spelling
feedback and ok tb@ jmc@ ok ratchov@

Add a DVACT_WAKEUP op to the *_activate() API. This is called after the
kernel resumes normal (non-cold, able to run processes, etc) operation.
Previously we were relying on specific DVACT_RESUME op

Add a DVACT_WAKEUP op to the *_activate() API. This is called after the
kernel resumes normal (non-cold, able to run processes, etc) operation.
Previously we were relying on specific DVACT_RESUME op's in drivers
creating callback/threads themselves, but that has become too common,
indicating the need for a built-in mechanism.
ok dlg kettenis, tested by a sufficient amount of people

show more ...

replace workqs with tasks for handling resume

from kimberley manning

After suspend/resume, reload the firmware. Skip firmware on
FXP_REV_82550_C with server extensions.

remove the powerhook code. All architectures now use the ca_activate tree
traversal code to suspend/resume
ok oga kettenis blambert

Since the pci-specific activate function does nothing, move it to fxp.c.
Then it is possible for fxp_powerhook to use it.

ca_activate for suspend/resume; tested by andrew@afresh1.com

Add detach support to a few more drivers, and in others do the neccessary
operations in the detach function in the right order. Also ensure that the
interrupt handlers not trust registers that go aw

Add detach support to a few more drivers, and in others do the neccessary
operations in the detach function in the right order. Also ensure that the
interrupt handlers not trust registers that go away.
read over very carefully by dms, tested by me

show more ...

More cases of shutdown hooks not needed after card is already stopped. In
these cases the xxstop function is a bit more complicated and has a flag of
some sort, but the use of that flag does not mat

More cases of shutdown hooks not needed after card is already stopped. In
these cases the xxstop function is a bit more complicated and has a flag of
some sort, but the use of that flag does not matter; DMA is already ceased
ok dlg

show more ...

Make fxp at pci detachable; untested.

Some fxp(4) chips do not like having their status/command registers accessed
as 8-bit reads/writes even though they are 16-bit. This can cause the
respective interfaces to lock up until the system is

Some fxp(4) chips do not like having their status/command registers accessed
as 8-bit reads/writes even though they are 16-bit. This can cause the
respective interfaces to lock up until the system is hard rebooted. Having
the driver read/write to these registers as 16-bit entities resolves the
instability.

From mickey@

ok krw@ sthen@ dlg@

show more ...

Gently nudge up the interrupt coalescing timeout to 128 microseconds. This
gives the firmware time to do its job. Makes my fxps go from 12000int/sec
while moving 100Mbps to 3000int/sec
ok beck claudio

put more foo_detach() routines where they are used

(fxp/re)_attach_common -> (fxp/re)_attach

Disable the minimum size frame bundling when using the CPUSaver microcode.
This could cause 2 interrupts per frame if the frame size is below 128 bytes.

From Dmitry Bogdan <dsb at imcs dot dvgu dot

Disable the minimum size frame bundling when using the CPUSaver microcode.
This could cause 2 interrupts per frame if the frame size is below 128 bytes.

From Dmitry Bogdan <dsb at imcs dot dvgu dot ru>

show more ...

- fix reception of VLAN sized frames on 82558 and up
- add support for reception of VLAN sized frames on 82557

rev 1.85

Newer chips do not need the receiver lock-up workaround; detect when it
is required.

From NetBSD

use PCI Memory Write and Invalidate; from bogdan@eastonline.ru, ok pascoe

change FXPF_UCODE to mean "firmware load attempted". if the first
loadfirmware() fails because the file is missing, we do not want to
try again when the first softclock -> fxp_stats_update -> fxp_in

change FXPF_UCODE to mean "firmware load attempted". if the first
loadfirmware() fails because the file is missing, we do not want to
try again when the first softclock -> fxp_stats_update -> fxp_init
happens later. calling namei in that context is really bad; tested by mcbride

show more ...

meant to commit this too, remove old resume workaround flags

interrupt coalescing on post 82557 chips; from freebsd via Dmitry Bogdan <dsb@poi.dvo.ru>

stop passing an empty enaddr around since we figure out the
MAC address in the common attach code and not the PCI/CardBus
specific portions, this was a side effect of the code being
split to support

stop passing an empty enaddr around since we figure out the
MAC address in the common attach code and not the PCI/CardBus
specific portions, this was a side effect of the code being
split to support CardBus.

ok beck@

show more ...

fix from netbsd to twiddle the fxp eeprom to disable dynamic standby mode
on cards that have issues with this creating pci errors in 10 mbps mode
tested by many, including me, nick, and pval.
ok jaso

fix from netbsd to twiddle the fxp eeprom to disable dynamic standby mode
on cards that have issues with this creating pci errors in 10 mbps mode
tested by many, including me, nick, and pval.
ok jason@, markus@

show more ...

Howe eye wysh eye cud spel.

Typos: defintion -> definition, proccess -> process

There are more occurences hiding in binutils, lynx and afs but I am too
lazy to report them upstream at the moment.