From: "Mintz, Yuval" <yuval.mi...@cavium.com> Date: Mon, 22 May 2017 17:44:12 +0000
>> >> Please. >> >> >> >> This is exactly the problem when such a terse commit message, or lack of >> >> any >> >> message at all as is the case here, occurs. >> >> >> >> I have no idea what removing this method assignment has to do with fixing >> >> sparse warnings. >> > >> > It's currently defined twice in the qede_dcbnl_ops struct - > >> >Ok, so what's important is that you don't just post a patch with no >> commit message, just saying "sparse warning fixes." > >> At a minimum, if you're too lazy to explain what exactly each thing >> is fixing, provide the sparse warnings you are eliminating in the >> commit message. > > Well, this time it was intentional lazines; Thought it would have been > wasted reading. But sure, here it is: > > drivers/net/ethernet/qlogic/qede/qede_fp.c:349:34: warning: invalid > assignment: |= > drivers/net/ethernet/qlogic/qede/qede_fp.c:349:34: left side has type > restricted __le16 > drivers/net/ethernet/qlogic/qede/qede_fp.c:349:34: right side has type int > drivers/net/ethernet/qlogic/qede/qede_fp.c:1516:50: warning: invalid > assignment: |= > drivers/net/ethernet/qlogic/qede/qede_fp.c:1516:50: left side has type > restricted __le16 > drivers/net/ethernet/qlogic/qede/qede_fp.c:1516:50: right side has type int > drivers/net/ethernet/qlogic/qede/qede_fp.c:1525:50: warning: invalid > assignment: ^= > drivers/net/ethernet/qlogic/qede/qede_fp.c:1525:50: left side has type > restricted __le16 > drivers/net/ethernet/qlogic/qede/qede_fp.c:1525:50: right side has type int > drivers/net/ethernet/qlogic/qede/qede_fp.c:1590:42: warning: invalid > assignment: |= > drivers/net/ethernet/qlogic/qede/qede_fp.c:1590:42: left side has type > restricted __le16 > drivers/net/ethernet/qlogic/qede/qede_fp.c:1590:42: right side has type > unsigned int > drivers/net/ethernet/qlogic/qede/qede_ethtool.c:1315:34: warning: invalid > assignment: |= > drivers/net/ethernet/qlogic/qede/qede_ethtool.c:1315:34: left side has > type restricted __le16 > drivers/net/ethernet/qlogic/qede/qede_ethtool.c:1315:34: right side has > type int > drivers/net/ethernet/qlogic/qede/qede_ethtool.c:1330:13: warning: incorrect > type in assignment (different base types) > drivers/net/ethernet/qlogic/qede/qede_ethtool.c:1330:13: expected int > [signed] [assigned] val > drivers/net/ethernet/qlogic/qede/qede_ethtool.c:1330:13: got restricted > __le16 [usertype] <noident> > drivers/net/ethernet/qlogic/qede/qede_ethtool.c:1331:33: warning: incorrect > type in assignment (different base types) > drivers/net/ethernet/qlogic/qede/qede_ethtool.c:1331:33: expected > restricted __le16 [usertype] bd_prod > drivers/net/ethernet/qlogic/qede/qede_ethtool.c:1331:33: got int [signed] > [assigned] val > drivers/net/ethernet/qlogic/qede/qede_roce.c:224:29: warning: symbol > 'qede_roce_get_free_event_node' was not declared. Should it be static? > > Do you want us to re-spin the series with the updated commit message? Where in that log above does it say that the method is defined twice in qede_dcbnl_ops? I hate to say this, but this is why I get frustrated with such across the board cleanups with empty commit messages.
