| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <4A305A1A.2010901@cn.fujitsu.com>
Date: Thu, 11 Jun 2009 09:12:58 +0800
From: Lai Jiangshan <laijs@...fujitsu.com>
To: Steven Rostedt <rostedt@...dmis.org>
CC: linux-kernel@...r.kernel.org, Ingo Molnar <mingo@...e.hu>,
Andrew Morton <akpm@...ux-foundation.org>,
Thomas Gleixner <tglx@...utronix.de>,
Peter Zijlstra <peterz@...radead.org>,
Frederic Weisbecker <fweisbec@...il.com>,
Theodore Tso <tytso@....edu>,
Arnaldo Carvalho de Melo <acme@...hat.com>,
Mathieu Desnoyers <compudj@...stal.dyndns.org>,
"Martin J. Bligh" <mbligh@...igh.org>,
Christoph Hellwig <hch@...radead.org>,
Li Zefan <lizf@...fujitsu.com>,
Huang Ying <ying.huang@...el.com>,
"H. Peter Anvin" <hpa@...or.com>,
Hidetoshi Seto <seto.hidetoshi@...fujitsu.com>,
Masami Hiramatsu <mhiramat@...hat.com>,
Steven Rostedt <srostedt@...hat.com>
Subject: Re: [PATCH 1/3] ring-buffer: make the buffer a true circular link
list
Steven Rostedt wrote:
> From: Steven Rostedt <srostedt@...hat.com>
>
> This patch changes the ring buffer data pages from using a link list
> head pointer, to making each buffer page point to another buffer page
> and never back to a "head".
>
> This makes the handling of the ring buffer less complex, since the
> traversing of the ring buffer pages no longer needs to account for the
> head pointer.
>
> This change also is needed to make the ring buffer lockless.
>
> [ Impact: simplify the ring buffer to help make it lockless ]
>
> Signed-off-by: Steven Rostedt <srostedt@...hat.com>
> ---
> kernel/trace/ring_buffer.c | 54 ++++++++++++++++++++++++++++++-------------
> 1 files changed, 37 insertions(+), 17 deletions(-)
>
> diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
> index 2e642b2..d1edd63 100644
> --- a/kernel/trace/ring_buffer.c
> +++ b/kernel/trace/ring_buffer.c
> @@ -404,7 +404,7 @@ struct ring_buffer_per_cpu {
> spinlock_t reader_lock; /* serialize readers */
> raw_spinlock_t lock;
> struct lock_class_key lock_key;
> - struct list_head pages;
> + struct list_head *pages;
> struct buffer_page *head_page; /* read from head */
> struct buffer_page *tail_page; /* write to tail */
> struct buffer_page *commit_page; /* committed pages */
> @@ -494,7 +494,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_normalize_time_stamp);
> */
> static int rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer)
> {
> - struct list_head *head = &cpu_buffer->pages;
> + struct list_head *head = cpu_buffer->pages;
> struct buffer_page *bpage, *tmp;
>
> if (RB_WARN_ON(cpu_buffer, head->next->prev != head))
> @@ -517,12 +517,13 @@ static int rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer)
> static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer,
> unsigned nr_pages)
> {
> - struct list_head *head = &cpu_buffer->pages;
> struct buffer_page *bpage, *tmp;
> unsigned long addr;
> LIST_HEAD(pages);
> unsigned i;
>
> + WARN_ON(!nr_pages);
> +
> for (i = 0; i < nr_pages; i++) {
> bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()),
> GFP_KERNEL, cpu_to_node(cpu_buffer->cpu));
> @@ -537,7 +538,18 @@ static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer,
> rb_init_page(bpage->page);
> }
>
> - list_splice(&pages, head);
> + /*
> + * The ring buffer page list is a circular list that does not
> + * start and end with a list head. All page list items point to
> + * other pages. Remove one of the pages, init its list head,
> + * and use list splice to move the rest of the pages to it.
> + */
> + bpage = list_entry(pages.next, struct buffer_page, list);
> + list_del_init(&bpage->list);
> + cpu_buffer->pages = &bpage->list;
> +
> + list_splice(&pages, cpu_buffer->pages);
> +
Is it proper?
cpu_buffer->pages = pages.next;
list_del(&pages);
>
> rb_check_pages(cpu_buffer);
>
> @@ -569,7 +581,6 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu)
> spin_lock_init(&cpu_buffer->reader_lock);
> lockdep_set_class(&cpu_buffer->reader_lock, buffer->reader_lock_key);
> cpu_buffer->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED;
> - INIT_LIST_HEAD(&cpu_buffer->pages);
>
> bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()),
> GFP_KERNEL, cpu_to_node(cpu));
> @@ -590,7 +601,7 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu)
> goto fail_free_reader;
>
> cpu_buffer->head_page
> - = list_entry(cpu_buffer->pages.next, struct buffer_page, list);
> + = list_entry(cpu_buffer->pages, struct buffer_page, list);
> cpu_buffer->tail_page = cpu_buffer->commit_page = cpu_buffer->head_page;
>
> return cpu_buffer;
> @@ -605,15 +616,20 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu)
>
> static void rb_free_cpu_buffer(struct ring_buffer_per_cpu *cpu_buffer)
> {
> - struct list_head *head = &cpu_buffer->pages;
> + struct list_head *head = cpu_buffer->pages;
> struct buffer_page *bpage, *tmp;
>
> free_buffer_page(cpu_buffer->reader_page);
>
> - list_for_each_entry_safe(bpage, tmp, head, list) {
> - list_del_init(&bpage->list);
> + if (head) {
> + list_for_each_entry_safe(bpage, tmp, head, list) {
> + list_del_init(&bpage->list);
> + free_buffer_page(bpage);
> + }
> + bpage = list_entry(head, struct buffer_page, list);
> free_buffer_page(bpage);
> }
> +
> kfree(cpu_buffer);
> }
>
> @@ -767,14 +783,14 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages)
> synchronize_sched();
>
> for (i = 0; i < nr_pages; i++) {
> - if (RB_WARN_ON(cpu_buffer, list_empty(&cpu_buffer->pages)))
> + if (RB_WARN_ON(cpu_buffer, list_empty(cpu_buffer->pages)))
> return;
> - p = cpu_buffer->pages.next;
> + p = cpu_buffer->pages->next;
> bpage = list_entry(p, struct buffer_page, list);
> list_del_init(&bpage->list);
> free_buffer_page(bpage);
> }
> - if (RB_WARN_ON(cpu_buffer, list_empty(&cpu_buffer->pages)))
> + if (RB_WARN_ON(cpu_buffer, list_empty(cpu_buffer->pages)))
> return;
>
> rb_reset_cpu(cpu_buffer);
> @@ -802,7 +818,7 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer,
> p = pages->next;
> bpage = list_entry(p, struct buffer_page, list);
> list_del_init(&bpage->list);
> - list_add_tail(&bpage->list, &cpu_buffer->pages);
> + list_add_tail(&bpage->list, cpu_buffer->pages);
> }
> rb_reset_cpu(cpu_buffer);
>
> @@ -999,9 +1015,6 @@ static inline void rb_inc_page(struct ring_buffer_per_cpu *cpu_buffer,
> {
> struct list_head *p = (*bpage)->list.next;
>
> - if (p == &cpu_buffer->pages)
> - p = p->next;
> -
> *bpage = list_entry(p, struct buffer_page, list);
> }
>
> @@ -2212,6 +2225,13 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer)
> cpu_buffer->reader_page->list.next = reader->list.next;
> cpu_buffer->reader_page->list.prev = reader->list.prev;
>
> + /*
> + * cpu_buffer->pages just needs to point to the buffer, it
> + * has no specific buffer page to point to. Lets move it out
> + * of our way so we don't accidently swap it.
> + */
> + cpu_buffer->pages = reader->list.prev;
> +
> local_set(&cpu_buffer->reader_page->write, 0);
> local_set(&cpu_buffer->reader_page->entries, 0);
> local_set(&cpu_buffer->reader_page->page->commit, 0);
> @@ -2656,7 +2676,7 @@ static void
> rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer)
> {
> cpu_buffer->head_page
> - = list_entry(cpu_buffer->pages.next, struct buffer_page, list);
> + = list_entry(cpu_buffer->pages, struct buffer_page, list);
> local_set(&cpu_buffer->head_page->write, 0);
> local_set(&cpu_buffer->head_page->entries, 0);
> local_set(&cpu_buffer->head_page->page->commit, 0);
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists