Bill Paul [Fri, 7 May 1999 03:28:54 +0000 (03:28 +0000)]
Add support to wicontrol(8) and wi(4) for enabling and configuring
power management. This will only work on newer firmware revisions; older
firmware will silently ignore the attempts to turn power management on.
Kirk McKusick [Fri, 7 May 1999 02:26:47 +0000 (02:26 +0000)]
Severe slowdowns have been reported when creating or removing many
files at once on a filesystem running soft updates. The root of
the problem is that soft updates limits the amount of memory that
may be allocated to dependency structures so as to avoid hogging
kernel memory. The original algorithm just waited for the disk I/O
to catch up and reduce the number of dependencies. This new code
takes a much more aggressive approach. Basically there are two
resources that routinely hit the limit. Inode dependencies during
periods with a high file creation rate and file and block removal
dependencies during periods with a high file removal rate. I have
attacked these problems from two fronts. When the inode dependency
limits are reached, I pick a random inode dependency, UFS_UPDATE
it together with all the other dirty inodes contained within its
disk block and then write that disk block. This trick usually
clears 5-50 inode dependencies in a single disk I/O. For block and
file removal dependencies, I pick a random directory page that has
at least one remove pending and VOP_FSYNC its directory. That
releases all its removal dependencies to the work queue. To further
hasten things along, I also immediately start the work queue process
rather than waiting for its next one second scheduled run.
Peter Wemm [Thu, 6 May 1999 22:21:31 +0000 (22:21 +0000)]
``fix'' the devfs_add_devswf() calls, the printf string wasn't factoring
in the unit number. I'm not so sure about this at all, the SVR4 systems
I have access to have a mixture of names and often %03d format units.
Add a number of interrelated CAM feature enhancements and bug fixes.
NOTE: These changes will require recompilation of any userland
applications, like cdrecord, xmcd, etc., that use the CAM passthrough
interface. A make world is recommended.
camcontrol.[c8]:
- We now support two new commands, "tags" and "negotiate".
- The tags commands allows users to view the number of tagged
openings for a device as well as a number of other related
parameters, and it allows users to set tagged openings for
a device.
- The negotiate command allows users to enable and disable
disconnection and tagged queueing, set sync rates, offsets
and bus width. Note that not all of those features are
available for all controllers. Only the adv, ahc, and ncr
drivers fully support all of the features at this point.
Some cards do not allow the setting of sync rates, offsets and
the like, and some of the drivers don't have any facilities to
do so. Some drivers, like the adw driver, only support enabling
or disabling sync negotiation, but do not support setting sync
rates.
- new description in the camcontrol man page of how to format a disk
- cleanup of the camcontrol inquiry command
- add support in the 'devlist' command for skipping unconfigured devices if
-v was not specified on the command line.
- make use of the new base_transfer_speed in the path inquiry CCB.
- fix CCB bzero cases
cam_xpt.c, cam_sim.[ch], cam_ccb.h:
- new flags on many CCB function codes to designate whether they're
non-immediate, use a user-supplied CCB, and can only be passed from
userland programs via the xpt device. Use these flags in the transport
layer and pass driver to categorize CCBs.
- new flag in the transport layer device matching code for device nodes
that indicates whether a device is unconfigured
- bump the CAM version from 0x10 to 0x11
- Change the CAM ioctls to use the version as their group code, so we can
force users to recompile code even when the CCB size doesn't change.
- add + fill in a new value in the path inquiry CCB, base_transfer_speed.
Remove a corresponding field from the cam_sim structure, and add code to
every SIM to set this field to the proper value.
- Fix the set transfer settings code in the transport layer.
scsi_cd.c:
- make some variables volatile instead of just casting them in various
places
- fix a race condition in the changer code
- attach unless we get a "logical unit not supported" error. This should
fix all of the cases where people have devices that return weird errors
when they don't have media in the drive.
scsi_da.c:
- attach unless we get a "logical unit not supported" error
scsi_pass.c:
- for immediate CCBs, just malloc a CCB to send the user request in. This
gets rid of the 'held' count problem in camcontrol tags.
scsi_pass.h:
- change the CAM ioctls to use the CAM version as their group code.
adv driver:
- Allow changing the sync rate and offset separately.
adw driver
- Allow changing the sync rate and offset separately.
aha driver:
- Don't return CAM_REQ_CMP for SET_TRAN_SETTINGS CCBs.
ahc driver:
- Allow setting offset and sync rate separately
bt driver:
- Don't return CAM_REQ_CMP for SET_TRAN_SETTINGS CCBs.
NCR driver:
- Fix the ultra/ultra 2 negotiation bug
- allow setting both the sync rate and offset separately
Other HBA drivers:
- Put code in to set the base_transfer_speed field for
XPT_GET_TRAN_SETTINGS CCBs.
Roger Hardiman [Thu, 6 May 1999 19:24:45 +0000 (19:24 +0000)]
Added details of PCI network cards which work by emulating the NE2000.
RealTek 8029, NetVin 5000, Winbond W89C940, Surecom NE-34, VIA VT86C926.
(checked with Bill Paul)
Mention the Brooktree Bt878 is supported by the Bt848 driver.
Peter Wemm [Thu, 6 May 1999 18:50:10 +0000 (18:50 +0000)]
Fix a precedence bug in the atapi tape driver. I think it could either
write a filemark where it wasn't needed, or neglect to write one at all,
depending on how the boolean converted to an int value for the &.
Peter Wemm [Thu, 6 May 1999 18:44:42 +0000 (18:44 +0000)]
Fix up a few easy 'assignment used as truth value' and 'suggest parens
around && within ||' type warnings. I'm pretty sure I have not masked
any problems here, I've committed real problem fixes seperately.
Bill Paul [Thu, 6 May 1999 16:32:45 +0000 (16:32 +0000)]
Modify wicontrol(8) and wi(4) to allow setting the frequency of the
WaveLAN's radio modem. The default is whatever the NIC uses since NICs
sold in different countries may default to different frequencies. (The
Lose95/LoseNT software doesn't let you select the channel so it's probably
not really meant to be changed.)
Bill Paul [Thu, 6 May 1999 15:32:52 +0000 (15:32 +0000)]
Tweak the Macronix driver to hopefully make it more reliable:
- Change to the same transmit scheme as the PNIC driver.
- Dynamically set the cache alignment, and set burst size the same as
the PNIC driver in mx_init().
- Enable 'store and forward' mode by default. This is the slowest option
and it does reduce 100Mbps performance somewhat, but it's the most
reliable setting I can find. I'm more interested in having the driver
work reliably than trying to squeeze the best performance out of it.
The reason I'm doing this is that on *some* systems you may see a lot
of transmit underruns (which I can't explain: these are *fast* test
systems) and these errors seem to cause unusual and decidedly
non-tulip-like behavior. In normal 10Mbps mode, performance is fine
(you can easily saturate a 10Mbps link).
Also tweak some of the other drivers:
- Increase the size of the TX ring for the Winbond, ASIX, VIA Rhine
and PNIC drivers.
- Set a larger value for ifq_maxlen in the ThunderLAN driver. The setting
of TL_TX_LIST_CNT - 1 is too low (the ThunderLAN driver only allocates
20 transmit descriptors, and I don't want to fiddle with that now
because the ThunderLAN's descriptor structure is an oddball size
compared to the others).
Peter Wemm [Thu, 6 May 1999 13:42:25 +0000 (13:42 +0000)]
Move the proc0 init before the driver probe/attach etc since machdep.c
doesn't set curproc anymore, and certain drivers like to tsleep() during
probes, usb for example.
Peter Wemm [Thu, 6 May 1999 12:47:21 +0000 (12:47 +0000)]
I'm not sure why the #ifdef SMP became #if 1 (this overrode the npx probe
and always succeeded as is required on SMP). Anyway, reverting this
still compiles and appears ok.
Bruce Evans [Thu, 6 May 1999 09:44:57 +0000 (09:44 +0000)]
Fixed profiling of elf kernels. Made high resolution profiling compile
for elf kernels (it is broken for all kernels due to lack of egcs support).
Renaming of many assembler labels is avoided by declaring by declaring
the labels that need to be visible to gprof as having type "function"
and depending on the elf version of gprof being zealous about discarding
the others. A few type declarations are still missing, mainly for SMP.
Bill Paul [Thu, 6 May 1999 03:34:02 +0000 (03:34 +0000)]
Fix some byte ordering problems; I was storing string lengths wrong,
which was causing wicontrol to crash when reading string parameters
from the WaveLAN.
Archie Cobbs [Thu, 6 May 1999 03:23:24 +0000 (03:23 +0000)]
Add the capability for traceroute(8) to send packets of any IP protocol
instead of just UDP; an alternate protocol is specified by '-P proto'.
This is useful for finding routers that are blocking packets based on
IP protocol. New handlers can be added fairly easily to do protocol-
specific things.
John Birrell [Wed, 5 May 1999 11:24:15 +0000 (11:24 +0000)]
Add an INIT_PATH option to allow the init_path (for sys/kern/init_main.c)
to be customised in an embedded system which doesn't want to
run either init of sysinstall.
Bill Paul [Wed, 5 May 1999 07:37:11 +0000 (07:37 +0000)]
Add device driver support for the Lucent WaveLAN/IEEE 802.11 PCMCIA
adapter (and some workalikes). Also add man pages and a wicontrol
utility to manipulate some of the card parameters.
This driver was written using information gleaned from the Lucent HCF Light
library, though it does not use any of the HCF Light code itself, mainly
because it's contaminated by the GPL (but also because it's pretty gross).
The HCF Light lacks certain featurs from the full (but proprietary) HCF
library, including 802.11 frame encapsulation support, however it has
just enough register information about the Hermes chip to allow someone
with enough spare time and energy to implement a proper driver. (I would
have prefered getting my hands on the Hermes manual, but that's proprietary
too. For those who are wondering, the Linux driver uses the proprietary
HCF library, but it's provided in object code form only.)
Note that I do not have access to a WavePOINT access point, so I have
only been able to test ad-hoc mode. The wicontrol utility can turn on
BSS mode, but I don't know for certain that the NIC will associate with
an access point correctly. Testers are encouraged to send their results
to me so that I can find out if I screwed up or not.
Bill Paul [Wed, 5 May 1999 07:11:38 +0000 (07:11 +0000)]
Add device driver support for the Lucent WaveLAN/IEEE 802.11 wireless
network adapters. These are all PCMCIA devices (the ISA version is a
PCMCIA to ISA bridge with a PCMCIA card plugged into it). Also add a
wicontrol utility to read and write some of the card's parameters.
Note: I do not have access to a WavePOINT access point, so I have only
been able to test this driver in ad-hoc (point to point) mode. The
wicontrol utility allows programming the desired service set name (SSID)
and enabling BSS mode, but I can't tell for sure if it works (I know the
card switches modes, but I can't verify that it joins a service set
correctly).
This driver was written using information gleaned from the Lucent HCF Light
library, which is an API library designed to simplify driver development
for devices based on the Lucent Hermes chip. Unfortunately, the HCF Light
is missing certain features (like 802.11 frame encapsulation!) which are
available only in the proprietary complete HCF code, which is not available
to the public. This driver uses none of the HCF Light code: it's very ugly
and contaminated by the GPL. IP and ARP packets are encapsulated as 802.11
frames, everything else is encapsulated as 802.3.
(It would be easier to just get the Hermes programming manual, but that's
not publically available either. For those who are wondering, the Linux
WaveLAN/IEEE driver uses the proprietary HCF code, which is provided in
object code form only. So much for supporting open source sofware.)
Multicast filter support is implemented, however it appears that the
filter doesn't work: programming in one IP mutlicast group enables them
all.
Doug Rabson [Tue, 4 May 1999 21:18:20 +0000 (21:18 +0000)]
Use unit, not device_id as an argument to an old-style ISA interrupt
handler. This fixes pnp interrupts and would have fixed pccard interrupts
but a workaround has been applied there.
This the sound driver problems which people have reported with new-bus.
Make `ed0' example match reality.
Merge `ed1' and `ed2' examples as we don't ship with an `ed1' any more we
yet, the example of extra flags could be useful to someone.