On Mon, Oct 29 2007, Vasily Averin wrote:
> i2o crashed when CONFIG_DEBUG_SG is enabled because i2o_block_request 
> structure
> includes array of scatterlists that should be initialised
> 
> Signed-off-by:        Vasily Averin <[EMAIL PROTECTED]>
> --- a/drivers/message/i2o/i2o_block.c
> +++ b/drivers/message/i2o/i2o_block.c
> @@ -1137,6 +1137,18 @@ static struct i2o_driver i2o_block_driver = {
>   *
>   *   Returns 0 on success or negative error code on failure.
>   */
> +
> +#ifdef CONFIG_DEBUG_SG
> +static void i2o_block_req_ctor(struct kmem_cache *cachep, void *objp) {
> +     sg_init_table(((struct i2o_block_request *)(objp))->sg_table,
> +                                              I2O_MAX_PHYS_SEGMENTS);
> +}
> +
> +#define I2O_BLK_CTOR &i2o_block_req_ctor
> +#else
> +#define I2O_BLK_CTOR NULL
> +#endif
> +
>  static int __init i2o_block_init(void)
>  {
>       int rc;
> @@ -1147,7 +1159,7 @@ static int __init i2o_block_init(void)
>       /* Allocate request mempool and slab */
>       size = sizeof(struct i2o_block_request);
>       i2o_blk_req_pool.slab = kmem_cache_create("i2o_block_req", size, 0,
> -                                               SLAB_HWCACHE_ALIGN, NULL);
> +                                       SLAB_HWCACHE_ALIGN, I2O_BLK_CTOR);
>       if (!i2o_blk_req_pool.slab) {
>               osm_err("can't init request slab\n");
>               rc = -ENOMEM;

This should already work, since Oct 24. Your i2o_block_request_alloc()
should look like this:

static inline struct i2o_block_request *i2o_block_request_alloc(void)
{
        struct i2o_block_request *ireq;

        ireq = mempool_alloc(i2o_blk_req_pool.pool, GFP_ATOMIC);
        if (!ireq)
                return ERR_PTR(-ENOMEM);

        INIT_LIST_HEAD(&ireq->queue);
        sg_init_table(ireq->sg_table, I2O_MAX_PHYS_SEGMENTS);

        return ireq;
}

Note that I also don't like your solution, no need to change this to be
a constructor setup and you definitely should not guard sg_init_table()
with CONFIG_DEBUG_SG. It needs to be done always.

-- 
Jens Axboe

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Reply via email to