This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[PATCH] Increase nptl test case buffer size so we really block on current Linux kernels.


So the minimum SO_SNDBUF size has increased in recent Linux kernels,
and the value is dependent upon the machine word size.

Essentially the current value is something like:

	2 * (2048 + roundup(sizeof(struct sk_buff), CACHE_LINE_SIZE))

and the aim is to make sure that two fully sized packets can always
be sent by TCP.  This minimum is enforced for all protocols, not just
TCP, because the enforcement occurs in the generic socket layer.

Any objections?  I get a clean sparc64 testsuite now with this change.

	* tst-cancel4.c (WRITE_BUFFER_SIZE): Increase to 16384.

diff --git a/nptl/ChangeLog b/nptl/ChangeLog
index fc2ef41..0e4747e 100644
--- a/nptl/ChangeLog
+++ b/nptl/ChangeLog
@@ -1,3 +1,7 @@
+2013-07-22  David S. Miller  <davem@davemloft.net>
+
+	* tst-cancel4.c (WRITE_BUFFER_SIZE): Increase to 16384.
+
 2013-07-19  Dominik Vogt  <vogt@de.ibm.com>
 
 	* pthread_mutex_lock.c: Fix whitespace.
diff --git a/nptl/tst-cancel4.c b/nptl/tst-cancel4.c
index 9ffd5d1..04f9282 100644
--- a/nptl/tst-cancel4.c
+++ b/nptl/tst-cancel4.c
@@ -83,7 +83,7 @@ static pthread_barrier_t b2;
 # define IPC_ADDVAL 0
 #endif
 
-#define WRITE_BUFFER_SIZE 4096
+#define WRITE_BUFFER_SIZE 16384
 
 /* Cleanup handling test.  */
 static int cl_called;


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]