On Tue, 2008-05-27 at 07:38 -0700, Roland Dreier wrote:
'bnx2id' is the user component in this solution. bnx2id daemon uses
socket calls to bind tcp ports in high range and hands them over to
driver. This is how iscsi driver tries to solve tcp port collision
issue. User daemon
On Tue, 2008-05-27 at 12:52 -0700, David Miller wrote:
From: Roland Dreier [EMAIL PROTECTED]
Date: Tue, 27 May 2008 07:38:19 -0700
So you are creating sockets just to reserve TCP ports to avoid host
stack clashes with your offload engine? Wasn't this approach strongly
rejected (in the
On Fri, 2008-05-23 at 13:23 -0700, Roland Dreier wrote:
Hi Michael, I was reading over the driver to try and figure out how you
handle allocating source ports for the offloaded TCP connections you
make so that they don't collide with the main network stack. It looks
like you have:
+/**
On Thu, 2008-05-22 at 11:15 -0400, Konrad Rzeszutek wrote:
A cursory glance..
.. snip..
+struct bnx2i_cleanup_request {
+#if defined(__BIG_ENDIAN)
+ u8 op_code;
+ u8 reserved1;
+ u16 reserved0;
+#elif defined(__LITTLE_ENDIAN)
+ u16 reserved0;
+ u8 reserved1;
+ u8
On Thu, 2008-05-22 at 17:15 -0400, Christoph Hellwig wrote:
+struct bnx2i_async_msg {
+#if defined(__BIG_ENDIAN)
+ u8 op_code;
+ u8 reserved1;
+ u16 reserved0;
+#elif defined(__LITTLE_ENDIAN)
+ u16 reserved0;
+ u8 reserved1;
+ u8 op_code;
+#endif
+ u32
Konrad Rzeszutek wrote:
+ * bnx2i_iscsi_license_error - displays iscsi license related error message
Doesn't look very license related. Just says 'not supported'.
Could be that some hardware strap pin is being used as a dongle. This
doesn't seem entirely in the spirit of GPL, and would be
+struct bnx2i_async_msg {
+#if defined(__BIG_ENDIAN)
+ u8 op_code;
+ u8 reserved1;
+ u16 reserved0;
+#elif defined(__LITTLE_ENDIAN)
+ u16 reserved0;
+ u8 reserved1;
+ u8 op_code;
+#endif
+ u32 reserved2;
+ u32 exp_cmd_sn;
+ u32 max_cmd_sn;
+