Thread (18 messages) 18 messages, 4 authors, 2019-03-20

Re: [PATCH v12 08/13] net, arm64: untag user pointers in tcp_zerocopy_receive

From: Eric Dumazet <edumazet@google.com>
Date: 2019-03-18 17:35:31
Also in: bpf, linux-arch, linux-arm-kernel, linux-doc, linux-kselftest, linux-mm, lkml
Subsystem: networking [general], networking [tcp], the rest · Maintainers: "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Neal Cardwell, Linus Torvalds

On Mon, Mar 18, 2019 at 10:18 AM Andrey Konovalov [off-list ref] wrote:
quoted hunk ↗ jump to hunk
This patch is a part of a series that extends arm64 kernel ABI to allow to
pass tagged user pointers (with the top byte set to something else other
than 0x00) as syscall arguments.

tcp_zerocopy_receive() uses provided user pointers for vma lookups, which
can only by done with untagged pointers.

Untag user pointers in this function.

Signed-off-by: Andrey Konovalov <redacted>
---
 net/ipv4/tcp.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 6baa6dc1b13b..e76beb5ff1ff 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1749,7 +1749,7 @@ EXPORT_SYMBOL(tcp_mmap);
 static int tcp_zerocopy_receive(struct sock *sk,
                                struct tcp_zerocopy_receive *zc)
 {
-       unsigned long address = (unsigned long)zc->address;
+       unsigned long address;
        const skb_frag_t *frags = NULL;
        u32 length = 0, seq, offset;
        struct vm_area_struct *vma;
@@ -1758,7 +1758,12 @@ static int tcp_zerocopy_receive(struct sock *sk,
        int inq;
        int ret;

-       if (address & (PAGE_SIZE - 1) || address != zc->address)
+       address = (unsigned long)untagged_addr(zc->address);
+
+       /* The second test in this if detects if the u64->unsigned long
+        * conversion had any truncated bits.
+        */
+       if (address & (PAGE_SIZE - 1) || address != untagged_addr(zc->address))
                return -EINVAL;

        if (sk->sk_state == TCP_LISTEN)

This is quite ugly, the comment does not really help nor belong to this patch.

What about using  untagged_addr()  only once ?
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 6baa6dc1b13b0b94b1da238668b93e167cf444fe..855a1f68c1ea9b0d07a92bd7f5e7c24840a99d3d
100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1761,6 +1761,8 @@ static int tcp_zerocopy_receive(struct sock *sk,
        if (address & (PAGE_SIZE - 1) || address != zc->address)
                return -EINVAL;

+       address = untagged_addr(address);
+
        if (sk->sk_state == TCP_LISTEN)
                return -ENOTCONN;
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help