idr: Fix handling of IDs above INT_MAX

Khalid reported that the kernel selftests are currently failing:

selftests: test_bpf.sh
========================================
test_bpf: [FAIL]
not ok 1..8 selftests:  test_bpf.sh [FAIL]

He bisected it to 6ce711f275 ("idr: Make
1-based IDRs more efficient").

The root cause is doing a signed comparison in idr_alloc_u32() instead
of an unsigned comparison.  I went looking for any similar problems and
found a couple (which would each result in the failure to warn in two
situations that aren't supposed to happen).

I knocked up a few test-cases to prove that I was right and added them
to the test-suite.

Reported-by: Khalid Aziz <khalid.aziz@oracle.com>
Tested-by: Khalid Aziz <khalid.aziz@oracle.com>
Signed-off-by: Matthew Wilcox <mawilcox@microsoft.com>
This commit is contained in:
Matthew Wilcox
2018-02-26 14:39:30 -05:00
parent 3d4d5d6186
commit 4b0ad07653
2 changed files with 59 additions and 6 deletions

View File

@@ -36,8 +36,8 @@ int idr_alloc_u32(struct idr *idr, void *ptr, u32 *nextid,
{
struct radix_tree_iter iter;
void __rcu **slot;
int base = idr->idr_base;
int id = *nextid;
unsigned int base = idr->idr_base;
unsigned int id = *nextid;
if (WARN_ON_ONCE(radix_tree_is_internal_node(ptr)))
return -EINVAL;
@@ -204,10 +204,11 @@ int idr_for_each(const struct idr *idr,
radix_tree_for_each_slot(slot, &idr->idr_rt, &iter, 0) {
int ret;
unsigned long id = iter.index + base;
if (WARN_ON_ONCE(iter.index > INT_MAX))
if (WARN_ON_ONCE(id > INT_MAX))
break;
ret = fn(iter.index + base, rcu_dereference_raw(*slot), data);
ret = fn(id, rcu_dereference_raw(*slot), data);
if (ret)
return ret;
}
@@ -230,8 +231,8 @@ void *idr_get_next(struct idr *idr, int *nextid)
{
struct radix_tree_iter iter;
void __rcu **slot;
int base = idr->idr_base;
int id = *nextid;
unsigned long base = idr->idr_base;
unsigned long id = *nextid;
id = (id < base) ? 0 : id - base;
slot = radix_tree_iter_find(&idr->idr_rt, &iter, id);