linuxkpi: fix iteration in __sg_alloc_table_from_pages

Commit 3f686532c9 tried to fix an issue with not properly starting
at the first page in the sg list to prevent a panic. This worked but
with the side effect of incrementing "s" during the final iteration
causing it to be NULL since the list had ended.
In cases non-DEBUG kernels this causes a panic with drm-5.15, since
"s" is NULL when we later pass it to sg_mark_end().
This change decouples the iteration sg from the return value so that
it is never incremented past the final page in the chain.

MFC after:	3 days
Reviewed by:	manu
Differential Revision: https://reviews.freebsd.org/D41574
This commit is contained in:
Austin Shafer 2023-09-06 14:08:05 +00:00 committed by Bjoern A. Zeeb
parent 22cf89c938
commit 09b0401e91

View file

@ -343,7 +343,7 @@ __sg_alloc_table_from_pages(struct sg_table *sgt,
{
unsigned int i, segs, cur, len;
int rc;
struct scatterlist *s;
struct scatterlist *s, *sg_iter;
#if defined(LINUXKPI_VERSION) && LINUXKPI_VERSION >= 51300
if (prv != NULL) {
@ -377,10 +377,18 @@ __sg_alloc_table_from_pages(struct sg_table *sgt,
#endif
cur = 0;
for (i = 0, s = sgt->sgl; i < sgt->orig_nents; i++) {
for_each_sg(sgt->sgl, sg_iter, sgt->orig_nents, i) {
unsigned long seg_size;
unsigned int j;
/*
* We need to make sure that when we exit this loop "s" has the
* last sg in the chain so we can call sg_mark_end() on it.
* Only set this inside the loop since sg_iter will be iterated
* until it is NULL.
*/
s = sg_iter;
len = 0;
for (j = cur + 1; j < count; ++j) {
len += PAGE_SIZE;
@ -394,8 +402,6 @@ __sg_alloc_table_from_pages(struct sg_table *sgt,
size -= seg_size;
off = 0;
cur = j;
s = sg_next(s);
}
KASSERT(s != NULL, ("s is NULL after loop in __sg_alloc_table_from_pages()"));