Commit graph

71 commits

Author SHA1 Message Date
Ben Hutchings
718cff1eec sfc: Match calls to netif_napi_add() and netif_napi_del()
sfc could call netif_napi_add() multiple times for the same
napi_struct, corrupting the list of napi_structs for the associated
device and leading to a busy-loop on device removal.  Move the call to
netif_napi_add() and add a call to netif_napi_del() in the obvious
places.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-14 19:47:46 -07:00
Ben Hutchings
7e765cb4a0 sfc: Call netif_napi_add() before registering our interrupt handler
If we use a shared legacy IRQ then our interrupt handler may be called
as soon as it is registered even though IRQs are disabled on the NIC.
Now that the legacy interrupt handler also checks for event delivery,
it may decide to schedule polling in this case.  Ensure that the NAPI
context is valid but disabled at this point.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-31 15:07:21 -07:00
Ingo Molnar
6e15cf0486 Merge branch 'core/percpu' into percpu-cpumask-x86-for-linus-2
Conflicts:
	arch/parisc/kernel/irq.c
	arch/x86/include/asm/fixmap_64.h
	arch/x86/include/asm/setup.h
	kernel/irq/handle.c

Semantic merge:
        arch/x86/include/asm/fixmap.h

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-03-27 17:28:43 +01:00
Ben Hutchings
6fb70fd1b5 sfc: Implement adaptive IRQ moderation
Calculate a score for each 1000 IRQs:
- TX completions are worth 1 point
- RX completions are worth 4 if merged using LRO or 2 otherwise

Reduce moderation if the score is less than 10000, down to a minimum
of 5 us.  Increase moderation if the score is more than 20000, up to
the specified maximum.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-21 19:06:58 -07:00
Ben Hutchings
115122afc5 sfc: Clean up properly on reset failure paths
If MAC switching fails, stop the port properly.

If PHY reinitialisation fails, clear the port_initialized flag.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-04 17:40:22 -08:00
Ben Hutchings
8129d2173e sfc: Clean up LED control
Reinitialise LEDs after overriding them for identification.

Rename set_fault_led method to set_id_led since we always use it for
NIC identification and not faults.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-02 03:15:05 -08:00
Ingo Molnar
249d51b53a Merge commit 'v2.6.29-rc4' into core/percpu
Conflicts:
	arch/x86/mach-voyager/voyager_smp.c
	arch/x86/mm/fault.c
2009-02-09 14:58:11 +01:00
David S. Miller
05bee47377 Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Conflicts:
	drivers/net/e1000/e1000_main.c
2009-01-30 14:31:07 -08:00
Ben Hutchings
1974cc205e sfc: Replace stats_enabled flag with a disable count
Currently we use a spin-lock to serialise statistics fetches and also
to inhibit them for short periods of time, plus a flag to
enable/disable statistics fetches for longer periods of time, during
online reset.  This was apparently insufficient to deal with the several
reasons for stats being disabled.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-30 14:06:33 -08:00
Steve Hodgson
4b988280be sfc: Reinitialise the PHY completely in case of a PHY or NIC reset
In particular, set pause advertising bits properly.

A PHY reset is not necessary to recover from the register self-test,
so use a "invisible" reset there instead.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-30 14:06:30 -08:00
Herbert Xu
da3bc07171 sfc: Replace LRO with GRO
This patch makes sfc invoke the GRO hooks instead of LRO.  As
GRO has a compatible external interface to LRO this is a very
straightforward replacement.

Everything should appear identical to the user except that the
offload is now controlled by the GRO ethtool option instead of
LRO.  I've kept the lro module parameter as is since that's for
compatibility only.

I have eliminated efx_rx_mk_skb as the GRO layer can take care
of all packets regardless of whether GRO is enabled or not.

So the only case where we don't call GRO is if the packet checksum
is absent.  This is to keep the behaviour changes of the patch to
a minimum.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-21 14:34:06 -08:00
Ben Hutchings
288379f050 net: Remove redundant NAPI functions
Following the removal of the unused struct net_device * parameter from
the NAPI functions named *netif_rx_* in commit 908a7a1, they are
exactly equivalent to the corresponding *napi_* functions and are
therefore redundant.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-21 14:33:50 -08:00
Rusty Russell
2f8975fbcf cpumask: convert drivers/net/sfc
Impact: reduce stack usage, use new cpumask API.

Remove a cpumask from the stack.  Ben Hutchings indicated that printing
a warning and returning 1 was acceptable for the corner case where allocation
fails.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Cc: linux-net-drivers@solarflare.com
2009-01-11 19:12:56 +01:00
Rusty Russell
fbd59a8d1f cpumask: Use topology_core_cpumask()/topology_thread_cpumask()
Impact: reduce stack usage, use new cpumask API.

This actually uses topology_core_cpumask() and
topology_thread_cpumask(), removing the only users of
topology_core_siblings() and topology_thread_siblings()

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Cc: linux-net-drivers@solarflare.com
2009-01-11 19:12:49 +01:00
David S. Miller
2d5451d261 net: Fix warning fallout from recent NAPI interface changes.
When we removed the network device argument from several
NAPI interfaces in 908a7a16b8
("net: Remove unused netdev arg from some NAPI interfaces.")
several drivers now started getting unused variable warnings.

This fixes those up.

Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 15:10:00 -08:00
Ben Hutchings
f4bd954e7c sfc: When disabling the NIC, close the device rather than unregistering it
This should reduce user confusion and may also aid recovery (ioctls
will still be available).

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 13:48:51 -08:00
Ben Hutchings
6977dc6309 sfc: Generate unique names for per-NIC workqueues
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 13:44:39 -08:00
Neil Horman
908a7a16b8 net: Remove unused netdev arg from some NAPI interfaces.
When the napi api was changed to separate its 1:1 binding to the net_device
struct, the netif_rx_[prep|schedule|complete] api failed to remove the now
vestigual net_device structure parameter.  This patch cleans up that api by
properly removing it..

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-22 20:43:12 -08:00
Ben Hutchings
7dde596ef7 sfc: Fix synchronisation of efx_mtd_{probe,rename,remove}
Currently efx_mtd_rename() can race with the probe() and remove()
functions.

Move probe() before device registration and remove() after
unregistration.  Move initialisation/update of all names based on the
netdev name into a new function and call it under the RTNL immediately
after registration.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 22:09:38 -08:00
Steve Hodgson
fa402b2ea9 sfc: Treat probe as unsuccessful if it scheduled a reset
From: Steve Hodgson <shodgson@solarflare.com>

efx_pci_probe_main() can return success despite a reset being scheduled.
Catch this and retry or abort probe depending on the reset type.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 22:08:16 -08:00
Ben Hutchings
766ca0fa6b sfc: Rework MAC, PHY and board event handling
From: Steve Hodgson <shodgson@solarflare.com>

MAC, PHY and board events may be separately enabled and signalled.
Our current arrangement of chaining the polling functions can result
in events being missed.  Change them to be more independent.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 21:59:24 -08:00
Ben Hutchings
177dfcd80f sfc: Add support for sub-10G speeds
The SFC4000 has a separate MAC for use at sub-10G speeds.  Introduce
an efx_mac_operations structure with implementations for the two MACs.
Switch between the MACs as necessary.

PHY settings are independent of the MAC, so add get_settings() and
set_settings() to efx_phy_operations.  Also add macs field to indicate
which MACs the PHY is connected to.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 21:58:17 -08:00
Ben Hutchings
06d5e19318 sfc: Add phy_type device attribute
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 21:57:59 -08:00
Ben Hutchings
f31a45d2f4 sfc: Remove MII extension cruft
Replace efx_nic::link_options bitfield with link_speed (speed in
Mbit/s) and link_fd (full duplex flag).

Remove broken auto-negotiation functions.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 21:57:57 -08:00
Ben Hutchings
56536e9ce7 sfc: Provide hints to irqbalance daemon
Allocate IRQs with the name format <device>[-<type>]-<number> so that
future versions of irqbalanced understand what we're doing.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 21:57:45 -08:00
Neil Turton
28b581ab0a sfc: Add option to use a separate channel for TX completions
In a bidirectional forwarding test, we find that the best performance
is achieved by sending the TX completion interrupts from one NIC to a
CPU which shares an L2 cache with RX completion interrupts from the
other NIC.  To facilitate this, add an option (through a module
parameter) to create separate channels for RX and TX completion with
separate IRQs when MSI-X is available.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 21:57:44 -08:00
Ben Hutchings
84ae48fe4c sfc: Restore phy_flash_cfg module parameter
This is needed for recovery in case a PHY firmware upgrade is aborted.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 21:57:44 -08:00
Steve Hodgson
1ab00629f0 sfc: Make reset_workqueue driver-global rather than per-NIC
Each reset is serialised by the rtnl_lock anyway, so there's no win
per-NIC.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 21:56:21 -08:00
Ben Hutchings
a17102b1ba sfc: Board support fixes
Set dummy monitor method for unrecognised boards.

Clean up board resources if efx_pci_probe_main() fails after board has
been initialised.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-12 21:28:20 -08:00
Stephen Hemminger
c3ecb9f323 sfc: convert to net_device_ops
Convert driver to new net_device_ops. Compile tested only.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-21 17:32:54 -08:00
Ben Hutchings
739bb23d72 sfc: Do not reset when hardware monitor detects a fault
The TX watchdog should trigger a reset, but a temperature/power alarm
should not as this is unlikely to solve the problem.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-11-06 00:50:15 -05:00
Ben Hutchings
f41507245e sfc: Expose flash region storing boot code as MTD
The boot code that appears as a PCI expansion ROM on the SFC4000 is
stored in flash.  Expose this as a standard MTD device to allow for
in-place upgrades.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-11-06 00:49:57 -05:00
Johannes Berg
e174961ca1 net: convert print_mac to %pM
This converts pretty much everything to print_mac. There were
a few things that had conflicts which I have just dropped for
now, no harm done.

I've built an allyesconfig with this and looked at the files
that weren't built very carefully, but it's a huge patch.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-10-27 17:06:18 -07:00
Ben Hutchings
6bc5d3a933 sfc: Make queue flushes more reliable
Increase the potential retry count for RX flushes from 5 to 100.

Stop polling the RX_DESC_PTR_TBL to infer that a flush might have
happened. Instead absolutely rely on the flush events, unless bug 7803
applies (Falcon rev A only).

To keep things quick, request flushes for every TX and RX queue up
front, and match up the events to requests.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-24 18:54:37 -04:00
Ben Hutchings
ef08af03ef sfc: Remove workaround for old firmware bug
There was a bug in XAUI synchronisation in early 10Xpress firmware
versions.  This is fixed in released firmware and we do not need to
work around it.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-24 18:54:35 -04:00
Ben Hutchings
e1074a0d96 sfc: Fix memory BAR release call on error path
Match pci_request_region() with pci_release_region(), not
release_mem_region().

From: Steve Hodgson <shodgson@solarflare.com>
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-24 18:54:35 -04:00
Ben Hutchings
a816f75ac5 sfc: Rework efx_set_multicast_hash()
When !port_enabled, defer the write to reconfigure_mac_wrapper.

Whilst here, simplify the logic now that efx_start_port() always calls
efx_reconfigure_port().

From: Steve Hodgson <shodgson@solarflare.com>
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:49 -04:00
Ben Hutchings
3c78708fe8 sfc: Remove the STATE_RESETTING flag
This was originally a kludge to fix broken locking, which has since
been fixed properly.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:49 -04:00
Ben Hutchings
8c8661e4ce sfc: Extend self-tests
Include PMA/PMD in loopback self-tests as intended.

Add NVRAM checksum validation and include it in self-tests.

Add register self-tests.

Run PHY self-tests where available.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:48 -04:00
Ben Hutchings
2467ca46b6 sfc: Cleanup reset code
Move more code from efx_reset() into efx_reset_down() and efx_reset_up().

Stop propagating MAC/PHY setting failures from efx_reset_down() and
efx_reset_up() as these should not be fatal.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:48 -04:00
Ben Hutchings
bc3c90a2b7 sfc: Remove some unreachable error paths
Some functions return an error code which is always 0.  Change their
return types to void and simplify their callers accordingly.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:48 -04:00
Ben Hutchings
01aad7b6ff sfc: Remove efx_nic_dummy_op_int() as redundant with efx_port_dummy_op_int()
Update comment on the dummy operation implementations.

Line up the board operation initialisers.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:47 -04:00
Ben Hutchings
740847dab1 sfc: Enable TSO for 802.1q VLAN devices
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:47 -04:00
Steve Hodgson
dacccc7411 sfc: Do not call netif_{stop,wake}_queue() before register_netdev
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:47 -04:00
Ben Hutchings
f8b87c1701 sfc: Make PHY flash mode a device attribute, not a module parameter
This allows updating PHY firmware for one interface without removing
all other interfaces handled by the driver.

Replace tx_disabled flags and 10Xpress status enumeration with flags in
enum efx_phy_mode.

Prevent an interface from being brought up while in PHY flash mode.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:47 -04:00
Ben Hutchings
42cbe2d73c sfc: Cleanup RX event processing
Make efx_process_channel() and falcon_process_eventq() return the
number of packets received rather than updating the quota, consistent
with new NAPI.

Since channels and RX queues are mapped one-to-one, remove return
value from falcon_handle_rx_event() and add a warning for events
with the wrong RX queue number.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:47 -04:00
Ben Hutchings
d307402534 sfc: Remove efx_channel::evqnum field
It is redundant with efx_channel::channel.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:46 -04:00
Ben Hutchings
8831da7b6c sfc: Cleanup RX queue information
Rename efx_nic::rss_queues to the more obvious n_rx_queues

Remove efx_rx_queue::used and other stuff that's redundant with it.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:46 -04:00
Ben Hutchings
64ee3120f7 sfc: Remove efx_channel::has_interrupt
efx_channel::has_interrupt is redundant with efx_channel::used_flags.

Remove efx_test_eventq() because it is now obviously unreachable.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:46 -04:00
Ben Hutchings
46123d043d sfc: Move CPU counting for RSS into a separate function, efx_wanted_rx_queues()
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-09-03 09:53:45 -04:00