2022-11-21 17:32 (UTC-0500), ok...@kernel.org:
> From: Sinan Kaya <ok...@kernel.org>
>
> In malloc_elem_find_max_iova_contig result of call to rte_mem_virt2memseg
> is dereferenced here and may be null.
>
> Signed-off-by: Sinan Kaya <ok...@kernel.org>
> ---
> lib/eal/common/malloc_elem.c | 11 ++++++++---
> lib/eal/common/malloc_heap.c | 2 +-
> 2 files changed, 9 insertions(+), 4 deletions(-)
>
> diff --git a/lib/eal/common/malloc_elem.c b/lib/eal/common/malloc_elem.c
> index 83f05497cc..8f49812846 100644
> --- a/lib/eal/common/malloc_elem.c
> +++ b/lib/eal/common/malloc_elem.c
> @@ -63,6 +63,8 @@ malloc_elem_find_max_iova_contig(struct malloc_elem *elem,
> size_t align)
>
> cur_page = RTE_PTR_ALIGN_FLOOR(contig_seg_start, page_sz);
> ms = rte_mem_virt2memseg(cur_page, elem->msl);
> + if (ms == NULL)
> + return 0;
>
> /* do first iteration outside the loop */
> page_end = RTE_PTR_ADD(cur_page, page_sz);
> @@ -91,9 +93,12 @@ malloc_elem_find_max_iova_contig(struct malloc_elem *elem,
> size_t align)
> * we're not blowing past data end.
> */
> ms = rte_mem_virt2memseg(contig_seg_start, elem->msl);
> - cur_page = ms->addr;
> - /* don't trigger another recalculation */
> - expected_iova = ms->iova;
> + if (ms != NULL) {
> + cur_page = ms->addr;
> +
> + /* don't trigger another recalculation */
> + expected_iova = ms->iova;
> + }
> continue;
> }
> /* cur_seg_end ends on a page boundary or on data end. if we're
> diff --git a/lib/eal/common/malloc_heap.c b/lib/eal/common/malloc_heap.c
> index 3f41430e42..88270ce4d2 100644
> --- a/lib/eal/common/malloc_heap.c
> +++ b/lib/eal/common/malloc_heap.c
> @@ -930,7 +930,7 @@ malloc_heap_free(struct malloc_elem *elem)
> const struct rte_memseg *tmp =
> rte_mem_virt2memseg(aligned_start, msl);
>
> - if (tmp->flags & RTE_MEMSEG_FLAG_DO_NOT_FREE) {
> + if ((tmp != NULL) && (tmp->flags &
> RTE_MEMSEG_FLAG_DO_NOT_FREE)) {
> /* this is an unfreeable segment, so move start */
> aligned_start = RTE_PTR_ADD(tmp->addr, tmp->len);
> }
In these three places "ms" or "tmp" are from the MSL by construction.
I think RTE_ASSERT() would be sufficient.
