files, so that SWAP_META_PAGES does not vary either.
swap_pager.c ended up with a value of 16, everybody else 8. Go with
the 16 for now.
This should only have any effect in the "kill processes because we
are out of swap" scenario, where it will make some sort of estimate
of something more precise.
1. There was a race condition between a thread unlocking
a umtx and the thread contesting it. If the unlocking
thread won the race it may try to wakeup a thread that
was not yet in msleep(). The contesting thread would then
go to sleep to await a wakeup that would never come. It's
not possible to close the race by using a lock because
calls to casuptr() may have to fault a page in from swap.
Instead, the race was closed by introducing a flag that
the unlocking thread will set when waking up a thread.
The contesting thread will check for this flag before
going to sleep. For now the flag is kept in td_flags,
but it may be better to use some other member or create
a new one because of the possible performance/contention
issues of having to own sched_lock. Thanks to jhb for
pointing me in the right direction on this one.
2. Once a umtx was contested all future locks and unlocks
were happening in the kernel, regardless of whether it
was contested or not. To prevent this from happening,
when a thread locks a umtx it checks the queue for that
umtx and unsets the contested bit if there are no other
threads waiting on it. Again, this is slightly more
complicated than it needs to be because we can't hold
a lock across casuptr(). So, the thread has to check
the queue again after unseting the bit, and reset the
contested bit if it finds that another thread has put
itself on the queue in the mean time.
3. Remove the if... block for unlocking an uncontested
umtx, and replace it with a KASSERT. The _only_ time
a thread should be unlocking a umtx in the kernel is
if it is contested.
- 5705 doesn't support jumbo frames
- Statistics must be read from registers
- RX return ring must be capped at 512 entries
- Omit initialization of certain device blocks
- Acknowledge link change interrupts by setting the 'link changed'
bit in the status register (used to have no effect)
- Remember to toggle the MI completion bit too
- Set the mbuf low watermark differently (on-chip memory buffers,
not BSD mbufs)
- Don't enable Ethernet@WireSpeed feature for certain 5705 chip revs
- Add additional PCI IDs for 5705 and 5782 parts
- Add a forgotten 5704 PCI ID
Most changes ripped kicking and screaming from the Broadcom linux driver.
Thanks to Paul Saab for sanity testing. (My lack of sanity has been
confirmed.)
code from i386. The code has a slight bogon that interrupts are counted
twice. Once on the ithread dispatch and once on the dispatch for the vector
vmstat -i and systat -vm now contains interrupt counts.
Reviewed by: jake
sync of the NetBSD code.
fix isochornous support for ohci. This gets webcams like my OV511
working on sparc64.
PR: kern/52589
Submitted by: Bruce R. Montague (isochonous support)
Reviewed by: joe among others
Set 31 is still special because rules belonging to it are not deleted
by the "ipfw flush" command, but must be deleted explicitly with
"ipfw delete set 31" or by individual rule numbers.
This implement a flexible form of "persistent rules" which you might
want to have available even after an "ipfw flush".
Note that this change does not violate POLA, because you could not
use set 31 in a ruleset before this change.
sbin/ipfw changes to allow manipulation of set 31 will follow shortly.
Suggested by: Paul Richards
lastest rev of the spec. Use an explicit flag for Fast Recovery. [1]
Fix bug with exiting Fast Recovery on a retransmit timeout
diagnosed by Lu Guohan. [2]
Reviewed by: Thomas Henderson <thomas.r.henderson@boeing.com>
Reported and tested by: Lu Guohan <lguohan00@mails.tsinghua.edu.cn> [2]
Approved by: Thomas Henderson <thomas.r.henderson@boeing.com>,
Sally Floyd <floyd@acm.org> [1]
callback will never be deferred. ATM needs to prevent cell and packet
ordering. Also use the default mutex and lock functions (those that
panic) for the tag creation.
with a ProATM-155 and an IDT evaluation board and should also work
with a ProATM-25 (it seems to work at least, I cannot really measure
what the card emits). The driver has been tested on i386 and sparc64,
but should work an other archs also. It supports UBR, CBR, ABR and VBR;
AAL0, AAL5 and AALraw. As an additional feature VCI/VPI 0/0 can be
opened for receiving in AALraw mode and receives all cells not claimed
by other open VCs (even cells with invalid GFC, VPI and VCI fields and
OAM cells).
Thanks to Christian Bucari from ProSum for lending two cards and answering
my questions.
implement the ATMIOCGVCCS ioctls. This routine handles changing
VCC tables (which can occure because we cannot hold the driver mutex
while allocating memory) with a loop and a re-allocation, should the
table not fit in the allocated memory.
large to huge amounts of small or medium sized receive buffers. The problem
with these situations is that they eat up the available DMA address space
very quickly when using mbufs or even mbuf clusters. Additionally this
facility provides a direct mapping between 32-bit integers and these buffers.
This is needed for devices originally designed for 32-bit systems. Ususally
the virtual address of the buffer is used as a handle to find the buffer as
soon as it is returned by the card. This does not work for 64-bit machines
and hence this mapping is needed.
MFNetBSD: revision 1.137
date: 2003/01/20 07:12:13; author: simonb;
Grrr. So much for my ability to use grep(1) effectively. Pointed out
by Stephen Degler in private mail.
date: 2002/12/10 14:07:37; author: toshii; state: Exp; lines: +6 -6
Add a couple of le32toh which were missing in the previous.
Pointed out by SOMEYA Yoshihiko.
date: 2002/12/07 07:33:20; author: toshii; state: Exp; lines: +50 -29
Update xfer->frlengths for input isoc transfer. Based on patches from
SOMEYA Yoshihiko.
Also fix error handling for isoc transfer somewhat; usb_transfer_complete
shouldn't be called for more than once.
date: 2002/12/07 07:14:28; author: toshii;
Fix several nits. Mostly from SOMEYA Yoshihiko.
- Call usbd_transfer_complete at splusb.
- Fix a botched for loop in ohci_rem_ed.
- In ohci_close_pipe, wait 1ms after removing an ED to avoid possible race
condition.
The splusb change is non-functional on FreeBSD.
The botched loop and race condition changes came from us.
This patch is non-functional.
date: 2002/09/29 20:58:25; author: augustss;
Add some spl calls to protect critical regions. From kern/18440,
Takeshi Nakayama.
(No functional change on FreeBSD).
date: 2002/12/31 02:21:31; author: dsainty;
Be somewhat more persuasive about enabling the port on a port reset.
USB protocol dictates that the port enable must be implied by the port
reset. To implement this on (at least) the VIA VT83C572 this means we
need to wait around tweaking the chip state until the port actually
transitions to enabled (or the device goes away). Likely fixes
kern/11018.
get a Hub descriptor, we have to set req.wValue to "Descriptor Type
and Descriptor Index". In this case, Descriptor Type is 0x29
(UDESC_HUB), Descriptor Index should be 0.
If I don't do a check (dev->address > 1 ... ), root hub fails.
A new Cytronix 4-port USB 2.0 Hub (Cypress CY7C65640 chip) now works
after this patch.
Submitted by: Alexander Pohoyda <alexander.pohoyda@gmx.net>
MFC after: 7 days
64-bit counters that wrap on overflow. They are collecte once per
second from the chips. Currently they can be retrieved via a sysctl phy_stats.
A write of an arbitrary value to the sysctl atomically retrieves the
statistics and clears them.
* Use ACPI_BUFFER as the type for AcpiGetObjectInfo
* Remove AcpiEnableEvent/AcpiClearEvent for ACPI_EVENT_FIXED (power/sleep
buttons) as they are no longer needed
* Change calls to use the new GPE functions
* Add AcpiOs*Lock functions
from the network interface earlier in ether_input(). At some point
(no fingers pointed), things were restructured and the labeling operation
moved later. This wasn't a problem as BPF_MTAP() relies on the ifnet
label not the mbuf label, but there might have been other problems.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
Even if we have no AT keyboard, an AT keyboard is registered because
it's probed with KB_CONF_PROBE_ONLY flag set during console initialization.
Unregister the keyboard if it doesn't present while second probe.
This should fix USB keyboard only case without 'kbdcontrol -k /dev/kbd1'.
them again afterwards. This fixes a disabled FP fault while in the FPSWA
handler.
While here, merge the FP fault and FP trap handling code to reduce code
duplication. Where code was different, it was not sure it should be.
Trigger case: ports/math/atlas
multiple mutex pools with different options and sizes. Mutex pools can
be created with either the default sleep mutexes or with spin mutexes.
A dynamically created mutex pool can now be destroyed if it is no longer
needed.
Create two pools by default, one that matches the existing pool that
uses the MTX_NOWITNESS option that should be used for building higher
level locks, and a new pool with witness checking enabled.
Modify the users of the existing mutex pool to use the appropriate pool
in the new implementation.
Reviewed by: jhb
Since we already had 'O_NOP' instructions which always match, all
I needed to do is allow the NOP command to have arbitrary length
(i.e. move its label in a different part of the switch() which
validates instructions).
The kernel must know nothing about comments, everything else is
done in userland (which will be described in the upcoming ipfw2.c
commit).
our unwind information for functions that are entry points into the
kernel. When stepping to the next frame, the unwinder will let us
know when sych a marker was encountered. We use this to stop the
current unwind session, query the trapframe and restart a new
unwind session based on the new trapframe.
The implementation is a bit sloppy, but at this time there are
bigger fish to fry.
default. There were 2 cases where the trace code was in the else-
part of an if-statement without any compound construct to protect
against nullification. The result was that 2 unconditional statements
turned into conditional statements and wrecked havoc.
This fix has been returned to the vendor and in anticipation of a
future import committed onto the vendor branch.
it to the bss section and skips the initialization. This causes all
sorts of havoc because the bogus bss zero code clobbered previously set
variables. All our supported boot loaders already zero the bss, even
kgzip for the elf case. Since we dont generate a.out kernels, the old
a.out bootblocks and the a.out kgzip are not a factor anymore.
modify vendor code (libuwx) with a specific include directive.
The second order advantage is that we can also enable verbosity
in the glue code (ia64/ia64/unwind.c).
(mainly the 3Com 3c996B/BCM5701).
For some reason that I don't fully understand, the 5701 signals PCS
encoding errors as though they were link change events, i.e. the 'link
state changed' bit in the status word of the status block is updated
and an interrupt is generated. This would cause the bge_tick() function
to be invoked and a "gigabit link up" message to be printed on the console.
To avoid this, the interrupt handler now checks the MAC status register
when a link change interrupt is triggered, and it will only call the
bge_tick() function if the 'PCS encoding error detected' bit is clear.
(This change should have no effect on copper NICs since this bit can
only ever be set in TBI mode. I do not know how it affects 5704 NICs
with a BCM8002 SERDES PHY.)
Special thanks to: Sherry Rogers at UCB for allowing me access to one
of their traffic monitor boxes so I could diagnose this problem.
compilation accordingly. The net effect is that tracing is not by
default present and that it can easily be compiled-in.
The tracer uses getenv() and printf(), which cannot be used in all
cases (ie from the debugger) and with this change we improved the
applicability of the unwinder.
This change is made on the vendor branch and given back to the
vendor for possible inclusion in future versions.
immediately after the kernel map has been sized, and is
the optimal place for the autosizing of memory allocations
which occur within the kernel map to occur.
Suggested by: bde
without Giant held.
A quick outline of the locking strategy:
Since all IOMMUs are synchronized, there is a single lock, iommu_mtx,
which protects the hardware registers (where needed) and the global and
per-IOMMU software states. As soon as the IOMMUs are divorced, each struct
iommu_state will have its own mutex (and the remaining global state
will be moved into the struct).
The dvma rman has its own internal mutex; the TSB slots may only be
accessed by the owner of the corresponding resource, so neither needs
extra protection.
Since there is a second access path to maps via LRU queues, the consumer-
provided locking is not sufficient; therefore, each map which is on a
queue is additionally protected by iommu_mtx (in part, there is one
member which only the map owner may access). Each map on a queue may
be accessed and removed from or repositioned in a queue in any context as
long as the lock is held; only the owner may insert a map.
To reduce lock contention, some bus_dma functions remove the map from
the queue temporarily (on behalf of the map owner) for some operations and
reinsert it when they are done. Shorter operations and operations which are
not done on behalf of the lock owner are completely covered by the lock.
To facilitate the locking, reorganize the streaming buffer handling;
while being there, fix an old oversight which would cause the streaming
buffer to always be flushed, regardless of whether streaming was enabled
in the TSB entry. The streaming buffer is still disabled for now, since
there are a number of drivers which lack critical bus_dmamp_sync() calls.
Additional testing by: jake
series, the 8139C+ has a descriptor-based DMA mechanism, and its
performance is actually pretty respectable. Note: the 8139D chip does
not support C+ mode. Only the 8139C+ and 8169 gigE chips support C+ mode.
Supported features:
- RX and TX checksum offload
- hardware VLAN tag insertion/extraction
- TX interrupt moderation using the 8139's on-board timer
Everything should be properly busdma'ed and endian-independent, so
things should work ok on non-x86 platforms. Unfortunately, my call
for testers on this code was met with deafening silence, and I don't
have access to any non-x86 FreeBSD boxes at the moment, so this is
speculation.
The device detection code has been cleaned up a little as well
(thanks to Michal Mertl) for the patches.
There are also updates to the rl(4) man page (which I accidentally
checked in before when I updated the dc(4) man page. Oops.)
Todo: finish support for the 8169 gigabit ethernet chip. This
mainly requires writing an rlgphy driver to handle the 8169's built-in
PHY. This will have to wait until I actually get my hands on an 8169
card for testing though. (I still can't find a source for one in the
U.S. Suggestions/pointers welcome.)
- MN-110 10/100 USB ethernet (ADMtek Pegasus II, if_aue)
- MN-120 10/100 cardbus (ADMtek Centaur-C, if_dc)
- MN-130 10/100 PCI (ADMtek Centaur-P, if_dc)
Also update dc(4) man page to mention support for MN-120 and MN-130.
* Always use polled mode. The intr approach did not work for many
controllers and required the hw.acpi.ec.event_driven workaround.
* Only use an edge (not level) triggered GPE handler
* Add sc->ec_mtx for locking operations to a single EC. There were
many race conditions earlier between an SCI event and EcRead/Write.
* Use 1 ms as the global lock timeout
* Only acquire global lock if _GLK != 0
* Update EcWaitEvent to use an incremental backoff delay in its
poll loop. Wait 50 ms max instead of 10. Most ECs respond
in < 5 us (50 us when heavily loaded). However, some time out
occasionally even with a 10 ms timeout. For delays past 1 ms, use
msleep instead of DELAY to give SCI interrupts a chance to occur.
* Add EcCommand to send a command and wait for the appropriate event.
* The hw.acpi.ec.event_driven tunable is no longer applicable and
has been removed.
Ideas from: Linux
bus_dma_tag_create. We need to be sure that our packets are
kept in-sequence (that's how ATM is supposed to work) and
therefor use BUS_DMA_NOWAIT in all calls to bus_dmamap_load.
For memory allocated with bus_dmamem_alloc the use of anything
other than NULL arguments for the locking is anyway bogus because
this memory never should need bouncing and hence the load should never
be defered.
Allow the receipt of OAM and RM cells on raw connections. Caveat: it seems
that RM cells are still processed by the hardware even when we open the
connection as UBR.
register, present only on 3c90xB and later NICs. This meant that you could
not use a 1500 byte MTU with VLANs on original 3c905/3c900 cards (boomerang
chipset). The boomerang chip does support large frames though, just not
in the same way: you can set the 'allow large frames' bit in the MAC
control register to receive frames up to 4K in size.
Changes:
- Set the 'allow large frames' bit for boomerang chips and increase
the packet size register for cyclone and later chips. This allows
us to use IFCAP_VLAN_MTU on all supported xl(4) NICs.
- Actually set the IFCAP_VLAN_MTU flag in the capabilities word
in xl_attach().
- Change the method used to detect older boomerang chips. My 3c575C
cardbus NIC was being incorrectly identified as 3c90x chip instead
of 3c90xB because the capabilities word in its EEPROM reports
a bizzare value. In addition to checking for the supportsNoTxLength
bit, also check for the absence of the supportsLargePackets bit.
Both of these cases denote a 3c90xB chip.
- Make RX and TX checksums configurable via the SIOCSIFCAP ioctl.
- Avoid an unecessary le32toh() in xl_rxeof(): we already have the
received frame size in the lower 16 bits of rxstat, no need to
read it again.
Tested with 3c905-TX, 3c900-TPO, 3c980C and 3c575C NICs.
on the implied sign extension. The single unified VADDR() macro was
not able to avoid sign extending the VM_MAXUSER_ADDRESS/USRSTACK values.
Be explicit about UVADDR() (positive address space) and KVADDR()
(kernel negative address space) to make mistakes show up more
spectacularly.
Increase user VM space from 1/2TB (512GB) to 128TB.
corresponding release code. This was preventing the use of more than
1/2TB of user VM. I also spent a week staring at this code only to
eventually find that I'd mistakenly typed a P as an R.
rather than a non-existing pte. There is code elsewhere in i386/amd64
pmap that neglects to handle the large page cases because it knows that
it will see PG_PS in the returned "pte".
