 @@ static bool bfq_may_be_close_cooperator(struct bfq_queue *bfqq, */ static struct bfq_queue * bfq_setup_cooperator(struct bfq_data *bfqd, struct bfq_queue *bfqq, - void *io_struct, bool request) + void *io_struct, bool request, struct bfq_io_cq *bic) { struct bfq_queue *in_service_bfqq, *new_bfqq; /* + * Check delayed stable merge for rotational or non-queueing + * devs. For this branch to be executed, bfqq must not be + * currently merged with some other queue (i.e., bfqq->bic + * must be non null). If we considered also merged queues, + * then we should also check whether bfqq has already been + * merged with bic->stable_merge_bfqq. But this would be + * costly and complicated. + */ + if (unlikely(!bfqd->nonrot_with_queueing)) { + if (bic->stable_merge_bfqq && + !bfq_bfqq_just_created(bfqq) && + time_is_after_jiffies(bfqq->split_time + + msecs_to_jiffies(200))) { + struct bfq_queue *stable_merge_bfqq = + bic->stable_merge_bfqq; + int proc_ref = min(bfqq_process_refs(bfqq), + bfqq_process_refs(stable_merge_bfqq)); + + /* deschedule stable merge, because done or aborted here */ + bfq_put_stable_ref(stable_merge_bfqq); + + bic->stable_merge_bfqq = NULL; + + if (!idling_boosts_thr_without_issues(bfqd, bfqq) && + proc_ref > 0) { + /* next function will take at least one ref */ + struct bfq_queue *new_bfqq = + bfq_setup_merge(bfqq, stable_merge_bfqq); + + bic->stably_merged = true; + if (new_bfqq && new_bfqq->bic) + new_bfqq->bic->stably_merged = true; + return new_bfqq; + } else + return NULL; + } + } + + /* * Do not perform queue merging if the device is non * rotational and performs internal queueing. In fact, such a * device reaches a high speed through internal parallelism @@ -2789,6 +2839,17 @@ static void bfq_bfqq_save_state(struct bfq_queue *bfqq) } } + +static void +bfq_reassign_last_bfqq(struct bfq_queue *cur_bfqq, struct bfq_queue *new_bfqq) +{ + if (cur_bfqq->entity.parent && + cur_bfqq->entity.parent->last_bfqq_created == cur_bfqq) + cur_bfqq->entity.parent->last_bfqq_created = new_bfqq; + else if (cur_bfqq->bfqd && cur_bfqq->bfqd->last_bfqq_created == cur_bfqq) + cur_bfqq->bfqd->last_bfqq_created = new_bfqq; +} + void bfq_release_process_ref(struct bfq_data *bfqd, struct bfq_queue *bfqq) { /* @@ -2806,6 +2867,8 @@ void bfq_release_process_ref(struct bfq_data *bfqd, struct bfq_queue *bfqq) bfqq != bfqd->in_service_queue) bfq_del_bfqq_busy(bfqd, bfqq, false); + bfq_reassign_last_bfqq(bfqq, NULL); + bfq_put_queue(bfqq); } @@ -2823,6 +2886,29 @@ bfq_merge_bfqqs(struct bfq_data *bfqd, struct bfq_io_cq *bic, bfq_clear_bfqq_IO_bound(bfqq); /* + * The processes associated with bfqq are cooperators of the + * processes associated with new_bfqq. So, if bfqq has a + * waker, then assume that all these processes will be happy + * to let bfqq's waker freely inject I/O when they have no + * I/O. + */ + if (bfqq->waker_bfqq && !new_bfqq->waker_bfqq && + bfqq->waker_bfqq != new_bfqq) { + new_bfqq->waker_bfqq = bfqq->waker_bfqq; + new_bfqq->tentative_waker_bfqq = NULL; + + /* + * If the waker queue disappears, then + * new_bfqq->waker_bfqq must be reset. So insert + * new_bfqq into the woken_list of the waker. See + * bfq_check_waker for details. + */ + hlist_add_head(&new_bfqq->woken_list_node, + &new_bfqq->waker_bfqq->woken_list); + + } + + /* * If bfqq is weight-raised, then let new_bfqq inherit * weight-raising. To reduce false positives, neglect the case * where bfqq has just been created, but has not yet made it @@ -2879,6 +2965,9 @@ bfq_merge_bfqqs(struct bfq_data *bfqd, struct bfq_io_cq *bic, */ new_bfqq->pid = -1; bfqq->bic = NULL; + + bfq_reassign_last_bfqq(bfqq, new_bfqq); + bfq_release_process_ref(bfqd, bfqq); } @@ -2906,7 +2995,7 @@ static bool bfq_allow_bio_merge(struct request_queue *q, struct request *rq, * We take advantage of this function to perform an early merge * of the queues of possible cooperating processes. */ - new_bfqq = bfq_setup_cooperator(bfqd, bfqq, bio, false); + new_bfqq = bfq_setup_cooperator(bfqd, bfqq, bio, false, bfqd->bio_bic); if (new_bfqq) { /* * bic still points to bfqq, then it has not yet been @@ -4491,9 +4580,15 @@ check_queue: bfq_bfqq_busy(bfqq->bic->bfqq[0]) && bfqq->bic->bfqq[0]->next_rq ? bfqq->bic->bfqq[0] : NULL; + struct bfq_queue *blocked_bfqq = + !hlist_empty(&bfqq->woken_list) ? + container_of(bfqq->woken_list.first, + struct bfq_queue, + woken_list_node) + : NULL; /* - * The next three mutually-exclusive ifs decide + * The next four mutually-exclusive ifs decide * whether to try injection, and choose the queue to * pick an I/O request from. * @@ -4526,7 +4621,15 @@ check_queue: * next bfqq's I/O is brought forward dramatically, * for it is not blocked for milliseconds. * - * The third if checks whether bfqq is a queue for + * The third if checks whether there is a queue woken + * by bfqq, and currently with pending I/O. Such a + * woken queue does not steal bandwidth from bfqq, + * because it remains soon without I/O if bfqq is not + * served. So there is virtually no risk of loss of + * bandwidth for bfqq if this woken queue has I/O + * dispatched while bfqq is waiting for new I/O. + * + * The fourth if checks whether bfqq is a queue for * which it is better to avoid injection. It is so if * bfqq delivers more throughput when served without * any further I/O from other queues in the middle, or @@ -4546,11 +4649,11 @@ check_queue: * bfq_update_has_short_ttime(), it is rather likely * that, if I/O is being plugged for bfqq and the * waker queue has pending I/O requests that are - * blocking bfqq's I/O, then the third alternative + * blocking bfqq's I/O, then the fourth alternative * above lets the waker queue get served before the * I/O-plugging timeout fires. So one may deem the * second alternative superfluous. It is not, because - * the third alternative may be way less effective in + * the fourth alternative may be way less effective in * case of a synchronization. For two main * reasons. First, throughput may be low because the * inject limit may be too low to guarantee the same @@ -4559,7 +4662,7 @@ check_queue: * guarantees (the second alternative unconditionally * injects a pending I/O request of the waker queue * for each bfq_dispatch_request()). Second, with the - * third alternative, the duration of the plugging, + * fourth alternative, the duration of the plugging, * i.e., the time before bfqq finally receives new I/O, * may not be minimized, because the waker queue may * happen to be served only after other queues. @@ -4577,6 +4680,14 @@ check_queue: bfq_bfqq_budget_left(bfqq->waker_bfqq) ) bfqq = bfqq->waker_bfqq; + else if (blocked_bfqq && + bfq_bfqq_busy(blocked_bfqq) && + blocked_bfqq->next_rq && + bfq_serv_to_charge(blocked_bfqq->next_rq, + blocked_bfqq) <= + bfq_bfqq_budget_left(blocked_bfqq) + ) + bfqq = blocked_bfqq; else if (!idling_boosts_thr_without_issues(bfqd, bfqq) && (bfqq->wr_coeff == 1 || bfqd->wr_busy_queues > 1 || !bfq_bfqq_has_short_ttime(bfqq))) @@ -4983,6 +5094,12 @@ void bfq_put_queue(struct bfq_queue *bfqq) bfqg_and_blkg_put(bfqg); } +static void bfq_put_stable_ref(struct bfq_queue *bfqq) +{ + bfqq->stable_ref--; + bfq_put_queue(bfqq); +} + static void bfq_put_cooperator(struct bfq_queue *bfqq) { struct bfq_queue *__bfqq, *next; @@ -5039,6 +5156,24 @@ static void bfq_exit_icq(struct io_cq *icq) { struct bfq_io_cq *bic = icq_to_bic(icq); + if (bic->stable_merge_bfqq) { + struct bfq_data *bfqd = bic->stable_merge_bfqq->bfqd; + + /* + * bfqd is NULL if scheduler already exited, and in + * that case this is the last time bfqq is accessed. + */ + if (bfqd) { + unsigned long flags; + + spin_lock_irqsave(&bfqd->lock, flags); + bfq_put_stable_ref(bic->stable_merge_bfqq); + spin_unlock_irqrestore(&bfqd->lock, flags); + } else { + bfq_put_stable_ref(bic->stable_merge_bfqq); + } + } + bfq_exit_icq_bfqq(bic, true); bfq_exit_icq_bfqq(bic, false); } @@ -5099,7 +5234,8 @@ bfq_set_next_ioprio_data(struct bfq_queue *bfqq, struct bfq_io_cq *bic) static struct bfq_queue *bfq_get_queue(struct bfq_data *bfqd, struct bio *bio, bool is_sync, - struct bfq_io_cq *bic); + struct bfq_io_cq *bic, + bool respawn); static void bfq_check_ioprio_change(struct bfq_io_cq *bic, struct bio *bio) { @@ -5119,7 +5255,7 @@ static void bfq_check_ioprio_change(struct bfq_io_cq *bic, struct bio *bio) bfqq = bic_to_bfqq(bic, false); if (bfqq) { bfq_release_process_ref(bfqd, bfqq); - bfqq = bfq_get_queue(bfqd, bio, BLK_RW_ASYNC, bic); + bfqq = bfq_get_queue(bfqd, bio, BLK_RW_ASYNC, bic, true); bic_set_bfqq(bic, bfqq, false); } @@ -5162,6 +5298,8 @@ static void bfq_init_bfqq(struct bfq_data *bfqd, struct bfq_queue *bfqq, /* set end request to minus infinity from now */ bfqq->ttime.last_end_request = now_ns + 1; + bfqq->creation_time = jiffies; + bfqq->io_start_time = now_ns; bfq_mark_bfqq_IO_bound(bfqq); @@ -5211,9 +5349,156 @@ static struct bfq_queue **bfq_async_queue_prio(struct bfq_data *bfqd, } } +static struct bfq_queue * +bfq_do_early_stable_merge(struct bfq_data *bfqd, struct bfq_queue *bfqq, + struct bfq_io_cq *bic, + struct bfq_queue *last_bfqq_created) +{ + struct bfq_queue *new_bfqq = + bfq_setup_merge(bfqq, last_bfqq_created); + + if (!new_bfqq) + return bfqq; + + if (new_bfqq->bic) + new_bfqq->bic->stably_merged = true; + bic->stably_merged = true; + + /* + * Reusing merge functions. This implies that + * bfqq->bic must be set too, for + * bfq_merge_bfqqs to correctly save bfqq's + * state before killing it. + */ + bfqq->bic = bic; + bfq_merge_bfqqs(bfqd, bic, bfqq, new_bfqq); + + return new_bfqq; +} + +/* + * Many throughput-sensitive workloads are made of several parallel + * I/O flows, with all flows generated by the same application, or + * more generically by the same task (e.g., system boot). The most + * counterproductive action with these workloads is plugging I/O + * dispatch when one of the bfq_queues associated with these flows + * remains temporarily empty. + * + * To avoid this plugging, BFQ has been using a burst-handling + * mechanism for years now. This mechanism has proven effective for + * throughput, and not detrimental for service guarantees. The + * following function pushes this mechanism a little bit further, + * basing on the following two facts. + * + * First, all the I/O flows of a the same application or task + * contribute to the execution/completion of that common application + * or task. So the performance figures that matter are total + * throughput of the flows and task-wide I/O latency. In particular, + * these flows do not need to be protected from each other, in terms + * of individual bandwidth or latency. + * + * Second, the above fact holds regardless of the number of flows. + * + * Putting these two facts together, this commits merges stably the + * bfq_queues associated with these I/O flows, i.e., with the + * processes that generate these IO/ flows, regardless of how many the + * involved processes are. + * + * To decide whether a set of bfq_queues is actually associated with + * the I/O flows of a common application or task, and to merge these + * queues stably, this function operates as follows: given a bfq_queue, + * say Q2, currently being created, and the last bfq_queue, say Q1, + * created before Q2, Q2 is merged stably with Q1 if + * - very little time has elapsed since when Q1 was created + * - Q2 has the same ioprio as Q1 + * - Q2 belongs to the same group as Q1 + * + * Merging bfq_queues also reduces scheduling overhead. A fio test + * with ten random readers on /dev/nullb shows a throughput boost of + * 40%, with a quadcore. Since BFQ's execution time amounts to ~50% of + * the total per-request processing time, the above throughput boost + * implies that BFQ's overhead is reduced by more than 50%. + * + * This new mechanism most certainly obsoletes the current + * burst-handling heuristics. We keep those heuristics for the moment. + */ +static struct bfq_queue *bfq_do_or_sched_stable_merge(struct bfq_data *bfqd, + struct bfq_queue *bfqq, + struct bfq_io_cq *bic) +{ + struct bfq_queue **source_bfqq = bfqq->entity.parent ? + &bfqq->entity.parent->last_bfqq_created : + &bfqd->last_bfqq_created; + + struct bfq_queue *last_bfqq_created = *source_bfqq; + + /* + * If last_bfqq_created has not been set yet, then init it. If + * it has been set already, but too long ago, then move it + * forward to bfqq. Finally, move also if bfqq belongs to a + * different group than last_bfqq_created, or if bfqq has a + * different ioprio or ioprio_class. If none of these + * conditions holds true, then try an early stable merge or + * schedule a delayed stable merge. + * + * A delayed merge is scheduled (instead of performing an + * early merge), in case bfqq might soon prove to be more + * throughput-beneficial if not merged. Currently this is + * possible only if bfqd is rotational with no queueing. For + * such a drive, not merging bfqq is better for throughput if + * bfqq happens to contain sequential I/O. So, we wait a + * little bit for enough I/O to flow through bfqq. After that, + * if such an I/O is sequential, then the merge is + * canceled. Otherwise the merge is finally performed. + */ + if (!last_bfqq_created || + time_before(last_bfqq_created->creation_time + + bfqd->bfq_burst_interval, + bfqq->creation_time) || + bfqq->entity.parent != last_bfqq_created->entity.parent || + bfqq->ioprio != last_bfqq_created->ioprio || + bfqq->ioprio_class != last_bfqq_created->ioprio_class) + *source_bfqq = bfqq; + else if (time_after_eq(last_bfqq_created->creation_time + + bfqd->bfq_burst_interval, + bfqq->creation_time)) { + if (likely(bfqd->nonrot_with_queueing)) + /* + * With this type of drive, leaving + * bfqq alone may provide no + * throughput benefits compared with + * merging bfqq. So merge bfqq now. + */ + bfqq = bfq_do_early_stable_merge(bfqd, bfqq, + bic, + last_bfqq_created); + else { /* schedule tentative stable merge */ + /* + * get reference on last_bfqq_created, + * to prevent it from being freed, + * until we decide whether to merge + */ + last_bfqq_created->ref++; + /* + * need to keep track of stable refs, to + * compute process refs correctly + */ + last_bfqq_created->stable_ref++; + /* + * Record the bfqq to merge to. + */ + bic->stable_merge_bfqq = last_bfqq_created; + } + } + + return bfqq; +} + + static struct bfq_queue *bfq_get_queue(struct bfq_data *bfqd, struct bio *bio, bool is_sync, - struct bfq_io_cq *bic) + struct bfq_io_cq *bic, + bool respawn) { const int ioprio = IOPRIO_PRIO_DATA(bic->ioprio); const int ioprio_class = IOPRIO_PRIO_CLASS(bic->ioprio); @@ -5271,7 +5556,10 @@ static struct bfq_queue *bfq_get_queue(struct bfq_data *bfqd, out: bfqq->ref++; /* get a process reference to this queue */ - bfq_log_bfqq(bfqd, bfqq, "get_queue, at end: %p, %d", bfqq, bfqq->ref); + + if (bfqq != &bfqd->oom_bfqq && is_sync && !respawn) + bfqq = bfq_do_or_sched_stable_merge(bfqd, bfqq, bic); + rcu_read_unlock(); return bfqq; } @@ -5521,7 +5809,8 @@ static void bfq_rq_enqueued(struct bfq_data *bfqd, struct bfq_queue *bfqq, static bool __bfq_insert_request(struct bfq_data *bfqd, struct request *rq) { struct bfq_queue *bfqq = RQ_BFQQ(rq), - *new_bfqq = bfq_setup_cooperator(bfqd, bfqq, rq, true); + *new_bfqq = bfq_setup_cooperator(bfqd, bfqq, rq, true, + RQ_BIC(rq)); bool waiting, idle_timer_disabled = false; if (new_bfqq) { @@ -5627,7 +5916,48 @@ static void bfq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, spin_lock_irq(&bfqd->lock); bfqq = bfq_init_rq(rq); - if (!bfqq || at_head || blk_rq_is_passthrough(rq)) { + + /* + * Reqs with at_head or passthrough flags set are to be put + * directly into dispatch list. Additional case for putting rq + * directly into the dispatch queue: the only active + * bfq_queues are bfqq and either its waker bfq_queue or one + * of its woken bfq_queues. The rationale behind this + * additional condition is as follows: + * - consider a bfq_queue, say Q1, detected as a waker of + * another bfq_queue, say Q2 + * - by definition of a waker, Q1 blocks the I/O of Q2, i.e., + * some I/O of Q1 needs to be completed for new I/O of Q2 + * to arrive. A notable example of waker is journald + * - so, Q1 and Q2 are in any respect the queues of two + * cooperating processes (or of two cooperating sets of + * processes): the goal of Q1's I/O is doing what needs to + * be done so that new Q2's I/O can finally be + * issued. Therefore, if the service of Q1's I/O is delayed, + * then Q2's I/O is delayed too. Conversely, if Q2's I/O is + * delayed, the goal of Q1's I/O is hindered. + * - as a consequence, if some I/O of Q1/Q2 arrives while + * Q2/Q1 is the only queue in service, there is absolutely + * no point in delaying the service of such an I/O. The + * only possible result is a throughput loss + * - so, when the above condition holds, the best option is to + * have the new I/O dispatched as soon as possible + * - the most effective and efficient way to attain the above + * goal is to put the new I/O directly in the dispatch + * list + * - as an additional restriction, Q1 and Q2 must be the only + * busy queues for this commit to put the I/O of Q2/Q1 in + * the dispatch list. This is necessary, because, if also + * other queues are waiting for service, then putting new + * I/O directly in the dispatch list may evidently cause a + * violation of service guarantees for the other queues + */ + if (!bfqq || + (bfqq != bfqd->in_service_queue && + bfqd->in_service_queue != NULL && + bfq_tot_busy_queues(bfqd) == 1 + bfq_bfqq_busy(bfqq) && + (bfqq->waker_bfqq == bfqd->in_service_queue || + bfqd->in_service_queue->waker_bfqq == bfqq)) || at_head) { if (at_head) list_add(&rq->queuelist, &bfqd->dispatch); else @@ -5767,7 +6097,17 @@ static void bfq_completed_request(struct bfq_queue *bfqq, struct bfq_data *bfqd) 1UL<<(BFQ_RATE_SHIFT - 10)) bfq_update_rate_reset(bfqd, NULL); bfqd->last_completion = now_ns; - bfqd->last_completed_rq_bfqq = bfqq; + /* + * Shared queues are likely to receive I/O at a high + * rate. This may deceptively let them be considered as wakers + * of other queues. But a false waker will unjustly steal + * bandwidth to its supposedly woken queue. So considering + * also shared queues in the waking mechanism may cause more + * control troubles than throughput benefits. Then do not set + * last_completed_rq_bfqq to bfqq if bfqq is a shared queue. + */ + if (!bfq_bfqq_coop(bfqq)) + bfqd->last_completed_rq_bfqq = bfqq; /* * If we are waiting to discover whether the request pattern @@ -6124,7 +6464,7 @@ static struct bfq_queue *bfq_get_bfqq_handle_split(struct bfq_data *bfqd, if (bfqq) bfq_put_queue(bfqq); - bfqq = bfq_get_queue(bfqd, bio, is_sync, bic); + bfqq = bfq_get_queue(bfqd, bio, is_sync, bic, split); bic_set_bfqq(bic, bfqq, is_sync); if (split && is_sync) { @@ -6245,8 +6585,9 @@ static struct bfq_queue *bfq_init_rq(struct request *rq) if (likely(!new_queue)) { /* If the queue was seeky for too long, break it apart. */ - if (bfq_bfqq_coop(bfqq) && bfq_bfqq_split_coop(bfqq)) { - bfq_log_bfqq(bfqd, bfqq, "breaking apart bfqq"); + if (bfq_bfqq_coop(bfqq) && bfq_bfqq_split_coop(bfqq) && + !bic->stably_merged) { + struct bfq_queue *old_bfqq = bfqq; /* Update bic before losing reference to bfqq */ if (bfq_bfqq_in_large_burst(bfqq)) @@ -6255,11 +6596,24 @@ static struct bfq_queue *bfq_init_rq(struct request *rq) bfqq = bfq_split_bfqq(bic, bfqq); split = true; - if (!bfqq) + if (!bfqq) { bfqq = bfq_get_bfqq_handle_split(bfqd, bic, bio, true, is_sync, NULL); - else + bfqq->waker_bfqq = old_bfqq->waker_bfqq; + bfqq->tentative_waker_bfqq = NULL; + + /* + * If the waker queue disappears, then + * new_bfqq->waker_bfqq must be + * reset. So insert new_bfqq into the + * woken_list of the waker. See + * bfq_check_waker for details. + */ + if (bfqq->waker_bfqq) + hlist_add_head(&bfqq->woken_list_node, + &bfqq->waker_bfqq->woken_list); + } else bfqq_already_existing = true; } } diff --git a/block/bfq-iosched.h b/block/bfq-iosched.h index b8e793c34ff1..99c2a3cb081e 100644 --- a/block/bfq-iosched.h +++ b/block/bfq-iosched.h @@ -197,6 +197,9 @@ struct bfq_entity { /* flag, set if the entity is counted in groups_with_pending_reqs */ bool in_groups_with_pending_reqs; + + /* last child queue of entity created (for non-leaf entities) */ + struct bfq_queue *last_bfqq_created; }; struct bfq_group; @@ -230,6 +233,8 @@ struct bfq_ttime { struct bfq_queue { /* reference counter */ int ref; + /* counter of references from other queues for delayed stable merge */ + int stable_ref; /* parent bfq_data */ struct bfq_data *bfqd; @@ -365,6 +370,8 @@ struct bfq_queue { unsigned long first_IO_time; /* time of first I/O for this queue */ + unsigned long creation_time; /* when this queue is created */ + /* max service rate measured so far */ u32 max_service_rate; @@ -454,6 +461,11 @@ struct bfq_io_cq { u64 saved_last_serv_time_ns; unsigned int saved_inject_limit; unsigned long saved_decrease_time_jif; + + /* candidate queue for a stable merge (due to close creation time) */ + struct bfq_queue *stable_merge_bfqq; + + bool stably_merged; /* non splittable if true */ }; /** @@ -578,6 +590,9 @@ struct bfq_data { /* bfqq owning the last completed rq */ struct bfq_queue *last_completed_rq_bfqq; + /* last bfqq created, among those in the root group */ + struct bfq_queue *last_bfqq_created; + /* time of last transition from empty to non-empty (ns) */ u64 last_empty_occupied_ns; diff --git a/block/bfq-wf2q.c b/block/bfq-wf2q.c index 070e34a7feb1..7a462df71f68 100644 --- a/block/bfq-wf2q.c +++ b/block/bfq-wf2q.c @@ -1706,4 +1706,12 @@ void bfq_add_bfqq_busy(struct bfq_data *bfqd, struct bfq_queue *bfqq) if (bfqq->wr_coeff > 1) bfqd->wr_busy_queues++; + + /* Move bfqq to the head of the woken list of its waker */ + if (!hlist_unhashed(&bfqq->woken_list_node) && + &bfqq->woken_list_node != bfqq->waker_bfqq->woken_list.first) { + hlist_del_init(&bfqq->woken_list_node); + hlist_add_head(&bfqq->woken_list_node, + &bfqq->waker_bfqq->woken_list); + } } diff --git a/block/bio-integrity.c b/block/bio-integrity.c index dfa652122a2d..4b4eb8964a6f 100644 --- a/block/bio-integrity.c +++ b/block/bio-integrity.c @@ -204,7 +204,6 @@ bool bio_integrity_prep(struct bio *bio) { struct bio_integrity_payload *bip; struct blk_integrity *bi = blk_get_integrity(bio->bi_bdev->bd_disk); - struct request_queue *q = bio->bi_bdev->bd_disk->queue; void *buf; unsigned long start, end; unsigned int len, nr_pages; @@ -238,7 +237,7 @@ bool bio_integrity_prep(struct bio *bio) /* Allocate kernel buffer for protection data */ len = intervals * bi->tuple_size; - buf = kmalloc(len, GFP_NOIO | q->bounce_gfp); + buf = kmalloc(len, GFP_NOIO); status = BLK_STS_RESOURCE; if (unlikely(buf == NULL)) { printk(KERN_ERR "could not allocate integrity buffer\n"); diff --git a/block/bio.c b/block/bio.c index 50e579088aca..44205dfb6b60 100644 --- a/block/bio.c +++ b/block/bio.c @@ -493,20 +493,20 @@ struct bio *bio_kmalloc(gfp_t gfp_mask, unsigned short nr_iovecs) } EXPORT_SYMBOL(bio_kmalloc); -void zero_fill_bio_iter(struct bio *bio, struct bvec_iter start) +void zero_fill_bio(struct bio *bio) { unsigned long flags; struct bio_vec bv; struct bvec_iter iter; - __bio_for_each_segment(bv, bio, iter, start) { + bio_for_each_segment(bv, bio, iter) { char *data = bvec_kmap_irq(&bv, &flags); memset(data, 0, bv.bv_len); flush_dcache_page(bv.bv_page); bvec_kunmap_irq(data, &flags); } } -EXPORT_SYMBOL(zero_fill_bio_iter); +EXPORT_SYMBOL(zero_fill_bio); /** * bio_truncate - truncate the bio to small size of @new_size @@ -1236,43 +1236,6 @@ void bio_copy_data(struct bio *dst, struct bio *src) } EXPORT_SYMBOL(bio_copy_data); -/** - * bio_list_copy_data - copy contents of data buffers from one chain of bios to - * another - * @src: source bio list - * @dst: destination bio list - * - * Stops when it reaches the end of either the @src list or @dst list - that is, - * copies min(src->bi_size, dst->bi_size) bytes (or the equivalent for lists of - * bios). - */ -void bio_list_copy_data(struct bio *dst, struct bio *src) -{ - struct bvec_iter src_iter = src->bi_iter; - struct bvec_iter dst_iter = dst->bi_iter; - - while (1) { - if (!src_iter.bi_size) { - src = src->bi_next; - if (!src) - break; - - src_iter = src->bi_iter; - } - - if (!dst_iter.bi_size) { - dst = dst->bi_next; - if (!dst) - break; - - dst_iter = dst->bi_iter; - } - - bio_copy_data_iter(dst, &dst_iter, src, &src_iter); - } -} -EXPORT_SYMBOL(bio_list_copy_data); - void bio_free_pages(struct bio *bio) { struct bio_vec *bvec; diff --git a/block/blk-core.c b/block/blk-core.c index fc60ff208497..9bcdae93f6d4 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -1161,10 +1161,8 @@ static blk_status_t blk_cloned_rq_check_limits(struct request_queue *q, } /* - * queue's settings related to segment counting like q->bounce_pfn - * may differ from that of other stacking queues. - * Recalculate it to check the request correctly on this queue's - * limitation. + * The queue settings related to segment counting may differ from the + * original queue. */ rq->nr_phys_segments = blk_recalc_rq_segments(rq); if (rq->nr_phys_segments > queue_max_segments(q)) { diff --git a/block/blk-iocost.c b/block/blk-iocost.c index 98d656bdb42b..e0c4baa01857 100644 --- a/block/blk-iocost.c +++ b/block/blk-iocost.c @@ -987,10 +987,6 @@ static void ioc_adjust_base_vrate(struct ioc *ioc, u32 rq_wait_pct, return; } - /* rq_wait signal is always reliable, ignore user vrate_min */ - if (rq_wait_pct > RQ_WAIT_BUSY_PCT) - vrate_min = VRATE_MIN; - /* * If vrate is out of bounds, apply clamp gradually as the * bounds can change abruptly. Otherwise, apply busy_level diff --git a/block/blk-map.c b/block/blk-map.c index 1ffef782fcf2..3743158ddaeb 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -123,7 +123,6 @@ static int bio_uncopy_user(struct bio *bio) bio_free_pages(bio); } kfree(bmd); - bio_put(bio); return ret; } @@ -132,7 +131,7 @@ static int bio_copy_user_iov(struct request *rq, struct rq_map_data *map_data, { struct bio_map_data *bmd; struct page *page; - struct bio *bio, *bounce_bio; + struct bio *bio; int i = 0, ret; int nr_pages; unsigned int len = iter->count; @@ -181,7 +180,7 @@ static int bio_copy_user_iov(struct request *rq, struct rq_map_data *map_data, i++; } else { - page = alloc_page(rq->q->bounce_gfp | gfp_mask); + page = alloc_page(GFP_NOIO | gfp_mask); if (!page) { ret = -ENOMEM; goto cleanup; @@ -218,16 +217,9 @@ static int bio_copy_user_iov(struct request *rq, struct rq_map_data *map_data, bio->bi_private = bmd; - bounce_bio = bio; - ret = blk_rq_append_bio(rq, &bounce_bio); + ret = blk_rq_append_bio(rq, bio); if (ret) goto cleanup; - - /* - * We link the bounce buffer in and could have to traverse it later, so - * we have to get a ref to prevent it from being freed - */ - bio_get(bounce_bio); return 0; cleanup: if (!map_data) @@ -242,7 +234,7 @@ static int bio_map_user_iov(struct request *rq, struct iov_iter *iter, gfp_t gfp_mask) { unsigned int max_sectors = queue_max_hw_sectors(rq->q); - struct bio *bio, *bounce_bio; + struct bio *bio; int ret; int j; @@ -304,49 +296,17 @@ static int bio_map_user_iov(struct request *rq, struct iov_iter *iter, break; } - /* - * Subtle: if we end up needing to bounce a bio, it would normally - * disappear when its bi_end_io is run. However, we need the original - * bio for the unmap, so grab an extra reference to it - */ - bio_get(bio); - - bounce_bio = bio; - ret = blk_rq_append_bio(rq, &bounce_bio); + ret = blk_rq_append_bio(rq, bio); if (ret) - goto out_put_orig; - - /* - * We link the bounce buffer in and could have to traverse it - * later, so we have to get a ref to prevent it from being freed - */ - bio_get(bounce_bio); + goto out_unmap; return 0; - out_put_orig: - bio_put(bio); out_unmap: bio_release_pages(bio, false); bio_put(bio); return ret; } -/** - * bio_unmap_user - unmap a bio - * @bio: the bio being unmapped - * - * Unmap a bio previously mapped by bio_map_user_iov(). Must be called from - * process context. - * - * bio_unmap_user() may sleep. - */ -static void bio_unmap_user(struct bio *bio) -{ - bio_release_pages(bio, bio_data_dir(bio) == READ); - bio_put(bio); - bio_put(bio); -} - static void bio_invalidate_vmalloc_pages(struct bio *bio) { #ifdef ARCH_HAS_FLUSH_KERNEL_DCACHE_PAGE @@ -486,7 +446,7 @@ static struct bio *bio_copy_kern(struct request_queue *q, void *data, if (bytes > len) bytes = len; - page = alloc_page(q->bounce_gfp | gfp_mask); + page = alloc_page(GFP_NOIO | gfp_mask); if (!page) goto cleanup; @@ -519,33 +479,24 @@ cleanup: * Append a bio to a passthrough request. Only works if the bio can be merged * into the request based on the driver constraints. */ -int blk_rq_append_bio(struct request *rq, struct bio **bio) +int blk_rq_append_bio(struct request *rq, struct bio *bio) { - struct bio *orig_bio = *bio; struct bvec_iter iter; struct bio_vec bv; unsigned int nr_segs = 0; - blk_queue_bounce(rq->q, bio); - - bio_for_each_bvec(bv, *bio, iter) + bio_for_each_bvec(bv, bio, iter) nr_segs++; if (!rq->bio) { - blk_rq_bio_prep(rq, *bio, nr_segs); + blk_rq_bio_prep(rq, bio, nr_segs); } else { - if (!ll_back_merge_fn(rq, *bio, nr_segs)) { - if (orig_bio != *bio) { - bio_put(*bio); - *bio = orig_bio; - } + if (!ll_back_merge_fn(rq, bio, nr_segs)) return -EINVAL; - } - - rq->biotail->bi_next = *bio; - rq->biotail = *bio; - rq->__data_len += (*bio)->bi_iter.bi_size; - bio_crypt_free_ctx(*bio); + rq->biotail->bi_next = bio; + rq->biotail = bio; + rq->__data_len += (bio)->bi_iter.bi_size; + bio_crypt_free_ctx(bio); } return 0; @@ -566,12 +517,6 @@ EXPORT_SYMBOL(blk_rq_append_bio); * * A matching blk_rq_unmap_user() must be issued at the end of I/O, while * still in process context. - * - * Note: The mapped bio may need to be bounced through blk_queue_bounce() - * before being submitted to the device, as pages mapped may be out of - * reach. It's the callers responsibility to make sure this happens. The - * original bio must be passed back in to blk_rq_unmap_user() for proper - * unmapping. */ int blk_rq_map_user_iov(struct request_queue *q, struct request *rq, struct rq_map_data *map_data, @@ -588,6 +533,8 @@ int blk_rq_map_user_iov(struct request_queue *q, struct request *rq, if (map_data) copy = true; + else if (blk_queue_may_bounce(q)) + copy = true; else if (iov_iter_alignment(iter) & align) copy = true; else if (queue_virt_boundary(q)) @@ -641,25 +588,21 @@ EXPORT_SYMBOL(blk_rq_map_user); */ int blk_rq_unmap_user(struct bio *bio) { - struct bio *mapped_bio; + struct bio *next_bio; int ret = 0, ret2; while (bio) { - mapped_bio = bio; - if (unlikely(bio_flagged(bio, BIO_BOUNCED))) - mapped_bio = bio->bi_private; - if (bio->bi_private) { - ret2 = bio_uncopy_user(mapped_bio); + ret2 = bio_uncopy_user(bio); if (ret2 && !ret) ret = ret2; } else { - bio_unmap_user(mapped_bio); + bio_release_pages(bio, bio_data_dir(bio) == READ); } - mapped_bio = bio; + next_bio = bio; bio = bio->bi_next; - bio_put(mapped_bio); + bio_put(next_bio); } return ret; @@ -684,7 +627,7 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf, { int reading = rq_data_dir(rq) == READ; unsigned long addr = (unsigned long) kbuf; - struct bio *bio, *orig_bio; + struct bio *bio; int ret; if (len > (queue_max_hw_sectors(q) << 9)) @@ -692,7 +635,8 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf, if (!len || !kbuf) return -EINVAL; - if (!blk_rq_aligned(q, addr, len) || object_is_on_stack(kbuf)) + if (!blk_rq_aligned(q, addr, len) || object_is_on_stack(kbuf) || + blk_queue_may_bounce(q)) bio = bio_copy_kern(q, kbuf, len, gfp_mask, reading); else bio = bio_map_kern(q, kbuf, len, gfp_mask); @@ -703,14 +647,9 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf, bio->bi_opf &= ~REQ_OP_MASK; bio->bi_opf |= req_op(rq); - orig_bio = bio; - ret = blk_rq_append_bio(rq, &bio); - if (unlikely(ret)) { - /* request is too big */ - bio_put(orig_bio); - return ret; - } - - return 0; + ret = blk_rq_append_bio(rq, bio); + if (unlikely(ret)) + bio_put(bio); + return ret; } EXPORT_SYMBOL(blk_rq_map_kern); diff --git a/block/blk-mq-debugfs.c b/block/blk-mq-debugfs.c index 271f6596435b..2a75bc7401df 100644 --- a/block/blk-mq-debugfs.c +++ b/block/blk-mq-debugfs.c @@ -972,6 +972,14 @@ void blk_mq_debugfs_register_sched_hctx(struct request_queue *q, { struct elevator_type *e = q->elevator->type; + /* + * If the parent debugfs directory has not been created yet, return; + * We will be called again later on with appropriate parent debugfs + * directory from blk_register_queue() + */ + if (!hctx->debugfs_dir) + return; + if (!e->hctx_debugfs_attrs) return; diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index 9c92053e704d..2a37731e8244 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -373,8 +373,8 @@ static bool blk_mq_tagset_count_completed_rqs(struct request *rq, } /** - * blk_mq_tagset_wait_completed_request - wait until all completed req's - * complete funtion is run + * blk_mq_tagset_wait_completed_request - Wait until all scheduled request + * completions have finished. * @tagset: Tag set to drain completed request * * Note: This function has to be run after all IO queues are shutdown @@ -517,7 +517,7 @@ struct blk_mq_tags *blk_mq_init_tags(unsigned int total_tags, tags->nr_tags = total_tags; tags->nr_reserved_tags = reserved_tags; - if (flags & BLK_MQ_F_TAG_HCTX_SHARED) + if (blk_mq_is_sbitmap_shared(flags)) return tags; if (blk_mq_init_bitmap_tags(tags, node, alloc_policy) < 0) { @@ -529,7 +529,7 @@ struct blk_mq_tags *blk_mq_init_tags(unsigned int total_tags, void blk_mq_free_tags(struct blk_mq_tags *tags, unsigned int flags) { - if (!(flags & BLK_MQ_F_TAG_HCTX_SHARED)) { + if (!blk_mq_is_sbitmap_shared(flags)) { sbitmap_queue_free(tags->bitmap_tags); sbitmap_queue_free(tags->breserved_tags); } diff --git a/block/blk-mq.c b/block/blk-mq.c index 4e3a70ab5be1..cd4ad58c88a5 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -361,11 +361,12 @@ static struct request *__blk_mq_alloc_request(struct blk_mq_alloc_data *data) if (e) { /* - * Flush requests are special and go directly to the + * Flush/passthrough requests are special and go directly to the * dispatch list. Don't include reserved tags in the * limiting, as it isn't useful. */ if (!op_is_flush(data->cmd_flags) && + !blk_op_is_passthrough(data->cmd_flags) && e->type->ops.limit_depth && !(data->flags & BLK_MQ_REQ_RESERVED)) e->type->ops.limit_depth(data->cmd_flags, data); diff --git a/block/blk-settings.c b/block/blk-settings.c index b4aa2f37fab6..9c009090c4b5 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -7,7 +7,6 @@ #include <linux/init.h> #include <linux/bio.h> #include <linux/blkdev.h> -#include <linux/memblock.h> /* for max_pfn/max_low_pfn */ #include <linux/gcd.h> #include <linux/lcm.h> #include <linux/jiffies.h> @@ -17,11 +16,6 @@ #include "blk.h" #include "blk-wbt.h" -unsigned long blk_max_low_pfn; -EXPORT_SYMBOL(blk_max_low_pfn); - -unsigned long blk_max_pfn; - void blk_queue_rq_timeout(struct request_queue *q, unsigned int timeout) { q->rq_timeout = timeout; @@ -55,7 +49,7 @@ void blk_set_default_limits(struct queue_limits *lim) lim->discard_alignment = 0; lim->discard_misaligned = 0; lim->logical_block_size = lim->physical_block_size = lim->io_min = 512; - lim->bounce_pfn = (unsigned long)(BLK_BOUNCE_ANY >> PAGE_SHIFT); + lim->bounce = BLK_BOUNCE_NONE; lim->alignment_offset = 0; lim->io_opt = 0; lim->misaligned = 0; @@ -92,39 +86,16 @@ EXPORT_SYMBOL(blk_set_stacking_limits); /** * blk_queue_bounce_limit - set bounce buffer limit for queue * @q: the request queue for the device - * @max_addr: the maximum address the device can handle + * @bounce: bounce limit to enforce * * Description: - * Different hardware can have different requirements as to what pages - * it can do I/O directly to. A low level driver can call - * blk_queue_bounce_limit to have lower memory pages allocated as bounce - * buffers for doing I/O to pages residing above @max_addr. + * Force bouncing for ISA DMA ranges or highmem. + * + * DEPRECATED, don't use in new code. **/ -void blk_queue_bounce_limit(struct request_queue *q, u64 max_addr) +void blk_queue_bounce_limit(struct request_queue *q, enum blk_bounce bounce) { - unsigned long b_pfn = max_addr >> PAGE_SHIFT; - int dma = 0; - - q->bounce_gfp = GFP_NOIO; -#if BITS_PER_LONG == 64 - /* - * Assume anything <= 4GB can be handled by IOMMU. Actually - * some IOMMUs can handle everything, but I don't know of a - * way to test this here. - */ - if (b_pfn < (min_t(u64, 0xffffffffUL, BLK_BOUNCE_HIGH) >> PAGE_SHIFT)) - dma = 1; - q->limits.bounce_pfn = max(max_low_pfn, b_pfn); -#else - if (b_pfn < blk_max_low_pfn) - dma = 1; - q->limits.bounce_pfn = b_pfn; -#endif - if (dma) { - init_emergency_isa_pool(); - q->bounce_gfp = GFP_NOIO | GFP_DMA; - q->limits.bounce_pfn = b_pfn; - } + q->limits.bounce = bounce; } EXPORT_SYMBOL(blk_queue_bounce_limit); @@ -547,7 +518,7 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b, b->max_write_zeroes_sectors); t->max_zone_append_sectors = min(t->max_zone_append_sectors, b->max_zone_append_sectors); - t->bounce_pfn = min_not_zero(t->bounce_pfn, b->bounce_pfn); + t->bounce = max(t->bounce, b->bounce); t->seg_boundary_mask = min_not_zero(t->seg_boundary_mask, b->seg_boundary_mask); @@ -927,11 +898,3 @@ void blk_queue_set_zoned(struct gendisk *disk, enum blk_zoned_model model) } } EXPORT_SYMBOL_GPL(blk_queue_set_zoned); - -static int __init blk_settings_init(void) -{ - blk_max_low_pfn = max_low_pfn - 1; - blk_max_pfn = max_pfn - 1; - return 0; -} -subsys_initcall(blk_settings_init); diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index 0f4f0c8a7825..e03bedf180ab 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -60,7 +60,7 @@ static ssize_t queue_var_store64(s64 *var, const char *page) static ssize_t queue_requests_show(struct request_queue *q, char *page) { - return queue_var_show(q->nr_requests, (page)); + return queue_var_show(q->nr_requests, page); } static ssize_t @@ -264,6 +264,11 @@ static ssize_t queue_max_hw_sectors_show(struct request_queue *q, char *page) return queue_var_show(max_hw_sectors_kb, (page)); } +static ssize_t queue_virt_boundary_mask_show(struct request_queue *q, char *page) +{ + return queue_var_show(q->limits.virt_boundary_mask, (page)); +} + #define QUEUE_SYSFS_BIT_FNS(name, flag, neg) \ static ssize_t \ queue_##name##_show(struct request_queue *q, char *page) \ @@ -610,6 +615,7 @@ QUEUE_RO_ENTRY(queue_fua, "fua"); QUEUE_RO_ENTRY(queue_dax, "dax"); QUEUE_RW_ENTRY(queue_io_timeout, "io_timeout"); QUEUE_RW_ENTRY(queue_wb_lat, "wbt_lat_usec"); +QUEUE_RO_ENTRY(queue_virt_boundary_mask, "virt_boundary_mask"); #ifdef CONFIG_BLK_DEV_THROTTLING_LOW QUEUE_RW_ENTRY(blk_throtl_sample_time, "throttle_sample_time"); @@ -670,6 +676,7 @@ static struct attribute *queue_attrs[] = { #ifdef CONFIG_BLK_DEV_THROTTLING_LOW &blk_throtl_sample_time_entry.attr, #endif + &queue_virt_boundary_mask_entry.attr, NULL, }; diff --git a/block/blk-zoned.c b/block/blk-zoned.c index c0276b42d9fb..250cb76ee615 100644 --- a/block/blk-zoned.c +++ b/block/blk-zoned.c @@ -52,14 +52,6 @@ const char *blk_zone_cond_str(enum blk_zone_cond zone_cond) } EXPORT_SYMBOL_GPL(blk_zone_cond_str); -static inline sector_t blk_zone_start(struct request_queue *q, - sector_t sector) -{ - sector_t zone_mask = blk_queue_zone_sectors(q) - 1; - - return sector & ~zone_mask; -} - /* * Return true if a request is a write requests that needs zone write locking. */ diff --git a/block/blk.h b/block/blk.h index 3b53e44b967e..8b3591aee0a5 100644 --- a/block/blk.h +++ b/block/blk.h @@ -6,6 +6,7 @@ #include <linux/blk-mq.h> #include <linux/part_stat.h> #include <linux/blk-crypto.h> +#include <linux/memblock.h> /* for max_pfn/max_low_pfn */ #include <xen/xen.h> #include "blk-crypto-internal.h" #include "blk-mq.h" @@ -311,18 +312,20 @@ static inline void blk_throtl_bio_endio(struct bio *bio) { } static inline void blk_throtl_stat_add(struct request *rq, u64 time) { } #endif -#ifdef CONFIG_BOUNCE -extern int init_emergency_isa_pool(void); -extern void blk_queue_bounce(struct request_queue *q, struct bio **bio); -#else -static inline int init_emergency_isa_pool(void) +void __blk_queue_bounce(struct request_queue *q, struct bio **bio); + +static inline bool blk_queue_may_bounce(struct request_queue *q) { - return 0; + return IS_ENABLED(CONFIG_BOUNCE) && + q->limits.bounce == BLK_BOUNCE_HIGH && + max_low_pfn >= max_pfn; } + static inline void blk_queue_bounce(struct request_queue *q, struct bio **bio) { + if (unlikely(blk_queue_may_bounce(q) && bio_has_data(*bio))) + __blk_queue_bounce(q, bio); } -#endif /* CONFIG_BOUNCE */ #ifdef CONFIG_BLK_CGROUP_IOLATENCY extern int blk_iolatency_init(struct request_queue *q); @@ -346,7 +349,6 @@ char *disk_name(struct gendisk *hd, int partno, char *buf); #define ADDPART_FLAG_NONE 0 #define ADDPART_FLAG_RAID 1 #define ADDPART_FLAG_WHOLEDISK 2 -void delete_partition(struct block_device *part); int bdev_add_partition(struct block_device *bdev, int partno, sector_t start, sector_t length); int bdev_del_partition(struct block_device *bdev, int partno); diff --git a/block/bounce.c b/block/bounce.c index 6c441f4f1cd4..94081e013c58 100644 --- a/block/bounce.c +++ b/block/bounce.c @@ -18,7 +18,6 @@ #include <linux/init.h> #include <linux/hash.h> #include <linux/highmem.h> -#include <linux/memblock.h> #include <linux/printk.h> #include <asm/tlbflush.h> @@ -29,7 +28,7 @@ #define ISA_POOL_SIZE 16 static struct bio_set bounce_bio_set, bounce_bio_split; -static mempool_t page_pool, isa_page_pool; +static mempool_t page_pool; static void init_bounce_bioset(void) { @@ -49,11 +48,11 @@ static void init_bounce_bioset(void) bounce_bs_setup = true; } -#if defined(CONFIG_HIGHMEM) static __init int init_emergency_pool(void) { int ret; -#if defined(CONFIG_HIGHMEM) && !defined(CONFIG_MEMORY_HOTPLUG) + +#ifndef CONFIG_MEMORY_HOTPLUG if (max_pfn <= max_low_pfn) return 0; #endif @@ -67,9 +66,7 @@ static __init int init_emergency_pool(void) } __initcall(init_emergency_pool); -#endif -#ifdef CONFIG_HIGHMEM /* * highmem version, map in to vec */ @@ -82,48 +79,6 @@ static void bounce_copy_vec(struct bio_vec *to, unsigned char *vfrom) kunmap_atomic(vto); } -#else /* CONFIG_HIGHMEM */ - -#define bounce_copy_vec(to, vfrom) \ - memcpy(page_address((to)->bv_page) + (to)->bv_offset, vfrom, (to)->bv_len) - -#endif /* CONFIG_HIGHMEM */ - -/* - * allocate pages in the DMA region for the ISA pool - */ -static void *mempool_alloc_pages_isa(gfp_t gfp_mask, void *data) -{ - return mempool_alloc_pages(gfp_mask | GFP_DMA, data); -} - -static DEFINE_MUTEX(isa_mutex); - -/* - * gets called "every" time someone init's a queue with BLK_BOUNCE_ISA - * as the max address, so check if the pool has already been created. - */ -int init_emergency_isa_pool(void) -{ - int ret; - - mutex_lock(&isa_mutex); - - if (mempool_initialized(&isa_page_pool)) { - mutex_unlock(&isa_mutex); - return 0; - } - - ret = mempool_init(&isa_page_pool, ISA_POOL_SIZE, mempool_alloc_pages_isa, - mempool_free_pages, (void *) 0); - BUG_ON(ret); - - pr_info("isa pool size: %d pages\n", ISA_POOL_SIZE); - init_bounce_bioset(); - mutex_unlock(&isa_mutex); - return 0; -} - /* * Simple bounce buffer support for highmem pages. Depending on the * queue gfp mask set, *to may or may not be a highmem page. kmap it @@ -159,7 +114,7 @@ static void copy_to_high_bio_irq(struct bio *to, struct bio *from) } } -static void bounce_end_io(struct bio *bio, mempool_t *pool) +static void bounce_end_io(struct bio *bio) { struct bio *bio_orig = bio->bi_private; struct bio_vec *bvec, orig_vec; @@ -173,7 +128,7 @@ static void bounce_end_io(struct bio *bio, mempool_t *pool) orig_vec = bio_iter_iovec(bio_orig, orig_iter); if (bvec->bv_page != orig_vec.bv_page) { dec_zone_page_state(bvec->bv_page, NR_BOUNCE); - mempool_free(bvec->bv_page, pool); + mempool_free(bvec->bv_page, &page_pool); } bio_advance_iter(bio_orig, &orig_iter, orig_vec.bv_len); } @@ -185,33 +140,17 @@ static void bounce_end_io(struct bio *bio, mempool_t *pool) static void bounce_end_io_write(struct bio *bio) { - bounce_end_io(bio, &page_pool); + bounce_end_io(bio); } -static void bounce_end_io_write_isa(struct bio *bio) -{ - - bounce_end_io(bio, &isa_page_pool); -} - -static void __bounce_end_io_read(struct bio *bio, mempool_t *pool) +static void bounce_end_io_read(struct bio *bio) { struct bio *bio_orig = bio->bi_private; if (!bio->bi_status) copy_to_high_bio_irq(bio_orig, bio); - bounce_end_io(bio, pool); -} - -static void bounce_end_io_read(struct bio *bio) -{ - __bounce_end_io_read(bio, &page_pool); -} - -static void bounce_end_io_read_isa(struct bio *bio) -{ - __bounce_end_io_read(bio, &isa_page_pool); + bounce_end_io(bio); } static struct bio *bounce_clone_bio(struct bio *bio_src) @@ -241,12 +180,8 @@ static struct bio *bounce_clone_bio(struct bio *bio_src) * asking for trouble and would force extra work on * __bio_clone_fast() anyways. */ - if (bio_is_passthrough(bio_src)) - bio = bio_kmalloc(GFP_NOIO | __GFP_NOFAIL, - bio_segments(bio_src)); - else - bio = bio_alloc_bioset(GFP_NOIO, bio_segments(bio_src), - &bounce_bio_set); + bio = bio_alloc_bioset(GFP_NOIO, bio_segments(bio_src), + &bounce_bio_set); bio->bi_bdev = bio_src->bi_bdev; if (bio_flagged(bio_src, BIO_REMAPPED)) bio_set_flag(bio, BIO_REMAPPED); @@ -287,8 +222,7 @@ err_put: return NULL; } -static void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig, - mempool_t *pool) +void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig) { struct bio *bio; int rw = bio_data_dir(*bio_orig); @@ -301,14 +235,13 @@ static void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig, bio_for_each_segment(from, *bio_orig, iter) { if (i++ < BIO_MAX_VECS) sectors += from.bv_len >> 9; - if (page_to_pfn(from.bv_page) > q->limits.bounce_pfn) + if (PageHighMem(from.bv_page)) bounce = true; } if (!bounce) return; - if (!bio_is_passthrough(*bio_orig) && - sectors < bio_sectors(*bio_orig)) { + if (sectors < bio_sectors(*bio_orig)) { bio = bio_split(*bio_orig, sectors, GFP_NOIO, &bounce_bio_split); bio_chain(bio, *bio_orig); submit_bio_noacct(*bio_orig); @@ -324,10 +257,10 @@ static void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig, for (i = 0, to = bio->bi_io_vec; i < bio->bi_vcnt; to++, i++) { struct page *page = to->bv_page; - if (page_to_pfn(page) <= q->limits.bounce_pfn) + if (!PageHighMem(page)) continue; - to->bv_page = mempool_alloc(pool, q->bounce_gfp); + to->bv_page = mempool_alloc(&page_pool, GFP_NOIO); inc_zone_page_state(to->bv_page, NR_BOUNCE); if (rw == WRITE) { @@ -346,46 +279,11 @@ static void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig, bio->bi_flags |= (1 << BIO_BOUNCED); - if (pool == &page_pool) { + if (rw == READ) + bio->bi_end_io = bounce_end_io_read; + else bio->bi_end_io = bounce_end_io_write; - if (rw == READ) - bio->bi_end_io = bounce_end_io_read; - } else { - bio->bi_end_io = bounce_end_io_write_isa; - if (rw == READ) - bio->bi_end_io = bounce_end_io_read_isa; - } bio->bi_private = *bio_orig; *bio_orig = bio; } - -void blk_queue_bounce(struct request_queue *q, struct bio **bio_orig) -{ - mempool_t *pool; - - /* - * Data-less bio, nothing to bounce - */ - if (!bio_has_data(*bio_orig)) - return; - - /* - * for non-isa bounce case, just check if the bounce pfn is equal - * to or bigger than the highest pfn in the system -- in that case, - * don't waste time iterating over bio segments - */ - if (!(q->bounce_gfp & GFP_DMA)) { - if (q->limits.bounce_pfn >= blk_max_pfn) - return; - pool = &page_pool; - } else { - BUG_ON(!mempool_initialized(&isa_page_pool)); - pool = &isa_page_pool; - } - - /* - * slow path - */ - __blk_queue_bounce(q, bio_orig, pool); -} diff --git a/block/elevator.c b/block/elevator.c index 293c5c81397a..440699c28119 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -621,7 +621,8 @@ static inline bool elv_support_iosched(struct request_queue *q) */ static struct elevator_type *elevator_get_default(struct request_queue *q) { - if (q->nr_hw_queues != 1) + if (q->nr_hw_queues != 1 && + !blk_mq_is_sbitmap_shared(q->tag_set->flags)) return NULL; return elevator_get(q, "mq-deadline", false); diff --git a/block/genhd.c b/block/genhd.c index 8c8f543572e6..39ca97b0edc6 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -161,81 +161,6 @@ static void part_in_flight_rw(struct block_device *part, inflight[1] = 0; } -/** - * disk_part_iter_init - initialize partition iterator - * @piter: iterator to initialize - * @disk: disk to iterate over - * @flags: DISK_PITER_* flags - * - * Initialize @piter so that it iterates over partitions of @disk. - * - * CONTEXT: - * Don't care. - */ -void disk_part_iter_init(struct disk_part_iter *piter, struct gendisk *disk, - unsigned int flags) -{ - piter->disk = disk; - piter->part = NULL; - if (flags & (DISK_PITER_INCL_PART0 | DISK_PITER_INCL_EMPTY_PART0)) - piter->idx = 0; - else - piter->idx = 1; - piter->flags = flags; -} - -/** - * disk_part_iter_next - proceed iterator to the next partition and return it - * @piter: iterator of interest - * - * Proceed @piter to the next partition and return it. - * - * CONTEXT: - * Don't care. - */ -struct block_device *disk_part_iter_next(struct disk_part_iter *piter) -{ - struct block_device *part; - unsigned long idx; - - /* put the last partition */ - disk_part_iter_exit(piter); - - rcu_read_lock(); - xa_for_each_start(&piter->disk->part_tbl, idx, part, piter->idx) { - if (!bdev_nr_sectors(part) && - !(piter->flags & DISK_PITER_INCL_EMPTY) && - !(piter->flags & DISK_PITER_INCL_EMPTY_PART0 && - piter->idx == 0)) - continue; - - piter->part = bdgrab(part); - if (!piter->part) - continue; - piter->idx = idx + 1; - break; - } - rcu_read_unlock(); - - return piter->part; -} - -/** - * disk_part_iter_exit - finish up partition iteration - * @piter: iter of interest - * - * Called when iteration is over. Cleans up @piter. - * - * CONTEXT: - * Don't care. - */ -void disk_part_iter_exit(struct disk_part_iter *piter) -{ - if (piter->part) - bdput(piter->part); - piter->part = NULL; -} - /* * Can be deleted altogether. Later. * @@ -472,13 +397,22 @@ static char *bdevt_str(dev_t devt, char *buf) void disk_uevent(struct gendisk *disk, enum kobject_action action) { - struct disk_part_iter piter; struct block_device *part; + unsigned long idx; + + rcu_read_lock(); + xa_for_each(&disk->part_tbl, idx, part) { + if (bdev_is_partition(part) && !bdev_nr_sectors(part)) + continue; + if (!bdgrab(part)) + continue; - disk_part_iter_init(&piter, disk, DISK_PITER_INCL_EMPTY_PART0); - while ((part = disk_part_iter_next(&piter))) + rcu_read_unlock(); kobject_uevent(bdev_kobj(part), action); - disk_part_iter_exit(&piter); + bdput(part); + rcu_read_lock(); + } + rcu_read_unlock(); } EXPORT_SYMBOL_GPL(disk_uevent); @@ -646,18 +580,6 @@ void device_add_disk_no_queue_reg(struct device *parent, struct gendisk *disk) } EXPORT_SYMBOL(device_add_disk_no_queue_reg); -static void invalidate_partition(struct block_device *bdev) -{ - fsync_bdev(bdev); - __invalidate_device(bdev, true); - - /* - * Unhash the bdev inode for this device so that it can't be looked - * up any more even if openers still hold references to it. - */ - remove_inode_hash(bdev->bd_inode); -} - /** * del_gendisk - remove the gendisk * @disk: the struct gendisk to remove @@ -679,9 +601,6 @@ static void invalidate_partition(struct block_device *bdev) */ void del_gendisk(struct gendisk *disk) { - struct disk_part_iter piter; - struct block_device *part; - might_sleep(); if (WARN_ON_ONCE(!disk->queue)) @@ -696,15 +615,19 @@ void del_gendisk(struct gendisk *disk) */ down_write(&bdev_lookup_sem); - /* invalidate stuff */ - disk_part_iter_init(&piter, disk, DISK_PITER_INCL_EMPTY); - while ((part = disk_part_iter_next(&piter))) { - invalidate_partition(part); - delete_partition(part); - } - disk_part_iter_exit(&piter); + mutex_lock(&disk->part0->bd_mutex); + blk_drop_partitions(disk); + mutex_unlock(&disk->part0->bd_mutex); + + fsync_bdev(disk->part0); + __invalidate_device(disk->part0, true); + + /* + * Unhash the bdev inode for this device so that it can't be looked + * up any more even if openers still hold references to it. + */ + remove_inode_hash(disk->part0->bd_inode); - invalidate_partition(disk->part0); set_capacity(disk, 0); disk->flags &= ~GENHD_FL_UP; up_write(&bdev_lookup_sem); @@ -817,10 +740,10 @@ void __init printk_all_partitions(void) class_dev_iter_init(&iter, &block_class, NULL, &disk_type); while ((dev = class_dev_iter_next(&iter))) { struct gendisk *disk = dev_to_disk(dev); - struct disk_part_iter piter; struct block_device *part; char name_buf[BDEVNAME_SIZE]; char devt_buf[BDEVT_SIZE]; + unsigned long idx; /* * Don't show empty devices or things that have been @@ -831,30 +754,29 @@ void __init printk_all_partitions(void) continue; /* - * Note, unlike /proc/partitions, I am showing the - * numbers in hex - the same format as the root= - * option takes. + * Note, unlike /proc/partitions, I am showing the numbers in + * hex - the same format as the root= option takes. */ - disk_part_iter_init(&piter, disk, DISK_PITER_INCL_PART0); - while ((part = disk_part_iter_next(&piter))) { - bool is_part0 = part == disk->part0; - - printk("%s%s %10llu %s %s", is_part0 ? "" : " ", + rcu_read_lock(); + xa_for_each(&disk->part_tbl, idx, part) { + if (!bdev_nr_sectors(part)) + continue; + printk("%s%s %10llu %s %s", + bdev_is_partition(part) ? " " : "", bdevt_str(part->bd_dev, devt_buf), bdev_nr_sectors(part) >> 1, disk_name(disk, part->bd_partno, name_buf), part->bd_meta_info ? part->bd_meta_info->uuid : ""); - if (is_part0) { - if (dev->parent && dev->parent->driver) - printk(" driver: %s\n", - dev->parent->driver->name); - else - printk(" (driver?)\n"); - } else + if (bdev_is_partition(part)) printk("\n"); + else if (dev->parent && dev->parent->driver) + printk(" driver: %s\n", + dev->parent->driver->name); + else + printk(" (driver?)\n"); } - disk_part_iter_exit(&piter); + rcu_read_unlock(); } class_dev_iter_exit(&iter); } @@ -919,8 +841,8 @@ static void *show_partition_start(struct seq_file *seqf, loff_t *pos) static int show_partition(struct seq_file *seqf, void *v) { struct gendisk *sgp = v; - struct disk_part_iter piter; struct block_device *part; + unsigned long idx; char buf[BDEVNAME_SIZE]; /* Don't show non-partitionable removeable devices or empty devices */ @@ -930,15 +852,16 @@ static int show_partition(struct seq_file *seqf, void *v) if (sgp->flags & GENHD_FL_SUPPRESS_PARTITION_INFO) return 0; - /* show the full disk and all non-0 size partitions of it */ - disk_part_iter_init(&piter, sgp, DISK_PITER_INCL_PART0); - while ((part = disk_part_iter_next(&piter))) + rcu_read_lock(); + xa_for_each(&sgp->part_tbl, idx, part) { + if (!bdev_nr_sectors(part)) + continue; seq_printf(seqf, "%4d %7d %10llu %s\n", MAJOR(part->bd_dev), MINOR(part->bd_dev), bdev_nr_sectors(part) >> 1, disk_name(sgp, part->bd_partno, buf)); - disk_part_iter_exit(&piter); - + } + rcu_read_unlock(); return 0; } @@ -1247,11 +1170,11 @@ const struct device_type disk_type = { static int diskstats_show(struct seq_file *seqf, void *v) { struct gendisk *gp = v; - struct disk_part_iter piter; struct block_device *hd; char buf[BDEVNAME_SIZE]; unsigned int inflight; struct disk_stats stat; + unsigned long idx; /* if (&disk_to_dev(gp)->kobj.entry == block_class.devices.next) @@ -1261,8 +1184,10 @@ static int diskstats_show(struct seq_file *seqf, void *v) "\n\n"); */ - disk_part_iter_init(&piter, gp, DISK_PITER_INCL_EMPTY_PART0); - while ((hd = disk_part_iter_next(&piter))) { + rcu_read_lock(); + xa_for_each(&gp->part_tbl, idx, hd) { + if (bdev_is_partition(hd) && !bdev_nr_sectors(hd)) + continue; part_stat_read_all(hd, &stat); if (queue_is_mq(gp->queue)) inflight = blk_mq_in_flight(gp->queue, hd); @@ -1305,7 +1230,7 @@ static int diskstats_show(struct seq_file *seqf, void *v) NSEC_PER_MSEC) ); } - disk_part_iter_exit(&piter); + rcu_read_unlock(); return 0; } diff --git a/block/ioprio.c b/block/ioprio.c index 364d2294ba90..bee628f9f1b2 100644 --- a/block/ioprio.c +++ b/block/ioprio.c @@ -119,11 +119,17 @@ SYSCALL_DEFINE3(ioprio_set, int, which, int, who, int, ioprio) pgrp = task_pgrp(current); else pgrp = find_vpid(who); + + read_lock(&tasklist_lock); do_each_pid_thread(pgrp, PIDTYPE_PGID, p) { ret = set_task_ioprio(p, ioprio); - if (ret) - break; + if (ret) { + read_unlock(&tasklist_lock); + goto out; + } } while_each_pid_thread(pgrp, PIDTYPE_PGID, p); + read_unlock(&tasklist_lock); + break; case IOPRIO_WHO_USER: uid = make_kuid(current_user_ns(), who); @@ -153,6 +159,7 @@ free_uid: ret = -EINVAL; } +out: rcu_read_unlock(); return ret; } diff --git a/block/mq-deadline.c b/block/mq-deadline.c index f3631a287466..04aded71ead2 100644 --- a/block/mq-deadline.c +++ b/block/mq-deadline.c @@ -500,11 +500,8 @@ static void dd_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, trace_block_rq_insert(rq); - if (at_head || blk_rq_is_passthrough(rq)) { - if (at_head) - list_add(&rq->queuelist, &dd->dispatch); - else - list_add_tail(&rq->queuelist, &dd->dispatch); + if (at_head) { + list_add(&rq->queuelist, &dd->dispatch); } else { deadline_add_rq_rb(dd, rq); diff --git a/block/partitions/core.c b/block/partitions/core.c index 46f055bc7ecb..dc60ecf46fe6 100644 --- a/block/partitions/core.c +++ b/block/partitions/core.c @@ -285,8 +285,11 @@ struct device_type part_type = { * Must be called either with bd_mutex held, before a disk can be opened or * after all disk users are gone. */ -void delete_partition(struct block_device *part) +static void delete_partition(struct block_device *part) { + fsync_bdev(part); + __invalidate_device(part, true); + xa_erase(&part->bd_disk->part_tbl, part->bd_partno); kobject_put(part->bd_holder_dir); device_del(&part->bd_device); @@ -424,21 +427,21 @@ out_put: static bool partition_overlaps(struct gendisk *disk, sector_t start, sector_t length, int skip_partno) { - struct disk_part_iter piter; struct block_device *part; bool overlap = false; - - disk_part_iter_init(&piter, disk, DISK_PITER_INCL_EMPTY); - while ((part = disk_part_iter_next(&piter))) { - if (part->bd_partno == skip_partno || - start >= part->bd_start_sect + bdev_nr_sectors(part) || - start + length <= part->bd_start_sect) - continue; - overlap = true; - break; + unsigned long idx; + + rcu_read_lock(); + xa_for_each_start(&disk->part_tbl, idx, part, 1) { + if (part->bd_partno != skip_partno && + start < part->bd_start_sect + bdev_nr_sectors(part) && + start + length > part->bd_start_sect) { + overlap = true; + break; + } } + rcu_read_unlock(); - disk_part_iter_exit(&piter); return overlap; } @@ -475,9 +478,6 @@ int bdev_del_partition(struct block_device *bdev, int partno) if (part->bd_openers) goto out_unlock; - sync_blockdev(part); - invalidate_bdev(part); - delete_partition(part); ret = 0; out_unlock: @@ -533,28 +533,20 @@ static bool disk_unlock_native_capacity(struct gendisk *disk) } } -int blk_drop_partitions(struct block_device *bdev) +void blk_drop_partitions(struct gendisk *disk) { - struct disk_part_iter piter; struct block_device *part; + unsigned long idx; - if (bdev->bd_part_count) - return -EBUSY; + lockdep_assert_held(&disk->part0->bd_mutex); - sync_blockdev(bdev); - invalidate_bdev(bdev); - - disk_part_iter_init(&piter, bdev->bd_disk, DISK_PITER_INCL_EMPTY); - while ((part = disk_part_iter_next(&piter))) + xa_for_each_start(&disk->part_tbl, idx, part, 1) { + if (!bdgrab(part)) + continue; delete_partition(part); - disk_part_iter_exit(&piter); - - return 0; + bdput(part); + } } -#ifdef CONFIG_S390 -/* for historic reasons in the DASD driver */ -EXPORT_SYMBOL_GPL(blk_drop_partitions); -#endif static bool blk_add_partition(struct gendisk *disk, struct block_device *bdev, struct parsed_partitions *state, int p) diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c index 6599bac0a78c..1b3fe99b83a6 100644 --- a/block/scsi_ioctl.c +++ b/block/scsi_ioctl.c @@ -353,10 +353,6 @@ static int sg_io(struct request_queue *q, struct gendisk *bd_disk, start_time = jiffies; - /* ignore return value. All information is passed back to caller - * (if he doesn't check that is his problem). - * N.B. a non-zero SCSI status is _not_ necessarily an error. - */ blk_execute_rq(bd_disk, rq, at_head); hdr->duration = jiffies_to_msecs(jiffies - start_time); @@ -431,7 +427,7 @@ int sg_scsi_ioctl(struct request_queue *q, struct gendisk *disk, fmode_t mode, bytes = max(in_len, out_len); if (bytes) { - buffer = kzalloc(bytes, q->bounce_gfp | GFP_USER| __GFP_NOWARN); + buffer = kzalloc(bytes, GFP_NOIO | GFP_USER | __GFP_NOWARN); if (!buffer) return -ENOMEM; diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index 48b8934970f3..fd8b6febbf70 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -1043,8 +1043,7 @@ int ata_scsi_dev_config(struct scsi_device *sdev, struct ata_device *dev) blk_queue_max_segments(q, queue_max_segments(q) - 1); sdev->dma_drain_len = ATAPI_MAX_DRAIN; - sdev->dma_drain_buf = kmalloc(sdev->dma_drain_len, - q->bounce_gfp | GFP_KERNEL); + sdev->dma_drain_buf = kmalloc(sdev->dma_drain_len, GFP_NOIO); if (!sdev->dma_drain_buf) { ata_dev_err(dev, "drain buffer allocation failed\n"); return -ENOMEM; diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index fc4b0f1aa86d..bd3556585122 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -1199,6 +1199,42 @@ try_next_bio: return 1; } +/** + * bio_list_copy_data - copy contents of data buffers from one chain of bios to + * another + * @src: source bio list + * @dst: destination bio list + * + * Stops when it reaches the end of either the @src list or @dst list - that is, + * copies min(src->bi_size, dst->bi_size) bytes (or the equivalent for lists of + * bios). + */ +static void bio_list_copy_data(struct bio *dst, struct bio *src) +{ + struct bvec_iter src_iter = src->bi_iter; + struct bvec_iter dst_iter = dst->bi_iter; + + while (1) { + if (!src_iter.bi_size) { + src = src->bi_next; + if (!src) + break; + + src_iter = src->bi_iter; + } + + if (!dst_iter.bi_size) { + dst = dst->bi_next; + if (!dst) + break; + + dst_iter = dst->bi_iter; + } + + bio_copy_data_iter(dst, &dst_iter, src, &src_iter); + } +} + /* * Assemble a bio to write one packet and queue the bio for processing * by the underlying block device. diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c index b705988629f2..f6ca2fbb711e 100644 --- a/drivers/nvme/host/lightnvm.c +++ b/drivers/nvme/host/lightnvm.c @@ -660,7 +660,7 @@ static struct request *nvme_nvm_alloc_request(struct request_queue *q, rq->cmd_flags &= ~REQ_FAILFAST_DRIVER; if (rqd->bio) - blk_rq_append_bio(rq, &rqd->bio); + blk_rq_append_bio(rq, rqd->bio); else rq->ioprio = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, IOPRIO_NORM); diff --git a/drivers/s390/block/dasd_genhd.c b/drivers/s390/block/dasd_genhd.c index a9698fba9b76..8d6587ec73e2 100644 --- a/drivers/s390/block/dasd_genhd.c +++ b/drivers/s390/block/dasd_genhd.c @@ -146,12 +146,11 @@ void dasd_destroy_partitions(struct dasd_block *block) block->bdev = NULL; mutex_lock(&bdev->bd_mutex); - blk_drop_partitions(bdev); + bdev_disk_changed(bdev, true); mutex_unlock(&bdev->bd_mutex); /* Matching blkdev_put to the blkdev_get in dasd_scan_partitions. */ blkdev_put(bdev, FMODE_READ); - set_capacity(block->gdp, 0); } int dasd_gendisk_init(void) diff --git a/drivers/scsi/BusLogic.c b/drivers/scsi/BusLogic.c index ccb061ab0a0a..69068081f439 100644 --- a/drivers/scsi/BusLogic.c +++ b/drivers/scsi/BusLogic.c @@ -562,60 +562,6 @@ done: /* - blogic_add_probeaddr_isa appends a single ISA I/O Address to the list - of I/O Address and Bus Probe Information to be checked for potential BusLogic - Host Adapters. -*/ - -static void __init blogic_add_probeaddr_isa(unsigned long io_addr) -{ - struct blogic_probeinfo *probeinfo; - if (blogic_probeinfo_count >= BLOGIC_MAX_ADAPTERS) - return; - probeinfo = &blogic_probeinfo_list[blogic_probeinfo_count++]; - probeinfo->adapter_type = BLOGIC_MULTIMASTER; - probeinfo->adapter_bus_type = BLOGIC_ISA_BUS; - probeinfo->io_addr = io_addr; - probeinfo->pci_device = NULL; -} - - -/* - blogic_init_probeinfo_isa initializes the list of I/O Address and - Bus Probe Information to be checked for potential BusLogic SCSI Host Adapters - only from the list of standard BusLogic MultiMaster ISA I/O Addresses. -*/ - -static void __init blogic_init_probeinfo_isa(struct blogic_adapter *adapter) -{ - /* - If BusLogic Driver Options specifications requested that ISA - Bus Probes be inhibited, do not proceed further. - */ - if (blogic_probe_options.noprobe_isa) - return; - /* - Append the list of standard BusLogic MultiMaster ISA I/O Addresses. - */ - if (!blogic_probe_options.limited_isa || blogic_probe_options.probe330) - blogic_add_probeaddr_isa(0x330); - if (!blogic_probe_options.limited_isa || blogic_probe_options.probe334) - blogic_add_probeaddr_isa(0x334); - if (!blogic_probe_options.limited_isa || blogic_probe_options.probe230) - blogic_add_probeaddr_isa(0x230); - if (!blogic_probe_options.limited_isa || blogic_probe_options.probe234) - blogic_add_probeaddr_isa(0x234); - if (!blogic_probe_options.limited_isa || blogic_probe_options.probe130) - blogic_add_probeaddr_isa(0x130); - if (!blogic_probe_options.limited_isa || blogic_probe_options.probe134) - blogic_add_probeaddr_isa(0x134); -} - - -#ifdef CONFIG_PCI - - -/* blogic_sort_probeinfo sorts a section of blogic_probeinfo_list in order of increasing PCI Bus and Device Number. */ @@ -667,14 +613,11 @@ static int __init blogic_init_mm_probeinfo(struct blogic_adapter *adapter) int nonpr_mmcount = 0, mmcount = 0; bool force_scan_order = false; bool force_scan_order_checked = false; - bool addr_seen[6]; struct pci_dev *pci_device = NULL; int i; if (blogic_probeinfo_count >= BLOGIC_MAX_ADAPTERS) return 0; blogic_probeinfo_count++; - for (i = 0; i < 6; i++) - addr_seen[i] = false; /* Iterate over the MultiMaster PCI Host Adapters. For each enumerated host adapter, determine whether its ISA Compatible @@ -744,11 +687,8 @@ static int __init blogic_init_mm_probeinfo(struct blogic_adapter *adapter) host_adapter->io_addr = io_addr; blogic_intreset(host_adapter); if (blogic_cmd(host_adapter, BLOGIC_INQ_PCI_INFO, NULL, 0, - &adapter_info, sizeof(adapter_info)) == - sizeof(adapter_info)) { - if (adapter_info.isa_port < 6) - addr_seen[adapter_info.isa_port] = true; - } else + &adapter_info, sizeof(adapter_info)) != + sizeof(adapter_info)) adapter_info.isa_port = BLOGIC_IO_DISABLE; /* Issue the Modify I/O Address command to disable the @@ -836,45 +776,6 @@ static int __init blogic_init_mm_probeinfo(struct blogic_adapter *adapter) blogic_sort_probeinfo(&blogic_probeinfo_list[nonpr_mmindex], nonpr_mmcount); /* - If no PCI MultiMaster Host Adapter is assigned the Primary - I/O Address, then the Primary I/O Address must be probed - explicitly before any PCI host adapters are probed. - */ - if (!blogic_probe_options.noprobe_isa) - if (pr_probeinfo->io_addr == 0 && - (!blogic_probe_options.limited_isa || - blogic_probe_options.probe330)) { - pr_probeinfo->adapter_type = BLOGIC_MULTIMASTER; - pr_probeinfo->adapter_bus_type = BLOGIC_ISA_BUS; - pr_probeinfo->io_addr = 0x330; - } - /* - Append the list of standard BusLogic MultiMaster ISA I/O Addresses, - omitting the Primary I/O Address which has already been handled. - */ - if (!blogic_probe_options.noprobe_isa) { - if (!addr_seen[1] && - (!blogic_probe_options.limited_isa || - blogic_probe_options.probe334)) - blogic_add_probeaddr_isa(0x334); - if (!addr_seen[2] && - (!blogic_probe_options.limited_isa || - blogic_probe_options.probe230)) - blogic_add_probeaddr_isa(0x230); - if (!addr_seen[3] && - (!blogic_probe_options.limited_isa || - blogic_probe_options.probe234)) - blogic_add_probeaddr_isa(0x234); - if (!addr_seen[4] && - (!blogic_probe_options.limited_isa || - blogic_probe_options.probe130)) - blogic_add_probeaddr_isa(0x130); - if (!addr_seen[5] && - (!blogic_probe_options.limited_isa || - blogic_probe_options.probe134)) - blogic_add_probeaddr_isa(0x134); - } - /* Iterate over the older non-compliant MultiMaster PCI Host Adapters, noting the PCI bus location and assigned IRQ Channel. */ @@ -1078,18 +979,10 @@ static void __init blogic_init_probeinfo_list(struct blogic_adapter *adapter) } } } - } else { - blogic_init_probeinfo_isa(adapter); } } -#else -#define blogic_init_probeinfo_list(adapter) \ - blogic_init_probeinfo_isa(adapter) -#endif /* CONFIG_PCI */ - - /* blogic_failure prints a standardized error message, and then returns false. */ @@ -1539,14 +1432,6 @@ static bool __init blogic_rdconfig(struct blogic_adapter *adapter) else if (config.irq_ch15) adapter->irq_ch = 15; } - if (adapter->adapter_bus_type == BLOGIC_ISA_BUS) { - if (config.dma_ch5) - adapter->dma_ch = 5; - else if (config.dma_ch6) - adapter->dma_ch = 6; - else if (config.dma_ch7) - adapter->dma_ch = 7; - } /* Determine whether Extended Translation is enabled and save it in the Host Adapter structure. @@ -1686,8 +1571,7 @@ static bool __init blogic_rdconfig(struct blogic_adapter *adapter) if (adapter->fw_ver[0] == '5') adapter->adapter_qdepth = 192; else if (adapter->fw_ver[0] == '4') - adapter->adapter_qdepth = (adapter->adapter_bus_type != - BLOGIC_ISA_BUS ? 100 : 50); + adapter->adapter_qdepth = 100; else adapter->adapter_qdepth = 30; if (strcmp(adapter->fw_ver, "3.31") >= 0) { @@ -1728,25 +1612,16 @@ static bool __init blogic_rdconfig(struct blogic_adapter *adapter) */ adapter->bios_addr = ext_setupinfo.bios_addr << 12; /* - ISA Host Adapters require Bounce Buffers if there is more than - 16MB memory. - */ - if (adapter->adapter_bus_type == BLOGIC_ISA_BUS && - (void *) high_memory > (void *) MAX_DMA_ADDRESS) - adapter->need_bouncebuf = true; - /* BusLogic BT-445S Host Adapters prior to board revision E have a hardware bug whereby when the BIOS is enabled, transfers to/from the same address range the BIOS occupies modulo 16MB are handled incorrectly. Only properly functioning BT-445S Host Adapters - have firmware version 3.37, so require that ISA Bounce Buffers - be used for the buggy BT-445S models if there is more than 16MB - memory. + have firmware version 3.37. */ - if (adapter->bios_addr > 0 && strcmp(adapter->model, "BT-445S") == 0 && - strcmp(adapter->fw_ver, "3.37") < 0 && - (void *) high_memory > (void *) MAX_DMA_ADDRESS) - adapter->need_bouncebuf = true; + if (adapter->bios_addr > 0 && + strcmp(adapter->model, "BT-445S") == 0 && + strcmp(adapter->fw_ver, "3.37") < 0) + return blogic_failure(adapter, "Too old firmware"); /* Initialize parameters common to MultiMaster and FlashPoint Host Adapters. @@ -1769,14 +1644,9 @@ common: if (adapter->drvr_opts != NULL && adapter->drvr_opts->qdepth[tgt_id] > 0) qdepth = adapter->drvr_opts->qdepth[tgt_id]; - else if (adapter->need_bouncebuf) - qdepth = BLOGIC_TAG_DEPTH_BB; adapter->qdepth[tgt_id] = qdepth; } - if (adapter->need_bouncebuf) - adapter->untag_qdepth = BLOGIC_UNTAG_DEPTH_BB; - else - adapter->untag_qdepth = BLOGIC_UNTAG_DEPTH; + adapter->untag_qdepth = BLOGIC_UNTAG_DEPTH; if (adapter->drvr_opts != NULL) adapter->common_qdepth = adapter->drvr_opts->common_qdepth; if (adapter->common_qdepth > 0 && @@ -1839,11 +1709,7 @@ static bool __init blogic_reportconfig(struct blogic_adapter *adapter) blogic_info("Configuring BusLogic Model %s %s%s%s%s SCSI Host Adapter\n", adapter, adapter->model, blogic_adapter_busnames[adapter->adapter_bus_type], (adapter->wide ? " Wide" : ""), (adapter->differential ? " Differential" : ""), (adapter->ultra ? " Ultra" : "")); blogic_info(" Firmware Version: %s, I/O Address: 0x%lX, IRQ Channel: %d/%s\n", adapter, adapter->fw_ver, adapter->io_addr, adapter->irq_ch, (adapter->level_int ? "Level" : "Edge")); if (adapter->adapter_bus_type != BLOGIC_PCI_BUS) { - blogic_info(" DMA Channel: ", adapter); - if (adapter->dma_ch > 0) - blogic_info("%d, ", adapter, adapter->dma_ch); - else - blogic_info("None, ", adapter); + blogic_info(" DMA Channel: None, ", adapter); if (adapter->bios_addr > 0) blogic_info("BIOS Address: 0x%lX, ", adapter, adapter->bios_addr); @@ -1996,18 +1862,6 @@ static bool __init blogic_getres(struct blogic_adapter *adapter) } adapter->irq_acquired = true; /* - Acquire exclusive access to the DMA Channel. - */ - if (adapter->dma_ch > 0) { - if (request_dma(adapter->dma_ch, adapter->full_model) < 0) { - blogic_err("UNABLE TO ACQUIRE DMA CHANNEL %d - DETACHING\n", adapter, adapter->dma_ch); - return false; - } - set_dma_mode(adapter->dma_ch, DMA_MODE_CASCADE); - enable_dma(adapter->dma_ch); - adapter->dma_chan_acquired = true; - } - /* Indicate the System Resource Acquisition completed successfully, */ return true; @@ -2027,11 +1881,6 @@ static void blogic_relres(struct blogic_adapter *adapter) if (adapter->irq_acquired) free_irq(adapter->irq_ch, adapter); /* - Release exclusive access to the DMA Channel. - */ - if (adapter->dma_chan_acquired) - free_dma(adapter->dma_ch); - /* Release any allocated memory structs not released elsewhere */ if (adapter->mbox_space) @@ -2299,7 +2148,6 @@ static void __init blogic_inithoststruct(struct blogic_adapter *adapter, host->this_id = adapter->scsi_id; host->can_queue = adapter->drvr_qdepth; host->sg_tablesize = adapter->drvr_sglimit; - host->unchecked_isa_dma = adapter->need_bouncebuf; host->cmd_per_lun = adapter->untag_qdepth; } @@ -3666,37 +3514,7 @@ static int __init blogic_parseopts(char *options) memset(drvr_opts, 0, sizeof(struct blogic_drvr_options)); while (*options != '\0' && *options != ';') { - /* Probing Options. */ - if (blogic_parse(&options, "IO:")) { - unsigned long io_addr = simple_strtoul(options, - &options, 0); - blogic_probe_options.limited_isa = true; - switch (io_addr) { - case 0x330: - blogic_probe_options.probe330 = true; - break; - case 0x334: - blogic_probe_options.probe334 = true; - break; - case 0x230: - blogic_probe_options.probe230 = true; - break; - case 0x234: - blogic_probe_options.probe234 = true; - break; - case 0x130: - blogic_probe_options.probe130 = true; - break; - case 0x134: - blogic_probe_options.probe134 = true; - break; - default: - blogic_err("BusLogic: Invalid Driver Options (invalid I/O Address 0x%lX)\n", NULL, io_addr); - return 0; - } - } else if (blogic_parse(&options, "NoProbeISA")) - blogic_probe_options.noprobe_isa = true; - else if (blogic_parse(&options, "NoProbePCI")) + if (blogic_parse(&options, "NoProbePCI")) blogic_probe_options.noprobe_pci = true; else if (blogic_parse(&options, "NoProbe")) blogic_probe_options.noprobe = true; @@ -3851,7 +3669,6 @@ static struct scsi_host_template blogic_template = { #if 0 .eh_abort_handler = blogic_abort, #endif - .unchecked_isa_dma = 1, .max_sectors = 128, }; diff --git a/drivers/scsi/BusLogic.h b/drivers/scsi/BusLogic.h index 6182cc8a0344..a8e4a19788a7 100644 --- a/drivers/scsi/BusLogic.h +++ b/drivers/scsi/BusLogic.h @@ -237,18 +237,10 @@ struct blogic_probeinfo { struct blogic_probe_options { bool noprobe:1; /* Bit 0 */ - bool noprobe_isa:1; /* Bit 1 */ bool noprobe_pci:1; /* Bit 2 */ bool nosort_pci:1; /* Bit 3 */ bool multimaster_first:1; /* Bit 4 */ bool flashpoint_first:1; /* Bit 5 */ - bool limited_isa:1; /* Bit 6 */ - bool probe330:1; /* Bit 7 */ - bool probe334:1; /* Bit 8 */ - bool probe230:1; /* Bit 9 */ - bool probe234:1; /* Bit 10 */ - bool probe130:1; /* Bit 11 */ - bool probe134:1; /* Bit 12 */ }; /* @@ -997,10 +989,8 @@ struct blogic_adapter { unsigned char bus; unsigned char dev; unsigned char irq_ch; - unsigned char dma_ch; unsigned char scsi_id; bool irq_acquired:1; - bool dma_chan_acquired:1; bool ext_trans_enable:1; bool parity:1; bool reset_enabled:1; @@ -1013,7 +1003,6 @@ struct blogic_adapter { bool terminfo_valid:1; bool low_term:1; bool high_term:1; - bool need_bouncebuf:1; bool strict_rr:1; bool scam_enabled:1; bool scam_lev2:1; diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index 06b87c7f6bab..3d114be5b662 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -497,7 +497,7 @@ config SCSI_HPTIOP config SCSI_BUSLOGIC tristate "BusLogic SCSI support" - depends on (PCI || ISA) && SCSI && ISA_DMA_API && VIRT_TO_BUS + depends on PCI && SCSI && VIRT_TO_BUS help This is support for BusLogic MultiMaster and FlashPoint SCSI Host Adapters. Consult the SCSI-HOWTO, available from diff --git a/drivers/scsi/advansys.c b/drivers/scsi/advansys.c index ec5627890809..993596d44a12 100644 --- a/drivers/scsi/advansys.c +++ b/drivers/scsi/advansys.c @@ -84,8 +84,6 @@ typedef unsigned char uchar; #define ASC_CS_TYPE unsigned short -#define ASC_IS_ISA (0x0001) -#define ASC_IS_ISAPNP (0x0081) #define ASC_IS_EISA (0x0002) #define ASC_IS_PCI (0x0004) #define ASC_IS_PCI_ULTRA (0x0104) @@ -101,11 +99,6 @@ typedef unsigned char uchar; #define ASC_CHIP_MIN_VER_PCI (0x09) #define ASC_CHIP_MAX_VER_PCI (0x0F) #define ASC_CHIP_VER_PCI_BIT (0x08) -#define ASC_CHIP_MIN_VER_ISA (0x11) -#define ASC_CHIP_MIN_VER_ISA_PNP (0x21) -#define ASC_CHIP_MAX_VER_ISA (0x27) -#define ASC_CHIP_VER_ISA_BIT (0x30) -#define ASC_CHIP_VER_ISAPNP_BIT (0x20) #define ASC_CHIP_VER_ASYN_BUG (0x21) #define ASC_CHIP_VER_PCI 0x08 #define ASC_CHIP_VER_PCI_ULTRA_3150 (ASC_CHIP_VER_PCI | 0x02) @@ -116,7 +109,6 @@ typedef unsigned char uchar; #define ASC_CHIP_LATEST_VER_EISA ((ASC_CHIP_MIN_VER_EISA - 1) + 3) #define ASC_MAX_VL_DMA_COUNT (0x07FFFFFFL) #define ASC_MAX_PCI_DMA_COUNT (0xFFFFFFFFL) -#define ASC_MAX_ISA_DMA_COUNT (0x00FFFFFFL) #define ASC_SCSI_ID_BITS 3 #define ASC_SCSI_TIX_TYPE uchar @@ -194,7 +186,6 @@ typedef unsigned char uchar; #define ASC_FLAG_SRB_LINEAR_ADDR 0x08 #define ASC_FLAG_WIN16 0x10 #define ASC_FLAG_WIN32 0x20 -#define ASC_FLAG_ISA_OVER_16MB 0x40 #define ASC_FLAG_DOS_VM_CALLBACK 0x80 #define ASC_TAG_FLAG_EXTRA_BYTES 0x10 #define ASC_TAG_FLAG_DISABLE_DISCONNECT 0x04 @@ -464,8 +455,6 @@ typedef struct asc_dvc_cfg { ASC_SCSI_BIT_ID_TYPE disc_enable; ASC_SCSI_BIT_ID_TYPE sdtr_enable; uchar chip_scsi_id; - uchar isa_dma_speed; - uchar isa_dma_channel; uchar chip_version; ushort mcode_date; ushort mcode_version; @@ -572,10 +561,8 @@ typedef struct asc_cap_info_array { #define ASC_EEP_MAX_RETRY 20 /* - * These macros keep the chip SCSI id and ISA DMA speed - * bitfields in board order. C bitfields aren't portable - * between big and little-endian platforms so they are - * not used. + * These macros keep the chip SCSI id bitfields in board order. C bitfields + * aren't portable between big and little-endian platforms so they are not used. */ #define ASC_EEP_GET_CHIP_ID(cfg) ((cfg)->id_speed & 0x0f) @@ -2340,9 +2327,8 @@ static void asc_prt_asc_dvc_cfg(ASC_DVC_CFG *h) printk(" disc_enable 0x%x, sdtr_enable 0x%x,\n", h->disc_enable, h->sdtr_enable); - printk(" chip_scsi_id %d, isa_dma_speed %d, isa_dma_channel %d, " - "chip_version %d,\n", h->chip_scsi_id, h->isa_dma_speed, - h->isa_dma_channel, h->chip_version); + printk(" chip_scsi_id %d, chip_version %d,\n", + h->chip_scsi_id, h->chip_version); printk(" mcode_date 0x%x, mcode_version %d\n", h->mcode_date, h->mcode_version); @@ -2415,8 +2401,8 @@ static void asc_prt_scsi_host(struct Scsi_Host *s) printk(" dma_channel %d, this_id %d, can_queue %d,\n", s->dma_channel, s->this_id, s->can_queue); - printk(" cmd_per_lun %d, sg_tablesize %d, unchecked_isa_dma %d\n", - s->cmd_per_lun, s->sg_tablesize, s->unchecked_isa_dma); + printk(" cmd_per_lun %d, sg_tablesize %d\n", + s->cmd_per_lun, s->sg_tablesize); if (ASC_NARROW_BOARD(boardp)) { asc_prt_asc_dvc_var(&boardp->dvc_var.asc_dvc_var); @@ -2632,42 +2618,28 @@ static const char *advansys_info(struct Scsi_Host *shost) if (ASC_NARROW_BOARD(boardp)) { asc_dvc_varp = &boardp->dvc_var.asc_dvc_var; ASC_DBG(1, "begin\n"); - if (asc_dvc_varp->bus_type & ASC_IS_ISA) { - if ((asc_dvc_varp->bus_type & ASC_IS_ISAPNP) == - ASC_IS_ISAPNP) { - busname = "ISA PnP"; + + if (asc_dvc_varp->bus_type & ASC_IS_VL) { + busname = "VL"; + } else if (asc_dvc_varp->bus_type & ASC_IS_EISA) { + busname = "EISA"; + } else if (asc_dvc_varp->bus_type & ASC_IS_PCI) { + if ((asc_dvc_varp->bus_type & ASC_IS_PCI_ULTRA) + == ASC_IS_PCI_ULTRA) { + busname = "PCI Ultra"; } else { - busname = "ISA"; + busname = "PCI"; } - sprintf(info, - "AdvanSys SCSI %s: %s: IO 0x%lX-0x%lX, IRQ 0x%X, DMA 0x%X", - ASC_VERSION, busname, - (ulong)shost->io_port, - (ulong)shost->io_port + ASC_IOADR_GAP - 1, - boardp->irq, shost->dma_channel); } else { - if (asc_dvc_varp->bus_type & ASC_IS_VL) { - busname = "VL"; - } else if (asc_dvc_varp->bus_type & ASC_IS_EISA) { - busname = "EISA"; - } else if (asc_dvc_varp->bus_type & ASC_IS_PCI) { - if ((asc_dvc_varp->bus_type & ASC_IS_PCI_ULTRA) - == ASC_IS_PCI_ULTRA) { - busname = "PCI Ultra"; - } else { - busname = "PCI"; - } - } else { - busname = "?"; - shost_printk(KERN_ERR, shost, "unknown bus " - "type %d\n", asc_dvc_varp->bus_type); - } - sprintf(info, - "AdvanSys SCSI %s: %s: IO 0x%lX-0x%lX, IRQ 0x%X", - ASC_VERSION, busname, (ulong)shost->io_port, - (ulong)shost->io_port + ASC_IOADR_GAP - 1, - boardp->irq); + busname = "?"; + shost_printk(KERN_ERR, shost, "unknown bus " + "type %d\n", asc_dvc_varp->bus_type); } + sprintf(info, + "AdvanSys SCSI %s: %s: IO 0x%lX-0x%lX, IRQ 0x%X", + ASC_VERSION, busname, (ulong)shost->io_port, + (ulong)shost->io_port + ASC_IOADR_GAP - 1, + boardp->irq); } else { /* * Wide Adapter Information @@ -2873,12 +2845,7 @@ static void asc_prt_asc_board_eeprom(struct seq_file *m, struct Scsi_Host *shost ASCEEP_CONFIG *ep; int i; uchar serialstr[13]; -#ifdef CONFIG_ISA - ASC_DVC_VAR *asc_dvc_varp; - int isa_dma_speed[] = { 10, 8, 7, 6, 5, 4, 3, 2 }; - asc_dvc_varp = &boardp->dvc_var.asc_dvc_var; -#endif /* CONFIG_ISA */ ep = &boardp->eep_config.asc_eep; seq_printf(m, @@ -2926,14 +2893,6 @@ static void asc_prt_asc_board_eeprom(struct seq_file *m, struct Scsi_Host *shost seq_printf(m, " %c", (ep->init_sdtr & ADV_TID_TO_TIDMASK(i)) ? 'Y' : 'N'); seq_putc(m, '\n'); - -#ifdef CONFIG_ISA - if (asc_dvc_varp->bus_type & ASC_IS_ISA) { - seq_printf(m, - " Host ISA DMA speed: %d MB/S\n", - isa_dma_speed[ASC_EEP_GET_DMA_SPD(ep)]); - } -#endif /* CONFIG_ISA */ } /* @@ -3181,10 +3140,6 @@ static void asc_prt_driver_conf(struct seq_file *m, struct Scsi_Host *shost) shost->sg_tablesize, shost->cmd_per_lun); seq_printf(m, - " unchecked_isa_dma %d\n", - shost->unchecked_isa_dma); - - seq_printf(m, " flags 0x%x, last_reset 0x%lx, jiffies 0x%lx, asc_n_io_port 0x%x\n", boardp->flags, shost->last_reset, jiffies, boardp->asc_n_io_port); @@ -8563,12 +8518,6 @@ static unsigned short AscGetChipBiosAddress(PortAddr iop_base, } cfg_lsw = AscGetChipCfgLsw(iop_base); - - /* - * ISA PnP uses the top bit as the 32K BIOS flag - */ - if (bus_type == ASC_IS_ISAPNP) - cfg_lsw &= 0x7FFF; bios_addr = ASC_BIOS_MIN_ADDR + (cfg_lsw >> 12) * ASC_BIOS_BANK_SIZE; return bios_addr; } @@ -8611,19 +8560,6 @@ static unsigned char AscGetChipVersion(PortAddr iop_base, return AscGetChipVerNo(iop_base); } -#ifdef CONFIG_ISA -static void AscEnableIsaDma(uchar dma_channel) -{ - if (dma_channel < 4) { - outp(0x000B, (ushort)(0xC0 | dma_channel)); - outp(0x000A, dma_channel); - } else if (dma_channel < 8) { - outp(0x00D6, (ushort)(0xC0 | (dma_channel - 4))); - outp(0x00D4, (ushort)(dma_channel - 4)); - } -} -#endif /* CONFIG_ISA */ - static int AscStopQueueExe(PortAddr iop_base) { int count = 0; @@ -8644,65 +8580,11 @@ static int AscStopQueueExe(PortAddr iop_base) static unsigned int AscGetMaxDmaCount(ushort bus_type) { - if (bus_type & ASC_IS_ISA) - return ASC_MAX_ISA_DMA_COUNT; - else if (bus_type & (ASC_IS_EISA | ASC_IS_VL)) + if (bus_type & (ASC_IS_EISA | ASC_IS_VL)) return ASC_MAX_VL_DMA_COUNT; return ASC_MAX_PCI_DMA_COUNT; } -#ifdef CONFIG_ISA -static ushort AscGetIsaDmaChannel(PortAddr iop_base) -{ - ushort channel; - - channel = AscGetChipCfgLsw(iop_base) & 0x0003; - if (channel == 0x03) - return (0); - else if (channel == 0x00) - return (7); - return (channel + 4); -} - -static ushort AscSetIsaDmaChannel(PortAddr iop_base, ushort dma_channel) -{ - ushort cfg_lsw; - uchar value; - - if ((dma_channel >= 5) && (dma_channel <= 7)) { - if (dma_channel == 7) - value = 0x00; - else - value = dma_channel - 4; - cfg_lsw = AscGetChipCfgLsw(iop_base) & 0xFFFC; - cfg_lsw |= value; - AscSetChipCfgLsw(iop_base, cfg_lsw); - return (AscGetIsaDmaChannel(iop_base)); - } - return 0; -} - -static uchar AscGetIsaDmaSpeed(PortAddr iop_base) -{ - uchar speed_value; - - AscSetBank(iop_base, 1); - speed_value = AscReadChipDmaSpeed(iop_base); - speed_value &= 0x07; - AscSetBank(iop_base, 0); - return speed_value; -} - -static uchar AscSetIsaDmaSpeed(PortAddr iop_base, uchar speed_value) -{ - speed_value &= 0x07; - AscSetBank(iop_base, 1); - AscWriteChipDmaSpeed(iop_base, speed_value); - AscSetBank(iop_base, 0); - return AscGetIsaDmaSpeed(iop_base); -} -#endif /* CONFIG_ISA */ - static void AscInitAscDvcVar(ASC_DVC_VAR *asc_dvc) { int i; @@ -8712,7 +8594,7 @@ static void AscInitAscDvcVar(ASC_DVC_VAR *asc_dvc) iop_base = asc_dvc->iop_base; asc_dvc->err_code = 0; if ((asc_dvc->bus_type & - (ASC_IS_ISA | ASC_IS_PCI | ASC_IS_EISA | ASC_IS_VL)) == 0) { + (ASC_IS_PCI | ASC_IS_EISA | ASC_IS_VL)) == 0) { asc_dvc->err_code |= ASC_IERR_NO_BUS_TYPE; } AscSetChipControl(iop_base, CC_HALT); @@ -8767,17 +8649,6 @@ static void AscInitAscDvcVar(ASC_DVC_VAR *asc_dvc) (SEC_ACTIVE_NEGATE | SEC_SLEW_RATE)); } - asc_dvc->cfg->isa_dma_speed = ASC_DEF_ISA_DMA_SPEED; -#ifdef CONFIG_ISA - if ((asc_dvc->bus_type & ASC_IS_ISA) != 0) { - if (chip_version >= ASC_CHIP_MIN_VER_ISA_PNP) { - AscSetChipIFC(iop_base, IFC_INIT_DEFAULT); - asc_dvc->bus_type = ASC_IS_ISAPNP; - } - asc_dvc->cfg->isa_dma_channel = - (uchar)AscGetIsaDmaChannel(iop_base); - } -#endif /* CONFIG_ISA */ for (i = 0; i <= ASC_MAX_TID; i++) { asc_dvc->cur_dvc_qng[i] = 0; asc_dvc->max_dvc_qng[i] = ASC_MAX_SCSI1_QNG; @@ -9141,7 +9012,6 @@ static int AscInitFromEEP(ASC_DVC_VAR *asc_dvc) asc_dvc->cfg->sdtr_enable = eep_config->init_sdtr; asc_dvc->cfg->disc_enable = eep_config->disc_enable; asc_dvc->cfg->cmd_qng_enabled = eep_config->use_cmd_qng; - asc_dvc->cfg->isa_dma_speed = ASC_EEP_GET_DMA_SPD(eep_config); asc_dvc->start_motor = eep_config->start_motor; asc_dvc->dvc_cntl = eep_config->cntl; asc_dvc->no_scam = eep_config->no_scam; @@ -9314,22 +9184,10 @@ static int AscInitSetConfig(struct pci_dev *pdev, struct Scsi_Host *shost) } } else #endif /* CONFIG_PCI */ - if (asc_dvc->bus_type == ASC_IS_ISAPNP) { - if (AscGetChipVersion(iop_base, asc_dvc->bus_type) - == ASC_CHIP_VER_ASYN_BUG) { - asc_dvc->bug_fix_cntl |= ASC_BUG_FIX_ASYN_USE_SYN; - } - } if (AscSetChipScsiID(iop_base, asc_dvc->cfg->chip_scsi_id) != asc_dvc->cfg->chip_scsi_id) { asc_dvc->err_code |= ASC_IERR_SET_SCSI_ID; } -#ifdef CONFIG_ISA - if (asc_dvc->bus_type & ASC_IS_ISA) { - AscSetIsaDmaChannel(iop_base, asc_dvc->cfg->isa_dma_channel); - AscSetIsaDmaSpeed(iop_base, asc_dvc->cfg->isa_dma_speed); - } -#endif /* CONFIG_ISA */ asc_dvc->init_state |= ASC_INIT_STATE_END_SET_CFG; @@ -10752,12 +10610,6 @@ static struct scsi_host_template advansys_template = { .eh_host_reset_handler = advansys_reset, .bios_param = advansys_biosparam, .slave_configure = advansys_slave_configure, - /* - * Because the driver may control an ISA adapter 'unchecked_isa_dma' - * must be set. The flag will be cleared in advansys_board_found - * for non-ISA adapters. - */ - .unchecked_isa_dma = true, }; static int advansys_wide_init_chip(struct Scsi_Host *shost) @@ -10923,29 +10775,21 @@ static int advansys_board_found(struct Scsi_Host *shost, unsigned int iop, */ switch (asc_dvc_varp->bus_type) { #ifdef CONFIG_ISA - case ASC_IS_ISA: - shost->unchecked_isa_dma = true; - share_irq = 0; - break; case ASC_IS_VL: - shost->unchecked_isa_dma = false; share_irq = 0; break; case ASC_IS_EISA: - shost->unchecked_isa_dma = false; share_irq = IRQF_SHARED; break; #endif /* CONFIG_ISA */ #ifdef CONFIG_PCI case ASC_IS_PCI: - shost->unchecked_isa_dma = false; share_irq = IRQF_SHARED; break; #endif /* CONFIG_PCI */ default: shost_printk(KERN_ERR, shost, "unknown adapter type: " "%d\n", asc_dvc_varp->bus_type); - shost->unchecked_isa_dma = false; share_irq = 0; break; } @@ -10964,7 +10808,6 @@ static int advansys_board_found(struct Scsi_Host *shost, unsigned int iop, * For Wide boards set PCI information before calling * AdvInitGetConfig(). */ - shost->unchecked_isa_dma = false; share_irq = IRQF_SHARED; ASC_DBG(2, "AdvInitGetConfig()\n"); @@ -11000,7 +10843,7 @@ static int advansys_board_found(struct Scsi_Host *shost, unsigned int iop, ep->init_sdtr = asc_dvc_varp->cfg->sdtr_enable; ep->disc_enable = asc_dvc_varp->cfg->disc_enable; ep->use_cmd_qng = asc_dvc_varp->cfg->cmd_qng_enabled; - ASC_EEP_SET_DMA_SPD(ep, asc_dvc_varp->cfg->isa_dma_speed); + ASC_EEP_SET_DMA_SPD(ep, ASC_DEF_ISA_DMA_SPEED); ep->start_motor = asc_dvc_varp->start_motor; ep->cntl = asc_dvc_varp->dvc_cntl; ep->no_scam = asc_dvc_varp->no_scam; @@ -11228,22 +11071,6 @@ static int advansys_board_found(struct Scsi_Host *shost, unsigned int iop, /* Register DMA Channel for Narrow boards. */ shost->dma_channel = NO_ISA_DMA; /* Default to no ISA DMA. */ -#ifdef CONFIG_ISA - if (ASC_NARROW_BOARD(boardp)) { - /* Register DMA channel for ISA bus. */ - if (asc_dvc_varp->bus_type & ASC_IS_ISA) { - shost->dma_channel = asc_dvc_varp->cfg->isa_dma_channel; - ret = request_dma(shost->dma_channel, DRV_NAME); - if (ret) { - shost_printk(KERN_ERR, shost, "request_dma() " - "%d failed %d\n", - shost->dma_channel, ret); - goto err_unmap; - } - AscEnableIsaDma(shost->dma_channel); - } - } -#endif /* CONFIG_ISA */ /* Register IRQ Number. */ ASC_DBG(2, "request_irq(%d, %p)\n", boardp->irq, shost); @@ -11262,7 +11089,7 @@ static int advansys_board_found(struct Scsi_Host *shost, unsigned int iop, shost_printk(KERN_ERR, shost, "request_irq(): IRQ 0x%x " "failed with %d\n", boardp->irq, ret); } - goto err_free_dma; + goto err_unmap; } /* @@ -11314,11 +11141,6 @@ static int advansys_board_found(struct Scsi_Host *shost, unsigned int iop, advansys_wide_free_mem(boardp); err_free_irq: free_irq(boardp->irq, shost); - err_free_dma: -#ifdef CONFIG_ISA - if (shost->dma_channel != NO_ISA_DMA) - free_dma(shost->dma_channel); -#endif err_unmap: if (boardp->ioremap_addr) iounmap(boardp->ioremap_addr); @@ -11339,12 +11161,7 @@ static int advansys_release(struct Scsi_Host *shost) ASC_DBG(1, "begin\n"); scsi_remove_host(shost); free_irq(board->irq, shost); -#ifdef CONFIG_ISA - if (shost->dma_channel != NO_ISA_DMA) { - ASC_DBG(1, "free_dma()\n"); - free_dma(shost->dma_channel); - } -#endif + if (ASC_NARROW_BOARD(board)) { dma_unmap_single(board->dev, board->dvc_var.asc_dvc_var.overrun_dma, @@ -11366,79 +11183,13 @@ static PortAddr _asc_def_iop_base[ASC_IOADR_TABLE_MAX_IX] = { 0x0210, 0x0230, 0x0250, 0x0330 }; -/* - * The ISA IRQ number is found in bits 2 and 3 of the CfgLsw. It decodes as: - * 00: 10 - * 01: 11 - * 10: 12 - * 11: 15 - */ -static unsigned int advansys_isa_irq_no(PortAddr iop_base) -{ - unsigned short cfg_lsw = AscGetChipCfgLsw(iop_base); - unsigned int chip_irq = ((cfg_lsw >> 2) & 0x03) + 10; - if (chip_irq == 13) - chip_irq = 15; - return chip_irq; -} - -static int advansys_isa_probe(struct device *dev, unsigned int id) -{ - int err = -ENODEV; - PortAddr iop_base = _asc_def_iop_base[id]; - struct Scsi_Host *shost; - struct asc_board *board; - - if (!request_region(iop_base, ASC_IOADR_GAP, DRV_NAME)) { - ASC_DBG(1, "I/O port 0x%x busy\n", iop_base); - return -ENODEV; - } - ASC_DBG(1, "probing I/O port 0x%x\n", iop_base); - if (!AscFindSignature(iop_base)) - goto release_region; - if (!(AscGetChipVersion(iop_base, ASC_IS_ISA) & ASC_CHIP_VER_ISA_BIT)) - goto release_region; - - err = -ENOMEM; - shost = scsi_host_alloc(&advansys_template, sizeof(*board)); - if (!shost) - goto release_region; - - board = shost_priv(shost); - board->irq = advansys_isa_irq_no(iop_base); - board->dev = dev; - board->shost = shost; - - err = advansys_board_found(shost, iop_base, ASC_IS_ISA); - if (err) - goto free_host; - - dev_set_drvdata(dev, shost); - return 0; - - free_host: - scsi_host_put(shost); - release_region: - release_region(iop_base, ASC_IOADR_GAP); - return err; -} - -static void advansys_isa_remove(struct device *dev, unsigned int id) +static void advansys_vlb_remove(struct device *dev, unsigned int id) { int ioport = _asc_def_iop_base[id]; advansys_release(dev_get_drvdata(dev)); release_region(ioport, ASC_IOADR_GAP); } -static struct isa_driver advansys_isa_driver = { - .probe = advansys_isa_probe, - .remove = advansys_isa_remove, - .driver = { - .owner = THIS_MODULE, - .name = DRV_NAME, - }, -}; - /* * The VLB IRQ number is found in bits 2 to 4 of the CfgLsw. It decodes as: * 000: invalid @@ -11507,7 +11258,7 @@ static int advansys_vlb_probe(struct device *dev, unsigned int id) static struct isa_driver advansys_vlb_driver = { .probe = advansys_vlb_probe, - .remove = advansys_isa_remove, + .remove = advansys_vlb_remove, .driver = { .owner = THIS_MODULE, .name = "advansys_vlb", @@ -11757,15 +11508,10 @@ static int __init advansys_init(void) { int error; - error = isa_register_driver(&advansys_isa_driver, - ASC_IOADR_TABLE_MAX_IX); - if (error) - goto fail; - error = isa_register_driver(&advansys_vlb_driver, ASC_IOADR_TABLE_MAX_IX); if (error) - goto unregister_isa; + goto fail; error = eisa_driver_register(&advansys_eisa_driver); if (error) @@ -11781,8 +11527,6 @@ static int __init advansys_init(void) eisa_driver_unregister(&advansys_eisa_driver); unregister_vlb: isa_unregister_driver(&advansys_vlb_driver); - unregister_isa: - isa_unregister_driver(&advansys_isa_driver); fail: return error; } @@ -11792,7 +11536,6 @@ static void __exit advansys_exit(void) pci_unregister_driver(&advansys_pci_driver); eisa_driver_unregister(&advansys_eisa_driver); isa_unregister_driver(&advansys_vlb_driver); - isa_unregister_driver(&advansys_isa_driver); } module_init(advansys_init); diff --git a/drivers/scsi/aha1542.c b/drivers/scsi/aha1542.c index 21aab9f5b117..1210e61afb18 100644 --- a/drivers/scsi/aha1542.c +++ b/drivers/scsi/aha1542.c @@ -65,9 +65,12 @@ struct aha1542_hostdata { dma_addr_t ccb_handle; }; +#define AHA1542_MAX_SECTORS 16 + struct aha1542_cmd { - struct chain *chain; - dma_addr_t chain_handle; + /* bounce buffer */ + void *data_buffer; + dma_addr_t data_buffer_handle; }; static inline void aha1542_intr_reset(u16 base) @@ -257,15 +260,19 @@ static int aha1542_test_port(struct Scsi_Host *sh) static void aha1542_free_cmd(struct scsi_cmnd *cmd) { struct aha1542_cmd *acmd = scsi_cmd_priv(cmd); - struct device *dev = cmd->device->host->dma_dev; - size_t len = scsi_sg_count(cmd) * sizeof(struct chain); - if (acmd->chain) { - dma_unmap_single(dev, acmd->chain_handle, len, DMA_TO_DEVICE); - kfree(acmd->chain); + if (cmd->sc_data_direction == DMA_FROM_DEVICE) { + void *buf = acmd->data_buffer; + struct req_iterator iter; + struct bio_vec bv; + + rq_for_each_segment(bv, cmd->request, iter) { + memcpy_to_page(bv.bv_page, bv.bv_offset, buf, + bv.bv_len); + buf += bv.bv_len; + } } - acmd->chain = NULL; scsi_dma_unmap(cmd); } @@ -416,7 +423,7 @@ static int aha1542_queuecommand(struct Scsi_Host *sh, struct scsi_cmnd *cmd) u8 lun = cmd->device->lun; unsigned long flags; int bufflen = scsi_bufflen(cmd); - int mbo, sg_count; + int mbo; struct mailbox *mb = aha1542->mb; struct ccb *ccb = aha1542->ccb; @@ -438,17 +445,17 @@ static int aha1542_queuecommand(struct Scsi_Host *sh, struct scsi_cmnd *cmd) print_hex_dump_bytes("command: ", DUMP_PREFIX_NONE, cmd->cmnd, cmd->cmd_len); } #endif - sg_count = scsi_dma_map(cmd); - if (sg_count) { - size_t len = sg_count * sizeof(struct chain); - - acmd->chain = kmalloc(len, GFP_DMA); - if (!acmd->chain) - goto out_unmap; - acmd->chain_handle = dma_map_single(sh->dma_dev, acmd->chain, - len, DMA_TO_DEVICE); - if (dma_mapping_error(sh->dma_dev, acmd->chain_handle)) - goto out_free_chain; + + if (cmd->sc_data_direction == DMA_TO_DEVICE) { + void *buf = acmd->data_buffer; + struct req_iterator iter; + struct bio_vec bv; + + rq_for_each_segment(bv, cmd->request, iter) { + memcpy_from_page(buf, bv.bv_page, bv.bv_offset, + bv.bv_len); + buf += bv.bv_len; + } } /* @@ -496,27 +503,12 @@ static int aha1542_queuecommand(struct Scsi_Host *sh, struct scsi_cmnd *cmd) direction = 16; memcpy(ccb[mbo].cdb, cmd->cmnd, ccb[mbo].cdblen); - - if (bufflen) { - struct scatterlist *sg; - int i; - - ccb[mbo].op = 2; /* SCSI Initiator Command w/scatter-gather */ - scsi_for_each_sg(cmd, sg, sg_count, i) { - any2scsi(acmd->chain[i].dataptr, sg_dma_address(sg)); - any2scsi(acmd->chain[i].datalen, sg_dma_len(sg)); - }; - any2scsi(ccb[mbo].datalen, sg_count * sizeof(struct chain)); - any2scsi(ccb[mbo].dataptr, acmd->chain_handle); -#ifdef DEBUG - shost_printk(KERN_DEBUG, sh, "cptr %p: ", acmd->chain); - print_hex_dump_bytes("cptr: ", DUMP_PREFIX_NONE, acmd->chain, 18); -#endif - } else { - ccb[mbo].op = 0; /* SCSI Initiator Command */ - any2scsi(ccb[mbo].datalen, 0); + ccb[mbo].op = 0; /* SCSI Initiator Command */ + any2scsi(ccb[mbo].datalen, bufflen); + if (bufflen) + any2scsi(ccb[mbo].dataptr, acmd->data_buffer_handle); + else any2scsi(ccb[mbo].dataptr, 0); - }; ccb[mbo].idlun = (target & 7) << 5 | direction | (lun & 7); /*SCSI Target Id */ ccb[mbo].rsalen = 16; ccb[mbo].linkptr[0] = ccb[mbo].linkptr[1] = ccb[mbo].linkptr[2] = 0; @@ -531,12 +523,6 @@ static int aha1542_queuecommand(struct Scsi_Host *sh, struct scsi_cmnd *cmd) spin_unlock_irqrestore(sh->host_lock, flags); return 0; -out_free_chain: - kfree(acmd->chain); - acmd->chain = NULL; -out_unmap: - scsi_dma_unmap(cmd); - return SCSI_MLQUEUE_HOST_BUSY; } /* Initialize mailboxes */ @@ -1027,6 +1013,27 @@ static int aha1542_biosparam(struct scsi_device *sdev, } MODULE_LICENSE("GPL"); +static int aha1542_init_cmd_priv(struct Scsi_Host *shost, struct scsi_cmnd *cmd) +{ + struct aha1542_cmd *acmd = scsi_cmd_priv(cmd); + + acmd->data_buffer = dma_alloc_coherent(shost->dma_dev, + SECTOR_SIZE * AHA1542_MAX_SECTORS, + &acmd->data_buffer_handle, GFP_KERNEL); + if (!acmd->data_buffer) + return -ENOMEM; + return 0; +} + +static int aha1542_exit_cmd_priv(struct Scsi_Host *shost, struct scsi_cmnd *cmd) +{ + struct aha1542_cmd *acmd = scsi_cmd_priv(cmd); + + dma_free_coherent(shost->dma_dev, SECTOR_SIZE * AHA1542_MAX_SECTORS, + acmd->data_buffer, acmd->data_buffer_handle); + return 0; +} + static struct scsi_host_template driver_template = { .module = THIS_MODULE, .proc_name = "aha1542", @@ -1037,10 +1044,12 @@ static struct scsi_host_template driver_template = { .eh_bus_reset_handler = aha1542_bus_reset, .eh_host_reset_handler = aha1542_host_reset, .bios_param = aha1542_biosparam, + .init_cmd_priv = aha1542_init_cmd_priv, + .exit_cmd_priv = aha1542_exit_cmd_priv, .can_queue = AHA1542_MAILBOXES, .this_id = 7, - .sg_tablesize = 16, - .unchecked_isa_dma = 1, + .max_sectors = AHA1542_MAX_SECTORS, + .sg_tablesize = SG_ALL, }; static int aha1542_isa_match(struct device *pdev, unsigned int ndev) diff --git a/drivers/scsi/esas2r/esas2r_main.c b/drivers/scsi/esas2r/esas2r_main.c index a9dd6345f064..5d9eeac6717a 100644 --- a/drivers/scsi/esas2r/esas2r_main.c +++ b/drivers/scsi/esas2r/esas2r_main.c @@ -249,7 +249,6 @@ static struct scsi_host_template driver_template = { .cmd_per_lun = ESAS2R_DEFAULT_CMD_PER_LUN, .present = 0, - .unchecked_isa_dma = 0, .emulated = 0, .proc_name = ESAS2R_DRVR_NAME, .change_queue_depth = scsi_change_queue_depth, diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index 2f162603876f..697c09ef259b 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -371,13 +371,9 @@ static struct device_type scsi_host_type = { struct Scsi_Host *scsi_host_alloc(struct scsi_host_template *sht, int privsize) { struct Scsi_Host *shost; - gfp_t gfp_mask = GFP_KERNEL; int index; - if (sht->unchecked_isa_dma && privsize) - gfp_mask |= __GFP_DMA; - - shost = kzalloc(sizeof(struct Scsi_Host) + privsize, gfp_mask); + shost = kzalloc(sizeof(struct Scsi_Host) + privsize, GFP_KERNEL); if (!shost) return NULL; @@ -419,7 +415,6 @@ struct Scsi_Host *scsi_host_alloc(struct scsi_host_template *sht, int privsize) shost->sg_tablesize = sht->sg_tablesize; shost->sg_prot_tablesize = sht->sg_prot_tablesize; shost->cmd_per_lun = sht->cmd_per_lun; - shost->unchecked_isa_dma = sht->unchecked_isa_dma; shost->no_write_same = sht->no_write_same; shost->host_tagset = sht->host_tagset; diff --git a/drivers/scsi/scsi_debugfs.c b/drivers/scsi/scsi_debugfs.c index c19ea7ab54cb..d9109771f274 100644 --- a/drivers/scsi/scsi_debugfs.c +++ b/drivers/scsi/scsi_debugfs.c @@ -8,7 +8,6 @@ #define SCSI_CMD_FLAG_NAME(name)[const_ilog2(SCMD_##name)] = #name static const char *const scsi_cmd_flags[] = { SCSI_CMD_FLAG_NAME(TAGGED), - SCSI_CMD_FLAG_NAME(UNCHECKED_ISA_DMA), SCSI_CMD_FLAG_NAME(INITIALIZED), }; #undef SCSI_CMD_FLAG_NAME diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 7d52a11e1b61..c289991ffaed 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -53,49 +53,16 @@ #endif static struct kmem_cache *scsi_sense_cache; -static struct kmem_cache *scsi_sense_isadma_cache; static DEFINE_MUTEX(scsi_sense_cache_mutex); static void scsi_mq_uninit_cmd(struct scsi_cmnd *cmd); -static inline struct kmem_cache * -scsi_select_sense_cache(bool unchecked_isa_dma) -{ - return unchecked_isa_dma ? scsi_sense_isadma_cache : scsi_sense_cache; -} - -static void scsi_free_sense_buffer(bool unchecked_isa_dma, - unsigned char *sense_buffer) -{ - kmem_cache_free(scsi_select_sense_cache(unchecked_isa_dma), - sense_buffer); -} - -static unsigned char *scsi_alloc_sense_buffer(bool unchecked_isa_dma, - gfp_t gfp_mask, int numa_node) -{ - return kmem_cache_alloc_node(scsi_select_sense_cache(unchecked_isa_dma), - gfp_mask, numa_node); -} - int scsi_init_sense_cache(struct Scsi_Host *shost) { - struct kmem_cache *cache; int ret = 0; mutex_lock(&scsi_sense_cache_mutex); - cache = scsi_select_sense_cache(shost->unchecked_isa_dma); - if (cache) - goto exit; - - if (shost->unchecked_isa_dma) { - scsi_sense_isadma_cache = - kmem_cache_create("scsi_sense_cache(DMA)", - SCSI_SENSE_BUFFERSIZE, 0, - SLAB_HWCACHE_ALIGN | SLAB_CACHE_DMA, NULL); - if (!scsi_sense_isadma_cache) - ret = -ENOMEM; - } else { + if (!scsi_sense_cache) { scsi_sense_cache = kmem_cache_create_usercopy("scsi_sense_cache", SCSI_SENSE_BUFFERSIZE, 0, SLAB_HWCACHE_ALIGN, @@ -103,7 +70,6 @@ int scsi_init_sense_cache(struct Scsi_Host *shost) if (!scsi_sense_cache) ret = -ENOMEM; } - exit: mutex_unlock(&scsi_sense_cache_mutex); return ret; } @@ -1748,15 +1714,12 @@ static int scsi_mq_init_request(struct blk_mq_tag_set *set, struct request *rq, unsigned int hctx_idx, unsigned int numa_node) { struct Scsi_Host *shost = set->driver_data; - const bool unchecked_isa_dma = shost->unchecked_isa_dma; struct scsi_cmnd *cmd = blk_mq_rq_to_pdu(rq); struct scatterlist *sg; int ret = 0; - if (unchecked_isa_dma) - cmd->flags |= SCMD_UNCHECKED_ISA_DMA; - cmd->sense_buffer = scsi_alloc_sense_buffer(unchecked_isa_dma, - GFP_KERNEL, numa_node); + cmd->sense_buffer = + kmem_cache_alloc_node(scsi_sense_cache, GFP_KERNEL, numa_node); if (!cmd->sense_buffer) return -ENOMEM; cmd->req.sense = cmd->sense_buffer; @@ -1770,8 +1733,7 @@ static int scsi_mq_init_request(struct blk_mq_tag_set *set, struct request *rq, if (shost->hostt->init_cmd_priv) { ret = shost->hostt->init_cmd_priv(shost, cmd); if (ret < 0) - scsi_free_sense_buffer(unchecked_isa_dma, - cmd->sense_buffer); + kmem_cache_free(scsi_sense_cache, cmd->sense_buffer); } return ret; @@ -1785,8 +1747,7 @@ static void scsi_mq_exit_request(struct blk_mq_tag_set *set, struct request *rq, if (shost->hostt->exit_cmd_priv) shost->hostt->exit_cmd_priv(shost, cmd); - scsi_free_sense_buffer(cmd->flags & SCMD_UNCHECKED_ISA_DMA, - cmd->sense_buffer); + kmem_cache_free(scsi_sense_cache, cmd->sense_buffer); } static int scsi_map_queues(struct blk_mq_tag_set *set) @@ -1821,8 +1782,6 @@ void __scsi_init_queue(struct Scsi_Host *shost, struct request_queue *q) dma_max_mapping_size(dev) >> SECTOR_SHIFT); } blk_queue_max_hw_sectors(q, shost->max_sectors); - if (shost->unchecked_isa_dma) - blk_queue_bounce_limit(q, BLK_BOUNCE_ISA); blk_queue_segment_boundary(q, shost->dma_boundary); dma_set_seg_boundary(dev, shost->dma_boundary); @@ -1988,7 +1947,6 @@ EXPORT_SYMBOL(scsi_unblock_requests); void scsi_exit_queue(void) { kmem_cache_destroy(scsi_sense_cache); - kmem_cache_destroy(scsi_sense_isadma_cache); } /** diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index 9af50e6f94c4..9b73aa506382 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -1078,8 +1078,7 @@ static int scsi_probe_and_add_lun(struct scsi_target *starget, if (!sdev) goto out; - result = kmalloc(result_len, GFP_KERNEL | - ((shost->unchecked_isa_dma) ? __GFP_DMA : 0)); + result = kmalloc(result_len, GFP_KERNEL); if (!result) goto out_free_sdev; @@ -1336,8 +1335,7 @@ static int scsi_report_lun_scan(struct scsi_target *starget, blist_flags_t bflag */ length = (511 + 1) * sizeof(struct scsi_lun); retry: - lun_data = kmalloc(length, GFP_KERNEL | - (sdev->host->unchecked_isa_dma ? __GFP_DMA : 0)); + lun_data = kmalloc(length, GFP_KERNEL); if (!lun_data) { printk(ALLOC_FAILURE_MSG, __func__); goto out; diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c index b6378c8ca783..b71ea1a69c8b 100644 --- a/drivers/scsi/scsi_sysfs.c +++ b/drivers/scsi/scsi_sysfs.c @@ -373,7 +373,6 @@ shost_rd_attr(cmd_per_lun, "%hd\n"); shost_rd_attr(can_queue, "%d\n"); shost_rd_attr(sg_tablesize, "%hu\n"); shost_rd_attr(sg_prot_tablesize, "%hu\n"); -shost_rd_attr(unchecked_isa_dma, "%d\n"); shost_rd_attr(prot_capabilities, "%u\n"); shost_rd_attr(prot_guard_type, "%hd\n"); shost_rd_attr2(proc_name, hostt->proc_name, "%s\n"); @@ -411,7 +410,6 @@ static struct attribute *scsi_sysfs_shost_attrs[] = { &dev_attr_can_queue.attr, &dev_attr_sg_tablesize.attr, &dev_attr_sg_prot_tablesize.attr, - &dev_attr_unchecked_isa_dma.attr, &dev_attr_proc_name.attr, &dev_attr_scan.attr, &dev_attr_hstate.attr, diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index 4383d93110f8..70f38715641e 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -974,7 +974,7 @@ sg_ioctl_common(struct file *filp, Sg_device *sdp, Sg_fd *sfp, */ return 0; case SG_GET_LOW_DMA: - return put_user((int) sdp->device->host->unchecked_isa_dma, ip); + return put_user(0, ip); case SG_GET_SCSI_ID: { sg_scsi_id_t v; @@ -1777,7 +1777,6 @@ sg_start_req(Sg_request *srp, unsigned char *cmd) if (sg_allow_dio && hp->flags & SG_FLAG_DIRECT_IO && dxfer_dir != SG_DXFER_UNKNOWN && !iov_count && - !sfp->parentdp->device->host->unchecked_isa_dma && blk_rq_aligned(q, (unsigned long)hp->dxferp, dxfer_len)) md = NULL; else @@ -1893,7 +1892,6 @@ sg_build_indirect(Sg_scatter_hold * schp, Sg_fd * sfp, int buff_size) int sg_tablesize = sfp->parentdp->sg_tablesize; int blk_size = buff_size, order; gfp_t gfp_mask = GFP_ATOMIC | __GFP_COMP | __GFP_NOWARN | __GFP_ZERO; - struct sg_device *sdp = sfp->parentdp; if (blk_size < 0) return -EFAULT; @@ -1919,9 +1917,6 @@ sg_build_indirect(Sg_scatter_hold * schp, Sg_fd * sfp, int buff_size) scatter_elem_sz_prev = num; } - if (sdp->device->host->unchecked_isa_dma) - gfp_mask |= GFP_DMA; - order = get_order(num); retry: ret_sz = 1 << (PAGE_SHIFT + order); @@ -2547,8 +2542,7 @@ static void sg_proc_debug_helper(struct seq_file *s, Sg_device * sdp) "(res)sgat=%d low_dma=%d\n", k, jiffies_to_msecs(fp->timeout), fp->reserve.bufflen, - (int) fp->reserve.k_use_sg, - (int) sdp->device->host->unchecked_isa_dma); + (int) fp->reserve.k_use_sg, 0); seq_printf(s, " cmd_q=%d f_packid=%d k_orphan=%d closed=0\n", (int) fp->cmd_q, (int) fp->force_packid, (int) fp->keep_orphan); diff --git a/drivers/scsi/sr_ioctl.c b/drivers/scsi/sr_ioctl.c index 5703f8400b73..15c305283b6c 100644 --- a/drivers/scsi/sr_ioctl.c +++ b/drivers/scsi/sr_ioctl.c @@ -33,10 +33,6 @@ static int xa_test = 0; module_param(xa_test, int, S_IRUGO | S_IWUSR); -/* primitive to determine whether we need to have GFP_DMA set based on - * the status of the unchecked_isa_dma flag in the host structure */ -#define SR_GFP_DMA(cd) (((cd)->device->host->unchecked_isa_dma) ? GFP_DMA : 0) - static int sr_read_tochdr(struct cdrom_device_info *cdi, struct cdrom_tochdr *tochdr) { @@ -45,7 +41,7 @@ static int sr_read_tochdr(struct cdrom_device_info *cdi, int result; unsigned char *buffer; - buffer = kmalloc(32, GFP_KERNEL | SR_GFP_DMA(cd)); + buffer = kmalloc(32, GFP_KERNEL); if (!buffer) return -ENOMEM; @@ -75,7 +71,7 @@ static int sr_read_tocentry(struct cdrom_device_info *cdi, int result; unsigned char *buffer; - buffer = kmalloc(32, GFP_KERNEL | SR_GFP_DMA(cd)); + buffer = kmalloc(32, GFP_KERNEL); if (!buffer) return -ENOMEM; @@ -384,7 +380,7 @@ int sr_get_mcn(struct cdrom_device_info *cdi, struct cdrom_mcn *mcn) { Scsi_CD *cd = cdi->handle; struct packet_command cgc; - char *buffer = kmalloc(32, GFP_KERNEL | SR_GFP_DMA(cd)); + char *buffer = kmalloc(32, GFP_KERNEL); int result; if (!buffer) @@ -567,7 +563,7 @@ int sr_is_xa(Scsi_CD *cd) if (!xa_test) return 0; - raw_sector = kmalloc(2048, GFP_KERNEL | SR_GFP_DMA(cd)); + raw_sector = kmalloc(2048, GFP_KERNEL); if (!raw_sector) return -ENOMEM; if (0 == sr_read_sector(cd, cd->ms_offset + 16, diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c index 9ca536aae784..3b1afe1d5b27 100644 --- a/drivers/scsi/st.c +++ b/drivers/scsi/st.c @@ -188,7 +188,7 @@ static int st_max_sg_segs = ST_MAX_SG; static int modes_defined; -static int enlarge_buffer(struct st_buffer *, int, int); +static int enlarge_buffer(struct st_buffer *, int); static void clear_buffer(struct st_buffer *); static void normalize_buffer(struct st_buffer *); static int append_to_buffer(const char __user *, struct st_buffer *, int); @@ -1289,7 +1289,7 @@ static int st_open(struct inode *inode, struct file *filp) } /* See that we have at least a one page buffer available */ - if (!enlarge_buffer(STp->buffer, PAGE_SIZE, STp->restr_dma)) { + if (!enlarge_buffer(STp->buffer, PAGE_SIZE)) { st_printk(KERN_WARNING, STp, "Can't allocate one page tape buffer.\n"); retval = (-EOVERFLOW); @@ -1586,7 +1586,7 @@ static int setup_buffering(struct scsi_tape *STp, const char __user *buf, } if (bufsize > STbp->buffer_size && - !enlarge_buffer(STbp, bufsize, STp->restr_dma)) { + !enlarge_buffer(STbp, bufsize)) { st_printk(KERN_WARNING, STp, "Can't allocate %d byte tape buffer.\n", bufsize); @@ -3894,7 +3894,7 @@ static long st_compat_ioctl(struct file *file, unsigned int cmd_in, unsigned lon /* Try to allocate a new tape buffer. Calling function must not hold dev_arr_lock. */ -static struct st_buffer *new_tape_buffer(int need_dma, int max_sg) +static struct st_buffer *new_tape_buffer(int max_sg) { struct st_buffer *tb; @@ -3905,7 +3905,6 @@ static struct st_buffer *new_tape_buffer(int need_dma, int max_sg) } tb->frp_segs = 0; tb->use_sg = max_sg; - tb->dma = need_dma; tb->buffer_size = 0; tb->reserved_pages = kcalloc(max_sg, sizeof(struct page *), @@ -3922,7 +3921,7 @@ static struct st_buffer *new_tape_buffer(int need_dma, int max_sg) /* Try to allocate enough space in the tape buffer */ #define ST_MAX_ORDER 6 -static int enlarge_buffer(struct st_buffer * STbuffer, int new_size, int need_dma) +static int enlarge_buffer(struct st_buffer * STbuffer, int new_size) { int segs, max_segs, b_size, order, got; gfp_t priority; @@ -3936,8 +3935,6 @@ static int enlarge_buffer(struct st_buffer * STbuffer, int new_size, int need_dm max_segs = STbuffer->use_sg; priority = GFP_KERNEL | __GFP_NOWARN; - if (need_dma) - priority |= GFP_DMA; if (STbuffer->cleared) priority |= __GFP_ZERO; @@ -3957,7 +3954,7 @@ static int enlarge_buffer(struct st_buffer * STbuffer, int new_size, int need_dm if (order == ST_MAX_ORDER) return 0; normalize_buffer(STbuffer); - return enlarge_buffer(STbuffer, new_size, need_dma); + return enlarge_buffer(STbuffer, new_size); } for (segs = STbuffer->frp_segs, got = STbuffer->buffer_size; @@ -4296,7 +4293,7 @@ static int st_probe(struct device *dev) i = queue_max_segments(SDp->request_queue); if (st_max_sg_segs < i) i = st_max_sg_segs; - buffer = new_tape_buffer((SDp->host)->unchecked_isa_dma, i); + buffer = new_tape_buffer(i); if (buffer == NULL) { sdev_printk(KERN_ERR, SDp, "st: Can't allocate new tape buffer. " @@ -4340,7 +4337,6 @@ static int st_probe(struct device *dev) tpnt->dirty = 0; tpnt->in_use = 0; tpnt->drv_buffer = 1; /* Try buffering if no mode sense */ - tpnt->restr_dma = (SDp->host)->unchecked_isa_dma; tpnt->use_pf = (SDp->scsi_level >= SCSI_2); tpnt->density = 0; tpnt->do_auto_lock = ST_AUTO_LOCK; @@ -4358,7 +4354,7 @@ static int st_probe(struct device *dev) tpnt->nbr_partitions = 0; blk_queue_rq_timeout(tpnt->device->request_queue, ST_TIMEOUT); tpnt->long_timeout = ST_LONG_TIMEOUT; - tpnt->try_dio = try_direct_io && !SDp->host->unchecked_isa_dma; + tpnt->try_dio = try_direct_io; for (i = 0; i < ST_NBR_MODES; i++) { STm = &(tpnt->modes[i]); diff --git a/drivers/scsi/st.h b/drivers/scsi/st.h index 95d2e7a7988d..9d3c38bb0794 100644 --- a/drivers/scsi/st.h +++ b/drivers/scsi/st.h @@ -35,7 +35,6 @@ struct st_request { /* The tape buffer descriptor. */ struct st_buffer { - unsigned char dma; /* DMA-able buffer */ unsigned char cleared; /* internal buffer cleared after open? */ unsigned short do_dio; /* direct i/o set up? */ int buffer_size; @@ -133,7 +132,6 @@ struct scsi_tape { unsigned char two_fm; unsigned char fast_mteom; unsigned char immediate; - unsigned char restr_dma; unsigned char scsi2_logical; unsigned char default_drvbuffer; /* 0xff = don't touch, value 3 bits */ unsigned char cln_mode; /* 0 = none, otherwise sense byte nbr */ diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c index 9ee797b8cb7e..d64c3ffdb52e 100644 --- a/drivers/target/target_core_pscsi.c +++ b/drivers/target/target_core_pscsi.c @@ -910,7 +910,7 @@ new_bio: " %d i: %d bio: %p, allocating another" " bio\n", bio->bi_vcnt, i, bio); - rc = blk_rq_append_bio(req, &bio); + rc = blk_rq_append_bio(req, bio); if (rc) { pr_err("pSCSI: failed to append bio\n"); goto fail; @@ -929,7 +929,7 @@ new_bio: } if (bio) { - rc = blk_rq_append_bio(req, &bio); + rc = blk_rq_append_bio(req, bio); if (rc) { pr_err("pSCSI: failed to append bio\n"); goto fail; diff --git a/fs/block_dev.c b/fs/block_dev.c index 09d6f7229db9..2ef24273069d 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1240,14 +1240,16 @@ static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part); int bdev_disk_changed(struct block_device *bdev, bool invalidate) { struct gendisk *disk = bdev->bd_disk; - int ret; + int ret = 0; lockdep_assert_held(&bdev->bd_mutex); rescan: - ret = blk_drop_partitions(bdev); - if (ret) - return ret; + if (bdev->bd_part_count) + return -EBUSY; + sync_blockdev(bdev); + invalidate_bdev(bdev); + blk_drop_partitions(disk); clear_bit(GD_NEED_PART_SCAN, &disk->state); diff --git a/include/linux/bio.h b/include/linux/bio.h index d0246c92a6e8..a0b4cfdf62a4 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -483,16 +483,10 @@ extern void bio_check_pages_dirty(struct bio *bio); extern void bio_copy_data_iter(struct bio *dst, struct bvec_iter *dst_iter, struct bio *src, struct bvec_iter *src_iter); extern void bio_copy_data(struct bio *dst, struct bio *src); -extern void bio_list_copy_data(struct bio *dst, struct bio *src); extern void bio_free_pages(struct bio *bio); -void zero_fill_bio_iter(struct bio *bio, struct bvec_iter iter); void bio_truncate(struct bio *bio, unsigned new_size); void guard_bio_eod(struct bio *bio); - -static inline void zero_fill_bio(struct bio *bio) -{ - zero_fill_bio_iter(bio, bio->bi_iter); -} +void zero_fill_bio(struct bio *bio); extern const char *bio_devname(struct bio *bio, char *buffer); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 158aefae1030..c1278f8f4f1f 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -272,6 +272,12 @@ static inline bool bio_is_passthrough(struct bio *bio) return blk_op_is_scsi(op) || blk_op_is_private(op); } +static inline bool blk_op_is_passthrough(unsigned int op) +{ + return (blk_op_is_scsi(op & REQ_OP_MASK) || + blk_op_is_private(op & REQ_OP_MASK)); +} + static inline unsigned short req_get_ioprio(struct request *req) { return req->ioprio; @@ -311,8 +317,17 @@ enum blk_zoned_model { BLK_ZONED_HM, /* Host-managed zoned block device */ }; +/* + * BLK_BOUNCE_NONE: never bounce (default) + * BLK_BOUNCE_HIGH: bounce all highmem pages + */ +enum blk_bounce { + BLK_BOUNCE_NONE, + BLK_BOUNCE_HIGH, +}; + struct queue_limits { - unsigned long bounce_pfn; + enum blk_bounce bounce; unsigned long seg_boundary_mask; unsigned long virt_boundary_mask; @@ -434,11 +449,6 @@ struct request_queue { */ int id; - /* - * queue needs bounce pages for pages above this limit - */ - gfp_t bounce_gfp; - spinlock_t queue_lock; /* @@ -683,6 +693,8 @@ static inline bool blk_account_rq(struct request *rq) dma_map_page_attrs(dev, (bv)->bv_page, (bv)->bv_offset, (bv)->bv_len, \ (dir), (attrs)) +#define queue_to_disk(q) (dev_to_disk(kobj_to_dev((q)->kobj.parent))) + static inline bool queue_is_mq(struct request_queue *q) { return q->mq_ops; @@ -838,24 +850,6 @@ static inline unsigned int blk_queue_depth(struct request_queue *q) return q->nr_requests; } -extern unsigned long blk_max_low_pfn, blk_max_pfn; - -/* - * standard bounce addresses: - * - * BLK_BOUNCE_HIGH : bounce all highmem pages - * BLK_BOUNCE_ANY : don't bounce anything - * BLK_BOUNCE_ISA : bounce pages above ISA DMA boundary - */ - -#if BITS_PER_LONG == 32 -#define BLK_BOUNCE_HIGH ((u64)blk_max_low_pfn << PAGE_SHIFT) -#else -#define BLK_BOUNCE_HIGH -1ULL -#endif -#define BLK_BOUNCE_ANY (-1ULL) -#define BLK_BOUNCE_ISA (DMA_BIT_MASK(24)) - /* * default timeout for SG_IO if none specified */ @@ -921,7 +915,7 @@ extern int blk_rq_prep_clone(struct request *rq, struct request *rq_src, extern void blk_rq_unprep_clone(struct request *rq); extern blk_status_t blk_insert_cloned_request(struct request_queue *q, struct request *rq); -extern int blk_rq_append_bio(struct request *rq, struct bio **bio); +int blk_rq_append_bio(struct request *rq, struct bio *bio); extern void blk_queue_split(struct bio **); extern int scsi_verify_blk_ioctl(struct block_device *, unsigned int); extern int scsi_cmd_blk_ioctl(struct block_device *, fmode_t, @@ -1139,7 +1133,7 @@ extern void blk_abort_request(struct request *); * Access functions for manipulating queue properties */ extern void blk_cleanup_queue(struct request_queue *); -extern void blk_queue_bounce_limit(struct request_queue *, u64); +void blk_queue_bounce_limit(struct request_queue *q, enum blk_bounce limit); extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int); extern void blk_queue_chunk_sectors(struct request_queue *, unsigned int); extern void blk_queue_max_segments(struct request_queue *, unsigned short); diff --git a/include/linux/genhd.h b/include/linux/genhd.h index f364619092cc..7e9660ea967d 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -204,25 +204,6 @@ static inline dev_t disk_devt(struct gendisk *disk) void disk_uevent(struct gendisk *disk, enum kobject_action action); -/* - * Smarter partition iterator without context limits. - */ -#define DISK_PITER_INCL_EMPTY (1 << 1) /* include 0-sized parts */ -#define DISK_PITER_INCL_PART0 (1 << 2) /* include partition 0 */ -#define DISK_PITER_INCL_EMPTY_PART0 (1 << 3) /* include empty partition 0 */ - -struct disk_part_iter { - struct gendisk *disk; - struct block_device *part; - unsigned long idx; - unsigned int flags; -}; - -extern void disk_part_iter_init(struct disk_part_iter *piter, - struct gendisk *disk, unsigned int flags); -struct block_device *disk_part_iter_next(struct disk_part_iter *piter); -extern void disk_part_iter_exit(struct disk_part_iter *piter); - /* block/genhd.c */ extern void device_add_disk(struct device *parent, struct gendisk *disk, const struct attribute_group **groups); @@ -273,7 +254,7 @@ static inline sector_t get_capacity(struct gendisk *disk) int bdev_disk_changed(struct block_device *bdev, bool invalidate); int blk_add_partitions(struct gendisk *disk, struct block_device *bdev); -int blk_drop_partitions(struct block_device *bdev); +void blk_drop_partitions(struct gendisk *disk); extern struct gendisk *__alloc_disk_node(int minors, int node_id); extern void put_disk(struct gendisk *disk); diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h index ace15b5dc956..0fd17a5344bd 100644 --- a/include/scsi/scsi_cmnd.h +++ b/include/scsi/scsi_cmnd.h @@ -55,11 +55,10 @@ struct scsi_pointer { /* for scmd->flags */ #define SCMD_TAGGED (1 << 0) -#define SCMD_UNCHECKED_ISA_DMA (1 << 1) -#define SCMD_INITIALIZED (1 << 2) -#define SCMD_LAST (1 << 3) +#define SCMD_INITIALIZED (1 << 1) +#define SCMD_LAST (1 << 2) /* flags preserved across unprep / reprep */ -#define SCMD_PRESERVED_FLAGS (SCMD_UNCHECKED_ISA_DMA | SCMD_INITIALIZED) +#define SCMD_PRESERVED_FLAGS (SCMD_INITIALIZED) /* for scmd->state */ #define SCMD_STATE_COMPLETE 0 diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h index e30fd963b97d..8343c6f9fec1 100644 --- a/include/scsi/scsi_host.h +++ b/include/scsi/scsi_host.h @@ -425,11 +425,6 @@ struct scsi_host_template { unsigned supported_mode:2; /* - * True if this host adapter uses unchecked DMA onto an ISA bus. - */ - unsigned unchecked_isa_dma:1; - - /* * True for emulated SCSI host adapters (e.g. ATAPI). */ unsigned emulated:1; @@ -617,7 +612,6 @@ struct Scsi_Host { */ unsigned nr_hw_queues; unsigned active_mode:2; - unsigned unchecked_isa_dma:1; /* * Host has requested that no further requests come through for the diff --git a/include/trace/events/kyber.h b/include/trace/events/kyber.h index c0e7d24ca256..f9802562edf6 100644 --- a/include/trace/events/kyber.h +++ b/include/trace/events/kyber.h @@ -30,7 +30,7 @@ TRACE_EVENT(kyber_latency, ), TP_fast_assign( - __entry->dev = disk_devt(dev_to_disk(kobj_to_dev(q->kobj.parent))); + __entry->dev = disk_devt(queue_to_disk(q)); strlcpy(__entry->domain, domain, sizeof(__entry->domain)); strlcpy(__entry->type, type, sizeof(__entry->type)); __entry->percentile = percentile; @@ -59,7 +59,7 @@ TRACE_EVENT(kyber_adjust, ), TP_fast_assign( - __entry->dev = disk_devt(dev_to_disk(kobj_to_dev(q->kobj.parent))); + __entry->dev = disk_devt(queue_to_disk(q)); strlcpy(__entry->domain, domain, sizeof(__entry->domain)); __entry->depth = depth; ), @@ -81,7 +81,7 @@ TRACE_EVENT(kyber_throttled, ), TP_fast_assign( - __entry->dev = disk_devt(dev_to_disk(kobj_to_dev(q->kobj.parent))); + __entry->dev = disk_devt(queue_to_disk(q)); strlcpy(__entry->domain, domain, sizeof(__entry->domain)); ), diff --git a/mm/Kconfig b/mm/Kconfig index 24c045b24b95..d0808a23e54b 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -283,12 +283,11 @@ config PHYS_ADDR_T_64BIT config BOUNCE bool "Enable bounce buffers" default y - depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM) + depends on BLOCK && MMU && HIGHMEM help - Enable bounce buffers for devices that cannot access - the full range of memory available to the CPU. Enabled - by default when ZONE_DMA or HIGHMEM is selected, but you - may say n to override this. + Enable bounce buffers for devices that cannot access the full range of + memory available to the CPU. Enabled by default when HIGHMEM is + selected, but you may say n to override this. config VIRT_TO_BUS bool |