There is an old patch that probably was missed. Can I commit it while putting the function call after the variable declarations ?
-----Original Message----- From: Smith, Stan [mailto:[email protected]] Sent: Monday, September 20, 2010 7:16 PM To: Alex Naslednikov; [email protected] Subject: RE: [ofw] [Patch][IPoIB_NDIS6_CM] Fix at __ipoib_port_down() -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Alex Naslednikov Sent: Sunday, September 19, 2010 6:07 AM To: [email protected] Subject: [ofw] [Patch][IPoIB_NDIS6_CM] Fix at __ipoib_port_down() >> We still have a difference between MLNX and OFW trunks. >> I continue to resend patches that were not applied for unknown reasons. This is a new (never seen before) patch. Why do you place the port_down() call before variable declarations? This is contrary to the ipoib coding style? -- Fixing the situation when adapter remains to be in connected state even after cable disconnect. That was because __ipoib_port_down called __ipoib_reset_all that disables and then enables the status Signed-off by: Alexander Naslednikov Index: ipoib_adapter.cpp =================================================================== --- ipoib_adapter.cpp (revision 5258) +++ ipoib_adapter.cpp (revision 5259) @@ -585,6 +585,7 @@ static inline void __ipoib_get_down( IN ipoib_adapter_t *p_adapter) { + ipoib_port_down( p_adapter->p_port ); NDIS_LINK_STATE link_state; NDIS_STATUS_INDICATION status_indication; @@ -608,7 +609,7 @@ IPOIB_PRINT( TRACE_LEVEL_INFORMATION, IPOIB_DBG_INIT, ("Link DOWN!\n") ); - ipoib_port_down( p_adapter->p_port ); + } static ib_api_status_t _______________________________________________ ofw mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ofw _______________________________________________ ofw mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ofw
