[LTP] [PATCH] sctp/test_1_to_1_sockopt.c: set SO_SNDBUF above the defined minimum

[LTP] [PATCH] sctp/test_1_to_1_sockopt.c: set SO_SNDBUF above the defined minimum

[Alexey Kodanev]

SO_SNDBUF have the following minimum value:
TCP_SKB_MIN_TRUESIZE (2048 + sizeof(struct sk_buff))
SOCK_MIN_SNDBUF (TCP_SKB_MIN_TRUESIZE * 2)

The test set 'SO_SNDBUF' to 2048, then it gets the value back
with getsockopt(). But the value was defaulted in the kernel
to the SOCK_MIN_SNDBUF.

Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
---
 utils/sctp/func_tests/test_1_to_1_sockopt.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/utils/sctp/func_tests/test_1_to_1_sockopt.c b/utils/sctp/func_tests/test_1_to_1_sockopt.c
index 4cd84da..436b344 100644
--- a/utils/sctp/func_tests/test_1_to_1_sockopt.c
+++ b/utils/sctp/func_tests/test_1_to_1_sockopt.c
@@ -340,7 +340,7 @@ main(void)
 			 "got value differs Set Value=%d Get Value=%d",
 			 (2*rcvbuf_val_set), rcvbuf_val_get);
 
-	sndbuf_val_set = 2048;
+	sndbuf_val_set = 5000;
 	/* TEST17: Test case for setsockopt SO_SNDBUF */
 	error = setsockopt(sk2, SOL_SOCKET, SO_SNDBUF, &sndbuf_val_set, len);
 	if (error < 0)
-- 
1.7.1


------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] [PATCH] sctp/test_1_to_1_sockopt.c: set SO_SNDBUF above the defined minimum

[Cyril Hrubis]

Hi!
> SO_SNDBUF have the following minimum value:
> TCP_SKB_MIN_TRUESIZE (2048 + sizeof(struct sk_buff))
> SOCK_MIN_SNDBUF (TCP_SKB_MIN_TRUESIZE * 2)
> 
> The test set 'SO_SNDBUF' to 2048, then it gets the value back
> with getsockopt(). But the value was defaulted in the kernel
> to the SOCK_MIN_SNDBUF.

I guess that the setsockopt() does that without failing and then
the comparsion of the value you get with getsockopt() fails, right?

I still do not get why the code does:

if ((2 * sndbuf_val_set) != sndbuf_val_get)

Shouldn't the sndbuf_val_set == sndbuf_val_get once the value is changed
to 5000?

-- 
Cyril Hrubis
chrubis@suse.cz

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] [PATCH] sctp/test_1_to_1_sockopt.c: set SO_SNDBUF above the defined minimum

[Cyril Hrubis]

Hi!
> This is what is written in kernel source (net/core/sock.c):
> 
> sk->sk_sndbuf = max_t(u32, val * 2, SOCK_MIN_SNDBUF);
> ...
> /*
>   * We double it on the way in to account for
>   * "struct sk_buff" etc. overhead.   Applications
>   * assume that the SO_RCVBUF setting they make will
>   * allow that much actual data to be received on that
>   * socket.
>   *
>   * Applications are unaware that "struct sk_buff" and
>   * other overheads allocate from the receive buffer
>   * during socket buffer allocation.
>   *
>   * And after considering the possible alternatives,
>   * returning the value we actually used in getsockopt
>   * is the most desirable behavior.
>   */
> 
> sk->sk_rcvbuf = max_t(u32, val * 2, SOCK_MIN_RCVBUF);
> 
> I'm not sure why it is not just 'val', because it chooses the maximum 
> value in any case and we have defined minimum.

Ok.

The patch is fine then.

-- 
Cyril Hrubis
chrubis@suse.cz

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list