written by Stuart Walsh and Duncan Barclay (with some kibbitzing by
me). I'm checking it in on Stuart's behalf.
The BCM4401 is built into several x86 laptop and desktop systems. For the
moment, I have only enabled it in the x86 kernel config because although
it's a PCI device, I haven't heard of any standalone NICs that use it. If
somebody knows of one, we can easily add it to the other arches.
This driver uses register/structure data gleaned from the Linux
driver released by Broadcom, but does not contain any of the code
from the Linux driver itself. It uses busdma.
rl(4) driver and put it in a new re(4) driver. The re(4) driver shares
the if_rlreg.h file with rl(4) but is a separate module. (Ultimately
I may change this. For now, it's convenient.)
rl(4) has been modified so that it will never attach to an 8139C+
chip, leaving it to re(4) instead. Only re(4) has the PCI IDs to
match the 8169/8169S/8110S gigE chips. if_re.c contains the same
basic code that was originally bolted onto if_rl.c, with the
following updates:
- Added support for jumbo frames. Currently, there seems to be
a limit of approximately 6200 bytes for jumbo frames on transmit.
(This was determined via experimentation.) The 8169S/8110S chips
apparently are limited to 7.5K frames on transmit. This may require
some more work, though the framework to handle jumbo frames on RX
is in place: the re_rxeof() routine will gather up frames than span
multiple 2K clusters into a single mbuf list.
- Fixed bug in re_txeof(): if we reap some of the TX buffers,
but there are still some pending, re-arm the timer before exiting
re_txeof() so that another timeout interrupt will be generated, just
in case re_start() doesn't do it for us.
- Handle the 'link state changed' interrupt
- Fix a detach bug. If re(4) is loaded as a module, and you do
tcpdump -i re0, then you do 'kldunload if_re,' the system will
panic after a few seconds. This happens because ether_ifdetach()
ends up calling the BPF detach code, which notices the interface
is in promiscuous mode and tries to switch promisc mode off while
detaching the BPF listner. This ultimately results in a call
to re_ioctl() (due to SIOCSIFFLAGS), which in turn calls re_init()
to handle the IFF_PROMISC flag change. Unfortunately, calling re_init()
here turns the chip back on and restarts the 1-second timeout loop
that drives re_tick(). By the time the timeout fires, if_re.ko
has been unloaded, which results in a call to invalid code and
blows up the system.
To fix this, I cleared the IFF_UP flag before calling ether_ifdetach(),
which stops the ioctl routine from trying to reset the chip.
- Modified comments in re_rxeof() relating to the difference in
RX descriptor status bit layout between the 8139C+ and the gigE
chips. The layout is different because the frame length field
was expanded from 12 bits to 13, and they got rid of one of the
status bits to make room.
- Add diagnostic code (re_diag()) to test for the case where a user
has installed a broken 32-bit 8169 PCI NIC in a 64-bit slot. Some
NICs have the REQ64# and ACK64# lines connected even though the
board is 32-bit only (in this case, they should be pulled high).
This fools the chip into doing 64-bit DMA transfers even though
there is no 64-bit data path. To detect this, re_diag() puts the
chip into digital loopback mode and sets the receiver to promiscuous
mode, then initiates a single 64-byte packet transmission. The
frame is echoed back to the host, and if the frame contents are
intact, we know DMA is working correctly, otherwise we complain
loudly on the console and abort the device attach. (At the moment,
I don't know of any way to work around the problem other than
physically modifying the board, so until/unless I can think of a
software workaround, this will have do to.)
- Created re(4) man page
- Modified rlphy.c to allow re(4) to attach as well as rl(4).
Note that this code works for the sample 8169/Marvell 88E1000 NIC
that I have, but probably won't work for the 8169S/8110S chips.
RealTek has sent me some sample NICs, but they haven't arrived yet.
I will probably need to add an rlgphy driver to handle the on-board
PHY in the 8169S/8110S (it needs special DSP initialization).
This option adds Postfix and Exim to the list, however, qmail is not added
due to license restrictions.
Collaborated with: Simon L. Nielsen <simon@nitro.dk>
Reviewed by: jhb, re@, -audit.
'use entire disk'. Neither for ia64 while I'm here - it needs a MBR if
its going to use fdisk+disklabel. The ia64 case is mostly academic though
because you'd be creating two partitions (dos + freebsd) rather than
a single freebsd-only partition.
PC98 boot blocks don't support UFS2. We keep newfs(8) defaulting to
UFS2.
Warn users that FreeBSD can only boot from a root file system smaller
than 1.5TB; hopefully this will get fixed by the patches currently
floating around on -CURRENT.
Reviewed by: nyan
ethernet controller. The driver has been tested with the LinkSys
USB200M adapter. I know for a fact that there are other devices out
there with this chip but don't have all the USB vendor/device IDs.
Note: I'm not sure if this will force the driver to end up in the
install kernel image or not. Special magic needs to be done to exclude
it to keep the boot floppies from bloating again, someone please
advise.
FreeBSD 5.1-RELEASE and later:
- newfs(8) will now create UFS2 file systems unless UFS1 is specifically
requested (-O1). To do this, I just twiddled the Oflag default.
- sysinstall(8) will now select UFS2 as the default layout for new
file systems unless specifically requested (use '1' and '2' to change
the file system layout in the disk labeler). To do this, I inverted
the ufs2 flag into a ufs1 flag, since ufs2 is now the default and
ufs1 is the edge case. There's a slight semantic change in the
key behavior: '2' no longer toggles, it changes the selection to UFS2.
This is very similar to a patch David O'Brien sent me at one point, and
that I couldn't find.
Approved by: re (telecon)
Reviewed by: mckusick, phk, bmah
physical memory. The default is still 2x physical memory. The nominal
calculation is used to back-off swap auto-allocation ('A'uto command)
when the disk is not large enough to accomodate all filesystem auto-defaults.
This gives other partitions (like /usr) more priority over swap on smaller
disks.
This should help solve reported auto-sizing failures on machines with small
hard drives and huge amounts of memory. For example, a machine with 2G of
disk and 4G of memory will fail to auto-size without this fix.
MFC after: 3 days
'base' dist rename.
- Rework struct dist to allow for different types of dists. There are
currently three types of dists: DT_TARBALL, the traditonal gzipped and
split tar file; DT_PACKAGE, a package; and DT_SUBDIST, a meta-dist in
the tree that has its own array of dists as its contents. For example,
the 'base' dist is a DT_TARBALL dist, the 'perl' dist is a DT_PACKAGE
dist, and the 'src' dist is a DT_SUBDIST dist with its own dist table
that contains 'sbase', 'ssys', etc.
- Add helper macros for defining array entries for the different types of
dists to try and make the statically defined dist table in dist.c more
readable.
- Split the logic to deal with a DT_TARBALL dist out of distExtract()
and into its own distExtractTarball() function. distExtract() now
calls other functions to extract each dist.
- Tweak the percentage complete calculation in distExtractTarball() to
do the multiply prior to the divide so it doesn't have to use floating
point.
- Axe the installPackage() function along with the special handling for
the perl and XFree86 dists in distExtractAll() since distExtract()
handles package dists directly now.
- Add back in subdists for the X packages based on the split up packages
that XFree86-4 uses that as closely map to the X dists we used with
X 3.3.x.
- Lots of things like distSetX() and the X dist masks are no longer
#ifndef X_AS_PKG since we use them in both cases now.
- Make the entire installFixupXFree() function #ifndef X_AS_PKG, we only
call it in that case anyways, and it's not suitable for the X_AS_PKG
case.
- Add in X dist menus for the X_AS_PKG case.
Approved by: re
load drivers from the driver floppy if the "driver_floppy" variable is set
in the kernel environment and call this function after probing devices but
before displaying the main menu.
X-MFC after: as soon as I finish committing to current
Approved by: re@ (blanket)
i386, remove the seatbelt preventing users from setting the UFS2 flag
on the root file system on i386. This seatbelt did not exist on
other platforms.
MFC candidate.
type of new slices and to change the type of existing slices. This also
has the advantage of moving a few #ifdef PC98's up to where the macros
are defined instead of in the middle of the code.
- Change the behavior of the 'T' option in the slice editor so that the
default value in the dialog box is the current type of the existing
slice rather than defaulting to changing the slice to a FreeBSD slice as
this is more intuitive.
Approved by: re
editor, in order to support specifying UFS2 as a newfs option.
(1) Support three different newfs types: NEWFS_UFS, NEWFS_MSDOS, and
NEWFS_CUSTOM. Don't mix up the arguments to them: you can't use
soft updates on an msdos file system.
(2) Distinguish adding new arguments to the newfs command line from
replacing it. Permit the addition of new arguments by the user for
NEWFS_UFS. If we entirely replace the command line provided by
sysinstall, call it NEWFS_CUSTOM. 'N' will now add additional
arguments; 'Z' will opt to replace the newfs command line entirely,
but will prompt the user with their current command line as a
starting point.
(3) Construct the newfs command line dynamically based on the options
provided by the user at label-time. Right now, this means selecting
UFS1 vs. UFS2, and the soft updates flag. Drop in some variables
to support ACLs and MAC Multilabel in the future also, but don't
expose them now.
This provides sysinstall with the ability to do more "in band" editing
of the newfs command line, so we can provide more support for the user,
but doesn't sacrifice the ability to entirely specify the newfs command
line of the user is willing to give up on the cushiness factor. It
also makes it easier for us to specify defaults in the future, and
define conditional behavior based on user configuration selections.
For now, we default to UFS1, and permit UFS2 to be used as the root
only on non-i386 systems.
While I was there, I dropped the default fragment and block sizes,
since newfs has much more sensible defaults now.
Reviewed by: jhb, marcel
Approved by: re
ia64 bits from: marcel
