Thread (29 messages) flat view 29 messages, 7 authors, 2020-03-23

Re: [PATCH RFC v1 09/15] xdp: clear grow memory in bpf_xdp_adjust_tail()

From: Toke Høiland-Jørgensen <hidden>
Date: 2020-03-18 09:15:45
Also in: bpf

Jesper Dangaard Brouer [off-list ref] writes:
quoted hunk ↗ jump to hunk
To reviewers: Need some opinions if this is needed?

(TODO: Squash patch)
---
 net/core/filter.c |    6 ++++++
 1 file changed, 6 insertions(+)
diff --git a/net/core/filter.c b/net/core/filter.c
index 0ceddee0c678..669f29992177 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3432,6 +3432,12 @@ BPF_CALL_2(bpf_xdp_adjust_tail, struct xdp_buff *, xdp, int, offset)
 	if (unlikely(data_end < xdp->data + ETH_HLEN))
 		return -EINVAL;
 
+	// XXX: To reviewers: How paranoid are we? Do we really need to
+	/* clear memory area on grow, as in-theory can contain uninit kmem */
+	if (offset > 0) {
+		memset(xdp->data_end, 0, offset);
+	}
This memory will usually be recycled through page_pool or equivalent,
right? So couldn't we clear the pages when they are first allocated?
That way, the only data that would be left there would be packet data
from previous packets...

-Toke
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help