It's been a useless no-op for long enough in 2.6 so I figured it's time to
remove it. The number of people that could object because they're
maintaining unified 2.4 and 2.6 drivers is probably rather small.
[ Handled drivers added by netdev tree and some missed IRDA cases... -DaveM ]
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Several devices have multiple independant RX queues per net
device, and some have a single interrupt doorbell for several
queues.
In either case, it's easier to support layouts like that if the
structure representing the poll is independant from the net
device itself.
The signature of the ->poll() call back goes from:
int foo_poll(struct net_device *dev, int *budget)
to
int foo_poll(struct napi_struct *napi, int budget)
The caller is returned the number of RX packets processed (or
the number of "NAPI credits" consumed if you want to get
abstract). The callee no longer messes around bumping
dev->quota, *budget, etc. because that is all handled in the
caller upon return.
The napi_struct is to be embedded in the device driver private data
structures.
Furthermore, it is the driver's responsibility to disable all NAPI
instances in it's ->stop() device close handler. Since the
napi_struct is privatized into the driver's private data structures,
only the driver knows how to get at all of the napi_struct instances
it may have per-device.
With lots of help and suggestions from Rusty Russell, Roland Dreier,
Michael Chan, Jeff Garzik, and Jamal Hadi Salim.
Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra,
Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan.
[ Ported to current tree and all drivers converted. Integrated
Stephen's follow-on kerneldoc additions, and restored poll_list
handling to the old style to fix mutual exclusion issues. -DaveM ]
Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This driver erroneously zeros dev->tx_queue_len, since
mp->tx_ring_size has not yet been initialized. Actually,
the driver shouldn't modify tx_queue_len at all and should
leave the value set by alloc_etherdev(), currently 1000.
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Commit 468d09f894 masked the "state"
interrupt (bit 20 of the cause register). This results in Radstone's
PPC7D repeatedly re-entering the interrupt routine, locking up the
board. The following patch returns the required handling for this
interrupt.
Signed-off-by: Martyn Welch <martyn.welch@radstone.co.uk>
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
There is no good reason for the asymmetry in the parameters of
eth_port_uc_addr_get() and eth_port_uc_addr_set(). Make them
symmetric. Remove some gratuitous block comments while we're here.
Signed-off-by: Gabriel Paubert <paubert@iram.es>
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
One less thing for drivers writers to worry about.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In this driver, the default ethernet address is first set by by calling
eth_port_uc_addr_get() which reads the relevant registers of the
corresponding port as initially set by firmware. However that function
used the port_num field accessed through the private area of net_dev
before it was set.
The result was that one board I have ended up with the unicast address
set to 00:00:00:00:00:00 (only port 1 is connected on this board). The
problem appeared after commit 84dd619e4d.
This patch fixes the bug by setting mp->port_num prior to calling
eth_port_uc_get_addr().
Signed-off-by: Gabriel Paubert <paubert@iram.es>
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
We were using the platform_device.id field to identify which ethernet
port is used for mv643xx_eth device. This is not generally correct.
It will be incorrect, for example, if a hardware platform uses a single
port but not the first port. Here, we add an explicit port_number field
to struct mv643xx_eth_platform_data.
This makes the mv643xx_eth_platform_data structure required, but that
isn't an issue since all users currently provide it already.
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
The information contained within platform_data should be self-contained.
Replace the pointer to a MAC address with the actual MAC address in
struct mv643xx_eth_platform_data.
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
The driver contains this little piece of candy:
#if defined(CONFIG_DMA_NONCOHERENT) || defined(CONFIG_NOT_COHERENT_CACHE)
#define ETH_DMA_ALIGN L1_CACHE_BYTES
#else
#define ETH_DMA_ALIGN 8
#endif
Any reason why we're not using dma_get_cache_alignment() instead?
Ralf
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
mv643xx_eth: Fix race condition in mv643xx_eth_free_tx_descs
This bug was found and isolated by Thibaut VARENE <T-Bone@parisc-linux.org>
and Jarek Poplawski <jarkao2@o2.pl>. This patch is a modification of their
fixes. We acquire and release the lock for each descriptor that is freed
to minimize the time the lock is held.
Signed-off-by: Jeff Garzik <jeff@garzik.org>
This reverts commit 4596c75c23 as
requested by Olaf Hering. It causes compile errors, and says Olaf:
"This change is also wrong, the autoloading works perfect with 2.6.18,
no need to add random PCI ids.
See commit a0245f7ad5, platform devices
have now a modalias entry in sysfs. The network card is not a PCI
device."
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Maintain a per-CPU global "struct pt_regs *" variable which can be used instead
of passing regs around manually through all ~1800 interrupt handlers in the
Linux kernel.
The regs pointer is used in few places, but it potentially costs both stack
space and code to pass it around. On the FRV arch, removing the regs parameter
from all the genirq function results in a 20% speed up of the IRQ exit path
(ie: from leaving timer_interrupt() to leaving do_IRQ()).
Where appropriate, an arch may override the generic storage facility and do
something different with the variable. On FRV, for instance, the address is
maintained in GR28 at all times inside the kernel as part of general exception
handling.
Having looked over the code, it appears that the parameter may be handed down
through up to twenty or so layers of functions. Consider a USB character
device attached to a USB hub, attached to a USB controller that posts its
interrupts through a cascaded auxiliary interrupt controller. A character
device driver may want to pass regs to the sysrq handler through the input
layer which adds another few layers of parameter passing.
I've build this code with allyesconfig for x86_64 and i386. I've runtested the
main part of the code on FRV and i386, though I can't test most of the drivers.
I've also done partial conversion for powerpc and MIPS - these at least compile
with minimal configurations.
This will affect all archs. Mostly the changes should be relatively easy.
Take do_IRQ(), store the regs pointer at the beginning, saving the old one:
struct pt_regs *old_regs = set_irq_regs(regs);
And put the old one back at the end:
set_irq_regs(old_regs);
Don't pass regs through to generic_handle_irq() or __do_IRQ().
In timer_interrupt(), this sort of change will be necessary:
- update_process_times(user_mode(regs));
- profile_tick(CPU_PROFILING, regs);
+ update_process_times(user_mode(get_irq_regs()));
+ profile_tick(CPU_PROFILING);
I'd like to move update_process_times()'s use of get_irq_regs() into itself,
except that i386, alone of the archs, uses something other than user_mode().
Some notes on the interrupt handling in the drivers:
(*) input_dev() is now gone entirely. The regs pointer is no longer stored in
the input_dev struct.
(*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does
something different depending on whether it's been supplied with a regs
pointer or not.
(*) Various IRQ handler function pointers have been moved to type
irq_handler_t.
Signed-Off-By: David Howells <dhowells@redhat.com>
(cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
Some stats reported by ethtool -S on mv643xx_eth device are cleared
between each call. This patch fixes it.
Signed-off-by: Maxime Bizon <mbizon@freebox.fr>
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
* 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6: (217 commits)
net/ieee80211: fix more crypto-related build breakage
[PATCH] Spidernet: add ethtool -S (show statistics)
[NET] GT96100: Delete bitrotting ethernet driver
[PATCH] mv643xx_eth: restrict to 32-bit PPC_MULTIPLATFORM
[PATCH] Cirrus Logic ep93xx ethernet driver
r8169: the MMIO region of the 8167 stands behin BAR#1
e1000, ixgb: Remove pointless wrappers
[PATCH] Remove powerpc specific parts of 3c509 driver
[PATCH] s2io: Switch to pci_get_device
[PATCH] gt96100: move to pci_get_device API
[PATCH] ehea: bugfix for register access functions
[PATCH] e1000 disable device on PCI error
drivers/net/phy/fixed: #if 0 some incomplete code
drivers/net: const-ify ethtool_ops declarations
[PATCH] ethtool: allow const ethtool_ops
[PATCH] sky2: big endian
[PATCH] sky2: fiber support
[PATCH] sky2: tx pause bug fix
drivers/net: Trim trailing whitespace
[PATCH] ehea: IBM eHEA Ethernet Device Driver
...
Manually resolved conflicts in drivers/net/ixgb/ixgb_main.c and
drivers/net/sky2.c related to CHECKSUM_HW/CHECKSUM_PARTIAL changes by
commit 84fa7933a3 that just happened to be
next to unrelated changes in this update.
Replace CHECKSUM_HW by CHECKSUM_PARTIAL (for outgoing packets, whose
checksum still needs to be completed) and CHECKSUM_COMPLETE (for
incoming packets, device supplied full checksum).
Patch originally from Herbert Xu, updated by myself for 2.6.18-rc3.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The last minute fix submitted by the author fixed a bug, but
broke the driver build.
Noticed by Al Viro, since I can't build on said platform.
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Fix a missing call to dma_unmap_single() in the receive path. Without
this call, errors have been observed on non-cache-coherent systems.
Signed-off-by Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
The linearisation operation doesn't need to be super-optimised. So we can
replace __skb_linearize with __pskb_pull_tail which does the same thing but
is more general.
Also, most users of skb_linearize end up testing whether the skb is linear
or not so it helps to make skb_linearize do just that.
Some callers of skb_linearize also use it to copy cloned data, so it's
useful to have a new function skb_linearize_cow to copy the data if it's
either non-linear or cloned.
Last but not least, I've removed the gfp argument since nobody uses it
anymore. If it's ever needed we can easily add it back.
Misc bugs fixed by this patch:
* via-velocity error handling (also, no SG => no frags)
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
On Sat, Mar 11, Olaf Hering wrote:
> Why is the /sys/class/net/eth0/device symlink not created for the
> mv643xx_eth driver? Does this work for other platform device drivers?
> Seems to work for the ps2 keyboard at least.
The SET_NETDEV_DEV has to be done before a call to register_netdev. With
the new patch below, the device symlink for the platform device was
created. Unfortunately, after the 4 ls commands, the network connection
died. No idea if the box crashed or if something else broke, lost remote
access.
Provide sysfs 'device' in /class/net/ethN Also, set module owner field,
like pcnet32 driver does.
Signed-off-by: Olaf Hering <olh@suse.de>
Acked-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
After resetting the hardware on a tx_timeout, call netif_wake_queue()
only if we have free tx descriptors.
Also, attempt to recover if mv643xx_eth_start_xmit() is called when
there are fewer free tx descriptors than expected.
The BUG_ON() call we are replacing was hit on a tx_timeout that
called netif_wake_queue(), indirectly via netif_device_attach(),
even though we did not have enough free tx descriptors.
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Fix the tx interrupt handler to free completed tx descriptors even
when NAPI is enabled. Otherwise, the tx queue would fill up resulting
in poor performance and "NETDEV WATCHDOG: <iface>: transmit timed out"
messages.
Signed-off-by: Brent Cook <bcook@bpointsys.com>
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
The task level rx queue refill feature hasn't ever worked
(at least in 2.6) and is of dubious value. Remove it.
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Simplify and remove redundant code for filling transmit descriptors.
No changes in features; it's just a code reorganization/cleanup.
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Use better terminology for HW queues. No functional changes.
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>