Re: [PATCH] dm: writecache: Use 2-factor allocator arguments

From: Kees Cook
Date: Mon Jun 18 2018 - 18:10:54 EST


On Mon, Jun 18, 2018 at 2:12 PM, Mikulas Patocka <mpatocka@xxxxxxxxxx> wrote:
>
>
> On Mon, 18 Jun 2018, Kees Cook wrote:
>
>> This adjusts the allocator calls to use the 2-factor argument style, as
>> already done treewide for better defense against allocator overflows.
>> Additionally adjusts style nit to avoid assignments in test expressions.
>>
>> Signed-off-by: Kees Cook <keescook@xxxxxxxxxxxx>
>> ---
>> drivers/md/dm-writecache.c | 16 ++++++++++------
>> 1 file changed, 10 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/md/dm-writecache.c b/drivers/md/dm-writecache.c
>> index 5961c7794ef3..7773f4c75701 100644
>> --- a/drivers/md/dm-writecache.c
>> +++ b/drivers/md/dm-writecache.c
>> @@ -259,7 +259,7 @@ static int persistent_memory_claim(struct dm_writecache *wc)
>> if (da != p) {
>> long i;
>> wc->memory_map = NULL;
>> - pages = kvmalloc(p * sizeof(struct page *), GFP_KERNEL);
>> + pages = kvmalloc_array(p, sizeof(struct page *), GFP_KERNEL);
>> if (!pages) {
>> r = -ENOMEM;
>> goto err2;
>> @@ -859,7 +859,8 @@ static int writecache_alloc_entries(struct dm_writecache *wc)
>>
>> if (wc->entries)
>> return 0;
>> - wc->entries = vmalloc(sizeof(struct wc_entry) * wc->n_blocks);
>> + wc->entries = vmalloc(array_size(sizeof(struct wc_entry),
>> + wc->n_blocks));
>> if (!wc->entries)
>> return -ENOMEM;
>> for (b = 0; b < wc->n_blocks; b++) {
>> @@ -1480,10 +1481,13 @@ static void __writecache_writeback_pmem(struct dm_writecache *wc, struct writeba
>> bio_set_dev(&wb->bio, wc->dev->bdev);
>> wb->bio.bi_iter.bi_sector = read_original_sector(wc, e);
>> wb->page_offset = PAGE_SIZE;
>> - if (max_pages <= WB_LIST_INLINE ||
>> - unlikely(!(wb->wc_list = kmalloc(max_pages * sizeof(struct wc_entry *),
>> - GFP_NOIO | __GFP_NORETRY |
>> - __GFP_NOMEMALLOC | __GFP_NOWARN)))) {
>> + if (max_pages > WB_LIST_INLINE)
>> + wb->wc_list = kmalloc_array(max_pages,
>> + sizeof(struct wc_entry *),
>> + GFP_NOIO | __GFP_NORETRY |
>> + __GFP_NOMEMALLOC |
>> + __GFP_NOWARN);
>> + if (max_pages <= WB_LIST_INLINE || !wb->wc_list) {
>
> The rest of patch is OK - but you shouldn't duplicate the comparison
> against WB_LIST_INLINE.

I couldn't find a better way to avoid an assignment in a test... open
to suggestions! :)

-Kees

--
Kees Cook
Pixel Security