CVS log for src/sys/dev/marvell/if_mvgbe.c

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

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

s/funcstions/functions/ in comment.

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@

fix tripple/quadruple "r" letter typos in comments and log messages.

Correct byte-order for bufsize field of RX descriptor.

XXX
This bug affected only armeb. Unfortunately, the systems including
armel still lock up eventually with NFS root on mvgbe(4)...

m_freem() *after* bus_dmamap_sync() and bus_dmamap_load() for
that mbuf. This is mandatory for some archs.

Add DPRINTF() to catch failure of m_defrag() in mvgbe_encap().

Raise threshold of debug level to 3 for DPRINTF() in non-error path of
mvgbe_encap(). Without this change, debug level 2 is almost unusable...

Add ARMEB support to mvgbe(4).

For ARMEB, peripheral is configured to little-endian mode, even if
CPU itself is in big-endian mode. Therefore, we need to configure
the device to little-endian mode, and byte-swap descriptor fields
(unlike the case of powerpc).

Merge thorpej-cfargs2.

Adapt to CFARGS().

Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
  actually needed.
- Don't be explicit about what interface attribute is attaching if
  the device only has one.  (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
  situations, making is visibly easier to see when indirect config is
  in play, and allowing for future change in semantics.  (As of now,
  this is just a wrapper around config_match(), but that is an
  implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance.  This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

Audit CFARG_IATTR in config_found() calls, and remove it in situations
where the interface attribute is not ambiguous.

Mechanical conversion of config_found_sm_loc() -> config_found().
CFARG_IATTR usage needs to be audited.

Mostly merge changes from HEAD upto 20200411

Merge changes from current as of 20200406

Sync with head.

Adopt <net/if_stats.h>

 if_flags is neither int nor short. It's unsigned short.

Sync with HEAD

 Use ETHER_LOCK()/ETHER_UNLOCK() for all ethernet drivers to protect ec_multi*.

 Whitespace fix (mainly tabify).

No functional change:
 - Simplify MII structure initialization and reference.
 - u_int*_t -> uint*_t.
 - KNF

 On 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.

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

Add intr_establish_xname support to arm and expose it to intrctl

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

 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.

update from HEAD

Sync with HEAD

remove checks for failure after memory allocation calls that cannot fail:

  kmem_alloc() with KM_SLEEP
  kmem_zalloc() with KM_SLEEP
  percpu_alloc()
  pserialize_create()
  psref_class_create()

all of these paths include an assertion that the allocation has not failed,
so callers should not assert that again.

Sync with HEAD

Sync with HEAD

Support Marvell Dove.

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

Apply deferred if_start framework

if_schedule_deferred_start checks if the if_snd queue contains packets,
so drivers don't need to check it by themselves.

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

Pass all unicast addresses if it is promisc mode.

Add more #warning message 'Please use mvxpe instead of this'.

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

add u-boot argument parser for Marvell ARMADA XP/370.
use 'ethaddr' and 'eth1addr' in u-boot argument to setup MAC address of mvgbe.

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

Rebase to HEAD as of a few days ago.

Merge tls-earlyentropy branch into HEAD.

sync with head.

for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.

this commit was splitted into small chunks to avoid
a limitation of cvs.  ("Protocol error: too many arguments")

sync with 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.

Support Armada 370 and DDR3 tags.

Ensure that the top level sysctl nodes (kern, vfs, net, ...) exist before
the sysctl link sets are processed, and remove redundancy.

Shaves >13kB off of an amd64 GENERIC, not to mention >1k duplicate
lines of code.

 Carefully set IFF_OACTIVE in mvgbe_start(). Even if mvgbe_encap()
returns other than 0, the TX ring might not full. Check whether
the TX ring has one or more packets. If the ring is empty,
dont' set IFF_OACTIVE because an TX complete interrupt never
occur and IFF_OACTIVE flags is left. The interface's timer
isn't reset, so a device timeout desn't occur.

 Fixes a bug that IFF_OACTIVE flag is left on heavy load.
Part of PR#48568.

Support Basic Mode for Armada XP.

resync with head

sync with head

sync with (a bit old) head

Pull up following revision(s) (requested by msaitoh in ticket #775):
	sys/dev/marvell/if_mvgbe.c: revision 1.34
Fix a bug that a multicast filter entry is incorrectly overrode
on some cases.

Pull up following revision(s) (requested by msaitoh in ticket #766):
	sys/dev/marvell/if_mvgbe.c: revision 1.33
Check wheter an MII PHY is really connected or not.
This change privents panic if MII PHY port is not used.

Fix a bug that a multicast filter entry is incorrectly overrode
on some cases.

Check wheter an MII PHY is really connected or not.
This change privents panic if MII PHY port is not used.

Apply patch (requested by msaitoh in ticket #671):

sys/dev/marvell/if_mvgbe.c		1.19-1.23, 1.26-1.31
sys/dev/marvell/mvgbereg.h		1.4-1.5, 1.7

Add missing bus_dmamap_sync() for the RX buffer.
Fix device timeout problem.
Fix broken hardware checksumming.
Fix a bug that kernel panics when the system get a packet while calling
mvgbe_stop (via ifconfig down).
Add missing mii_tick() call.
Change style a bit.
Don't use M_HASFCS flag.
When an interrput is link change, notify link change to mii layer using
mii_pollstat().
Fix a bug that the alignment of jumbo buffer(MVGBE_JLEN) is miscalculated.

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

Add MVGBE_EVENT_COUNTERS for debugging. Count rxoverrrun and wdogsoft.

Fix a bug that MVGBE_JLEN is miscalculated.

When an interrput is link change, notify link change to mii layer using
mii_pollstat().

Don't use M_HASFCS flag.

No functional change.
- Change style a bit.
- use "csc" for the variable name of struct mvgbec_softc in mvgbec_attach()
  for consistency.

sync with head

Add missing mii_tick() call.

Fix a bug that kernel panics when the system get a packet while calling
mvgbe_stop (via ifconfig down).

Add code to modify interrupt coalescing with sysctl.
 - both TX side an RX side.
 - different setting for each port
 - TX side is hw.mvgbe.mvgbe*.ipginttx
 - RX side is hw.mvgbe.mvgbe*.ipgintrx
 - The default value is 768.
 - The lowest value is 0
 - For highest value, 0x3777 is used for V1, and 0xffff is used for V2.

Add Interrupt Coalescing setting by Robert Swindells.

- The MVGBE_RX_MAX_FRAME_LEN_ERROR bit is used to check whether a packet
  is fragmented or not, so define new MVGBE_RX_IP_FRAGMENT with the same
  value and use it.
- Remove the checking whether a packet length is lower than 72 octet.
  This check is not used in Linux and FreeBSD. Tested with me (for Kirkwood)
  and Kiyohara (for DiscoveryII).

Fix udpcsum-rx bug on fragmented case this time.
  - Fix the usage of a local variable for csum_flags.
  - It seemd that sometimes MVGBE_RX_L4_CHECKSUM_OK bit were set to 0
    even if the checksum is correct and the packet was not fragmented.
    So we don't set M_CSUM_TCP_UDP_BAD even if csum bit is 0.

- Fix a bug that a fragmented packet is marked as bad checksum. It causes
  a real bug when HW checksum offload function is used. It was easy to
  reproduce with NFS UDP mount.
- Fix a potential bug that a packet other than TCP and UDP is marked as bad
  checksum.

Fix device timeout problem.
- Change the synching order of descriptors. First, sync descriptors except
  first and then sync the first descriptor.
- To recover from an race condition, reduce the if_timer from 5 to 1 and
  when timeout occur write MVGBE_TQC_ENQ bit again.

Add missing bus_dmamap_sync() for the RX buffer.

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.

Add 88F6282.  tested only OpenBlockS A6.
Also fix irq 11 to 15.  irq of 2nd NIC is 15.
Remove white-space.

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.

First step of random number subsystem rework described in
<20111022023242.BA26F14A158@mail.netbsd.org>.  This change includes
the following:

	An initial cleanup and minor reorganization of the entropy pool
	code in sys/dev/rnd.c and sys/dev/rndpool.c.  Several bugs are
	fixed.  Some effort is made to accumulate entropy more quickly at
	boot time.

	A generic interface, "rndsink", is added, for stream generators to
	request that they be re-keyed with good quality entropy from the pool
	as soon as it is available.

	The arc4random()/arc4randbytes() implementation in libkern is
	adjusted to use the rndsink interface for rekeying, which helps
	address the problem of low-quality keys at boot time.

	An implementation of the FIPS 140-2 statistical tests for random
	number generator quality is provided (libkern/rngtest.c).  This
	is based on Greg Rose's implementation from Qualcomm.

	A new random stream generator, nist_ctr_drbg, is provided.  It is
	based on an implementation of the NIST SP800-90 CTR_DRBG by
	Henric Jungheim.  This generator users AES in a modified counter
	mode to generate a backtracking-resistant random stream.

	An abstraction layer, "cprng", is provided for in-kernel consumers
	of randomness.  The arc4random/arc4randbytes API is deprecated for
	in-kernel use.  It is replaced by "cprng_strong".  The current
	cprng_fast implementation wraps the existing arc4random
	implementation.  The current cprng_strong implementation wraps the
	new CTR_DRBG implementation.  Both interfaces are rekeyed from
	the entropy pool automatically at intervals justifiable from best
	current cryptographic practice.

	In some quick tests, cprng_fast() is about the same speed as
	the old arc4randbytes(), and cprng_strong() is about 20% faster
	than rnd_extract_data().  Performance is expected to improve.

	The AES code in src/crypto/rijndael is no longer an optional
	kernel component, as it is required by cprng_strong, which is
	not an optional kernel component.

	The entropy pool output is subjected to the rngtest tests at
	startup time; if it fails, the system will reboot.  There is
	approximately a 3/10000 chance of a false positive from these
	tests.  Entropy pool _input_ from hardware random numbers is
	subjected to the rngtest tests at attach time, as well as the
	FIPS continuous-output test, to detect bad or stuck hardware
	RNGs; if any are detected, they are detached, but the system
	continues to run.

	A problem with rndctl(8) is fixed -- datastructures with
	pointers in arrays are no longer passed to userspace (this
	was not a security problem, but rather a major issue for
	compat32).  A new kernel will require a new rndctl.

	The sysctl kern.arandom() and kern.urandom() nodes are hooked
	up to the new generators, but the /dev/*random pseudodevices
	are not, yet.

	Manual pages for the new kernel interfaces are forthcoming.

Add workaround for infrequently encountered DMA engine limitation.

Only zero MVGBE_MTU on Orion platforms.

We occasionally will try to set up the multicast filter in a context that
kmem(9) is unusable (such as transmit timeout). Thus, put the filter table
on the stack instead, all 528 bytes of it.

Initialize MVGBE_MTU to 0, as suggested by the Orion databook.

If the platform firmware did not initialize this to 0 for us,
we were unable to transmit frames.

Use new style BPF calls.

Use cfdata_t

Sync with HEAD.

sync with head

fix typo, and use the parent to determine the offset. From Dave Mills

don't abuse cf_unit, requested by phone.

sync with head (and fix few botches with this)

sync with head

From Dave Mills: It would appear that the two PHYs can only be
accessed from the registers of controller0 (mvgbec0) hence why both
PHYs are being detected on controller0. I have made the assumption
that PHY0 => controller0 (mvgbec0), PHY1 => controller1(mvgbec1)
and developed a patch accordingly. Also the irq is 15 for controller1
not 14.

Sync with HEAD

Various improvements to mvgbe(4).  Notable is improved multicast filtering.

Use macros for ring position advances, assuming a power of two ring size.
(Also, spelling correction in comment.)

Sync with HEAD (-D20101022).

sync with head

Support Kirkwood's GbE.

Sync with HEAD.

sync with head.

file if_mvgbe.c was added on branch uebayasi-xip on 2010-08-17 06:46:17 +0000

file if_mvgbe.c was added on branch yamt-nfs-mp on 2010-08-11 22:53:38 +0000

Add macro MVA_{UNIT,OFFSET,IRQ}_DEFAULT.

sync with head

file if_mvgbe.c was added on branch rmind-uvmplock on 2010-07-03 01:19:36 +0000

Add Marvell Gigabit Ethernet Controller on SC/SoC.
  Will support on ofppc soon.
  Tested on G4+PegasosII (kiyohara@)
	    G3+PegasosII (phx@)