Fix KASAN: slab-out-of-bounds Read in sk_psock_get
Subsystems (2)
cover Fix KASAN: slab-out-of-bounds Read in sk_psock_get
Using size of sk_psock as the size for kcm_psock_cache size no longer reproduces the issue. There might be a better way to solve this issue though so I would like to ask for feedback.
Khalid Masum (1):
net: kcm: Use sk_psock size for kcm_psock_cache
net/kcm/kcmsock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
Signature · 1 line
2.36.1
RFC PATCH 1/1 net: kcm: Use sk_psock size for kcm_psock_cache
struct sock has a member sk_user_data, which got its memory allocated in kcm_attach by kcm_psock_cache with the size of kcm_psock. Which is not enough when the member is used as sk_psock causing out of bound read.
Use sk_psock size to allocate memory instead for sk_user_data.
--- net/kcm/kcmsock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
net/kcm/kcmsock.c +2 -1
@@ -23,6 +23,7 @@#include <linux/workqueue.h> #include <linux/syscalls.h> #include <linux/sched/signal.h> +#include <linux/skmsg.h> #include <net/kcm.h> #include <net/netns/generic.h>
@@ -2041,7 +2042,7 @@ static int __init kcm_init(void) goto fail; kcm_psockp =kmem_cache_create ("kcm_psock_cache" , -sizeof (struct kcm_psock ),0 , +sizeof (struct sk_psock ),0 ,SLAB_HWCACHE_ALIGN ,NULL );if (!kcm_psockp) goto fail;
Signature · 1 line
2.36.1
Greg KH
maintainerRe: [RFC PATCH 1/1] net: kcm: Use sk_psock size for kcm_psock_cache
Quoted 5 lines · Khalid Masum
struct sockhas a membersk_user_data, which got its memory allocated inkcm_attachbykcm_psock_cachewith the size ofkcm_psock. Which is not enough when the member is used assk_psockcausing out of bound read.
Quoted 7 lines
Use
sk_psocksize to allocate memory instead forsk_user_data.Reported-by [email protected]Signed-off-by Khalid Masum <[email protected]>--- net/kcm/kcmsock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
What commit id does this fix?
thanks,
greg k-h
Jakub Kicinski
maintainerRe: [RFC PATCH 1/1] net: kcm: Use sk_psock size for kcm_psock_cache
In reply to Khalid MasumQuoted 9 lines · On Wed, 13 Jul 2022 12:32:04 +0600 Khalid Masum wrote:
struct sockhas a membersk_user_data, which got its memory allocated inkcm_attachbykcm_psock_cachewith the size ofkcm_psock. Which is not enough when the member is used assk_psockcausing out of bound read.Use
sk_psocksize to allocate memory instead forsk_user_data.Reported-by [email protected]Signed-off-by Khalid Masum <[email protected]>
Odd CC list, it looks like a netdev problem but netdev@ is not copied.
We had an other parson looking at this on netdev. Is this happening when legit kernel socket gets mixed with BPF sockmap?
Does this fix work?
Re: [RFC PATCH 1/1] net: kcm: Use sk_psock size for kcm_psock_cache
Quoted 7 lines · Jakub Kicinski
Quoted 5 lines · On Wed, 13 Jul 2022 12:32:04 +0600 Khalid Masum wrote:
struct sockhas a membersk_user_data, which got its memory allocated inkcm_attachbykcm_psock_cachewith the size ofkcm_psock. Which is not enough when the member is used assk_psockcausing out of bound read.
Quoted 6 lines
Quoted 4 lines
Use
sk_psocksize to allocate memory instead forsk_user_data.Reported-by [email protected]Signed-off-by Khalid Masum <[email protected]>Odd CC list, it looks like a netdev problem but netdev@ is not copied.
I am sorry for the inconvenience. I resent the patch with netdev in CC here:
https://lore.kernel.org/netdev/[email protected]/
Quoted 4 lines
We had an other parson looking at this on netdev. Is this happening when legit kernel socket gets mixed with BPF sockmap?
Does this fix work?
I will take a look at it.
Quoted 1 line
Thanks,
Khalid Masum