On (08/07/15 15:37), Sergey Senozhatsky wrote:
[..]
> we now have errors like
> 'zram: Cannot initialise lzo compressing backend'
>
> and they will transform into
>
> 'block zram0: Cannot initialise lzo compressing backend'
>
> note the prefix 'zram:' became 'block zram0:'
but it doesn't come for free. where we had clean and nice
pr_err("Decompression failed!...
pr_info("Unable to allocate temp memory\n"...
etc...
now we have monsters
dev_err(disk_to_dev(zram->disk), "Decompression failed!...
dev_info(disk_to_dev(zram->disk), "Unable to allocate temp memory\n"...
etc.
other changes are very questionable... for example
pr_info("Added device: %s\n", zram->disk->disk_name);
becomes
dev_info(disk_to_dev(zram->disk), "Added device: %s\n",
zram->disk->disk_name);
why? there is no reason to do this!
and messages are converted in a bit random manner, shall I say. so now
we have a mix of dev_* and pr_* errors. to convert them all to dev_* (which
is not possible in all the cases) we would need to change some function
prototypes and start passing zram pointer, or disk...
example:
static struct zram_meta *zram_meta_alloc(int device_id, u64 disksize)
{
size_t num_pages;
char pool_name[8];
struct zram_meta *meta = kmalloc(sizeof(*meta), GFP_KERNEL);
if (!meta)
return NULL;
num_pages = disksize >> PAGE_SHIFT;
meta->table = vzalloc(num_pages * sizeof(*meta->table));
if (!meta->table) {
pr_err("Error allocating zram address table\n");
goto out_error;
}
snprintf(pool_name, sizeof(pool_name), "zram%d", device_id);
meta->mem_pool = zs_create_pool(pool_name, GFP_NOIO | __GFP_HIGHMEM);
if (!meta->mem_pool) {
pr_err("Error creating memory pool\n");
goto out_error;
}
...
so I see a little value. really. too much things to change.
-ss
--
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/
