*/int ring_buffer_swap_cpu(struct ring_buffer *buffer_a, struct ring_buffer *buffer_b, int cpu){ struct ring_buffer_per_cpu *cpu_buffer_a; struct ring_buffer_per_cpu *cpu_buffer_b; int ret = -EINVAL; if (!cpumask_test_cpu(cpu, buffer_a->cpumask) || !cpumask_test_cpu(cpu, buffer_b->cpumask)) goto out; cpu_buffer_a = buffer_a->buffers[cpu]; cpu_buffer_b = buffer_b->buffers[cpu]; /* At least make sure the two buffers are somewhat the same */ if (cpu_buffer_a->nr_pages != cpu_buffer_b->nr_pages) goto out; ret = -EAGAIN; if (ring_buffer_flags != RB_BUFFERS_ON) goto out; if (atomic_read(&buffer_a->record_disabled)) goto out; if (atomic_read(&buffer_b->record_disabled)) goto out; if (atomic_read(&cpu_buffer_a->record_disabled)) goto out; if (atomic_read(&cpu_buffer_b->record_disabled)) goto out; /* * We can't do a synchronize_sched here because this * function can be called in atomic context. * Normally this will be called from the same CPU as cpu. * If not it's up to the caller to protect this. */ atomic_inc(&cpu_buffer_a->record_disabled); atomic_inc(&cpu_buffer_b->record_disabled); ret = -EBUSY; if (local_read(&cpu_buffer_a->committing)) goto out_dec; if (local_read(&cpu_buffer_b->committing)) goto out_dec; buffer_a->buffers[cpu] = cpu_buffer_b; buffer_b->buffers[cpu] = cpu_buffer_a; cpu_buffer_b->buffer = buffer_a; cpu_buffer_a->buffer = buffer_b; ret = 0; out_dec: atomic_dec(&cpu_buffer_a->record_disabled); atomic_dec(&cpu_buffer_b->record_disabled);out: return ret;}EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu);#endif /* CONFIG_RING_BUFFER_ALLOW_SWAP */ /** * ring_buffer_alloc_read_page - allocate a page to read from buffer * @buffer: the buffer to allocate for. * * This function is used in conjunction with ring_buffer_read_page. * When reading a full page from the ring buffer, these functions * can be used to speed up the process. The calling function should * allocate a few pages first with this function. Then when it * needs to get pages from the ring buffer, it passes the result * of this function into ring_buffer_read_page, which will swap * the page that was allocated, with the read page of the buffer. * * Returns: * The page allocated, or NULL on error. */void *ring_buffer_alloc_read_page(struct ring_buffer *buffer, int cpu){ struct buffer_data_page *bpage; struct page *page; page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL | __GFP_NORETRY, 0); if (!page) return NULL; bpage = page_address(page); rb_init_page(bpage); return bpage;}EXPORT_SYMBOL_GPL(ring_buffer_alloc_read_page); /** * ring_buffer_free_read_page - free an allocated read page * @buffer: the buffer the page was allocate for * @data: the page to free * * Free a page allocated from ring_buffer_alloc_read_page. */void ring_buffer_free_read_page(struct ring_buffer *buffer, void *data){ free_page((unsigned long)data);}EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); /** * ring_buffer_read_page - extract a page from the ring buffer * @buffer: buffer to extract from * @data_page: the page to use allocated from ring_buffer_alloc_read_page * @len: amount to extract * @cpu: the cpu of the buffer to extract * @full: should the extraction only happen when the page is full. * * This function will pull out a page from the ring buffer and consume it. * @data_page must be the address of the variable that was returned * from ring_buffer_alloc_read_page. This is because the page might be used * to swap with a page in the ring buffer. * * for example: * rpage = ring_buffer_alloc_read_page(buffer); * if (!rpage) * return error; * ret = ring_buffer_read_page(buffer, &rpage, len, cpu, 0); * if (ret >= 0) * process_page(rpage, ret); * * When @full is set, the function will not return true unless * the writer is off the reader page. * * Note: it is up to the calling functions to handle sleeps and wakeups. * The ring buffer can be used anywhere in the kernel and can not * blindly call wake_up. The layer that uses the ring buffer must be * responsible for that. * * Returns: * >=0 if data has been transferred, returns the offset of consumed data. * <0 if no data has been transferred. */int ring_buffer_read_page(struct ring_buffer *buffer, void **data_page, size_t len, int cpu, int full){ struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; struct ring_buffer_event *event; struct buffer_data_page *bpage; struct buffer_page *reader; unsigned long missed_events; unsigned long flags; unsigned int commit; unsigned int read; u64 save_timestamp; int ret = -1; if (!cpumask_test_cpu(cpu, buffer->cpumask)) goto out; /* * If len is not big enough to hold the page header, then * we can not copy anything. */ if (len <= BUF_PAGE_HDR_SIZE) goto out; len -= BUF_PAGE_HDR_SIZE; if (!data_page) goto out; bpage = *data_page; if (!bpage) goto out; raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); reader = rb_get_reader_page(cpu_buffer); if (!reader) goto out_unlock; event = rb_reader_event(cpu_buffer); read = reader->read; commit = rb_page_commit(reader); /* Check if any events were dropped */ missed_events = cpu_buffer->lost_events; /* * If this page has been partially read or * if len is not big enough to read the rest of the page or * a writer is still on the page, then * we must copy the data from the page to the buffer. * Otherwise, we can simply swap the page with the one passed in. */ if (read || (len < (commit - read)) || cpu_buffer->reader_page == cpu_buffer->commit_page) { struct buffer_data_page *rpage = cpu_buffer->reader_page->page; unsigned int rpos = read; unsigned int pos = 0; unsigned int size; if (full) goto out_unlock; if (len > (commit - read)) len = (commit - read); /* Always keep the time extend and data together */ size = rb_event_ts_length(event); if (len < size) goto out_unlock; /* save the current timestamp, since the user will need it */ save_timestamp = cpu_buffer->read_stamp; /* Need to copy one event at a time */ do { /* We need the size of one event, because * rb_advance_reader only advances by one event, * whereas rb_event_ts_length may include the size of * one or two events. * We have already ensured there's enough space if this * is a time extend. */ size = rb_event_length(event); memcpy(bpage->data + pos, rpage->data + rpos, size); len -= size; rb_advance_reader(cpu_buffer); rpos = reader->read; pos += size; if (rpos >= commit) break; event = rb_reader_event(cpu_buffer); /* Always keep the time extend and data together */ size = rb_event_ts_length(event); } while (len >= size); /* update bpage */ local_set(&bpage->commit, pos); bpage->time_stamp = save_timestamp; /* we copied everything to the beginning */ read = 0; } else { /* update the entry counter */ cpu_buffer->read += rb_page_entries(reader); cpu_buffer->read_bytes += BUF_PAGE_SIZE; /* swap the pages */ rb_init_page(bpage); bpage = reader->page; reader->page = *data_page; local_set(&reader->write, 0); local_set(&reader->entries, 0); reader->read = 0; *data_page = bpage; /* * Use the real_end for the data size, * This gives us a chance to store the lost events * on the page. */ if (reader->real_end) local_set(&bpage->commit, reader->real_end); } ret = read; cpu_buffer->lost_events = 0; commit = local_read(&bpage->commit); /* * Set a flag in the commit field if we lost events */ if (missed_events) { /* If there is room at the end of the page to save the * missed events, then record it there. */ if (BUF_PAGE_SIZE - commit >= sizeof(missed_events)) { memcpy(&bpage->data[commit], &missed_events, sizeof(missed_events)); local_add(RB_MISSED_STORED, &bpage->commit); commit += sizeof(missed_events); } local_add(RB_MISSED_EVENTS, &bpage->commit); } /* * This page may be off to user land. Zero it out here. */ if (commit < BUF_PAGE_SIZE) memset(&bpage->data[commit], 0, BUF_PAGE_SIZE - commit); out_unlock: raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); out: return ret;}EXPORT_SYMBOL_GPL(ring_buffer_read_page); #ifdef CONFIG_HOTPLUG_CPUstatic int rb_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu){ struct ring_buffer *buffer = container_of(self, struct ring_buffer, cpu_notify); long cpu = (long)hcpu; int cpu_i, nr_pages_same; unsigned int nr_pages; switch (action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: if (cpumask_test_cpu(cpu, buffer->cpumask)) return NOTIFY_OK; nr_pages = 0; nr_pages_same = 1; /* check if all cpu sizes are same */ for_each_buffer_cpu(buffer, cpu_i) { /* fill in the size from first enabled cpu */ if (nr_pages == 0) nr_pages = buffer->buffers[cpu_i]->nr_pages; if (nr_pages != buffer->buffers[cpu_i]->nr_pages) { nr_pages_same = 0; break; } } /* allocate minimum pages, user can later expand it */ if (!nr_pages_same) nr_pages = 2; buffer->buffers[cpu] = rb_allocate_cpu_buffer(buffer, nr_pages, cpu); if (!buffer->buffers[cpu]) { WARN(1, "failed to allocate ring buffer on CPU %ld\n", cpu); return NOTIFY_OK; } smp_wmb(); cpumask_set_cpu(cpu, buffer->cpumask); break; case CPU_DOWN_PREPARE: case CPU_DOWN_PREPARE_FROZEN: /* * Do nothing. * If we were to free the buffer, then the user would * lose any trace that was in the buffer. */ break; default: break; } return NOTIFY_OK;}#endifLines 4001–4366 of 4366. Previous lines
*/ int ring_buffer_swap_cpu(struct ring_buffer *buffer_a, struct ring_buffer *buffer_b, int cpu) { struct ring_buffer_per_cpu *cpu_buffer_a; struct ring_buffer_per_cpu *cpu_buffer_b; int ret = -EINVAL; if (!cpumask_test_cpu(cpu, buffer_a->cpumask) || !cpumask_test_cpu(cpu, buffer_b->cpumask)) goto out; cpu_buffer_a = buffer_a->buffers[cpu]; cpu_buffer_b = buffer_b->buffers[cpu]; /* At least make sure the two buffers are somewhat the same */ if (cpu_buffer_a->nr_pages != cpu_buffer_b->nr_pages) goto out; ret = -EAGAIN; if (ring_buffer_flags != RB_BUFFERS_ON) goto out; if (atomic_read(&buffer_a->record_disabled)) goto out; if (atomic_read(&buffer_b->record_disabled)) goto out; if (atomic_read(&cpu_buffer_a->record_disabled)) goto out; if (atomic_read(&cpu_buffer_b->record_disabled)) goto out; /* * We can't do a synchronize_sched here because this * function can be called in atomic context. * Normally this will be called from the same CPU as cpu. * If not it's up to the caller to protect this. */ atomic_inc(&cpu_buffer_a->record_disabled); atomic_inc(&cpu_buffer_b->record_disabled); ret = -EBUSY; if (local_read(&cpu_buffer_a->committing)) goto out_dec; if (local_read(&cpu_buffer_b->committing)) goto out_dec; buffer_a->buffers[cpu] = cpu_buffer_b; buffer_b->buffers[cpu] = cpu_buffer_a; cpu_buffer_b->buffer = buffer_a; cpu_buffer_a->buffer = buffer_b; ret = 0; out_dec: atomic_dec(&cpu_buffer_a->record_disabled); atomic_dec(&cpu_buffer_b->record_disabled); out: return ret; } EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu); #endif /* CONFIG_RING_BUFFER_ALLOW_SWAP */ /** * ring_buffer_alloc_read_page - allocate a page to read from buffer * @buffer: the buffer to allocate for. * * This function is used in conjunction with ring_buffer_read_page. * When reading a full page from the ring buffer, these functions * can be used to speed up the process. The calling function should * allocate a few pages first with this function. Then when it * needs to get pages from the ring buffer, it passes the result * of this function into ring_buffer_read_page, which will swap * the page that was allocated, with the read page of the buffer. * * Returns: * The page allocated, or NULL on error. */ void *ring_buffer_alloc_read_page(struct ring_buffer *buffer, int cpu) { struct buffer_data_page *bpage; struct page *page; page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL | __GFP_NORETRY, 0); if (!page) return NULL; bpage = page_address(page); rb_init_page(bpage); return bpage; } EXPORT_SYMBOL_GPL(ring_buffer_alloc_read_page); /** * ring_buffer_free_read_page - free an allocated read page * @buffer: the buffer the page was allocate for * @data: the page to free * * Free a page allocated from ring_buffer_alloc_read_page. */ void ring_buffer_free_read_page(struct ring_buffer *buffer, void *data) { free_page((unsigned long)data); } EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); /** * ring_buffer_read_page - extract a page from the ring buffer * @buffer: buffer to extract from * @data_page: the page to use allocated from ring_buffer_alloc_read_page * @len: amount to extract * @cpu: the cpu of the buffer to extract * @full: should the extraction only happen when the page is full. * * This function will pull out a page from the ring buffer and consume it. * @data_page must be the address of the variable that was returned * from ring_buffer_alloc_read_page. This is because the page might be used * to swap with a page in the ring buffer. * * for example: * rpage = ring_buffer_alloc_read_page(buffer); * if (!rpage) * return error; * ret = ring_buffer_read_page(buffer, &rpage, len, cpu, 0); * if (ret >= 0) * process_page(rpage, ret); * * When @full is set, the function will not return true unless * the writer is off the reader page. * * Note: it is up to the calling functions to handle sleeps and wakeups. * The ring buffer can be used anywhere in the kernel and can not * blindly call wake_up. The layer that uses the ring buffer must be * responsible for that. * * Returns: * >=0 if data has been transferred, returns the offset of consumed data. * <0 if no data has been transferred. */ int ring_buffer_read_page(struct ring_buffer *buffer, void **data_page, size_t len, int cpu, int full) { struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; struct ring_buffer_event *event; struct buffer_data_page *bpage; struct buffer_page *reader; unsigned long missed_events; unsigned long flags; unsigned int commit; unsigned int read; u64 save_timestamp; int ret = -1; if (!cpumask_test_cpu(cpu, buffer->cpumask)) goto out; /* * If len is not big enough to hold the page header, then * we can not copy anything. */ if (len <= BUF_PAGE_HDR_SIZE) goto out; len -= BUF_PAGE_HDR_SIZE; if (!data_page) goto out; bpage = *data_page; if (!bpage) goto out; raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); reader = rb_get_reader_page(cpu_buffer); if (!reader) goto out_unlock; event = rb_reader_event(cpu_buffer); read = reader->read; commit = rb_page_commit(reader); /* Check if any events were dropped */ missed_events = cpu_buffer->lost_events; /* * If this page has been partially read or * if len is not big enough to read the rest of the page or * a writer is still on the page, then * we must copy the data from the page to the buffer. * Otherwise, we can simply swap the page with the one passed in. */ if (read || (len < (commit - read)) || cpu_buffer->reader_page == cpu_buffer->commit_page) { struct buffer_data_page *rpage = cpu_buffer->reader_page->page; unsigned int rpos = read; unsigned int pos = 0; unsigned int size; if (full) goto out_unlock; if (len > (commit - read)) len = (commit - read); /* Always keep the time extend and data together */ size = rb_event_ts_length(event); if (len < size) goto out_unlock; /* save the current timestamp, since the user will need it */ save_timestamp = cpu_buffer->read_stamp; /* Need to copy one event at a time */ do { /* We need the size of one event, because * rb_advance_reader only advances by one event, * whereas rb_event_ts_length may include the size of * one or two events. * We have already ensured there's enough space if this * is a time extend. */ size = rb_event_length(event); memcpy(bpage->data + pos, rpage->data + rpos, size); len -= size; rb_advance_reader(cpu_buffer); rpos = reader->read; pos += size; if (rpos >= commit) break; event = rb_reader_event(cpu_buffer); /* Always keep the time extend and data together */ size = rb_event_ts_length(event); } while (len >= size); /* update bpage */ local_set(&bpage->commit, pos); bpage->time_stamp = save_timestamp; /* we copied everything to the beginning */ read = 0; } else { /* update the entry counter */ cpu_buffer->read += rb_page_entries(reader); cpu_buffer->read_bytes += BUF_PAGE_SIZE; /* swap the pages */ rb_init_page(bpage); bpage = reader->page; reader->page = *data_page; local_set(&reader->write, 0); local_set(&reader->entries, 0); reader->read = 0; *data_page = bpage; /* * Use the real_end for the data size, * This gives us a chance to store the lost events * on the page. */ if (reader->real_end) local_set(&bpage->commit, reader->real_end); } ret = read; cpu_buffer->lost_events = 0; commit = local_read(&bpage->commit); /* * Set a flag in the commit field if we lost events */ if (missed_events) { /* If there is room at the end of the page to save the * missed events, then record it there. */ if (BUF_PAGE_SIZE - commit >= sizeof(missed_events)) { memcpy(&bpage->data[commit], &missed_events, sizeof(missed_events)); local_add(RB_MISSED_STORED, &bpage->commit); commit += sizeof(missed_events); } local_add(RB_MISSED_EVENTS, &bpage->commit); } /* * This page may be off to user land. Zero it out here. */ if (commit < BUF_PAGE_SIZE) memset(&bpage->data[commit], 0, BUF_PAGE_SIZE - commit); out_unlock: raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); out: return ret; } EXPORT_SYMBOL_GPL(ring_buffer_read_page); #ifdef CONFIG_HOTPLUG_CPU static int rb_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { struct ring_buffer *buffer = container_of(self, struct ring_buffer, cpu_notify); long cpu = (long)hcpu; int cpu_i, nr_pages_same; unsigned int nr_pages; switch (action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: if (cpumask_test_cpu(cpu, buffer->cpumask)) return NOTIFY_OK; nr_pages = 0; nr_pages_same = 1; /* check if all cpu sizes are same */ for_each_buffer_cpu(buffer, cpu_i) { /* fill in the size from first enabled cpu */ if (nr_pages == 0) nr_pages = buffer->buffers[cpu_i]->nr_pages; if (nr_pages != buffer->buffers[cpu_i]->nr_pages) { nr_pages_same = 0; break; } } /* allocate minimum pages, user can later expand it */ if (!nr_pages_same) nr_pages = 2; buffer->buffers[cpu] = rb_allocate_cpu_buffer(buffer, nr_pages, cpu); if (!buffer->buffers[cpu]) { WARN(1, "failed to allocate ring buffer on CPU %ld\n", cpu); return NOTIFY_OK; } smp_wmb(); cpumask_set_cpu(cpu, buffer->cpumask); break; case CPU_DOWN_PREPARE: case CPU_DOWN_PREPARE_FROZEN: /* * Do nothing. * If we were to free the buffer, then the user would * lose any trace that was in the buffer. */ break; default: break; } return NOTIFY_OK; } #endif