CVS log for src/sys/dev/ic/elink3.c

Up to [cvs.NetBSD.org] / src / sys / dev / ic

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

sys: Drop redundant NULL check before m_freem(9)

m_freem(9) safely has accepted NULL argument at least since 4.2BSD:
https://www.tuhs.org/cgi-bin/utree.pl?file=4.2BSD/usr/src/sys/sys/uipc_mbuf.c

Compile-tested on amd64/ALL.

Suggested by knakahara@

epget: Tiny clean up for 16-bit buses.

- (offset & 1) must be zero, since IP header is aligned.

- Use bus_space_read_1(9) instead of bus_space_read_multi_1(9) for
  single-byte read.

Merge changes from current as of 20200406

Sync with head.

Use callout_setfunc() / callout_schedule().

Use ifmedia_fini().

Adopt <net/if_stats.h>.

Sync with HEAD

No functional change:
 - Simplify MII structure initialization and reference.
 - KNF

 Whilespace fix. No functional change.

 In drivers which use MII(4) and have hook SIOC[GS]IFMEDIA which just pass to
ifmedia_ioctl(), the hook is not required because ether_ioctl has it
(if_ethersubr.c rev. 1.160). These drivers don't return ENETRESET in
ifmedia_ioctl(), so no functional change.

 Remove very old IFF_NOTRAILERS flag.

- add or adjust /* FALLTHROUGH */ where appropriate
- add __unreachable() after functions that can return but won't in
  this case, and thus can't be marked __dead easily

Sync with HEAD

 Change MII PHY read/write API from:

	int (*mii_readreg_t)(device_t, int, int);
	void (*mii_writereg_t)(device_t, int, int, int);
to:

	int (*mii_readreg_t)(device_t, int, int, uint16_t *);
	int (*mii_writereg_t)(device_t, int, int, uint16_t);

Now we can test if a read/write operation failed or not by the return value.

 In 802.3 spec says that the PHY shall not respond to read/write transaction
to the unimplemented register(22.2.4.3). Detecting timeout can be used to
check whether a register is implemented or not (if the register conforms to
the spec). ukphy(4) can be used this for MII_MMDACR and MII_MMDAADR.

 Note that I noticed that the following code do infinite loop in the
read/wirte function. If it accesses unimplemented PHY register, it will hang.
It should be fixed:

	arm/at91/at91emac.c
	arm/ep93xx/epe.c
	arm/omap/omapl1x_emac.c
	mips/ralink/ralink_eth.c
	arch/powerpc/booke/dev/pq3etsec.c(read)
	dev/cadence/if_cemac.c <- hkenken
	dev/ic/lan9118.c


Tested with the following device:

	axe+ukphy
	axe+rgephy
	axen+rgephy	(tested by Andrius V)
	wm+atphy
	wm+ukphy
	wm+igphy
	wm+ihphy
	wm+makphy
	sk+makphy
	sk+brgphy
	sk+gentbi
	msk+makphy
	sip+icsphy
	sip+ukphy
	re+rgephy
	bge+brgphy
	bnx+brgphy
	gsip+gphyter
	rtk+rlphy
	fxp+inphy	(tested by Andrius V)
	tlp+acphy
	ex+exphy
	epic+qsphy
	vge+ciphy	(tested by Andrius V)
	vr+ukphy	(tested by Andrius V)
	vte+ukphy	(tested by Andrius V)

Not tested (MAC):
	arm:at91emac
	arm:cemac
	arm:epe
	arm:geminigmac
	arm:enet
	arm:cpsw
	arm:emac(omac)
	arm:emac(sunxi)
	arm:npe
	evbppc:temac
	macppc:bm
	macppc:gm
	mips:aumac
	mips:ae
	mips:cnmac
	mips:reth
	mips:sbmac
	playstation2:smap
	powerpc:tsec
	powerpc:emac(ibm4xx)
	sgimips:mec
	sparc:be
	sf
	ne(ax88190, dl10019)
	awge
	ep
	gem
	hme
	smsh
	mtd
	sm
	age
	alc
	ale
	bce
	cas
	et
	jme
	lii
	nfe
	pcn
	ste
	stge
	tl
	xi
	aue
	mue
	smsc
	udav
	url

Not tested (PHY):
	amhphy
	bmtphy
	dmphy
	etphy
	glxtphy
	ikphy
	iophy
	lxtphy
	nsphyter
	pnaphy
	rdcphy
	sqphy
	tlphy
	tqphy
	urlphy

Sync with HEAD

Resolve a couple of conflicts (result of the uimin/uimax changes)

Rename min/max -> uimin/uimax for better honesty.

These functions are defined on unsigned int.  The generic name
min/max should not silently truncate to 32 bits on 64-bit systems.
This is purely a name change -- no functional change intended.

HOWEVER!  Some subsystems have

	#define min(a, b)	((a) < (b) ? (a) : (b))
	#define max(a, b)	((a) > (b) ? (a) : (b))

even though our standard name for that is MIN/MAX.  Although these
may invite multiple evaluation bugs, these do _not_ cause integer
truncation.

To avoid `fixing' these cases, I first changed the name in libkern,
and then compile-tested every file where min/max occurred in order to
confirm that it failed -- and thus confirm that nothing shadowed
min/max -- before changing it.

I have left a handful of bootloaders that are too annoying to
compile-test, and some dead code:

cobalt ews4800mips hp300 hppa ia64 luna68k vax
acorn32/if_ie.c (not included in any kernels)
macppc/if_gm.c (superseded by gem(4))

It should be easy to fix the fallout once identified -- this way of
doing things fails safe, and the goal here, after all, is to _avoid_
silent integer truncations, not introduce them.

Maybe one day we can reintroduce min/max as type-generic things that
never silently truncate.  But we should avoid doing that for a while,
so that existing code has a chance to be detected by the compiler for
conversion to uimin/uimax without changing the semantics until we can
properly audit it all.  (Who knows, maybe in some cases integer
truncation is actually intended!)

Sync with HEAD

Pull up following revision(s) (requested by msaitoh in ticket #938):
	sys/arch/acorn32/podulebus/if_ie.c: revision 1.41
	sys/arch/amiga/dev/if_es.c: revision 1.58
	sys/arch/amiga/dev/if_qn.c: revision 1.45
	sys/arch/arm/at91/at91emac.c: revision 1.20
	sys/arch/arm/ep93xx/epe.c: revision 1.37
	sys/arch/emips/ebus/if_le_ebus.c: revision 1.14
	sys/arch/emips/ebus/if_le_ebus.c: revision 1.15
	sys/arch/mac68k/dev/if_mc.c: revision 1.46
	sys/arch/macppc/dev/am79c950.c: revision 1.39
	sys/arch/newsmips/apbus/if_sn.c: revision 1.40
	sys/arch/next68k/dev/mb8795.c: revision 1.59
	sys/arch/playstation2/dev/if_smap.c: revision 1.25
	sys/arch/playstation2/dev/if_smap.c: revision 1.26
	sys/arch/sun2/dev/if_ec.c: revision 1.28
	sys/arch/sun3/dev/if_ie.c: revision 1.63
	sys/arch/x68k/dev/if_ne_intio.c: revision 1.19
	sys/arch/xen/xen/if_xennet_xenbus.c: revision 1.75
	sys/arch/xen/xen/xennetback_xenbus.c: revision 1.63
	sys/dev/bi/if_ni.c: revision 1.45
	sys/dev/cadence/if_cemac.c: revision 1.12
	sys/dev/ic/am7990.c: revision 1.78
	sys/dev/ic/am79900.c: revision 1.27
	sys/dev/ic/an.c: revision 1.67
	sys/dev/ic/cs89x0.c: revision 1.40
	sys/dev/ic/dm9000.c: revision 1.13
	sys/dev/ic/dm9000.c: revision 1.14
	sys/dev/ic/dp8390.c: revision 1.88
	sys/dev/ic/elink3.c: revision 1.141
	sys/dev/ic/elinkxl.c: revision 1.122
	sys/dev/ic/hme.c: revision 1.98
	sys/dev/ic/i82586.c: revision 1.77
	sys/dev/ic/lance.c: revision 1.53
	sys/dev/ic/mb86950.c: revision 1.27
	sys/dev/ic/mb86960.c: revision 1.86
	sys/dev/ic/mtd803.c: revision 1.34
	sys/dev/ic/pdq_ifsubr.c: revision 1.59
	sys/dev/ic/rrunner.c: revision 1.86
	sys/dev/ic/seeq8005.c: revision 1.58
	sys/dev/ic/sgec.c: revision 1.47
	sys/dev/ic/smc90cx6.c: revision 1.72
	sys/dev/ic/smc91cxx.c: revision 1.96
	sys/dev/ic/tropic.c: revision 1.49
	sys/dev/ic/wi.c: revision 1.245
	sys/dev/isa/if_eg.c: revision 1.93
	sys/dev/isa/if_el.c: revision 1.95
	sys/dev/isa/if_iy.c: revision 1.101
	sys/dev/ofw/ofnet.c: revision 1.58
	sys/dev/pci/if_alc.c: revision 1.27
	sys/dev/pci/if_de.c: revision 1.152
	sys/dev/pci/if_fpa.c: revision 1.61
	sys/dev/pci/if_jme.c: revision 1.34
	sys/dev/pci/if_tl.c: revision 1.108
	sys/dev/pci/if_vte.c: revision 1.19
	sys/dev/pci/ixgbe/ixgbe.h: revision 1.50
	sys/dev/pcmcia/if_cnw.c: revision 1.62
	sys/dev/pcmcia/if_malo_pcmcia.c: revision 1.17
	sys/dev/pcmcia/if_ray.c: revision 1.89
	sys/dev/pcmcia/if_xi.c: revision 1.81
	sys/dev/pcmcia/mhzc.c: revision 1.51
	sys/dev/pcmcia/xirc.c: revision 1.34
	sys/dev/qbus/if_de.c: revision 1.33
	sys/dev/qbus/if_qe.c: revision 1.78
	sys/dev/qbus/if_qt.c: revision 1.22
	sys/dev/sbus/be.c: revision 1.87
	sys/dev/sbus/qe.c: revision 1.68
	sys/dev/scsipi/if_se.c: revision 1.96
	sys/dev/usb/if_atu.c: revision 1.59
	sys/net/if_l2tp.c: revision 1.28 via patch
	sys/net/if_ppp.c: revision 1.160
It's not required to include net/bpfdesc.h. Remove it.
--
Simplify like other drivers. NULL check of ifp->if_bpf is done in
bpf_mtap(), so it's not required to do it here.
--
Remove duplicated inclusion of net/bpf.h.
--
Remove duplicated inclusion of net/bpf.h.
--
Simplify bpf_mtap() call. No functional change.

 Implement the BPF direction filter (BIOC[GS]DIRECTION). It provides backward
compatibility with BIOC[GS]SEESENT ioctl. The userland interface is the same
as FreeBSD.

 This change also fixes a bug that the direction is misunderstand on some
environment by passing the direction to bpf_mtap*() instead of checking
m->m_pkthdr.rcvif.

Sync with HEAD

 It's not required to include net/bpfdesc.h. Remove it.

update from HEAD

Sync with HEAD

Sync with HEAD.  (Note that most of these changes are simply $NetBSD$
tag issues.)

Move bpf_mtap and if_ipackets++ on Rx of each driver to percpuq if_input

The benefits of the change are:
- We can reduce codes
- We can provide the same behavior between drivers
  - Where/When if_ipackets is counted up
  - Note that some drivers still update packet statistics in their own
    way (periodical update)
- Moved bpf_mtap run in softint
  - This makes it easy to MP-ify bpf

Proposed on tech-kern and tech-net

Sync with HEAD

Sync with HEAD

MFREE -> m_free

Sync with HEAD

Introduce m_set_rcvif and m_reset_rcvif

The API is used to set (or reset) a received interface of a mbuf.
They are counterpart of m_get_rcvif, which will come in another
commit, hide internal of rcvif operation, and reduce the diff of
the upcoming change.

No functional change.

Sync with HEAD

Introduce softint-based if_input

This change intends to run the whole network stack in softint context
(or normal LWP), not hardware interrupt context. Note that the work is
still incomplete by this change; to that end, we also have to softint-ify
if_link_state_change (and bpf) which can still run in hardware interrupt.

This change softint-ifies at ifp->if_input that is called from
each device driver (and ieee80211_input) to ensure Layer 2 runs
in softint (e.g., ether_input and bridge_input). To this end,
we provide a framework (called percpuq) that utlizes softint(9)
and percpu ifqueues. With this patch, rxintr of most drivers just
queues received packets and schedules a softint, and the softint
dequeues packets and does rest packet processing.

To minimize changes to each driver, percpuq is allocated in struct
ifnet for now and that is initialized by default (in if_attach).
We probably have to move percpuq to softc of each driver, but it's
future work. At this point, only wm(4) has percpuq in its softc
as a reference implementation.

Additional information including performance numbers can be found
in the thread at tech-kern@ and tech-net@:
http://mail-index.netbsd.org/tech-kern/2016/01/14/msg019997.html

Acknowledgment: riastradh@ greatly helped this work.
Thank you very much!

Sync with HEAD

Convert sys/dev to use <sys/rndsource.h>.

Rebase to HEAD as of a few days ago.

Merge tls-earlyentropy branch into HEAD.

Be a little more clear and consistent about harvesting entropy from devices:

1) deprecate RND_FLAG_NO_ESTIMATE

2) define RND_FLAG_COLLECT_TIME, RND_FLAG_COLLECT_VALUE

3) define RND_FLAG_ESTIMATE_TIME, RND_FLAG_ESTIMATE_VALUE

4) define RND_FLAG_DEFAULT: RND_FLAG_COLLECT_TIME|
   RND_FLAG_COLLECT_VALUE|RND_FLAG_ESTIMATE_TIME

5) Make entropy harvesting from environmental sensors a little more generic
   and remove it from individual sensor drivers.

6) Remove individual open-coded delta-estimators for values from a few
   places in the tree (uvm, environmental drivers).

7) 0 -> RND_FLAG_DEFAULT, actually gather entropy from various drivers
   that had stubbed out code, other minor cleanups.

Resync to 2012-11-19 00:00:00 UTC

sync with head

split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

Fix mii_statchg to take a 'struct ifnet *' instead of device_t.  This fixes
problem with a common MDIO bus used for multiple interfaces.
Some drivers converted to CFATTACL_DECL_NEW.

sync with head

merge to -current.

Entropy-pool implementation move and cleanup.

1) Move core entropy-pool code and source/sink/sample management code
   to sys/kern from sys/dev.

2) Remove use of NRND as test for presence of entropy-pool code throughout
   source tree.

3) Remove use of RND_ENABLED in device drivers as microoptimization to
   avoid expensive operations on disabled entropy sources; make the
   rnd_add calls do this directly so all callers benefit.

4) Fix bug in recent rnd_add_data()/rnd_add_uint32() changes that might
   have lead to slight entropy overestimation for some sources.

5) Add new source types for environmental sensors, power sensors, VM
   system events, and skew between clocks, with a sample implementation
   for each.

ok releng to go in before the branch due to the difficulty of later
pullup (widespread #ifdef removal and moved files).  Tested with release
builds on amd64 and evbarm and live testing on amd64.

sync with head.

sync with head

Sync with HEAD.

Push the bpf_ops usage back into bpf.h. Push the common ifp->if_bpf
check into the inline functions as well the fourth argument for
bpf_attach.

sync with head

Redefine bpf linkage through an always present op vector, i.e.
#if NBPFILTER is no longer required in the client.  This change
doesn't yet add support for loading bpf as a module, since drivers
can register before bpf is attached.  However, callers of bpf can
now be modularized.

Dynamically loadable bpf could probably be done fairly easily with
coordination from the stub driver and the real driver by registering
attachments in the stub before the real driver is loaded and doing
a handoff.  ... and I'm not going to ponder the depths of unload
here.

Tested with i386/MONOLITHIC, modified MONOLITHIC without bpf and rump.

sync with head

Simplify activation routines: don't call mii_activate(), it's a
no-op.  Don't block interrupts, if_deactivate() does it.
Compile-tested, only.

Replace shutdownhook_establish(9) with pmf_device_register1(9) in
MI epconfig() and ep_detach(), and remove pmf(9) calls in pcmcia attachment.
Tested on ep0 at pci, but pcmcia attachment is untested.

sync with head.

Sync with HEAD.

Sync with HEAD.

Sync with wrstuden-revivesa-base-2.

device_t changes, pmf, minor knf.

Sync with HEAD.

sync with head.

sync with head.

Remove clause 3 and 4 from TNF licenses

use aprint_*_dev and device_xname

sync with HEAD

sync with head.

Sync with HEAD.

Follow the merge of pmap.c on i386 and amd64 and move
pmap_init_tmp_pgtbl into arch/x86/x86/pmap.c. Modify the ACPI wakeup
code to restore CR4 before jumping back into kernel space as the large
page option might cover that.

Sync with HEAD.

Sync with head.

machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

Sync with head.

Sync with HEAD.

sync with head.

Sync with HEAD.

Constify: use CLLADDR() instead of LLADDR() wherever possible.

Sync with head.

Merge some of the less invasive changes from the vmlocking branch:

- kthread, callout, devsw API changes
- select()/poll() improvements
- miscellaneous MT safety improvements

Adapt to callout API change.

Sync with HEAD.

Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

sync with head.

sync with head.

Sync with head.

__unused removal on arguments; approved by core.

sync with head

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

sync with head.

sync with head

remove more vestiges of CCITT, LLC, HDLC, NS, and NSIP.

sync with head

Merge from HEAD.

Add power hooks for "ep* at pcmcia?".

ok christos@

sync with head.

sync with head.

Sync with head.

Sync with head.

Add void cast to functions whose return values are ignored.

Sync with head.

sync with head.

Use device_is_active() rather than testing dv_flags for DVF_ACTIVE
directly.

Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

merge ktrace-lwp.

Sync with HEAD. Here we go again...

- add missing const
- fix variable shadowing
- remove unneeded casts

sync with -current

sync with head.  xen and whitespace.  xen part is not finished.

Sync with HEAD.

Hi Perry!

nuke trailing whitespace

sync with head.

Sync with HEAD.

de-__P

Pull up revision 1.108 (requested by thorpej in ticket #939):
  When adding or deleting multicast addresses, only change
  the address filter if the interface is marked RUNNING.
  Fixes PR#27678.

Sync with HEAD.

In epshutdown(), do not attempt to disable the interface when stopping it
because we're about to write to it again (with a global reset).  Disable
the interface explicitly after the reset is written.

Sync with HEAD.

When adding/deleting multicast addresses, only whack the address
filter if the interface is marked RUNNING.

Fixes kern/27678.

Fix the sync with head I botched.

Sync with HEAD.

Sync with HEAD.

Eliminate most 8-bit access.  Just the data in/out path remains.

Use aprint_*().

Remove variables that are only assigned too but not referenced.

Catch up to -current

Change the "dontcare bits" argument to ifmedia_init() to IFM_IMASK
so that it is possible select PHY instances other than the first
one (instance zero), if there is more than one PHY attached.

catch up with -current on kqueue branch

catch up with -current.

Catch up to -current.

make auto const variables static.

Sync kqueue branch with -current.

Catch up to -current.

Clean up the C_ and S_ constants. We don't really need two sets of them.
Also make the interrupt handler for older 3com cards look like the xl one.
I.e. don't ack the interrupt latch bit before checking if it is set.
At the same time, introduce constants for the watched interrupts, so that
we don't copy them all over the place.

Use bus_size_t's for bus space read/write offsets.

Catch up to -current.

add/cleanup RCSID

Sync the thorpej-mips-cache branch with -current.

trivially rename some macro parameters to avoid lint warnings.

Catch up with -current.

Catch up to -current.

Spell 'occurred' with two 'r's.

Catch up with -current.

update to -current

All CORKSCREWs have MII.

The EEPROM registers/bits are different on the Corkscrew (3c515).  Adjust
for this.

bzero -> memset

bcopy -> strcpy

Catch up to -current.

Make sure to allocate big enough mbuf data area before elevating
interrupt priority, to avoid allocating additional mbufs on
splsched(). Move some other misc bits out of splched()-guarded
section. This should make the code a bit more friendly wrt e.g.
clock interrupts. Sligh performance improvement was observed with
this change (like 1-2% on slow machine), even.

This also fixes ancient kern/2478 by Bill Sommerfeld.

epintr(): eliminate write-only variable 'addrandom'

remove unneeded #includes, fix typo

do not use the synchronous read mode for MCA 3c529

Catch up with -current.

Sync with HEAD.

Add MCA ep(4) attachment, for 3Com 3c529-type cards. Tested with 3c529-TP,
provided to me by David Brownlee (thanks!).

Performance of this card is quite poor on my PS/2 with 386DX, like 100KB/s
at best, but as low as 5KB/s when transferring bigger files due to
packet overruns. It would be good to revisit this later, probably by
teaching the ic code to use RX Early.

convert to use ether_ioctl()
fix bunch of typos

Sync with HEAD.

make some more constant arrays 'const'

Sync with HEAD

ALTQ'ify.

Sync with HEAD.

Update thorpej_scsipi to -current as of a month ago
A i386 GENERIC kernel compiles without the siop, ahc and bha drivers
(will be updated later). i386 IDE/ATAPI and ncr work, as well as
sparc/esp_sbus. alpha should work as well (untested yet).
siop, ahc and bha will be updated once I've updated the branch to current
-current, as well as machine-dependant code.

Move bpfattach()/bpfdetach() calls into ether_ifattach()/ether_ifdetach().

Move the check for "promisc + unicast + not for us" into ether_input(),
and change Ethernet drivers to always pass all received frames to
ether_input() (with a few exceptions, which are documented in the
code).

6 -> ETHER_ADDR_LEN

Use splsched() instead of splhigh(), and explain why in a comment,
as well as document a pitfall of the approach being used.

Sync w/ netbsd-1-5-base.

For all network drivers that call ether_ifattach(), and also
have _detach() functions:
  Ensure that softc keeps state about whether the attach succeeded,
  and make the detach function return immediately if the attach did
  not complete.

Pull up revision 1.80 (requested by martin):
  Make the 3Com 3C597 Fast Etherlink TX EISA card work.  Fixes PR#7067.

Avoid stomping on the W0_CONFIG_CTRL register which seems not to be
implemented on EISA cards. This makes a 3Com 3C597 Fast Etherlink TX
work for me at last, fixing PR kern/7067.

Remove register declarations.

New callout mechanism with two major improvements over the old
timeout()/untimeout() API:
- Clients supply callout handle storage, thus eliminating problems of
  resource allocation.
- Insertion and removal of callouts is constant time, important as
  this facility is used quite a lot in the kernel.

The old timeout()/untimeout() API has been removed from the kernel.

No longer necessary to futz with ifp->if_baudrate here.

when change mbuf type, clear m_pkthdr so that we do not get panic with
old data in pointer fields.  NOTE: we'll extend m_pkthdr field soon.

Wondering: do mbuf pre-allocation code in elink3.c really buy us much
performance?  it comes with code duplicate/synchronization problem with
sys/mbuf.h...

- Bail out from attach stage if failed to enable card, number of iospace
or memspace is unexpected, failed to map iospace or failed to configure
the controller properly.
- Disestablish an interrupt handler if failed to enable card.

Pull up revisions 1.59-1.63,1.65 (requested by enami):
  Add support for the 3Com 3c574.  Partially based on PR#8331.

Detach MII if there is.

Don't dry to diving MIIF_NOISOLATE in the PHY drivers.  Instead, pass
flags down from the parent to child vi mii_attach().

Detach rnd source in ep_detach().  Check for DVF_ACTIVE in epintr();

Make sure to untimeout() things on detach.

Move detach operations around a little to simplify (and perhaps get it
right?).  Prompted by Jason, itojun and myself.

Bring some order to the chaos which was the MII code function naming
"conventions".

Make sure to get rid of shutdown hook and ttach media on detach.

Pull up to last week's -current.

Use the common MII bit-bang module.

Sync with -current

Call mii_down() as appropriate.

Pull up 1.65 manually, to skip over API changes.

Revert this part of rev 1.59:
- Be more conservative when resetting the interface after a transmit error.

It caused the interface to totally wedge on the (somewhat common) transmit
FIFO underrun condition.

Adapt to mii_phy_probe() change.

remmove unncecessary empty line.

Call epstop at the beginning of epinit, especially to make sure that
ep_tick is untimeouted before new one is scheduled.  This makes 3C574-TX
works at least in 10Mb/s mode.

Sync w/ trunk.

fix typo in previous commit.

cosmetic changes before i do some functional changes.

- Encapsulate the EEPROM reading code into ep_read_eeprom(), and use
  the correct "read eeprom" opcode on the RoadRunner (which has a larger
  EEPROM, and thus needs a different opcode to make room for the larger
  offsets).
- Reset and enable the MII before probing for PHYs, and reset and enable
  the MII in epinit().
- Be more conservative when resetting the interface after a transmit error.

Based on PR #8331, from Ryoji KATO.

Closer on 3c574, probably will work fine with the 3CCFEM556BI (which
won't have an older rev TDK Semi PHY, and which was tested by the author
of the PR).

Sync w/ -current.

Rework layer 2 protocol input routines.  Instead of calling e.g. ether_input()
directly, call the function pointer (*if_input)(ifp, m).  The input routine
expects the packet header to be at the head of the packet, and will adjust
as necessary.  Privatize the layer 2 input and output routines, allowing
*_ifattach() to set them up as appropriate.

pullup 1.54->1.57 (Jonathan Stone)

Calling DELAY() during polling for command completion of the normal
per-packet discard of every received packet increases latency and
decrease throughput.  INstead, Spin on S_COMMAND_IN_PROGESS for enough
iterations as the expected time to completion.  Only do DELAY()/poll
loops (as suggested by Jason Thorpe) if spinning fails.
Take PCI as fastest case and compute worst-case estimate.

Shorten DELAY() in S_COMMAND_COMPLETE polling-loop up 10, loop-count
down.  to speed up epreset() completion (m/c filter change, ifup/ipdown, etc).

* Make doubly sure that the ep shutdownhooks are run at splnet() or above.
* Clear the `enabled' bit in the softc so late hardware interrupts
  (e.g., just after the full reset) done as part of shutdown) are dropped.
* Eliminate loops that  poll forever on S_COMMAND_IN_PROGRESS to complete.
  Add inline function with bounded loopcount plus small delay, to avoid
  bugs in EISA hardware which never sets S_COMMAND_IN_PROGRESS.
  Use for both TX_RESET/RX_RESET/GLOBAL_RESET cmds, and polling for
  discard-Rx completion

Endian-ness bugfix for elink3 on big-endian hosts.
  Big-endian CPUs should use bus_space_{read,write}_multi_stream methods
  to PIO packet data to NIC in 16-bit or 32-bit chunks.
  Make it so.
  Since we don't have stream methods on all ports yet (esp. i386), use
  the bus_space_{read,write}_multi methods if the compile-time
  stream-method feature-test   macro is not set.
Originally from tsubai@netbsd.org, via HAYAKAWA koichi's cardbus patches.

Use ETHER_*_LEN constants from <net/if_ether.h> instead of defining them
ourselves.

Update to slightly altered rnd_attach_source() api

Avoid initializing the interface in the SIOCSIFFLAGS if ! UP and ! RUNNING.

Simplify argument to MGET(), to generate better code.

The beginnings of interface detach support.  Still some bugs, but mostly
works for me.

This work was originally by Bill Studenmund, and cleaned up by me.

Now that the *_activate() functions don't do things which are attachment
specific, move them into the chipset drivers.

Expose the "enable" and "disable" functions so that the front-ends can
access them.

Fix some whitespace lossage, canonicalize some media names.

Add the 'ex' driver, a DMA driver for 3Com 90x and 90xB cards. Rename
constants from EP_ to ELINK_ since they're now used in the ex driver as well.

When determining the default media on the Vortex/Boomerang, don't pick
the full-duplex variant.  XXX We should check the EEPROM for "default to
full-duplex".

For non-MII Vortex/Boomerang, set full-duplex in the MAC CONTROL register
in epsetmedia() if full-duplex is selected (new feature on these cards!)

On MII Boomerangs, set the full-duplex bit in MAC CONTROL in the statchg
callback from the MII layer.  Should fix full-duplex on these cards.

First-crack at RoadRunner (3c574-TX 10/100Mbps PCMCIA).  Doens't yet use
any of the RoadRunner speed hacks.

Oops, back out something that I didn't intend to commit.

The 3c515 (Corkscrew) and 3c574 (RoadRunner) have their Window 1 registers
offset!  Define an inline function to compute the Window 1 register offset
and use it.

Completely rewrite media selection.  Add support for the MII/PHY on
the 3Com 3c905-TX and 3c905-T4.  Fix handling of the MII on the
3c595-MII and 3c597-MII (can't talk to the PHYs on these cards; just
use "manual" for the external MII port).

Fixes kern/4782 (Chris Demetriou).

defopt NS, NSIP.

defopt INET, NETATALK.

For the Rx:Tx split message change "(undefined)" to "3:5".

Adjust for config changes.

Protect #ifdef DEBUG printfs with a run-time teson a patchable variable
(epdebug), as per Jason Thorpe's comments.

Add and/or update rnd_*() hooks

Pull down from marc-pcmcia branch:

Minor tweaks to deal with PCMCIA differences, and support for power
management.

Garbage-collect a hack no longer needed since the compiler has been
fixed.

Rearrange the disable functions slightly so that an interface that
interfaces that don't have a disable hook will never be marked "disabled".

Now, once an interface has been marked "enabled" by the enable function,
it will remain marked as such for the duration of the kernel's lifetime.

Fix a brain-o in my last power management changes.  Specifically, allow
the media callbacks to be used while the interface is powered off.  If
the interface is powered off, return "none" in response to a "get media",
and in the case of "set media", just return success, since the init routine
will properly program the hardware when power is enabled.  If we return
an error in the powered off case, ifconfig(8) will lose because it performs
the "set address" (which will enable power on a disabled interface) after
parsing the entire command line.

Fix a few bugs in the enable/disable hook logic:
- Place the code to enable/disable in functions and use them, rather than
  duplicating it in a few places.  Make sure the "enabled" flag is set
  properly even if there is no hardware enable hook.
- Add additional checks for the "enabled" flag so that we don't attempt to
  write hardware registers when the card power is turned off.

Make the "enable" and "disable" hooks take pointers to the softc, rather
than void * and a separately specified argument.

add calls to enable/disable hooks

Remove some highly machine-dependent debugging code.

Update marc-pcmcia branch from trunk.

Initialize local variable 'j' in  eeprombusy() to pacify gcc -O (not -O2)
on i386.

added pcmcia infrastructure and a few devices

Make sure we're in register window 0 before attempting EEPROM access,
pointed out by Matthias Drochner.

Set TX-available threshhold in epinit() just before enabling interrupts,
so TX_AVAIL intrs are enabled for root-on-NFS kernels.
Tested (and required) on 3c595TX with root-on-NFS kernel booted from floppy.

Oops; forgot to GC the last mbuf allocated when out of clusters.

* Tidy PHY [sic] selection code in epsetmedia().

* epinit() had both explicit xcvr selection for 3c589 and a call
  to epsetmedia().  The first is redundant; delete it, and EP_COAX_DEFAULT.

* Update comments to reflect 3c589 and 3c509B fixes.  Fix typos.

- Replace the check for EP_BUS_PCMCIA in epsetmedia with a check for
  EP_CHIPSET_3C509 and change it a bit so it works now with the 3c509
  and still works with the 3c589.
- KNF

Fix typo in previous.

If we fail to allocate a cluster to hold a large packet, simply
drop it rather than using a chain of tiny mbufs.

Smaller, cleaner code for aligning struct ip on receipt.

Align struct ip on incoming packets for all types of cards, not just
32-bit ones.

Remove IFF_LINK[01] "connector" support.
Add if_media support.   Still missing a software autosense FSM.

Add ifmedia hooks: include <ifmedia.h> and add sc_media to softc.

Pull up fix for `sniper bug', which can cause packets to be accepted erroneously.

New ARP system, supports IPv4 over any hardware link.

Some of the stuff (e.g., rarpd, bootpd, dhcpd etc., libsa) still will
only support Ethernet. Tcpdump itself should be ok, but libpcap needs
lot of work.

For the detailed change history, look at the commit log entries for
the is-newarp branch.

netinet/if_ether.h => netinet/if_inarp.h

Recognize 3c900 and 3c905 cards.

Removed #include <route.h> and #include <netisr.h>. They're not needed, and
the latter will go away soon.

Merged in recent changes from Trunk.

Move if_ep media-sense code from front ends to inside back end,
to prepare for if_media, and to support EISA 3c59x cards:

  * change epconfig() to take a short that encodes the chipset
    type (3c509, or Demon/Vortex/Boomerang).
  * add  distinct 3c509 and Demon/Vortex/Boomerang media-sense
    functions to back end.
  * Add EISA match/attach support for the 3c592 and 3c597
    (Demon) cards as well as the 3c509-comatibl EISA 3c509/3c579.

  * Assume that ISA and ISA PnP cards are 3c509-style, until
    proof to the contrary (e.g. large packet support).

Tested on 3c595, 3c590, and 3c509b. Not tested on 3c515 or Demon.

* Add ep_complete_cmd() function, to issue commands that perform a reset
  (RX_RESET, TX_RESET, GLOBAL_RESET).
  Commands issued while a reset is in progress may be dropped. The driver
  must wait until resets are complete; ep_complete_cmd() does so.
* Change RX_RESET,  TX_RESET, GLOBAL_RESET to use ep_complete_cmd().

* Revert epconfig() to use inline RX_RESET, TX_RESET.
  That avoids accidentally setting IFF_UP, which  breaks both diskless boot,
  and "halt" commands issued before the interface is brought up.

Snapshot of new ARP code.

Our old ARP code was hardwired for 6-byte length medium
addresses, while the protocol is designed for any size.

This snapshot contains a first hack at getting rid of
Ethernet specific data structures. The ep driver is updated
(and tested on the PCI bus), the iy and fpa drivers have been
updated, but not real life tested yet.

If you want to test this with other drivers, you have to update
them first yourself, and probably tag the relevant directories.
Better contact me if you want to do this.

Use DELAY() after doing a TX_RESET or RX_RESET command.
Newer boards (e.g., Vortex) take  time to reset, and wedge on some machines.
Polling on S_COMMAND_IN_PROGRESS would be better, but that doesn't
work with xx_RESET commands on 3c509s.

*  make sure 100mbit media-present bits don't collied with 3c509
   CONFIG_CNTRL bits by making the "conn" argument to epconfig()
   a u_int, not a u_int16_t, and by defining 100mbit media
   bits in the upper 16 bits.

A redesigned interface that fixes this properly is in the pipeline.

* add epshutdown(), and  establish as shutdown hook: see multiboot
  comments in pr port-i386/2358: "if_ep/3C590 no longer working"
* Only call Vortex media-config code on PCI-attached devices, until
  we know if it works on ISA and EISA large-packet devices also.
* Fix typo in printf of MAC address

* Add 100Mbit xcvr flags to epconfig():
  look for 100mbit Xcvrs, print their names, add to ep_connectors.

* Print 10baseT as 10baseT, since media name (utp) is no longer a unique
  transceiver type.

* Redo FIFO memory size printing, also show Tx/Rx split (xxx isa 3c509?).

* Print MAC address on a separate line,  with 100Mbit media and fifo-ram
  info it no longer fits on the same line as media types.

* Comment the fact that the "connection" media-selection code just doesn't
  grok 100Mbit media.

* add code to look for 3c59x/3c90x media-select and autoconfig bits, though
  we don't yet dynamically probe media when the autoconfig bit is set
  (see previous bullet).

* XXX the whole connector/media code should be redone, perhaps a' la
  the BSDI ifmedia design.

Print the memory size on startup.
Add newline to `Interface disabled' printf and clean up formatting.
Change packet read/write code to align all reads and writes into
memory to an appropriate boundary for the I/O size. This also fixes
a bug where the trailing unaligned bytes (1-3 of them) would be
put in an additional mbuf even if there was enough space for them
in the one we were currently filling.
This has been tested on a 3c905 in an Alpha (32-bit I/O, alignment
required), a 3c905 in an i386 (32-bit I/O, alignment not required)
and a 3c509 in an i386 (16-bit I/O, alignment not required). It
should be tested on a 3c509 in an Alpha with an ISA bus, if somone
can manage to get the two items together.
There are still some bugs in the driver relating to initialisation;
on my Alpha with a 3c905 you need to do an `ifconfig ep0 down;
ifconfig ep0 up' after the initial ifconfig to make it start working.
Apparently there are also some problems with a 3c579 in an i386,
although a 3c905 in an i386 works just fine.

* 3c509s  drop the low-oder bits of EP_THRESH_DISABLE (2047) on a write,
  and the result reads back as zero.  Mask probe code accordingly.

Fix for hanging 3com ``large packet'' (100Mbit) Etherlink-III devices:
  * Add a slot in the softc, ep_pktlenshift, to indicate that
    a board implements ``large packets'' (FDDI-sized frames) and
    therefore implicitly shifts the argument to packet-length commands
    by two bits to fit a 13-bit length in  an 11-bit command argument field.

 *  Change  the reset of station-mask filterto be independent of bus-type
    (e.g.,. for 3c515)

 * Replace  the "2044" constant used to disable TX threshold interrupts
   on 3c509-generation cards with a #define. Use 2047 instead,
   since that works on the large-packet cards (Vortex, Demon, Boomerang)
   and 2044 doesn't.

 *  Add code to  epconfing to probe for large-packet support,
    set the softc appropriately, and  shift packet sizes in commands
    by ep_pktlenshift to compensate  for the implicit shift.

 *  Tweak the epconfig() code to avoid hanging on startup.
    Set the Tx threshold to interrupt when there's space for
    a maximal-sized ether frame, then  do epinit() then epstop(),
    instead of just the Tx and Rx reset. Required on at least a 3c595.

Tested on  a 3c595, but not on 3c509-generation hardware.

New bus.h implementation/interface:
	- No more distinction between i/o-mapped and memory-mapped
	  devices.  It's all "bus space" now, and space tags
	  differentiate the space with finer grain than the
	  bus chipset tag.
	- Add memory barrier methods.
	- Implement space alloc/free methods.
	- Implement region read/write methods (like memcpy to/from
	  bus space).
This interface provides a better abstraction for dealing with
machine-independent chipset drivers.

backout kprintf changes

- printf -> kprintf, sprintf -> ksprintf

Apply PR/2661 [from Jonathan Stone]

Be more careful with types, and pull in <machine/intr.h>

Update Herb's e-mail address; beer.org feels a lot more friendly.

Changed struct ifnet to have a pointer to the softc of the underlying
device and a printable "external name" (name + unit number), thus eliminating
if_name and if_unit.  Updated interface to (*if_watchdog)() and (*if_start)()
to take a struct ifnet *, rather than a unit number.

Move epstop prototype to the header file.

- include <sys/systm.h>
- fix prototype of epmbuffill()
- fix typo of epxstat -> eptxstat

Convert the Etherlink III (ep) driver to use <machine/bus.h>.

Break up the ep driver into isa, eisa, and pci front-ends and a
bus-independent core driver.  Tested on all three bus types, including
an isa 3c509 masquerading as an eisa device (use ep* at eisa? slot ? in
your kernel config file to catch this one).
XXX Driver still needs to be converted to <machine/bus.h>