This consolidates the checksum offload code that was duplicated
over the two qeth subdrivers.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

struct net_device contains a dev_port field. Store the OSA port number
in this field.
Signed-off-by: NUrsula Braun <ubraun@linux.ibm.com>
Reviewed-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As the vid_list is only accessed from process context, there's no need to
protect it with a spinlock.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Both qeth sub drivers use the same QDIO queue handlers, there's no need
to expose them via the driver's discipline. No functional change.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Submitting a cmd IO request (usually on the WRITE device, but for IDX
also on the READ device) is currently done with ccw_device_start()
and a manual timeout in the caller.
On timeout, the caller cleans up the related resources (eg. IO buffer).
But 1) the IO might still be active and utilize those resources, and
    2) when the IO completes, qeth_irq() will attempt to clean up the
       same resources again.

Instead of introducing additional resource locking, switch to
ccw_device_start_timeout() to ensure IO termination after timeout, and
let the IRQ handler alone deal with cleaning up after a request.

This also removes a stray write->irq_pending reset from
clear_ipacmd_list(). The routine doesn't terminate any pending IO on
the WRITE device, so this should be handled properly via IO timeout
in the IRQ handler.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When changing the MAC address on a L2 qeth device, current code first
unregisters the old address, then registers the new one.
If HW rejects the new address (or the IO fails), the device ends up with
no operable address at all.

Re-order the code flow so that the old address only gets dropped if the
new address was registered successfully. While at it, add logic to catch
some corner-cases.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Removing couple of duplicate includes, found by "make includecheck".
That leaves 1 duplicate include in arch/s390/kernel/entry.S, which is
there for a reason (it includes generated asm/syscall_table.h twice).
Signed-off-by: NVasily Gorbik <gor@linux.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

On removal, a qeth card's netdevice is currently not properly freed
because the call chain looks as follows:

qeth_core_remove_device(card)
	lx_remove_device(card)
		unregister_netdev(card->dev)
		card->dev = NULL			!!!
	qeth_core_free_card(card)
		if (card->dev)				!!!
			free_netdev(card->dev)

Fix it by free'ing the netdev straight after unregistering. This also
fixes the sysfs-driven layer switch case (qeth_dev_layer2_store()),
where the need to free the current netdevice was not considered at all.

Note that free_netdev() takes care of the netif_napi_del() for us too.

Fixes: 4a71df50 ("qeth: new qeth device driver")
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Reviewed-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

init_qdio_queues() resets the Input Queue's overall QDIO state, and
positions the buffer cursor back to 0. So this is the obvious place to
also reset the queue's NAPI context (in contrast to doing it rather
randomly in the middle of the big set_online() path).
No functional change.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Newly-allocated skbs default to PACKET_HOST, and eth_type_trans() is
smart enough to determine any other packet type from the frame's
destination address.
So except for the IQD sniffer case, there is no need to set up
skb->pkt_type manually.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

napi_alloc_skb() doesn't need to disable IRQs during the allocation,
and thus may save us a few cycles.
Doing so requires a small fix-up in the HiperTransport path, which
currently assumes a fixed NET_SKB_PAD headroom padding. napi_alloc_skb()
adds an additional NET_IP_ALIGN padding, so use the proper helper for
setting up the mac_header offset.

Use this opportunity to convert the non-NAPI path to netdev_alloc_skb(),
which means that skb->dev is now always set-up during allocation and
doesn't need to be assigned manually.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

qeth implements HW-based Unicast Filtering (via SETVMAC) on L2 devices.
Tell the stack, so it knows that receiving traffic for secondary
addresses doesn't require full-blown promiscuous mode.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

NETIF_F_SG support is currently limited to OSA (and for L2 even OSD)
devices. Advertise it for some more device types (OSM, L2 OSX, z/VM OSA)
that share the same code paths. For now, keep it switched off by
default on these devices.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

"s390/qeth: fix SETIP command handling" introduced a new helper, apply
it driver-wide.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Once all of qeth_l3_set_rx_mode()'s single-use helpers are folded back
in, the two implementations actually look quite similar. So improve the
readability by converting both set_rx_mode() routines to a common
format.

This also allows us to walk ip_mc_htable just once, instead of three
times.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Be a little more self-documenting, and get rid of OSA_ADDR_LEN.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For adding/removing a MAC address, use just one helper each that
handles both unicast and multicast.
Saves one level of indirection for multicast addresses, while improving
the error reporting for unicast addresses.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Instead of tracking the uc/mc state in each MAC address object, just
check the multicast bit in the address itself.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current GSO skb size limit was copy&pasted over from the L3 path,
where it is needed due to a TSO limitation.
As L2 devices don't offer TSO support (and thus all GSO skbs are
segmented before they reach the driver), there's no reason to restrict
the stack in how large it may build the GSO skbs.

Fixes: d52aec97 ("qeth: enable scatter/gather in layer 2 mode")
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Using GSO with small MTUs currently results in a substantial throughput
regression - which is caused by how qeth needs to map non-linear skbs
into its IO buffer elements:
compared to a linear skb, each GSO-segmented skb effectively consumes
twice as many buffer elements (ie two instead of one) due to the
additional header-only part. This causes the Output Queue to be
congested with low-utilized IO buffers.

Fix this as follows:
If the MSS is low enough so that a non-SG GSO segmentation produces
order-0 skbs (currently ~3500 byte), opt out from NETIF_F_SG. This is
where we anticipate the biggest savings, since an SG-enabled
GSO segmentation produces skbs that always consume at least two
buffer elements.

Larger MSS values continue to get a SG-enabled GSO segmentation, since
1) the relative overhead of the additional header-only buffer element
becomes less noticeable, and
2) the linearization overhead increases.

With the throughput regression fixed, re-enable NETIF_F_SG by default to
reap the significant CPU savings of GSO.

Fixes: 5722963a ("qeth: do not turn on SG per default")
Reported-by: NNils Hoppmann <niho@de.ibm.com>
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It's good to have SPDX identifiers in all files to make it easier to
audit the kernel tree for correct licenses.

Update the drivers/s390/net/ files with the correct SPDX license
identifier based on the license text in the file itself.  The SPDX
identifier is a legally binding shorthand, which can be used instead of
the full boiler plate text.

This work is based on a script and data from Thomas Gleixner, Philippe
Ombredanne, and Kate Stewart.

Cc: Julian Wiedmann <jwi@linux.vnet.ibm.com>
Cc: Ursula Braun <ubraun@linux.vnet.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Kate Stewart <kstewart@linuxfoundation.org>
Cc: Philippe Ombredanne <pombredanne@nexb.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Instead of silently discarding VLAN registration requests on OSM,
just indicate that this card type doesn't support VLAN.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There's no reason why l2_set_mac_address() should ever be called for
a netdevice that's not owned by qeth. It's certainly not required for
VLAN devices, which have their own netdev_ops.

Also:
1) we don't do such validation for any of the other netdev_ops routines.
2) the code in question clearly has never been actually exercised;
   it's broken. After determining that the device is not owned
   by qeth, it would still use dev->ml_priv to write a qeth trace entry.

Remove the check, and its helper that walked the global card list.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Properly return any error encountered during VLAN processing to the
the caller.
Resulting change in behaviour: if SETVLAN fails while registering a
new VLAN ID, the stack no longer creates the corresponding vlan device.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Reviewed-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Locking the output queue prior to TX is needed on OSA devices,
to synchronize against a packing flush from the TX completion code
(via qeth_check_outbound_queue()).
But send_packet_fast() is only used for IQDs, which don't do packing.
So remove the locking, and apply some easy cleanups.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Reviewed-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

HiperSockets allow configuring so called VNIC Characteristics (VNICC)
that influence how the underlying hardware handles packets. For VNICCs,
additional commands for getting and setting timeouts are available.
Currently, the learning VNICC uses these commands.

* Learning VNICC: If learning is enabled on a qeth device, the device
  learns the source MAC addresses of outgoing packets and incoming
  packets to those learned MAC addresses are received.

For learning, the timeout specifies the idle period in seconds, after
which the underlying hardware removes a learned MAC address again.

This patch adds support for the IPA commands that are required to get
and set the current timeout values for the learning VNIC characteristic.
Also, it introduces the sysfs interface that allows users to configure
the timeout.
Signed-off-by: NHans Wippel <hwippel@linux.vnet.ibm.com>
Reviewed-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

HiperSocket devices allow enabling and disabling so called VNIC
Characteristics (VNICC) that influence how the underlying hardware
handles packets. These VNICCs are:

* Flooding VNICC: Flooding allows specifying if packets to unknown
  destination MAC addresses are received by the qeth device.

* Multicast flooding VNICC: Multicast flooding allows specifying if
  packets to multicast MAC addresses are received by the qeth device.

* Learning VNICC: If learning is enabled on a qeth device, the device
  learns the source MAC addresses of outgoing packets and incoming
  packets to those learned MAC addresses are received.

* Takeover setvmac VNICC: If takeover setvmac is configured on a qeth
  device, the MAC address of this device can be configured on a
  different qeth device with the setvmac IPA command.

* Takeover by learning VNICC: If takeover learning is enabled on a qeth
  device, the MAC address of this device can be learned (learning VNICC)
  on a different qeth device.

* BridgePort invisible VNICC: If BridgePort invisible is enabled on a
  qeth device, (1) packets from this device are not sent to a BridgePort
  enabled qeth device and (2) packets coming from a BridgePort enabled
  qeth device are not received by this device.

* Receive broadcast VNICC: Receive broadcast allows configuring if a
  qeth device receives packets with the broadcast destination MAC
  address.

This patch adds support for the IPA commands that are required to enable
and disable these VNIC characteristics on qeth devices. As a
prerequisite, it also adds the query commands IPA command.

The query commands IPA command allows requesting the supported commands
for each characteristic from the underlying hardware.

Additionally, this patch provides users with a sysfs user interface to
enable/disable the VNICCs mentioned above.
Signed-off-by: NHans Wippel <hwippel@linux.vnet.ibm.com>
Reviewed-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

VNIC Characteristics (VNICC) are features of HiperSockets that define
how packets are handled by the underlying network hardware. For example,
if the VNICC flooding is configured on a qeth device, ethernet frames to
unknown destination MAC addresses are received.

Currently, there is support for seven VNICCs: flooding, multicast
flooding, receive broadcast, learning, takeover learning, takeover
setvmac, bridge invisible. Also, six IPA commands exist for configuring
VNICCs on a qeth device: query characteristics, query commands, enable
characteristic, disable characteristic, set timeout, get timeout.

This patch adds the basic code infrastructure for VNICC support to qeth.
It allows querying VNICC support from the underlying hardware. To this
end, it adds:

* basic message formats for IPA commands
* basic data structures
* basic error handling
* query characteristics IPA command support

The query characteristics IPA command allows requesting the currently
supported and currently enabled VNIC characteristics from the underlying
hardware.

Support for the other IPA commands and for the configuration of VNICCs
is added in follow-up patches together with the respective user
interface functions.
Signed-off-by: NHans Wippel <hwippel@linux.vnet.ibm.com>
Reviewed-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Taking a full copy via skb_realloc_headroom() on every xmit is overkill
and wastes CPU time; all we actually need is to push on the qeth_hdr.
So rework the L2 OSA TX path to avoid the copy.
Minor complications arise because struct qeth_hdr must not cross a page
boundary. So add a new helper qeth_push_hdr() that catches this, and
falls back to the hdr cache that we already use for IQDs.

This change uncovered that qeth's TX completion takes rather long.
Now that we no longer free the original skb straight away and thus call
skb->destructor later than before, throughput regresses significantly.
For now, restore old behaviour by adding an explicit skb_orphan(),
and a big TODO to improve the TX completion time.
Tested-by: NNils Hoppmann <niho@de.ibm.com>
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is a prerequisite for unifying the code to build header elements.
The TSO header has a different size, so we can no longer rely on implicitly
adding the size of a normal qeth_hdr.

No functional change.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For TSO we need to skip the skb's qeth/IP/TCP headers when mapping
it into buffer elements. Instead of (mis)using skb_pull(), pass a
corresponding offset to fill_buffer() like we already do for IQDs.

No actual change in the resulting TSO buffers.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The TSO code already calculates the length of its header element,
no need to duplicate this in the low-level code again.

Use this opportunity to make hd_len unsigned, and for TSO match
its calculation to what tso_fill_header() does.

No functional change.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For IQD we already need to fix up the qeth_hdr's length field, and
future changes will require more flexibility for OSA as well. The
device-specific path knows best what header length it requires, so just
pass it from there.
While at it, remove the unused qeth_card parameter.

No functional change.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

l2_hard_start_xmit() actually doesn't contain much shared code,
and having device-specific paths makes isolated changes a lot easier.
So split it into three routines for IQD, OSN and OSD/OSM/OSX.

No functional change.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Consolidation of duplicated code, no functional change.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Clean up the inline cruft in s390 net drivers. Many of the inlined
functions had only one caller anyway.
Suggested-by: NJoe Perches <joe@perches.com>
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Replace some open-coded parts with their proper API calls.

Also remove two skb_[re]set_mac_header() calls in the L2
xmit paths that are clearly no longer required, since at least
commit 6d1ccff6 ("net: reset mac header in dev_start_xmit()").
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Acked-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

qeth_l2_request_initial_mac() queries the device for its supported
adapterparms, even though they already have been queried as part of the
device's high-level setup. Remove that extra call.

The only call chain for qeth_l2_request_initial_mac() is
__qeth_l2_set_online()
	qeth_core_hardsetup_card()
		qeth_query_setadapterparms()
	qeth_l2_setup_netdev()
		qeth_l2_request_initial_mac()
			qeth_query_setadapterparms()

, and we only reach qeth_l2_request_initial_mac() if the first
adapterparms query succeeds. Hence removing the second query results in
no loss of functionality.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Acked-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

After transmitting a skb via send_packet[_fast](), the statistics
code accesses the skb once more to account for transmitted page frags.
This has a (theoretical?) race against the TX completion - if the TX
completion is processed and frees the skb before hard_start_xmit()
gets to the statistics part, we access random memory.

Fix this by caching the # of page frags, before the skb is transmitted.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Acked-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a s390 guest runs on a z/VM host that's part of a SSI cluster,
it can be migrated to a different host. In this case, the MAC address
it originally obtained on the old host may be re-assigned to a new
guest. This would result in address conflicts between the two guests.

When running as z/VM guest, use the diag26c MAC Service to obtain
a hypervisor-managed MAC address. The MAC Service is SSI-aware, and
won't re-assign the address after the guest is migrated to a new host.

This patch adds support for the z/VM MAC Service on L2 devices.
Signed-off-by: NJulian Wiedmann <jwi@linux.vnet.ibm.com>
Acked-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>