We received two reports of BUG_ON in bnad_txcmpl_process() where
hw_consumer_index appeared to be ahead of producer_index. Out of order
write/read of these variables could explain these reports.
bnad_start_xmit(), as a producer of tx descriptors, has a few memory
barriers sprinkled around writes to producer_index and the device's
doorbell but they're not paired with anything in bnad_txcmpl_process(), a
consumer.
Since we are synchronizing with a device, we must use mandatory barriers,
not smp_*. Also, I didn't see the purpose of the last smp_mb() in
bnad_start_xmit().
Signed-off-by: Benjamin Poirier <bpoirier@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 6e7333d "net: add rx_nohandler stat counter" added the new entry
rx_nohandler into struct rtnl_link_stats64. Unfortunately the bna
driver foolishly depends on the structure. It uses part of it for
ethtool statistics and it's not bad but the driver assumes its size
is constant as it defines string for each existing entry. The problem
occurs when the structure is extended because you need to modify bna
driver as well. If not any attempt to retrieve ethtool statistics results
in crash in bnad_get_strings().
The patch changes BNAD_ETHTOOL_STATS_NUM so it counts real number of
strings in the array and also removes rtnl_link_stats64 entries that
are not used in output and are always zero.
Fixes: 6e7333d "net: add rx_nohandler stat counter"
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit ba5ca784 "bna: check for dma mapping errors" added besides other
things a statistic that counts number of DMA buffer mapping failures
per each Rx queue. This counter is not included in ethtool stats output.
Fixes: ba5ca784 "bna: check for dma mapping errors"
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove global bnad_list_mutex as it is not used anymore. This makes
bnad_add_to_list() and bnad_remove_from_list() empty so remove them too.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change type of bna_id to atomic_t. The bnad_list_mutex is used to prevent
a race when bna_id is incremented. After the change the mutex can be
removed in the next step.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove global variable bnad_list and bnad->list_entry that are used
as list of bna driver instances. It is not necessary and useless.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
add and val are read with
sscanf(kern_buf, "%x:%x", &addr, &val);
and used as arguments for bna_reg_offset_check and writel
so they have to be unsigned.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
addr and len are read with
sscanf(kern_buf, "%x:%x", &addr, &len);
and used as arguments for
bna_reg_offset_check.
So they have to be unsigned.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use list_move_tail() to move MAC address entry from list of pending
to list of active entries. Simple list_add_tail() leaves the entry
also in the first list, this leads to list corruption.
Cc: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Acked-by: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The multi-buffer Rx mode implemented in the past introduced
a regression that causes a data corruption for received VLAN
traffic when VLAN tag stripping is enabled. This mode is supported
only be newer chipsets (1860) and is enabled when MTU > 4096.
When this mode is enabled Rx queue contains buffers with fixed size
2048 bytes. Any incoming packet larger than 2048 is divided into
multiple buffers that are attached as skb frags in polling routine.
The driver assumes that all buffers associated with a packet except
the last one is fully used (e.g. packet with size 5000 are divided
into 3 buffers 2048 + 2048 + 904 bytes) and ignores true size reported
in completions. This assumption is usually true but not when VLAN
packet is received and VLAN tag stripping is enabled. In this case
the first buffer is 2044 bytes long but as the driver always assumes
2048 bytes then 4 extra random bytes are included between the first
and the second frag. Additionally the driver sets checksum as correct
so the packet is properly processed by the core.
The driver needs to check the size of used space in each Rx buffer
reported by FW and not blindly use the fixed value.
Cc: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Reviewed-by: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Several functions can return negative value in case of error,
so their return type should be fixed as well as type of variables
to which this value is assigned.
The problem has been detected using proposed semantic patch
scripts/coccinelle/tests/assign_signed_to_unsigned.cocci [1].
[1]: http://permalink.gmane.org/gmane.linux.kernel/2046107
Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Check for DMA mapping errors, recover from them and register them in
ethtool stats like other errors.
Cc: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Acked-by: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The commit "e29aa33 bna: Enable Multi Buffer RX" moved packets counter
increment from the beginning of the NAPI processing loop after the check
for erroneous packets so they are never accounted. This counter is used
to inform firmware about number of processed completions (packets).
As these packets are never acked the firmware fires IRQs for them again
and again.
Fixes: e29aa33 ("bna: Enable Multi Buffer RX")
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Acked-by: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
BIT value is already unsigned so casting is not necessary.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
...and remove some of them. It is not necessary to log when .probe() and
.remove() are called or when TxQ is started or stopped. Also log level
of some of them was changed to more appropriate one (link up/down,
firmware loading failure.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Timeout functions are defined with 'void *' ptr argument. They should
be defined directly with 'struct bfa_ioc *' type to avoid type conversions.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove macros for manipulation with struct list_head and replace them
with standard ones.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pointer cmpl used to iterate through completion entries is updated at
the beginning of while loop as well as at the end. The update at the end
of the loop is useless.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Patch converts kzalloc->copy_from_user sequence to memdup_user. There
is also one useless assignment of NULL to bnad->regdata as it is followed
by assignment of kzalloc output.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TX_E_PRIO_CHANGE event is never sent for bna_tx so it doesn't need to be
handled. After this change bna_tx->flags cannot contain
BNA_TX_F_PRIO_CHANGED flag and it can be also eliminated.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The bna_rx_config struct member paused can be removed as it is never
written and as it cannot have non-zero value the bna_rxf struct member
flags also cannot have BNA_RXF_F_PAUSED value and is always zero.
So the flags member can be removed as well as bna_rxf_flags enum and
the code-paths that needs to have non-zero bna_rxf->flags.
This clean-up makes bna_rxf_sm_paused state unsed and can be also removed.
Signed-off-by: David S. Miller <davem@davemloft.net>
RXF_E_PAUSE & RXF_E_RESUME events are never sent for bna_rxf object so
they needn't to be handled. The bna_rxf's state bna_rxf_sm_fltr_clr_wait
and function bna_rxf_fltr_clear are unused after this so remove them also.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
removed:
bna_rx_ucast_add
bna_rx_ucast_del
simplified:
bna_enet_pause_config
bna_rx_mcast_delall
bna_rx_mcast_listset
bna_rx_mode_set
bna_rx_ucast_listset
bna_rx_ucast_set
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The patch converts mac_t type to widely used 'u8 [ETH_ALEN]'.
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Parameters of all ether_addr_copy instances were checked for proper
alignment. Alignment of bnad_bcast_addr is forced to 2 as the implicit
alignment is 1.
I have also renamed address parameter of bnad_set_mac_address() to addr.
The name mac_addr was a little bit confusing as the real parameter is
struct sockaddr *.
v2: added __aligned directive to bnad_bcast_addr, renamed parameter of
bnad_set_mac_address() (thx joe@perches.com)
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
EXTRA_CFLAGS should be used on the command line only.
Since EXTRA_CFLAGS here add only a non-existant path to compiler
include paths (by -I), remove EXTRA_CFLAGS completely.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>
Bug in the driver initialization causes soft-lockup if firmware
initialization timeout is reached. Polling function bfa_ioc_poll_fwinit()
incorrectly calls bfa_nw_iocpf_timeout() when the timeout is reached.
The problem is that bfa_nw_iocpf_timeout() calls again
bfa_ioc_poll_fwinit()... etc. The bfa_ioc_poll_fwinit() should directly
send timeout event for iocpf and the same should be done if firmware
download into HW fails.
Cc: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Driver starts iocpf timer prior bnad_ioceth_enable() call and this is
unreasonable. This piece of code probably originates from Brocade/Qlogic
out-of-box driver during initial import into upstream. This driver uses
only one timer and queue to implement multiple timers and this timer is
started at this place. The upstream driver uses multiple timers instead
of this.
Cc: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Firmware required by bna is stored in appropriate files as sequence
of LE32 integers. After loading by request_firmware() they need to be
byte-swapped on big-endian arches. Without this conversion the NIC
is unusable on big-endian machines.
Cc: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To test a checkpatch spelling patch, I ran codespell against
drivers/net/ethernet/.
$ git ls-files drivers/net/ethernet/ | \
while read file ; do \
codespell -w $file; \
done
I removed a false positive in e1000_hw.h
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch updates the BNA driver version to 3.2.25.1 and the firmware version
to 3.2.5.1
Signed-off-by: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The same macros are used for rx as well. So rename it.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
The second init_completion call should be a reinit_completion here.
patch is against 3.18.0 linux-next
Signed-off-by: Nicholas Mc Guire <der.herr@hofr.at>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use netdev_rss_key_fill() helper, as it provides better support for some
bonding setups.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb->truesize is not meant to be tracking amount of used bytes
in an skb, but amount of reserved/consumed bytes in memory.
For instance, if we use a single byte in last page fragment,
we have to account the full size of the fragment.
skb->truesize can be very different from skb->len, that has
a very specific safety purpose.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When Tx VLAN offloading is disabled frames with size ~ MTU are not
transmitted as the driver does not account 4 bytes of VLAN header added
by stack. It should use VLAN_ETH_HLEN instead of ETH_HLEN.
The second problem is with newer BNA chips (BNA 1860). These chips filter
out any VLAN tagged frames in Tx path. This is a problem when Tx VLAN
offloading is disabled and frames are tagged by stack. Older chips like
1010/1020 are not affected as they probably don't do such filtering.
Cc: Rasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use container_of instead of casting first structure member.
Compiled but untested.
Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: David S. Miller <davem@davemloft.net>
