^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /* Watch queue and general notification mechanism, built on pipes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) * Copyright (C) 2020 Red Hat, Inc. All Rights Reserved.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * Written by David Howells (dhowells@redhat.com)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) * See Documentation/watch_queue.rst
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #define pr_fmt(fmt) "watchq: " fmt
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/module.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/init.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/sched.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/printk.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/miscdevice.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/fs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <linux/pagemap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <linux/poll.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <linux/uaccess.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include <linux/vmalloc.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <linux/file.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include <linux/security.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #include <linux/cred.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) #include <linux/sched/signal.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #include <linux/watch_queue.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) #include <linux/pipe_fs_i.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) MODULE_DESCRIPTION("Watch queue");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) MODULE_AUTHOR("Red Hat, Inc.");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) MODULE_LICENSE("GPL");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) #define WATCH_QUEUE_NOTE_SIZE 128
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) #define WATCH_QUEUE_NOTES_PER_PAGE (PAGE_SIZE / WATCH_QUEUE_NOTE_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) static void watch_queue_pipe_buf_release(struct pipe_inode_info *pipe,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) struct pipe_buffer *buf)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) struct watch_queue *wqueue = (struct watch_queue *)buf->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) unsigned int bit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) /* We need to work out which note within the page this refers to, but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) * the note might have been maximum size, so merely ANDing the offset
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) * off doesn't work. OTOH, the note must've been more than zero size.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) bit = buf->offset + buf->len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) if ((bit & (WATCH_QUEUE_NOTE_SIZE - 1)) == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) bit -= WATCH_QUEUE_NOTE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) bit /= WATCH_QUEUE_NOTE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) page = buf->page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) bit += page->index;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) set_bit(bit, wqueue->notes_bitmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) generic_pipe_buf_release(pipe, buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) // No try_steal function => no stealing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) #define watch_queue_pipe_buf_try_steal NULL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) /* New data written to a pipe may be appended to a buffer with this type. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) static const struct pipe_buf_operations watch_queue_pipe_buf_ops = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) .release = watch_queue_pipe_buf_release,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) .try_steal = watch_queue_pipe_buf_try_steal,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) .get = generic_pipe_buf_get,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) * Post a notification to a watch queue.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) static bool post_one_notification(struct watch_queue *wqueue,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) struct watch_notification *n)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) void *p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) struct pipe_inode_info *pipe = wqueue->pipe;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) struct pipe_buffer *buf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) unsigned int head, tail, mask, note, offset, len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) bool done = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) if (!pipe)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) spin_lock_irq(&pipe->rd_wait.lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) if (wqueue->defunct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) mask = pipe->ring_size - 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) head = pipe->head;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) tail = pipe->tail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) if (pipe_full(head, tail, pipe->ring_size))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) goto lost;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) note = find_first_bit(wqueue->notes_bitmap, wqueue->nr_notes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) if (note >= wqueue->nr_notes)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) goto lost;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) page = wqueue->notes[note / WATCH_QUEUE_NOTES_PER_PAGE];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) offset = note % WATCH_QUEUE_NOTES_PER_PAGE * WATCH_QUEUE_NOTE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) get_page(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) len = n->info & WATCH_INFO_LENGTH;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) p = kmap_atomic(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) memcpy(p + offset, n, len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) kunmap_atomic(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) buf = &pipe->bufs[head & mask];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) buf->page = page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) buf->private = (unsigned long)wqueue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) buf->ops = &watch_queue_pipe_buf_ops;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) buf->offset = offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) buf->len = len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) buf->flags = PIPE_BUF_FLAG_WHOLE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) smp_store_release(&pipe->head, head + 1); /* vs pipe_read() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) if (!test_and_clear_bit(note, wqueue->notes_bitmap)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) spin_unlock_irq(&pipe->rd_wait.lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) wake_up_interruptible_sync_poll_locked(&pipe->rd_wait, EPOLLIN | EPOLLRDNORM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) done = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) spin_unlock_irq(&pipe->rd_wait.lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) if (done)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) return done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) lost:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) buf = &pipe->bufs[(head - 1) & mask];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) buf->flags |= PIPE_BUF_FLAG_LOSS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) * Apply filter rules to a notification.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) static bool filter_watch_notification(const struct watch_filter *wf,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) const struct watch_notification *n)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) const struct watch_type_filter *wt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) unsigned int st_bits = sizeof(wt->subtype_filter[0]) * 8;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) unsigned int st_index = n->subtype / st_bits;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) unsigned int st_bit = 1U << (n->subtype % st_bits);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) if (!test_bit(n->type, wf->type_filter))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) for (i = 0; i < wf->nr_filters; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) wt = &wf->filters[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) if (n->type == wt->type &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) (wt->subtype_filter[st_index] & st_bit) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) (n->info & wt->info_mask) == wt->info_filter)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) return false; /* If there is a filter, the default is to reject. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) * __post_watch_notification - Post an event notification
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) * @wlist: The watch list to post the event to.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) * @n: The notification record to post.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) * @cred: The creds of the process that triggered the notification.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) * @id: The ID to match on the watch.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) * Post a notification of an event into a set of watch queues and let the users
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) * know.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) * The size of the notification should be set in n->info & WATCH_INFO_LENGTH and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) * should be in units of sizeof(*n).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) void __post_watch_notification(struct watch_list *wlist,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) struct watch_notification *n,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) const struct cred *cred,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) u64 id)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) const struct watch_filter *wf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) struct watch_queue *wqueue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) struct watch *watch;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) if (((n->info & WATCH_INFO_LENGTH) >> WATCH_INFO_LENGTH__SHIFT) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) WARN_ON(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) rcu_read_lock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) hlist_for_each_entry_rcu(watch, &wlist->watchers, list_node) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) if (watch->id != id)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) n->info &= ~WATCH_INFO_ID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) n->info |= watch->info_id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) wqueue = rcu_dereference(watch->queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) wf = rcu_dereference(wqueue->filter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) if (wf && !filter_watch_notification(wf, n))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) if (security_post_notification(watch->cred, cred, n) < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) post_one_notification(wqueue, n);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) rcu_read_unlock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) EXPORT_SYMBOL(__post_watch_notification);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) * Allocate sufficient pages to preallocation for the requested number of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) * notifications.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) long watch_queue_set_size(struct pipe_inode_info *pipe, unsigned int nr_notes)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) struct watch_queue *wqueue = pipe->watch_queue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) struct page **pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) unsigned long *bitmap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) unsigned long user_bufs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) unsigned int bmsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) int ret, i, nr_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) if (!wqueue)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) if (wqueue->notes)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) return -EBUSY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) if (nr_notes < 1 ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) nr_notes > 512) /* TODO: choose a better hard limit */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) nr_pages = (nr_notes + WATCH_QUEUE_NOTES_PER_PAGE - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) nr_pages /= WATCH_QUEUE_NOTES_PER_PAGE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) user_bufs = account_pipe_buffers(pipe->user, pipe->nr_accounted, nr_pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) if (nr_pages > pipe->max_usage &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) (too_many_pipe_buffers_hard(user_bufs) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) too_many_pipe_buffers_soft(user_bufs)) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) pipe_is_unprivileged_user()) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) ret = -EPERM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) goto error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) nr_notes = nr_pages * WATCH_QUEUE_NOTES_PER_PAGE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) ret = pipe_resize_ring(pipe, roundup_pow_of_two(nr_notes));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) if (ret < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) goto error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) pages = kcalloc(sizeof(struct page *), nr_pages, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) if (!pages)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) goto error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) for (i = 0; i < nr_pages; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) pages[i] = alloc_page(GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) if (!pages[i])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) goto error_p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) pages[i]->index = i * WATCH_QUEUE_NOTES_PER_PAGE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) bmsize = (nr_notes + BITS_PER_LONG - 1) / BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) bmsize *= sizeof(unsigned long);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) bitmap = kmalloc(bmsize, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) if (!bitmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) goto error_p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) memset(bitmap, 0xff, bmsize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) wqueue->notes = pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) wqueue->notes_bitmap = bitmap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) wqueue->nr_pages = nr_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) wqueue->nr_notes = nr_notes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) error_p:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) while (--i >= 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) __free_page(pages[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) kfree(pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) error:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) (void) account_pipe_buffers(pipe->user, nr_pages, pipe->nr_accounted);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) * Set the filter on a watch queue.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) long watch_queue_set_filter(struct pipe_inode_info *pipe,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) struct watch_notification_filter __user *_filter)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) struct watch_notification_type_filter *tf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) struct watch_notification_filter filter;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) struct watch_type_filter *q;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) struct watch_filter *wfilter;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) struct watch_queue *wqueue = pipe->watch_queue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) int ret, nr_filter = 0, i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) if (!wqueue)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) if (!_filter) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) /* Remove the old filter */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) wfilter = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) goto set;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) /* Grab the user's filter specification */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) if (copy_from_user(&filter, _filter, sizeof(filter)) != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) return -EFAULT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) if (filter.nr_filters == 0 ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) filter.nr_filters > 16 ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) filter.__reserved != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) tf = memdup_user(_filter->filters, filter.nr_filters * sizeof(*tf));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) if (IS_ERR(tf))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) return PTR_ERR(tf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) for (i = 0; i < filter.nr_filters; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) if ((tf[i].info_filter & ~tf[i].info_mask) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) tf[i].info_mask & WATCH_INFO_LENGTH)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) goto err_filter;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) /* Ignore any unknown types */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) if (tf[i].type >= WATCH_TYPE__NR)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) nr_filter++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) /* Now we need to build the internal filter from only the relevant
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) * user-specified filters.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) ret = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) wfilter = kzalloc(struct_size(wfilter, filters, nr_filter), GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) if (!wfilter)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) goto err_filter;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) wfilter->nr_filters = nr_filter;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) q = wfilter->filters;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) for (i = 0; i < filter.nr_filters; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) if (tf[i].type >= WATCH_TYPE__NR)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) q->type = tf[i].type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) q->info_filter = tf[i].info_filter;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) q->info_mask = tf[i].info_mask;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) q->subtype_filter[0] = tf[i].subtype_filter[0];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) __set_bit(q->type, wfilter->type_filter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) q++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) kfree(tf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) set:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) pipe_lock(pipe);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) wfilter = rcu_replace_pointer(wqueue->filter, wfilter,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) lockdep_is_held(&pipe->mutex));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) pipe_unlock(pipe);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) if (wfilter)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) kfree_rcu(wfilter, rcu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) err_filter:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) kfree(tf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) static void __put_watch_queue(struct kref *kref)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) struct watch_queue *wqueue =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) container_of(kref, struct watch_queue, usage);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) struct watch_filter *wfilter;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) for (i = 0; i < wqueue->nr_pages; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) __free_page(wqueue->notes[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) kfree(wqueue->notes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) bitmap_free(wqueue->notes_bitmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) wfilter = rcu_access_pointer(wqueue->filter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) if (wfilter)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) kfree_rcu(wfilter, rcu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) kfree_rcu(wqueue, rcu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) * put_watch_queue - Dispose of a ref on a watchqueue.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) * @wqueue: The watch queue to unref.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) void put_watch_queue(struct watch_queue *wqueue)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) kref_put(&wqueue->usage, __put_watch_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) EXPORT_SYMBOL(put_watch_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) static void free_watch(struct rcu_head *rcu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) struct watch *watch = container_of(rcu, struct watch, rcu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) put_watch_queue(rcu_access_pointer(watch->queue));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) atomic_dec(&watch->cred->user->nr_watches);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) put_cred(watch->cred);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) kfree(watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) static void __put_watch(struct kref *kref)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) struct watch *watch = container_of(kref, struct watch, usage);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) call_rcu(&watch->rcu, free_watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) * Discard a watch.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) static void put_watch(struct watch *watch)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) kref_put(&watch->usage, __put_watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) * init_watch_queue - Initialise a watch
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) * @watch: The watch to initialise.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) * @wqueue: The queue to assign.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) * Initialise a watch and set the watch queue.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) void init_watch(struct watch *watch, struct watch_queue *wqueue)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) kref_init(&watch->usage);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) INIT_HLIST_NODE(&watch->list_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) INIT_HLIST_NODE(&watch->queue_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) rcu_assign_pointer(watch->queue, wqueue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) * add_watch_to_object - Add a watch on an object to a watch list
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) * @watch: The watch to add
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) * @wlist: The watch list to add to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) * @watch->queue must have been set to point to the queue to post notifications
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) * to and the watch list of the object to be watched. @watch->cred must also
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) * have been set to the appropriate credentials and a ref taken on them.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) * The caller must pin the queue and the list both and must hold the list
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) * locked against racing watch additions/removals.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) int add_watch_to_object(struct watch *watch, struct watch_list *wlist)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) struct watch_queue *wqueue = rcu_access_pointer(watch->queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) struct watch *w;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) hlist_for_each_entry(w, &wlist->watchers, list_node) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) struct watch_queue *wq = rcu_access_pointer(w->queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) if (wqueue == wq && watch->id == w->id)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) return -EBUSY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) watch->cred = get_current_cred();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) rcu_assign_pointer(watch->watch_list, wlist);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) if (atomic_inc_return(&watch->cred->user->nr_watches) >
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) task_rlimit(current, RLIMIT_NOFILE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) atomic_dec(&watch->cred->user->nr_watches);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) put_cred(watch->cred);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) return -EAGAIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) spin_lock_bh(&wqueue->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) kref_get(&wqueue->usage);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) kref_get(&watch->usage);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) hlist_add_head(&watch->queue_node, &wqueue->watches);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) spin_unlock_bh(&wqueue->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) hlist_add_head(&watch->list_node, &wlist->watchers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) EXPORT_SYMBOL(add_watch_to_object);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) * remove_watch_from_object - Remove a watch or all watches from an object.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) * @wlist: The watch list to remove from
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) * @wq: The watch queue of interest (ignored if @all is true)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) * @id: The ID of the watch to remove (ignored if @all is true)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) * @all: True to remove all objects
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) * Remove a specific watch or all watches from an object. A notification is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) * sent to the watcher to tell them that this happened.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) int remove_watch_from_object(struct watch_list *wlist, struct watch_queue *wq,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) u64 id, bool all)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) struct watch_notification_removal n;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) struct watch_queue *wqueue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) struct watch *watch;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) int ret = -EBADSLT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) rcu_read_lock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) again:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) spin_lock(&wlist->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) hlist_for_each_entry(watch, &wlist->watchers, list_node) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) if (all ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) (watch->id == id && rcu_access_pointer(watch->queue) == wq))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) goto found;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) spin_unlock(&wlist->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) found:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) hlist_del_init_rcu(&watch->list_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) rcu_assign_pointer(watch->watch_list, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) spin_unlock(&wlist->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) /* We now own the reference on watch that used to belong to wlist. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) n.watch.type = WATCH_TYPE_META;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) n.watch.subtype = WATCH_META_REMOVAL_NOTIFICATION;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) n.watch.info = watch->info_id | watch_sizeof(n.watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) n.id = id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) if (id != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) n.watch.info = watch->info_id | watch_sizeof(n);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) wqueue = rcu_dereference(watch->queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) /* We don't need the watch list lock for the next bit as RCU is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) * protecting *wqueue from deallocation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) if (wqueue) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) post_one_notification(wqueue, &n.watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) spin_lock_bh(&wqueue->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) if (!hlist_unhashed(&watch->queue_node)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) hlist_del_init_rcu(&watch->queue_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) put_watch(watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) spin_unlock_bh(&wqueue->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) if (wlist->release_watch) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) void (*release_watch)(struct watch *);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) release_watch = wlist->release_watch;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) rcu_read_unlock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) (*release_watch)(watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) rcu_read_lock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) put_watch(watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) if (all && !hlist_empty(&wlist->watchers))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) goto again;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) rcu_read_unlock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) EXPORT_SYMBOL(remove_watch_from_object);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) * Remove all the watches that are contributory to a queue. This has the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) * potential to race with removal of the watches by the destruction of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) * objects being watched or with the distribution of notifications.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) void watch_queue_clear(struct watch_queue *wqueue)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) struct watch_list *wlist;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) struct watch *watch;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) bool release;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) rcu_read_lock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) spin_lock_bh(&wqueue->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) /* Prevent new notifications from being stored. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) wqueue->defunct = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) while (!hlist_empty(&wqueue->watches)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) watch = hlist_entry(wqueue->watches.first, struct watch, queue_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) hlist_del_init_rcu(&watch->queue_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) /* We now own a ref on the watch. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) spin_unlock_bh(&wqueue->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) /* We can't do the next bit under the queue lock as we need to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) * get the list lock - which would cause a deadlock if someone
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) * was removing from the opposite direction at the same time or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) * posting a notification.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) wlist = rcu_dereference(watch->watch_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) if (wlist) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) void (*release_watch)(struct watch *);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) spin_lock(&wlist->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) release = !hlist_unhashed(&watch->list_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) if (release) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) hlist_del_init_rcu(&watch->list_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) rcu_assign_pointer(watch->watch_list, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) /* We now own a second ref on the watch. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) release_watch = wlist->release_watch;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) spin_unlock(&wlist->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) if (release) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) if (release_watch) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) rcu_read_unlock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) /* This might need to call dput(), so
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) * we have to drop all the locks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) (*release_watch)(watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) rcu_read_lock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) put_watch(watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) put_watch(watch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) spin_lock_bh(&wqueue->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) spin_unlock_bh(&wqueue->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) rcu_read_unlock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) * get_watch_queue - Get a watch queue from its file descriptor.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) * @fd: The fd to query.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) struct watch_queue *get_watch_queue(int fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) struct pipe_inode_info *pipe;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) struct watch_queue *wqueue = ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) struct fd f;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) f = fdget(fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) if (f.file) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) pipe = get_pipe_info(f.file, false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) if (pipe && pipe->watch_queue) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) wqueue = pipe->watch_queue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) kref_get(&wqueue->usage);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) fdput(f);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) return wqueue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) EXPORT_SYMBOL(get_watch_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) * Initialise a watch queue
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) int watch_queue_init(struct pipe_inode_info *pipe)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) struct watch_queue *wqueue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) wqueue = kzalloc(sizeof(*wqueue), GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) if (!wqueue)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) wqueue->pipe = pipe;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) kref_init(&wqueue->usage);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) spin_lock_init(&wqueue->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) INIT_HLIST_HEAD(&wqueue->watches);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666) pipe->watch_queue = wqueue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) }
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags