* [PATCH] kcpuid: Fix potential dereferencing of null pointers
@ 2024-09-26 22:35 Remington Brasga
2024-09-29 10:52 ` Thomas Gleixner
0 siblings, 1 reply; 2+ messages in thread
From: Remington Brasga @ 2024-09-26 22:35 UTC (permalink / raw)
To: Christian Heusel, Shuah Khan, Thomas Gleixner, Ahmed S . Darwish
Cc: linux-kernel, linux-kernel-mentees, Remington Brasga
Clang reported "clang-analyzer-core.NullDereference" on the `leaf` and `range`
variables in kcpuid.c, which makes sense if malloc/realloc fail.
These changes will ensure that the variables are not dereferenced while null.
Signed-off-by: Remington Brasga <rbrasga@uci.edu>
---
tools/arch/x86/kcpuid/kcpuid.c | 23 +++++++++++++++++++----
1 file changed, 19 insertions(+), 4 deletions(-)
diff --git a/tools/arch/x86/kcpuid/kcpuid.c b/tools/arch/x86/kcpuid/kcpuid.c
index 1b25c0a95d3f..c05226d105b6 100644
--- a/tools/arch/x86/kcpuid/kcpuid.c
+++ b/tools/arch/x86/kcpuid/kcpuid.c
@@ -144,19 +144,29 @@ static bool cpuid_store(struct cpuid_range *range, u32 f, int subleaf,
if (!func->leafs) {
func->leafs = malloc(sizeof(struct subleaf));
- if (!func->leafs)
+ if (!func->leafs) {
perror("malloc func leaf");
+ return false; // On malloc failure
+ }
func->nr = 1;
} else {
s = func->nr;
func->leafs = realloc(func->leafs, (s + 1) * sizeof(*leaf));
- if (!func->leafs)
+ if (!func->leafs) {
perror("realloc f->leafs");
+ return false; // On realloc failure
+ }
func->nr++;
}
+ // Check for valid index
+ if (s >= func->nr) {
+ fprintf(stderr, "Error: Invalid index for leaf\n");
+ return false;
+ }
+
leaf = &func->leafs[s];
leaf->index = f;
@@ -210,8 +220,10 @@ struct cpuid_range *setup_cpuid_range(u32 input_eax)
idx_func = (max_func & 0xffff) + 1;
range = malloc(sizeof(struct cpuid_range));
- if (!range)
+ if (!range) {
perror("malloc range");
+ return NULL; // On malloc failure
+ }
if (input_eax & 0x80000000)
range->is_ext = true;
@@ -219,8 +231,11 @@ struct cpuid_range *setup_cpuid_range(u32 input_eax)
range->is_ext = false;
range->funcs = malloc(sizeof(struct cpuid_func) * idx_func);
- if (!range->funcs)
+ if (!range->funcs) {
perror("malloc range->funcs");
+ free(range);
+ return NULL; // On malloc failure
+ }
range->nr = idx_func;
memset(range->funcs, 0, sizeof(struct cpuid_func) * idx_func);
--
2.34.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] kcpuid: Fix potential dereferencing of null pointers
2024-09-26 22:35 [PATCH] kcpuid: Fix potential dereferencing of null pointers Remington Brasga
@ 2024-09-29 10:52 ` Thomas Gleixner
0 siblings, 0 replies; 2+ messages in thread
From: Thomas Gleixner @ 2024-09-29 10:52 UTC (permalink / raw)
To: Remington Brasga, Christian Heusel, Shuah Khan, Ahmed S . Darwish
Cc: linux-kernel, linux-kernel-mentees, Remington Brasga
On Thu, Sep 26 2024 at 22:35, Remington Brasga wrote:
> if (!func->leafs) {
> func->leafs = malloc(sizeof(struct subleaf));
> - if (!func->leafs)
> + if (!func->leafs) {
> perror("malloc func leaf");
> + return false; // On malloc failure
Please get rid of these horrible and pointless tail comments.
Returning false here does not make sense. This simply should terminate
the program.
> + }
>
> func->nr = 1;
> } else {
> s = func->nr;
> func->leafs = realloc(func->leafs, (s + 1) * sizeof(*leaf));
> - if (!func->leafs)
> + if (!func->leafs) {
> perror("realloc f->leafs");
> + return false; // On realloc failure
> + }
>
> func->nr++;
> }
>
> + // Check for valid index
> + if (s >= func->nr) {
What's the point of this? s is guaranteed to be < func->nr, no?
Thanks,
tglx
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2024-09-29 11:00 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-09-26 22:35 [PATCH] kcpuid: Fix potential dereferencing of null pointers Remington Brasga
2024-09-29 10:52 ` Thomas Gleixner
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox