Re: [PATCH v2 1/4] printk/NMI: Handle continuous lines and missing newline

From: Steven Rostedt
Date: Fri Nov 11 2016 - 12:28:58 EST


On Wed, 9 Nov 2016 13:41:28 +0100
Petr Mladek <pmladek@xxxxxxxx> wrote:


> /*
> @@ -135,8 +170,8 @@ static void __printk_nmi_flush(struct irq_work *work)
> __RAW_SPIN_LOCK_INITIALIZER(read_lock);
> struct nmi_seq_buf *s = container_of(work, struct nmi_seq_buf, work);
> unsigned long flags;
> - size_t len, size;
> - int i, last_i;
> + size_t len;
> + int i;
>
> /*
> * The lock has two functions. First, one reader has to flush all
> @@ -154,12 +189,14 @@ static void __printk_nmi_flush(struct irq_work *work)
> /*
> * This is just a paranoid check that nobody has manipulated
> * the buffer an unexpected way. If we printed something then
> - * @len must only increase.
> + * @len must only increase. Also it should never overflow the
> + * buffer size.
> */
> - if (i && i >= len) {
> + if ((i && i >= len) || len > sizeof(s->buffer)) {

What's wrong with using s->len? Isn't that what is inside the buffer?
Couldn't just checking against the buffer size print garbage?

-- Steve


> const char *msg = "printk_nmi_flush: internal error\n";
>
> printk_nmi_flush_line(msg, strlen(msg));
> + len = 0;
> }
>
> if (!len)
> @@ -167,22 +204,7 @@ static void __printk_nmi_flush(struct irq_work *work)
>
> /* Make sure that data has been written up to the @len */
> smp_rmb();
> -
> - size = min(len, sizeof(s->buffer));
> - last_i = i;
> -
> - /* Print line by line. */
> - for (; i < size; i++) {
> - if (s->buffer[i] == '\n') {
> - printk_nmi_flush_seq_line(s, last_i, i);
> - last_i = i + 1;
> - }
> - }
> - /* Check if there was a partial line. */
> - if (last_i < size) {
> - printk_nmi_flush_seq_line(s, last_i, size - 1);
> - printk_nmi_flush_line("\n", strlen("\n"));
> - }
> + i += printk_nmi_flush_buffer(s->buffer + i, len - i);
>
> /*
> * Check that nothing has got added in the meantime and truncate