bna: avoid writing uninitialized data into hw registers
commit102a8a1634496d53fcf0b5fae92f7729a1d17f76
authorArnd Bergmann <arnd@arndb.de>
Thu, 23 Mar 2017 16:07:26 +0000 (23 17:07 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Mon, 25 Dec 2017 13:23:40 +0000 (25 14:23 +0100)
tree60fef1a7850300468cec9a131ee59c553e5fd28e
parent51533c4bf1442afd9b54747cd3d47d698940d624
bna: avoid writing uninitialized data into hw registers

[ Upstream commit a5af83925363eb85d467933e3d6ec5a87001eb7c ]

The latest gcc-7 snapshot warns about bfa_ioc_send_enable/bfa_ioc_send_disable
writing undefined values into the hardware registers:

drivers/net/ethernet/brocade/bna/bfa_ioc.c: In function 'bfa_iocpf_sm_disabling_entry':
arch/arm/include/asm/io.h:109:22: error: '*((void *)&disable_req+4)' is used uninitialized in this function [-Werror=uninitialized]
arch/arm/include/asm/io.h:109:22: error: '*((void *)&disable_req+8)' is used uninitialized in this function [-Werror=uninitialized]

The two functions look like they should do the same thing, but only one
of them initializes the time stamp and clscode field. The fact that we
only get a warning for one of the two functions seems to be arbitrary,
based on the inlining decisions in the compiler.

To address this, I'm making both functions do the same thing:

- set the clscode from the ioc structure in both
- set the time stamp from ktime_get_real_seconds (which also
  avoids the signed-integer overflow in 2038 and extends the
  well-defined behavior until 2106).
- zero-fill the reserved field

Fixes: 8b230ed8ec96 ("bna: Brocade 10Gb Ethernet device driver")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/net/ethernet/brocade/bna/bfa_ioc.c