30 Oct, 2009
2 commits
-
Program the receive pause thresholds differently depending on
chip version. This cloned from from the vendor (GPL) driver.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
This adds infrastructure for the newer chip versions and workarounds.
Extracted from the vendor (GPL) driver.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller
02 Oct, 2009
1 commit
-
This is based on Michal Schmidt fix for skge.
Most network drivers request their IRQ when the interface is activated.
sky2 does it in ->probe() instead, because it can work with two-port
cards where the two net_devices use the same IRQ. This works fine most
of the time, except in some situations when the interface gets renamed.
Consider this example:1. modprobe sky2
The card is detected as eth0 and requests IRQ 17. Directory
/proc/irq/17/eth0 is created.
2. There is an udev rule which says this interface should be called
eth1, so udev renames eth0 -> eth1.
3. modprobe 8139too
The Realtek card is detected as eth0. It will be using IRQ 17 too.
4. ip link set eth0 up
Now 8139too requests IRQ 17.The result is:
WARNING: at fs/proc/generic.c:590 proc_register ...
proc_dir_entry '17/eth0' already registeredThe fix is for sky2 to name the irq based on the pci device, as is done
by some other devices DRM, infiniband, ... ie. sky2@pci:0000:00:00Signed-off-by: Stephen Hemminger
Reviewed-by: Michal Schmidt
Signed-off-by: David S. Miller
02 Sep, 2009
2 commits
-
B0_CTST is a 24bit register according to the vendor driver (sk98lin).
A 16bit read on B0_CTST will always return 0 for Y2_VAUX_AVAIL (1<
Acked-by: Stephen Hemminger
Signed-off-by: David S. Miller -
Observed by Mike McCormack.
The LED bit here is just a software controlled value used to
turn on one of the LED's on some boards. The register value was wrong,
which could have been causing some power control issues.
Get rid of problematic define use the correct mask.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller
19 Aug, 2009
4 commits
-
Recycling turns out to be a bad idea! For most use cases, the
packet can not be reused: TCP packets are cloned. Even for the ideal
case of forwarding, it hurts performance because of CPU ping/pong.
On a multi-core system forwarding of 64 byte packets is worse
much worse: recycling = 24% forwarded vs no recycling = 42% forwardedSigned-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
Don't reference the list element in hardware transmit ring on transmit
completion. The list element is updated by hardware, therefore
it causes a cache miss. Do book keeping in software structure.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
Allocate and size transmit ring based on parameters. Saves excess
space and allows configuring larger rings for testing.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
This patch saves elements on transmit ring by only updating the upper
64 bit address when it changes. With many workloads skb's are located
in same region, so it saves space.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller
15 Aug, 2009
2 commits
-
The whole restarting flag was introduced by Mike McCormack
and was a temporary duct tape patch around issues with transmits
inflight during restart. The problems it was covering are now
fixed and the code should have been reverted.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
The sky2 driver combines auto speed negotiation with automatic negotiation
of pause parameters; but the ethtool interface expects them to be
split. This patch allows autonegotiation to be used for speed, but
manually disable flow control.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller
04 Aug, 2009
1 commit
-
This patch supersedes my previous patch "sky2: Avoid transmitting
during sky2_restart".I have reworked the patch to avoid crashes during both sky2_restart()
and sky2_set_ringparam().Without this patch, the sky2 driver can be crashed by doing:
# pktgen eth1 & (transmit many packets on eth1)
# ethtool -G eth1 tx 510I am aware you object to storing extra state, but I can't see a way
around this. Without remembering that we're restarting,
netif_wake_queue() is called in the ISR from sky2_tx_complete(), and
netif_tx_lock() is used in sky2_tx_done(). If anybody can see a way
around this, please let me know.Signed-off-by: Mike McCormack
Signed-off-by: David S. Miller
18 Jun, 2009
1 commit
-
This patch implements skb recycling. It reclaims transmitted skb's
for use in the receive ring.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller
07 Aug, 2008
1 commit
-
Fix the problems reported for 2.6.27-rc1 caused by over aggressive
power management. Turning clock off on PCI Express is problematic for WOL,
and when doing multi-booting.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
18 Jun, 2008
1 commit
-
Add support for Yukon 2 Ultra 2 chip set (88E8057) based on code in latest
version of vendor driver (sk98lin 10.60.2.3). Untested on real hardware.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
31 May, 2008
2 commits
-
Turn on special bits to save more power when device is shutdown.
Tested on a limited range of hardware, some of the bits are for hardware
that probably isn't even in production (like Yukon Supreme) and was ported
from the vendor driver.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik -
Put PHY int sleep mode (from vendor sk98lin 10.50 driver) when the
network device is brought down.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
13 May, 2008
1 commit
-
noticed while browsing code, apparent thinko. compile tested only.
Signed-off-by: Jesse Brandeburg
CC: Stephen Hemminger
Signed-off-by: Jeff Garzik
24 Feb, 2008
1 commit
-
Fix problems in LED management, so ethtool -p works correctly on Yukon-EC
and other chips. The driver was incorrectly setting the PHY LED overide bits.
Moral: read the spec sheet, not the vendor driver.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
06 Feb, 2008
1 commit
-
The Yukon FE chip has a ram buffer therefore it needs the alignment
restriction and hang check workarounds.Therefore:
* Autodetect the prescence/absence of ram buffer
* Rename the flag value to reflect this
* Use it consistently (ie don't reread register)Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
29 Jan, 2008
1 commit
-
Add support from sk98lin vendor driver 10.50.1.3 for 88E8055 and
88E8075 chips. I don't have this hardware to test, so this changes
are untested.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
13 Jan, 2008
1 commit
-
This patch might fix problems with 4G or more of memory.
It stops the driver from doing a small optimization for Tx and Rx,
and instead always sets the high-page on tx/rx descriptors.Fixes-bug: http://bugzilla.kernel.org/show_bug.cgi?id=9725
Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
02 Dec, 2007
1 commit
-
Using the hardware window into PCI config space is more reliable
and smaller/faster than using the pci_config routines. It avoids issues
with MMCONFIG etc.Reverts: 167f53d05fccb47b6eeadac7f6705b3f2f042d03
Please apply for 2.6.24
Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
10 Nov, 2007
1 commit
-
The PCI AER support may not work for a couple of reasons.
It may not be configured into the kernel or there may be a BIOS
bug that prevents MMCONFIG from working. If MMCONFIG doesn't work
then the PCI registers that control AER will not be accessible via
pci_read_config functions; luckly there is another window to access
PCI space in the device, so use that.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
12 Oct, 2007
1 commit
-
Use builtin statistics structure from net device.
Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller
11 Oct, 2007
4 commits
-
Use the PCI layer config access functions. The driver was using the
memory mapped window in device, to workaround issues accessing the
advanced error reporting registers.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
Signed-off-by: David S. Miller -
Use the kernel interfaces for advanced error reporting.
This should be cleaner and clear up errors on boot.For those systems with busted BIOS's that don't correctly
support mmconfig, advanced error reporting will be disabled.
The PCI registers for advanced error reporting start at 0x100 which
is too large to be accessed by legacy functions.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik -
Add documentation of GPHY_CTRL register bits even if driver
is not using them (yet).Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik -
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
Signed-off-by: David S. Miller
25 Sep, 2007
1 commit
-
Be more selective about when to enable the ram buffer watchdog code.
It is unnecessary on XL A3 or later revs, and with Yukon FE
the buffer is so small (4K) that the watchdog detects false positives.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
21 Sep, 2007
4 commits
-
A driver writer from another operating system hinted that
the versions of Yukon 2 chip with rambuffer (EC and XL) have
a hardware bug that if the FIFO ever gets completely full it
will hang. Sounds like a classic ring full vs ring empty wrap around
bug.As a workaround, use the existing watchdog timer to check for
ring full lockup.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik -
Add support for newest Marvell chips.
The Yukon FE plus chip is found in some not yet released laptops.
Tested on hardware evaluation boards.This version of the patch is for 2.6.23. It supersedes
the two previous patches that are sitting in netdev-2.6 (upstream branch).Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik -
This patch should cause no functional changes in driver behaviour.
There are (too) many revisions of the Yukon 2 chip now. Instead of
adding more conditionals based on chip revision; rerganize into a
set of feature flags so adding new versions is less problematic.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik -
The length check for truncated frames was not correctly handling
the case where VLAN acceleration had already read the tag.
Also, the Yukon EX has some features that use high bit of status
as security tag.Signed-off-by: Pierre-Yves Ritschard
Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
25 Aug, 2007
1 commit
-
This fixes the extra timer overhead that people were whining about
as a 2.6.23 regression.Running the watchdog timer all the time is unneeded. Change it
to run only if link is up, and reduce frequency to save power.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
11 Jul, 2007
1 commit
-
Add an optional debug interface for displaying state of transmit/receive
rings. Creates a file debugfs/sky2/ethX for each device that is up.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik
09 Jul, 2007
4 commits
-
Enable support for Yukon EX chipset (88e8071).
Most of changes are related to new commands to chip for transmit,
and change in status and checksumming.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik -
The General Purpose I/O register is yet another hardware workaround
catchall. Enable workaround that vendor driver does to stay
but for bug compatiable.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik -
This register is more of a test and control register on Yukon2.
So rename it to Q_TEST and give some bit definitions.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik -
Need to setup more PCI control control registers are on Yukon EX.
Some of these also exist on Yukon EC-U as well.Signed-off-by: Stephen Hemminger
Signed-off-by: Jeff Garzik