* [PATCH] binfmt_elf: Fix core dump memory corruption
@ 2018-07-17 16:25 Yu-cheng Yu
2018-09-25 17:39 ` Yu-cheng Yu
0 siblings, 1 reply; 2+ messages in thread
From: Yu-cheng Yu @ 2018-07-17 16:25 UTC (permalink / raw)
To: linux-kernel, x86, H. Peter Anvin, Thomas Gleixner, Ingo Molnar, H.J. Lu
Cc: Yu-cheng Yu
In fill_note_info(), we kzalloc elf_thread_core_info.notes[] only
for (core_note_type != 0) regsets. However, in
fill_thread_core_info(), we still leave empty notes and go beyond
the allocated size. Fix it.
Signed-off-by: Yu-cheng Yu <yu-cheng.yu@intel.com>
---
fs/binfmt_elf.c | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 816cc921cf36..6f42e05d2833 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1723,7 +1723,8 @@ static int fill_thread_core_info(struct elf_thread_core_info *t,
const struct user_regset_view *view,
long signr, size_t *total)
{
- unsigned int i;
+ unsigned int i; /* index to regsets */
+ unsigned int j; /* index to notes */
unsigned int regset0_size = regset_size(t->task, &view->regsets[0]);
/*
@@ -1744,9 +1745,9 @@ static int fill_thread_core_info(struct elf_thread_core_info *t,
/*
* Each other regset might generate a note too. For each regset
- * that has no core_note_type or is inactive, we leave t->notes[i]
- * all zero and we'll know to skip writing it later.
+ * that has no core_note_type or is inactive, we skip it.
*/
+ j = 1;
for (i = 1; i < view->n; ++i) {
const struct user_regset *regset = &view->regsets[i];
do_thread_regset_writeback(t->task, regset);
@@ -1763,17 +1764,18 @@ static int fill_thread_core_info(struct elf_thread_core_info *t,
kfree(data);
else {
if (regset->core_note_type != NT_PRFPREG)
- fill_note(&t->notes[i], "LINUX",
+ fill_note(&t->notes[j], "LINUX",
regset->core_note_type,
size, data);
else {
SET_PR_FPVALID(&t->prstatus,
1, regset0_size);
- fill_note(&t->notes[i], "CORE",
+ fill_note(&t->notes[j], "CORE",
NT_PRFPREG, size, data);
}
- *total += notesize(&t->notes[i]);
+ *total += notesize(&t->notes[j]);
}
+ j++;
}
}
--
2.17.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] binfmt_elf: Fix core dump memory corruption
2018-07-17 16:25 [PATCH] binfmt_elf: Fix core dump memory corruption Yu-cheng Yu
@ 2018-09-25 17:39 ` Yu-cheng Yu
0 siblings, 0 replies; 2+ messages in thread
From: Yu-cheng Yu @ 2018-09-25 17:39 UTC (permalink / raw)
To: linux-kernel, x86, H. Peter Anvin, Thomas Gleixner, Ingo Molnar,
H.J. Lu, Oleg Nesterov
On Tue, 2018-07-17 at 09:25 -0700, Yu, Yu-cheng wrote:
> In fill_note_info(), we kzalloc elf_thread_core_info.notes[] only
> for (core_note_type != 0) regsets. However, in
> fill_thread_core_info(), we still leave empty notes and go beyond
> the allocated size. Fix it.
>
> Signed-off-by: Yu-cheng Yu <yu-cheng.yu@intel.com>
> ---
> fs/binfmt_elf.c | 14 ++++++++------
> 1 file changed, 8 insertions(+), 6 deletions(-)
>
> diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
> index 816cc921cf36..6f42e05d2833 100644
> --- a/fs/binfmt_elf.c
> +++ b/fs/binfmt_elf.c
> @@ -1723,7 +1723,8 @@ static int fill_thread_core_info(struct
> elf_thread_core_info *t,
> const struct user_regset_view *view,
> long signr, size_t *total)
> {
> - unsigned int i;
> + unsigned int i; /* index to regsets */
> + unsigned int j; /* index to notes */
> unsigned int regset0_size = regset_size(t->task, &view->regsets[0]);
>
> /*
> @@ -1744,9 +1745,9 @@ static int fill_thread_core_info(struct
> elf_thread_core_info *t,
>
> /*
> * Each other regset might generate a note too. For each regset
> - * that has no core_note_type or is inactive, we leave t->notes[i]
> - * all zero and we'll know to skip writing it later.
> + * that has no core_note_type or is inactive, we skip it.
> */
> + j = 1;
> for (i = 1; i < view->n; ++i) {
> const struct user_regset *regset = &view->regsets[i];
> do_thread_regset_writeback(t->task, regset);
> @@ -1763,17 +1764,18 @@ static int fill_thread_core_info(struct
> elf_thread_core_info *t,
> kfree(data);
> else {
> if (regset->core_note_type != NT_PRFPREG)
> - fill_note(&t->notes[i], "LINUX",
> + fill_note(&t->notes[j], "LINUX",
> regset->core_note_type,
> size, data);
> else {
> SET_PR_FPVALID(&t->prstatus,
> 1, regset0_size);
> - fill_note(&t->notes[i], "CORE",
> + fill_note(&t->notes[j], "CORE",
> NT_PRFPREG, size, data);
> }
> - *total += notesize(&t->notes[i]);
> + *total += notesize(&t->notes[j]);
> }
> + j++;
> }
> }
>
> --
> 2.17.1
>
Hi All,
Any comments on this?
Thanks,
Yu-cheng
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2018-09-25 17:45 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-07-17 16:25 [PATCH] binfmt_elf: Fix core dump memory corruption Yu-cheng Yu
2018-09-25 17:39 ` Yu-cheng Yu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).