On the new 57500 chips, we need to allocate one RSS context for every
64 RX rings. In previous chips, only one RSS context per vnic is
required regardless of the number of RX rings. So increase the max
RSS context array count to 8.
Hardware ring groups are not used on the new chips. Note that the
software ring group structure is still maintained in the driver to
keep track of the rings associated with the vnic.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On the new 57500 chips, we allocate/free one CP ring for each RX ring or
TX ring separately. Using separate CP rings for RX/TX is an improvement
as TX events will no longer be stuck behind RX events.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Firmware ring allocation semantics are slightly different for most
ring types on 57500 chips. Allocation/deallocation for NQ rings are
also added for the new chips.
A CP ring handle is also added so that from the NQ interrupt event,
we can locate the CP ring.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On the new 57500 chips, getting the associated CP ring ID associated with
an RX ring or TX ring is different than before. On the legacy chips,
we find the associated ring group and look up the CP ring ID. On the
57500 chips, each RX ring and TX ring has a dedicated CP ring even if
they share the MSIX. Use these helper functions at appropriate places
to get the CP ring ID.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On 57500 chips, the original bnxt_cp_ring_info struct now refers to the
NQ. bp->cp_nr_rings refer to the number of NQs on 57500 chips. There
are now 2 pointers for the CP rings associated with RX and TX rings.
Modify bnxt_alloc_cp_rings() and bnxt_free_cp_rings() accordingly.
With multiple CP rings per NAPI, we need to add a pointer in
bnxt_cp_ring_info struct to point back to the bnxt_napi struct.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ring reservation functions have to be modified for P5 chips in the
following ways:
- bnxt_cp_ring_info structs map to internal NQs as well as CP rings.
- Ring groups are not used.
- 1 CP ring must be available for each RX or TX ring.
- number of RSS contexts to reserve is multiples of 64 RX rings.
- RFS currently not supported.
Also, RX AGG rings are only used for jumbo frames, so we need to
unconditionally call bnxt_reserve_rings() in __bnxt_open_nic()
to see if we need to reserve AGG rings in case MTU has changed.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Store the maximum MSIX capability in PCIe config. space earlier. When
we call firmware to query capability, we need to compare the PCIe
MSIX max count with the firmware count and use the smaller one as
the MSIX count for 57500 (P5) chips.
The new chips don't use ring groups. But previous chips do and
the existing logic limits the available rings based on resource
calculations including ring groups. Setting the max ring groups to
the max rx rings will work on the new chips without changing the
existing logic.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The 57500 series chips have a new 64-bit doorbell format. Use a new
bnxt_db_info structure to unify the new and the old 32-bit doorbells.
Add a new bnxt_set_db() function to set up the doorbell addreses and
doorbell keys ahead of time. Modify and introduce new doorbell
helpers to help abstract and unify the old and new doorbells.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
57500 series is a new chip class (P5) that requires some driver changes
in the next several patches. This adds basic chip ID, doorbells, and
the notification queue (NQ) structures. Each MSIX is associated with an
NQ instead of a CP ring in legacy chips. Each NQ has up to 2 associated
CP rings for RX and TX. The same bnxt_cp_ring_info struct will be used
for the NQ.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Call firmware to configure the DMA addresses of all context memory
pages on new devices requiring context memory.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
New device requires host context memory as a backing store. Call
firmware to check for context memory requirements and store the
parameters. Allocate host pages accordingly.
We also need to move the call bnxt_hwrm_queue_qportcfg() earlier
so that all the supported hardware queues and the IDs are known
before checking and allocating context memory.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Newer chips require the PTU_PTE_VALID bit to be set for every page
table entry for context memory and rings. Additional bits are also
required for page table entries for all rings. Add a flags field to
bnxt_ring_mem_info struct to specify these additional bits to be used
when setting up the pages tables as needed.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move the DMA page table and vmem fields in bnxt_ring_struct to a new
bnxt_ring_mem_info struct. This will allow context memory management
for a new device to re-use some of the existing infrastructure.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
New firmware spec. allows interrupt coalescing parameters, such as
maximums, timer units, supported features to be queried. Update
the driver to make use of the new call to query these parameters
and provide the legacy defaults if the call is not available.
Replace the hard-coded values with these parameters.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Support the max_ext_req_len field from the HWRM_VER_GET_RESPONSE.
If this field is valid and greater than the mailbox size, use the
short command format to send firmware messages greater than the
mailbox size. Newer devices use this method to send larger messages
to the firmware.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Latest firmware spec. has some additional rx extended port stats and new
tx extended port stats added. We now need to check the size of the
returned rx and tx extended stats and determine how many counters are
valid. New counters added include CoS byte and packet counts for rx
and tx.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Among the new changes are trusted VF support, 200Gbps support, and new
API to dump ring information on the new chips.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio says:
====================
selftests: pmtu: Add test choice and captures
This series adds a couple of features useful for debugging: 1/2
allows selecting single tests and 2/2 adds optional traffic
captures.
Semantics for current invocation of test script are preserved.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
If --trace is passed as an option and tcpdump is available,
capture traffic for all relevant interfaces to per-test pcap
files named <test>_<interface>.pcap.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
As number of tests is growing, it's quite convenient to allow
single tests to be run.
Display usage when the script is run with any invalid argument,
keep existing semantics when no arguments are passed so that
automated runs won't break.
Instead of just looping on the list of requested tests, if any,
check first that they exist, and go through them in a nested
loop to keep the existing way to display test descriptions.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
netif_device_detach() stops all tx queues already, so we don't need
this call.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Simplify this function, no functional change intended.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The newly added driver causes a warning about a function that is
not used anywhere:
drivers/net/ethernet/marvell/octeontx2/af/cgx.c:320:12: error: 'cgx_fwi_link_change' defined but not used [-Werror=unused-function]
Remove it for now, until a user gets added. If we want to use this
function from another module, we also need a declaration in a header
file, which is currently missing, so it would have to change anyway.
Fixes: 1463f382f5 ("octeontx2-af: Add support for CGX link management")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit makes it possible to use devlink to split the 100G CXP
Netronome into two 40G interfaces. Currently when you ask for 2
interfaces, the math in src/nfp_devlink.c:nfp_devlink_port_split
calculates that you want 5 lanes per port because for some reason
eth_port.port_lanes=10 (shouldn't this be 12 for CXP?). What we really
want when asking for 2 breakout interfaces is 4 lanes per port. This
commit makes that happen by calculating based on 8 lanes if 10 are
present.
Signed-off-by: Ryan C Goodfellow <rgoodfel@isi.edu>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Greg Weeks <greg.weeks@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ioana Ciornei says:
====================
dpaa2-eth: code cleanup
There are no functional changes in this patch set, only some cleanup
changes such as: unused parameters, uninitialized variables and
unnecessary Kconfig dependencies.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
According to the hardware ArchDef, the PTV1 field in FD[CTRL]
is ignored by WRIOP, so setting it for Tx FDs is pointless.
Remove all references to it from the code.
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ch parameter is never used in the dpaa2_eth_tx_conf function but
since its prototype must match the type defined in the consume field of
struct dpaa2_eth_fq, just mark it as __always_unused.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The priv parameter is never used in the build_linear_skb and
drain_channel function. Remove it from the function definitions.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All 3 cases of possible uninitialized variables are false
positives since they are used only as output parameters.
Nonetheless, fix the warnings.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The dpaa2_eth_set_dist_key function is only used in a single file.
Make it static.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Both ARCH_LAYERSCAPE and COMPILE_TEST dependencies are already implied
through the FSL_MC_BUS dep, so there's no need to state it explicitly.
Also, the fsl-mc bus depends on COMPILE_TEST only for some
architectures (arm, arm64, ppc, x86), so it's not correct to
claim build support unconditionally.
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In dual-emac mode the cpsw driver sends directed packets, that means
that packets go to the directed port, but an ALE lookup is performed
to determine untagged egress only. It means that on tx side no need
to add port bit for ALE mcast entry mask, and basically ALE entry
for port identification is needed only on rx side.
So, add only host port in dual_emac mode as used directed
transmission, and no need in one more port. For single port boards
and switch mode all ports used, as usual, so no changes for them.
Also it simplifies farther changes.
In other words, mcast entries for dual-emac should behave exactly
like unicast. It also can help avoid leaking packets between ports
with same vlan on h/w level if ports could became members of same vid.
So now, for instance, if mcast address 33:33:00:00:00:01 is added then
entries in ALE table:
vid = 1, addr = 33:33:00:00:00:01, port_mask = 0x1
vid = 2, addr = 33:33:00:00:00:01, port_mask = 0x1
Instead of:
vid = 1, addr = 33:33:00:00:00:01, port_mask = 0x3
vid = 2, addr = 33:33:00:00:00:01, port_mask = 0x5
With the same considerations, set only host port for unregistered
mcast for dual-emac mode in case of IFF_ALLMULTI is set, exactly like
it's done in cpsw_ale_set_allmulti().
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ivan Khoronzhuk says:
====================
net: ethernet: ti: cpsw fix mcast packet lost
The patchset omits redundant refresh of mcast address table and
prevents mcast packet lost.
Based on net-next/master
tested on am572x evm
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Whenever kernel or user decides to call rx mode update, it clears
every multicast entry from forwarding table and in some time adds
it again. This time can be enough to drop incoming multicast packets.
That's why clear only staled multicast entries and update or add new
one afterwards.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
It allows to use function under callbacks with same const qualifier of
mac address for farther changes.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit says:
====================
net: phy: improve and simplify state machine
Improve / simplify handling of states PHY_RUNNING and PHY_RESUMING in
phylib state machine.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Simplify code for handling state PHY_RESUMING, no functional change
intended.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Handling of state PHY_RUNNING seems to be more complex than it needs
to be. If not polling, then we don't have to do anything, we'll
receive an interrupt and go to state PHY_CHANGELINK once the link
goes down. If polling and link is down, we don't have to go the
extra mile over PHY_CHANGELINK and call phy_read_status() again
but can set status PHY_NOLINK directly.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This makes use of NTF_USE in vxlan driver consistent
with bridge driver.
Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The new command (NCSI_CMD_SEND_CMD) is added to allow user space application
to send NC-SI command to the network card.
Also, add a new attribute (NCSI_ATTR_DATA) for transferring request and response.
The work flow is as below.
Request:
User space application
-> Netlink interface (msg)
-> new Netlink handler - ncsi_send_cmd_nl()
-> ncsi_xmit_cmd()
Response:
Response received - ncsi_rcv_rsp()
-> internal response handler - ncsi_rsp_handler_xxx()
-> ncsi_rsp_handler_netlink()
-> ncsi_send_netlink_rsp ()
-> Netlink interface (msg)
-> user space application
Command timeout - ncsi_request_timeout()
-> ncsi_send_netlink_timeout ()
-> Netlink interface (msg with zero data length)
-> user space application
Error:
Error detected
-> ncsi_send_netlink_err ()
-> Netlink interface (err msg)
-> user space application
Signed-off-by: Justin Lee <justin.lee1@dell.com>
Reviewed-by: Samuel Mendoza-Jonas <sam@mendozajonas.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When starting the state machine there may be work to be done
immediately, e.g. if the initial state is PHY_UP then the state
machine may trigger an autonegotiation. Having said that I see no need
to wait a second until the state machine is run first time.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Toshiaki Makita says:
====================
veth: XDP stats improvement
ndo_xdp_xmit in veth did not update packet counters as described in [1].
Also, current implementation only updates counters on tx side so rx side
events like XDP_DROP were not collected.
This series implements the missing accounting as well as support for
ethtool per-queue stats in veth.
Patch 1: Update drop counter in ndo_xdp_xmit.
Patch 2: Update packet and byte counters for all XDP path, and drop
counter on XDP_DROP.
Patch 3: Support per-queue ethtool stats for XDP counters.
Note that counters are maintained on per-queue basis for XDP but not
otherwise (per-cpu and atomic as before). This is because 1) tx path in
veth is essentially lockless so we cannot update per-queue stats on tx,
and 2) rx path is net core routine (process_backlog) which cannot update
per-queue based stats when XDP is disabled. On the other hand there are
real rxqs and napi handlers for veth XDP, so update per-queue stats on
rx for XDP packets, and use them to calculate tx counters as well,
contrary to the existing non-XDP counters.
[1] https://patchwork.ozlabs.org/cover/953071/#1967449
====================
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
Expose per-queue stats for ethtool -S.
As there are only rx queues, and rx queues are used only when XDP is
used, per-queue counters are only rx XDP ones.
Example:
$ ethtool -S veth0
NIC statistics:
peer_ifindex: 11
rx_queue_0_xdp_packets: 28601434
rx_queue_0_xdp_bytes: 1716086040
rx_queue_0_xdp_drops: 28601434
rx_queue_1_xdp_packets: 17873050
rx_queue_1_xdp_bytes: 1072383000
rx_queue_1_xdp_drops: 17873050
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
On XDP path veth has napi handler so we can collect statistics on
per-queue basis for XDP.
By this change now we can collect XDP_DROP drop count as well as packets
and bytes coming through ndo_xdp_xmit. Packet counters shown by
"ip -s link", sysfs stats or /proc/net/dev is now correct for XDP.
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use existing atomic drop counter. Since drop path is really an
exceptional case here, I'm thinking atomic ops would not hurt the
performance.
XDP packets and bytes are not counted in ndo_xdp_xmit, but will be
accounted on rx side by the following commit.
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
INADDR_ANY is hard-coded when activating UDP bearer. So, we could not
bind to a specific IP address even with replicast mode using - given
remote ip address instead of using multicast ip address.
In this commit, we fixed it by checking and switch to use appropriate
local ip address.
before:
$netstat -plu
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address Foreign Address
udp 0 0 **0.0.0.0:6118** 0.0.0.0:*
after:
$netstat -plu
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address Foreign Address
udp 0 0 **10.0.0.2:6118** 0.0.0.0:*
Acked-by: Ying Xue <ying.xue@windriver.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Hoang Le <hoang.h.le@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
IPoIB netlink support and mlx5e pre-allocated netdevice initialization
IP link was broken due to the changes in IPoIB for the rdma_netdev
support after commit cd565b4b51
("IB/IPoIB: Support acceleration options callbacks").
This patchset fixes IPoIB pkey creation and removal using rtnetlink by
adding support in both IPoIB ULP layer and mlx5 layer:
From Jason and Denis:
1) Introduces changes in the RDMA netdev code in order to
allow allocation of the netdev to be done by the rtnl netdev code.
2) Reworks IPoIB initialization to use the two step rdma_netdev
creation.
From Feras and Saeed, mlx5e netdev layer refactoring to allow accepting
pre-allocated netdevs:
3) Adds support to initialize/cleanup netdevs that are not created
by mlx5 driver.
4) Change mlx5e netdevice layer to accept the pre-allocated netdevice
queue number.
5) Initialize mlx5e generic structures in one place to be used for all
netdevs types NIC/representors/IPoIB (both mlx5 allocated and
pre-allocted).
-----BEGIN PGP SIGNATURE-----
iQEcBAABAgAGBQJbvqHhAAoJEEg/ir3gV/o+TRQH+wYWgvi5AbqYlGpT8xSho8Dg
5tuFE9AbYDGLLWptZgtYXsBsTaltGVqKnwWy//dAuuSI7LvqtjeKqq0G1JKau70L
/49+vv8NTQ6vov2yY95yzzEo5B1/zVcYEl9dmJl4y6Xlwc+YIteewReVqRj+tucR
xO7ghLWc1o4Pl7gWBAcOULyOsZc+CtG8ElwEaBROXTtYRpsR7FKn7vN+WdWZ2VOG
MjtCUaG0vZlK1vaF76J3P9bL2V3y6o6i6S8RZwg1kPxO4jnrzyGrkxWMOX6f32KB
JAUcLVlJW5wckcRAKfTwLxsFMrc9vLBPHyj4XneXVWGKh8/dGUGY5gdGIJV5Jlc=
=m0tM
-----END PGP SIGNATURE-----
Merge tag 'mlx5e-updates-2018-10-10' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
Saeed Mahameed says:
====================
mlx5e-updates-2018-10-10
IPoIB netlink support and mlx5e pre-allocated netdevice initialization
IP link was broken due to the changes in IPoIB for the rdma_netdev
support after commit cd565b4b51
("IB/IPoIB: Support acceleration options callbacks").
This patchset fixes IPoIB pkey creation and removal using rtnetlink by
adding support in both IPoIB ULP layer and mlx5 layer:
From Jason and Denis:
1) Introduces changes in the RDMA netdev code in order to
allow allocation of the netdev to be done by the rtnl netdev code.
2) Reworks IPoIB initialization to use the two step rdma_netdev
creation.
From Feras and Saeed, mlx5e netdev layer refactoring to allow accepting
pre-allocated netdevs:
3) Adds support to initialize/cleanup netdevs that are not created
by mlx5 driver.
4) Change mlx5e netdevice layer to accept the pre-allocated netdevice
queue number.
5) Initialize mlx5e generic structures in one place to be used for all
netdevs types NIC/representors/IPoIB (both mlx5 allocated and
pre-allocted).
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Maciej W. Rozycki says:
====================
FDDI: DEC FDDIcontroller 700 TURBOchannel adapter support
This is an update to <http://patchwork.ozlabs.org/patch/342737/>. I
believe I have addressed all the requests made in the previous review
round.
There is still one `checkpatch.pl' warning remaining:
WARNING: quoted string split across lines
+ pr_info("%s: ROM rev. %.4s, firmware rev. %.4s, RMC rev. %.4s, "
+ "SMT ver. %u\n", fp->name, rom_rev, fw_rev, rmc_rev, smt_ver);
total: 0 errors, 1 warnings, 2458 lines checked
however I think the value of staying within 80 columns is higher than the
value of having the string on a single line. This is because with all the
formatting specifiers there it is not directly greppable based on the
final output produced to the kernel log on one hand, e.g.:
tc2: ROM rev. 1.0, firmware rev. 1.2, RMC rev. A, SMT ver. 1
while it can be easily tracked down by grepping for an obvious substring
such as "RMC rev" on the other.
The issue with MMIO barriers I discussed in the course of the original
review turned out mostly irrelevant to this driver, because as I have
learnt in a recent Alpha/Linux discussion starting here:
<https://marc.info/?i=alpine.LRH.2.02.1808161556450.13597%20()%20file01%20!%20intranet%20!%20prod%20!%20int%20!%20rdu2%20!%20redhat%20!%20com>
our MMIO API mandates the `readX' and `writeX' accessors to be strongly
ordered with respect to each other, even if that is not implicitly
enforced by hardware.
Consequently I have removed all the explicit ordering barriers and
instead submitted a fix for MIPS MMIO implementation, which currently does
not guarantee strong ordering (the MIPS architecture does not define bus
ordering rules except in terms of SYNC barriers), as recorded here:
<https://patchwork.linux-mips.org/project/linux-mips/list/?series=1538>.
Enforcing strong MMIO ordering can be costly however and is often
unnecessary, e.g. when using PIO to access network frame data in onboard
packet memory. I have therefore retained the information that would be
lost by the removal of barriers, by defining accessor wrappers suffixed by
`_o' and `_u', for accesses that have to be ordered and can be unordered
respectively.
If we ever have an API defined for weakly-ordered MMIO accesses, then
these wrappers can be redefined accordingly. Right now they all expand to
the respective `_relaxed' accessors, because, again, enforcing the
ordering WRT DMA transfers can be costly and we don't need it here except
in one place, where I chose to use explicit `dma_rmb' instead.
Similarly I have replaced the completion barriers with a read back from
the respective MMIO location (all adapter MMIO registers can be read with
no side effects incurred), which will serve its purpose on the basis of
MMIO being strongly ordered (although a read from TURBOchannel is going to
be slower than `iob', making the delay incurred unnecessarily longer).
And last but not least, I have split off the SMT Tx network tap support
to a separate change, 2/2 in this series, so that it does not block the
driver proper and can be discussed separately.
I think it has value in that it makes the view of the outgoing network
traffic complete, as if one actually physically tapped into the outgoing
line of the ring, between the station being examined and its downstream
neighbour. Without this part only traffic passed from applications
through the whole protocol stack can be captured and this is only a part
of the view.
With the `dev_queue_xmit_nit' interface now exported it's only
`ptype_all' that remains private, and to define a properly abstracted API
I propose to provide am exported `dev_nit_active' predicate that tells
whether any taps are active. This predicate is then used accordingly.
NB if there is a long-term maintenance concern about the `dev_nit_active'
predicate, then well, corresponding inline code currently present in
`xmit_one' has to be maintained anyway, and if the resulting changes
require `defza' to be updated accordingly, then I am going to handle it;
after some 20 years with Linux it's not that I am going to disappear
anywhere anytime. And once I am dead, which is inevitably going to happen
sooner or later, then the driver can simply be ripped from the kernel.
Though I suspect that at that point no DECstation Linux users may survive
anymore, even though hardware, being as sturdy as it is, likely will.
I have a patch for `tcpdump' to actually decode SMT frames, which I plan
to upstream sometime. Here's a sample of SMT traffic captured through the
`defza' driver in a small network of 4 stations and no concentrators,
printed in the most verbose mode:
01:16:59.138381 4f 00:60:b0:58:41:e7 00:60:b0:58:41:e7 73: SMT NIF ann vid:1 tid:00000270 sid:00-00-00-60-b0-58-41-e7 len:40: UNA: 00 00 00 06 0d 1a 02 ae StationDescr: 00 01 02 00 StationState: 00 00 30 00 MACFrameStatusFunctions.3: 00 00 00 01
01:17:00.332750 4f 08:00:2b:a3:a3:29 08:00:2b:a3:a3:29 73: SMT NIF ann vid:1 tid:0000013b sid:00-00-08-00-2b-a3-a3-29 len:40: UNA: 00 00 00 06 0d 1a 82 e7 StationDescr: 00 01 02 00 StationState: 00 00 30 00 MACFrameStatusFunctions.3: 00 00 00 01
01:17:00.354479 4f 00:60:b0:58:40:75 00:60:b0:58:40:75 73: SMT NIF ann vid:1 tid:0000029c sid:00-00-00-60-b0-58-40-75 len:40: UNA: 00 00 10 00 d4 74 b6 ae StationDescr: 00 01 02 00 StationState: 00 00 31 00 MACFrameStatusFunctions.3: 00 00 00 01
01:17:00.442175 4f 00:60:b0:58:41:e7 Broadcast 73: SMT NIF req vid:1 tid:00000271 sid:00-00-00-60-b0-58-41-e7 len:40: UNA: 00 00 00 06 0d 1a 02 ae StationDescr: 00 01 02 00 StationState: 00 00 30 00 MACFrameStatusFunctions.3: 00 00 00 01
01:17:00.448657 41 08:00:2b:a3:a3:29 00:60:b0:58:41:e7 73: SMT NIF rsp vid:1 tid:00000271 sid:00-00-08-00-2b-a3-a3-29 len:40: UNA: 00 00 00 06 0d 1a 82 e7 StationDescr: 00 01 02 00 StationState: 00 00 30 00 MACFrameStatusFunctions.3: 00 00 00 01
01:17:01.015152 4f 08:00:2b:a3:a3:29 Broadcast 73: SMT NIF req vid:1 tid:0000013c sid:00-00-08-00-2b-a3-a3-29 len:40: UNA: 00 00 00 06 0d 1a 82 e7 StationDescr: 00 01 02 00 StationState: 00 00 30 00 MACFrameStatusFunctions.3: 00 00 00 01
01:17:01.111644 41 08:00:2b:2e:6d:75 08:00:2b:a3:a3:29 73: SMT NIF rsp vid:1 tid:0000013c sid:00-00-08-00-2b-2e-6d-75 len:40: UNA: 00 00 10 00 d4 c5 c5 94 StationDescr: 00 01 01 00 StationState: 00 00 11 00 MACFrameStatusFunctions.2: 00 00 00 01
01:17:04.814603 4f 08:00:2b:2e:6d:75 Broadcast 73: SMT NIF req vid:1 tid:0000013c sid:00-00-08-00-2b-2e-6d-75 len:40: UNA: 00 00 10 00 d4 c5 c5 94 StationDescr: 00 01 01 00 StationState: 00 00 11 00 MACFrameStatusFunctions.2: 00 00 00 01
01:17:04.814939 4f 08:00:2b:2e:6d:75 Broadcast 73: SMT NIF req vid:1 tid:0000013c sid:00-00-08-00-2b-2e-6d-75 len:40: UNA: 00 00 10 00 d4 c5 c5 94 StationDescr: 00 01 01 00 StationState: 00 00 11 00 MACFrameStatusFunctions.2: 00 00 00 01
01:17:04.820960 4f 08:00:2b:2e:6d:75 08:00:2b:2e:6d:75 73: SMT NIF ann vid:1 tid:0000013b sid:00-00-08-00-2b-2e-6d-75 len:40: UNA: 00 00 10 00 d4 c5 c5 94 StationDescr: 00 01 01 00 StationState: 00 00 11 00 MACFrameStatusFunctions.2: 00 00 00 01
Questions, comments? Otherwise, please apply.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
DEC FDDIcontroller 700 (DEFZA) uses a Tx/Rx queue pair to communicate
SMT frames with adapter's firmware. Any SMT frame received from the RMC
via the Rx queue is queued back by the driver to the SMT Rx queue for
the firmware to process. Similarly the firmware uses the SMT Tx queue
to supply the driver with SMT frames which are queued back to the Tx
queue for the RMC to send to the ring.
When a network tap is attached to an FDDI interface handled by `defza'
any incoming SMT frames captured are queued to our usual processing of
network data received, which in turn delivers them to any listening
taps.
However the outgoing SMT frames produced by the firmware bypass our
network protocol stack and are therefore not delivered to taps. This in
turn means that taps are missing a part of network traffic sent by the
adapter, which may make it more difficult to track down network problems
or do general traffic analysis.
Call `dev_queue_xmit_nit' then in the SMT Tx path, having checked that
a network tap is attached, with a newly-created `dev_nit_active' helper
wrapping the usual condition used in the transmit path.
Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for the DEC FDDIcontroller 700 (DEFZA), Digital Equipment
Corporation's first-generation FDDI network interface adapter, made for
TURBOchannel and based on a discrete version of what eventually became
Motorola's widely used CAMEL chipset.
The CAMEL chipset is present for example in the DEC FDDIcontroller
TURBOchannel, EISA and PCI adapters (DEFTA/DEFEA/DEFPA) that we support
with the `defxx' driver, however the host bus interface logic and the
firmware API are different in the DEFZA and hence a separate driver is
required.
There isn't much to say about the driver except that it works, but there
is one peculiarity to mention. The adapter implements two Tx/Rx queue
pairs.
Of these one pair is the usual network Tx/Rx queue pair, in this case
used by the adapter to exchange frames with the ring, via the RMC (Ring
Memory Controller) chip. The Tx queue is handled directly by the RMC
chip and resides in onboard packet memory. The Rx queue is maintained
via DMA in host memory by adapter's firmware copying received data
stored by the RMC in onboard packet memory.
The other pair is used to communicate SMT frames with adapter's
firmware. Any SMT frame received from the RMC via the Rx queue must be
queued back by the driver to the SMT Rx queue for the firmware to
process. Similarly the firmware uses the SMT Tx queue to supply the
driver with SMT frames that must be queued back to the Tx queue for the
RMC to send to the ring.
This solution was chosen because the designers ran out of PCB space and
could not squeeze in more logic onto the board that would be required to
handle this SMT frame traffic without the need to involve the driver, as
with the later DEFTA/DEFEA/DEFPA adapters.
Finally the driver does some Frame Control byte decoding, so to avoid
magic numbers some macros are added to <linux/if_fddi.h>.
Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
