Remove evil allocation macros from machdep.c (why was that there???) and
use malloc() instead.
Move paramters out of param.h and into the code itself.
Move a bunch of internal definitions from public sys/*.h headers (without
#ifdef _KERNEL even) into the code itself.
I had hoped to make some of this more dynamic, but the cost of doing
wakeups on all sleeping processes on old arrays was too frightening.
The other possibility is to initialize on the first use, and allow
dynamic sysctl changes to parameters right until that point. That would
allow /etc/rc.sysctl to change SEM* and MSG* defaults as we presently
do with SHM*, but without the nightmare of changing a running system.
coming later this week. Mitsuru IWASAKI provided a patch to -mobile which
I used to make sure I was doing the right thing but only a small part of
the actual patch was used.
that fails to proberly close the disk.
The problem seems to be that the HP burners sometimes return
ready when they actually are not, the solution is to not use
immediate mode on the closing commands. This is suboptimal
for real burners, in that they now hog the ATA bus for possibly
minutes, where its really not nessesary, *sigh*.
was added accidentally, and although not terrible, it would improperly
hide the bug of calling M_PREPEND with a NULL mbuf argument.
Submitted by: jlemon (ISTR)
wrong for many years that negative niceness would lower the priority
of a process below PUSER, and once below PUSER, there were conditionals
in the code that are required to test for whether a process was in
the kernel which would break.
The breakage could (and did) cause lock-ups, basically nothing else
but the least nice program being able to run in some conditions. The
algorithm which adjusts the priority now subtracts PRIO_MIN to do
things properly, and the ESTCPULIM() algorithm was updated to use
PRIO_TOTAL (PRIO_MAX - PRIO_MIN) to calculate the estcpu.
NICE_WEIGHT is now 1 to accomodate the full range of priorities better
(a -20 process with full CPU time has the priority of a +0 process with
no CPU time). There are now 20 queues (exactly; 80 priorities) for
use in user processes' scheduling, and PUSER has been lowered to 48
to accomplish this.
This means, to the user, that things will be scheduled more correctly
(noticeable), there is no lock-up anymore WRT a niced -20 process
never releasing the CPU time for other processes. In this fair system,
tsleep()ed < PUSER processes now will get the proper higher priority
than priority >= PUSER user processes.
The detective work of this was done by me, along with part of the
solution. Luoqi Chen has provided most of the solution, and really
helped me understand what was happening better, to boot :)
Submitted by: luoqi
Concept reviewed by: bde
not u_long. On i386's with 64-bit longs, returning u_longs indirectly
in (more than) the space reserved for uintptr_t's tended to corrupt the
previous frame pointer in the stack frame, so it was not easy to debug.
The type mismatches are hidden by the bogus cast in DEVMETHOD().
declarations of structs for use in prototypes are only necessary if
the struct is not otherwise declared in scope.
Removed prototypes for fdissequential() and fdsequential(). These
functions never existed in FreeBSD.
Fixed most style bugs in FreeBSD changes (mainly disordered prototypes
and prototypes without parameter names).
Note that if_aue doesn't strictly depend on usb because it uses the
method interface for calls rather than using internal symbols, and
because it's a child driver of usb and therefore will not try and do
anything unless the parent usb code is loaded at some point. if_aue does
strictly depend on miibus as it will fail to link if it is missing.
bus/driver/kobj system. I am not 100% sure that this is the correct fix,
but it is harmless and does seem to solve the problem. At worst, it could
cause a tiny memory leak at unload time - this is better than a free(NULL)
and subsequent panic. I'm waiting for comments from Doug about this.
This may yet be backed out and fixed differently.
The change itself is to increment the reference count on drivers in one
case where it appears to have been missed. When everything is unloaded,
kobj_class_free() was being called twice in some cases, and panicing the
second time.
version dependency system. This isn't quite finished, but it is at a
useful stage to do a functional checkpoint.
Highlights:
- version and dependency metadata is gathered via linker sets, so things
are handled the same for static kernels and code built to live in a kld.
- The dependencies are at module level (versus at file level).
- Dependencies determine kld symbol search order - this means that you
cannot link against symbols in another file unless you depend on it. This
is so that you cannot accidently unload the target out from underneath
the ones referencing it.
- It is flexible enough that we can put tags in #include files and macros
so that we can get decent hooks for enforcing recompiles on incompatable
ABI changes. eg: if we change struct proc, we could force a recompile
for all kld's that reference the proc struct.
- Tangled dependency references at boot time are sorted. Files are
relocated once all their dependencies are already relocated.
Caveats:
- Loader support is incomplete, but has been worked on seperately.
- Actual enforcement of the version number tags is not active yet - just
the module dependencies are live. The actual structure of versioning
hasn't been agreed on yet. (eg: major.minor, or whatever)
- There is some backwards compatability for old modules without metadata
but I'm not sure how good it is.
This is based on work originally done by Boris Popov (bp@freebsd.org),
but I'm not sure he'd recognize much of it now. Don't blame him. :-)
Also, ideas have been borrowed from Mike Smith.
53C810 non 'A', 53C815 and 53C825 non 'A' are now
attached by the driver (by default).
The driver uses a different SCRIPTS set based on
MEMORY MOVE instructions for these chips.
2 SCRIPTS sets (firmwares) numbered #1 and #2 are
used for the whole support of the 53C8XX family
to get possible:
- FW #1 : Only based on MEMORY MOVE instructions.
Selected for 810, 815, 825.
- FW #2 : LOAD/STORE based. This is the firmware
also used by previous driver versions.
Selected for other chips.
When both `ncr' and `sym' are configured, `sym'
will now attach all the 53C8XX devices by default.
Previous balancing between `ncr' and `sym' can be
preserved by:
- Either editing sym_conf.h and commenting the
following compile option:
#define SYM_CONF_GENERIC_SUPPORT
(This also saves about 3.5Kb of kernel memory).
- Or setting kernel config option
SYM_SETUP_LP_PROBE_MAP to 64 (bit 0x40)
to PPTP) with more generic PacketAliasRedirectProto().
Major number is not bumped because it is believed that noone
has started using PacketAliasRedirectPptp() yet.
reset their grace timer as their ownership crossed the soft limit
threshhold. Thus if they had been over their limit in the past,
they were suddenly penalized as if they had been over their limit
ever since. The fix is to check when root gives away files, that
when the receiving user crosses their soft limit, their grace timer
is reset. See the PR report for a detailed method of reproducing
the bug.
PR: kern/17128
Submitted by: Andre Albsmeier <andre.albsmeier@mchp.siemens.de>
Reviewed by: Kirk McKusick <mckusick@mckusick.com>
first. This will fix a few cards that hang on the WD probe. He tells
me that PAO went one step farther and removed the WD proble completely
and none of the cards in the 2.x database broke in PAO3. Since I'm
more conservative in this code, I'm just swapping the order, which he
said also fixed his problem.
Reviewed by: mdodd, iwasaki
Submitted by: sanpai@sanpai.org
The makefile contains a reference to /sys/dev/ppbus. What really should
be done is copy the header files to /usr/include/sys/dev/ppbus.
PR: kern/16767
Submitted by: Jin Guojun (FTG staff) <jin@gracie.lbl.gov>
LSNAT links are first created by either PacketAliasRedirectPort() or
PacketAliasRedirectAddress() and then set up by one or more calls to
PacketAliasAddServer().
program running under linux emulation, the script binary is checked for
in /compat/linux first. Without this patch the wrong script binary
(i.e. the FreeBSD binary) will be run instead of the linux binary.
For example, #!/bin/sh, thus breaking out of linux compatibility mode.
This solves a number of problems people have had installing linux
software on FreeBSD boxes.
This driver should support both the SSI (V.35 etc) E1/T1 unchannelized,
DS3 and HSSI cards. Only tested on the SSI card.
More info at: http://www.lanmedia.com
Thanks to LanMedia for donating two LMC1000P cards.
if_de.c driver modified by: LanMedia
NetGraphification by: Stephen Kiernan <sk-ports@vegamuse.org>
- Fixed bogus CIS tuple dumping (Network node ID, IRQ modes and etc.)
- Include telling drivers ethernet address if Network node ID
tuple is available. This is usefull for some bogus ehter cards which
can't get correct ethernet address from CIS tupple.
Obtained from: PAO3
There's no excuse to have code in synthetic filestores that allows direct
references to the textvp anymore.
Feature requested by: msmith
Feature agreed to by: warner
Move requested by: phk
Move agreed to by: bde
below). This did not work previously because interrupts were
disabled when PXE calls were being made, and they must be enabled.
This should also allow us to be compliant with all newer PXE rom's
from Intel.
For PXE 0.99, this has been tested using the Intel N440BX motherboard
and I am confident it will work on the Intel L440GX motherboard.
Lots of help/information from: jhb, peter
I would like to thank Michael Johnston <michael.johnston@intel.com>,
Mike Henry <mike.henry@intel.com>, and all the other PXE developers
at Intel for their help, and information in helping solve this
problem.
from user mode. Don't disable interrupts when returning from vm86 mode
to user mode either. Now, we only disable interrupts before calling a
hardware interrupt handler, which is the only time we _should_ be
disabling interrupts.
Because of this, err, feature, any routine that one called in vm86 mode
had to re-enable interrupts by setting the interrupt flag or interrupts
would remain disabled even after the routine returned. For example, I
have a simple debugging routine that uses a vm86 mode function to dump
any arbitrary memory word that I use to read the BIOS timer or any other
memory location. This function does 1 load instruction from memory and
then returns. Since it didn't re-enable interrupts, the first time I
called it to read the BIOS timer, it disabled interrupts. This also
affected the PXE bootstrap as it needs interrupts enabled while it is
processing. This patch fixes both of those situations so that those
functions do not worry about having to enable interrupts. Hardware
interrupt handlers worked fine with the old code because they always
enable interrupts as part of their routine.
If you have any problems with the loader after this commit, please
let me know. I'd like to MFC it in a week or two since PXE support
needs it.
Noticed by: ps, Michael Johnston <michael.johnston@intel.com>
in struct bio. Eventually, bio_offset will probably obsolete the
bio_blkno and bio_pblkno fields.
Remove the special hack in atapi-cd.c to determine of bio_offset was valid.
that seems to be working (I have a MF card that has a 336 modem and
ethernet that the probe routine finds, but the attach fails on because
pccardd doesn't do what you'd like with MF cards all the time).
serial gdb: interrupts were causing either overruns or stealing
characters. Put splhigh() around the routines which transfer packets
across the line. Since this happens when the system is halted in
debug, this doesn't cause any particular problem. Now it is possible
to run the link at 115,200 bps.
PR: (not assigned yet, must be in limbo somewhere)
Add partial support for detecting non-existent gdb devices.
Add $FreeBSD$ tag.
* Report link errors to stdout with uprintf() so that the user can see
what went wrong (PR kern/9214).
* Add support code to allow module symbols to be loaded into GDB using
the debugger's "sharedlibrary" command.
Removed NETBSDTX spares.
Renamed a load of functions - specifcally
start -> tx
init has download, sj and assoc as sub-functions
report_params and update params move to repparams/upparams
Tidied up old #define's
Got rid of old DPRINTF, made printf's RAY_PRINTF and panic's RAY_PANIC
All code KNF
Removed checking the ECF_TO_HOST area in ray_init - this gets hosed
when someone updates/read a parameter from the card so causes
unneeded grief. This required moving the tib check into ray_attach.
Changed handling of interface flags in ioctl so that promiscous mode
changes are only done as needed.
Sequences of comq entries can be added to an array and automatically
dealt with - used in ray_init_user and others.
Moved IFF_RUNNING checks from the comq commands to com_runq - still
not sure what to do so we PANIC - will be fixed.
ray_sj now checks to see if any of the parameters it can update can be
updated.
ray_sj_done now updates parameters to the current n/w set if we changed
them earlier. I was being a bit thick in earlier comments as to why this
check was done - the ECF never changes the parameters.
Assocication with APs is handled outside of ray_sj - need to add
WEP stuff if I ever get my hands on an AP.
ray_stop, ray_unload and ray_reset are currently broken - reset isn't used anyway.
ray_tx_XXX and ray_rx only have cosmetic changes.
Interrupt handler now gets command out of ccs. This is so ray_intr_ccs and
ray_intr_rcs don't need to remap CM (I'm trying to roll things up a bit as
it must be slowing us down).
ray_intr_ccs just vectors commands to functions or error checks - a
jump table could replace it in ray_intr.
Ditto for ray_intr_rcs.
mcast is currently broken - I need to do more work for ALLMULTI etc. This is reasonably easy to fix becuase of the comq array stuff.
Added a few more checks into repparams so that it is firmware version aware. It also uses the new comq framework.
Parameter updating is also much better with the update parameter return routine re-worked.
Added a couple of runq entry helper functions - one that malloc's a new
entry and fills it in with sensible defaults and another that adds and
runs a seqencue of entries.
search routine, and scratching our heads over why it was so obfuscated.
This delta fixes a number of confusing style bugs and renames several
structure members to have more meaningful names. There remain a number
of odd control-flow structures. These changes do not affect the generated
code.
shared memory objects are regular files; the shm_open(3) routine
uses fcntl(2) to set a flag on the descriptor which tells mmap(2)
to automatically apply MAP_NOSYNC.
Not objected to by: bde, dillon, dufault, jasone
booting on a RAID volume.
- Change 'id_foo' and 'idfoo' to 'idad_foo' and 'idadfoo'. This makes
names more consistent with the devices that the code belongs to (more
in line with the style used in the amr/mlx driver.)
Reviewed by: jlemon
- Break out the support for the XMAC II's PHY into an miibus driver.
- Reorganize the probe/attach stuff using newbus. Each XMAC is now
attached to the parent GEnesis controller using newbus. This is
necessary since each XMAC must also have an attached miibus, and
the miibus read/write register routines need to be able to get
at the softc struct for each XMAC, not the one for the parent
controller. This allows me to get rid of the grotty code I added
for selecting the unit numbers for the ifnet interfaces: the unit
numbers are now derived from the newbus-assigned unit numbers,
which should track with the ifnet interface numbers. I think.
At the very least, there should never be any collisions.
- Add support for the SK-9821 and SK-9822 1000baseTX adapters. Special
thanks to SysKonnect for loaning me two adapters for testing.
would be returned with a wrong value.
While we're here, get rid of unnecessary panic call.
PR: 17311, 12996, 14457
Submitted by: Patrick Bihan-Faou <patrick@mindstep.com>,
Kris Kennaway <kris@FreeBSD.org>
Working download and start/join network. What's nice is that I didn't touch
TX or RX code at all.
Not fully done yet - stubs for assoc, cast, promisc and user param updating. I
can't work out the "nicest" way to get the mcast and promisc interactions
sorted without needing to abort runq commands half way through. And, as
mcast/promisc used elsewhere I want to get this right.
Other stuff
Fancy new debugging with __FUNCTION__ and __LINE__.
Lots of panic's inserted until I check some stuff out.
Removed all of the old scheduler.
Removed start join timeout as I don't need it.
Removed the promiscuous broadcast code that waas ifdef'd out.
New CCS allocation
where we can have targets (based on topology).
Much more importantly, make sure all mods to isp_sendmarker or |= so
we don't lose the marking of a bus that needs to have a marker sent for it.
require full logins after a LIP, which always led to loop resets, and
various other perturbations.
Update 2200 f/w from 2.01.00 release to 2.01.09 release.
- Free resources in all the possible attachment failure cases.
One critical bugfix:
- Fix a race where it was possible to get out of synch with the log
messages from the controller, if the controller hung up for any
period of time (eg. while resetting a SCSI bus). This would result
in an endless string of console errors, bringing the machine to a
halt.
o Modify xe driver to use this.
There's still some issues with this code, so xe can't map the cis just
yet. I'm thinking about how to resolve the issue. pccard_nbk's
pccard_alloc_resource is getting in the way.
Without this fix, all IPv6 TCP RST packet has wrong cksum value,
so IPv6 connect() trial to 5.0 machine won't fail until tcp connect timeout,
when they should fail soon.
Thanks to haro@tk.kubota.co.jp (Munehiro Matsuda) for his much debugging
help and detailed info.
I've done this by having requests to allocate memory propigate up the
tree. We'll see how well this works and reevaluate if it isn't
working well. Also initialize ptr in the tuple. As well as minor
reorg of memory allocation. Likely need to do similar things for I/O
when the time comes.
I've move all defines from pccardchip.h into pccardvar.h and
eliminated pccardchip.h.
chances of consistency with other file/directory meta-data in a
write. In the current set of extended attribute applications,
this does not hurt much. This should be discussed again later when
it comes time to optimize performance of attributes.
o Include an inode generation number in the per-attribute header
information. This allows consistency verification to catch when
a crash occurs, or an inode is recycled while attributes are not
properly configured. For now, an irritating error message is
displayed when an inconsistency occurs. At some point, may introduce
an ``extattrctl check ...'' which catches these before attributes
are enabled. Not today. If you get this message, it means you
somehow managed to get your attribute backing file out of synch
with the file system. When this occurs, attribute not found is
returned (== undefined). Writes will overwrite the value there
correcting the problem. Might want to think about introducing
a new errno or two to handle this kind of situation.
Discussed with: kris
maintainers.
After we established our branding method of writing upto 8 characters of
the OS name into the ELF header in the padding; the Binutils maintainers
and/or SCO (as USL) decided that instead the ELF header should grow two new
fields -- EI_OSABI and EI_ABIVERSION. Each of these are an 8-bit unsigned
integer. SCO has assigned official values for the EI_OSABI field. In
addition to this, the Binutils maintainers and NetBSD decided that a better
ELF branding method was to include ABI information in a ".note" ELF
section.
With this set of changes, we will now create ELF binaries branded using
both "official" methods. Due to the complexity of adding a section to a
binary, binaries branded with ``brandelf'' will only brand using the
EI_OSABI method. Also due to the complexity of pulling a section out of an
ELF file vs. poking around in the ELF header, our image activator only
looks at the EI_OSABI header field.
Note that a new kernel can still properly load old binaries except for
Linux static binaries branded in our old method.
*
* For a short period of time, ``ld'' will also brand ELF binaries
* using our old method. This is so people can still use kernel.old
* with a new world. This support will be removed before 5.0-RELEASE,
* and may not last anywhere upto the actual release. My expiration
* time for this is about 6mo.
*
source address when receiving frames (and keep using address 2 when in
pseudo-IBSS mode). This is apparently necessary in order to obtain the
true MAC address of the sending station which is needed for PPPoE.
Patch supplied by: Blaz Zupan <blaz@amis.net>
- Add support for using the PCI BIOS functions for configuration space
accesses, and make this the default.
- Make PNPBIOS the default (obsoletes the PNPBIOS config option).
- Add two new boot-time tunables to disable each of the above.
multi-firmware support. This patch just changes
numerous names in the driver sources and is actually
nilpotent. This has been checked by comparing the
generated assembly code.
The names that have been changed are related to the
script names. They were named `script' and `script H'.
They are now named respectively `script A' and
`script B'.
o Put back in {} removed during over-zealous cleanup of gratuitous
debugging output during preparation for the commit. Due to the
missing {}, writes on extended attributes always silently failed.
Doh.
o Don't unlock the target vnode if it's the backing vnode, as we
don't lock the target vnode if it's the backing vnode.
before I rip out the scheduler - whilst v. nice 'n all, it is doing the
wrong job. We need something that sends commands to the card atomically
so dhcp etc. works right.
I've renamed and moved a lot of the scheduler code so that it is all
in one place and all starts with ray_cmd_
ray_stop has some debugging crap left in - to be deleted rsn
extattr.h to be included. This fixes the broken ext2fs build as of
the import of extattr code.
Also added $FreeBSD: $ to a couple of files that didn't have them,
without which I couldn't commit this fix.
Reported by: "George W. Dinolt" <gdinolt@pacbell.net>
support. Changes are rather simplifications of the SCRIPTS
interface (prior to complexifying it again;) ), dead code
removes and comment fixes.
Code removed:
- Handling of kernel variables referenced from SCRIPTS.
- Handling of selection without ATN.
Slightly rewritten:
- Handling of illegal phase (4/5) and data overrun conditions.
Simplifications:
- Extended error flag and bits now only set from the C code.
- Move the extended error status (xerr_status) and nego
status (nego_status) outside the data structure accessed
by SCRIPTS (struct dsb).
- Get rid of the script status field (scr_st).
- Only patch SCR_NO_OP SCRIPTS instructions to adapt SCRIPTS
to actual chip capabilities.
Cosmetic changes:
- Miscellaneous comments in SCRIPTS.
- FreeBSD_4_Bus define replaced by FreeBSD_Bus_Io_Abstraction.
Exceptions:
Vinum untouched. This means that it cannot be compiled.
Greg Lehey is on the case.
CCD not converted yet, casts to struct buf (still safe)
atapi-cd casts to struct buf to examine B_PHYS
(name, value) pairs to be associated with inodes. This support is
used for ACLs, MAC labels, and Capabilities in the TrustedBSD
security extensions, which are currently under development.
In this implementation, attributes are backed to data vnodes in the
style of the quota support in FFS. Support for FFS extended
attributes may be enabled using the FFS_EXTATTR kernel option
(disabled by default). Userland utilities and man pages will be
committed in the next batch. VFS interfaces and man pages have
been in the repo since 4.0-RELEASE and are unchanged.
o ufs/ufs/extattr.h: UFS-specific extattr defines
o ufs/ufs/ufs_extattr.c: bulk of support routines
o ufs/{ufs,ffs,mfs}/*.[ch]: hooks and extattr.h includes
o contrib/softupdates/ffs_softdep.c: extattr.h includes
o conf/options, conf/files, i386/conf/LINT: added FFS_EXTATTR
o coda/coda_vfsops.c: XXX required extattr.h due to ufsmount.h
(This should not be the case, and will be fixed in a future commit)
Currently attributes are not supported in MFS. This will be fixed.
Reviewed by: adrian, bp, freebsd-fs, other unthanked souls
Obtained from: TrustedBSD Project
connections, after SYN packets were seen from both ends. Before this,
it would get applied right after the first SYN packet was seen (either
from client or server). With broken TCP connection attempts, when the
remote end does not respond with SYNACK nor with RST, this resulted in
having a useless (ie, no actual TCP connection associated with it) TCP
link with 86400 seconds TTL, wasting system memory. With high rate of
such broken connection attempts (for example, remote end simply blocks
these connection attempts with ipfw(8) without sending RST back), this
could result in a denial-of-service.
PR: bin/17963
incorrectly attach itself to ThunderLAN adapters which happen to have
a PHY who's model number happens out to be 0.
Problem reported by: Peter L. Thomas <Pete@painless-computing.com>
(which is actually a CATC Netmate).
Gee. I don't know why people make such a fuss over supporting USB ethernet
NICs. This is easier than collecting graft.
in attach.
- Change a EISA_CHANNEL_CLEAR to EISA_CHANNEL_BUSY in ida_v1_submit().
This may fix the problem with EISA IDA adapters though we have not heard
back from testers yet.
Reviewed by: jlemon
o Eliminate cross calls between the devices. Instead move to using the
newbus messaging system. Added three new card calls: attach_card,
detach_card, get_type.
o Eliminate interrupt routine in pccard we never use.
o Move from deactivate to detach for removing cards.
o Start mapping CIS memory, but it is broken and causes panics. At least
it is closer to working than before.
o Eliminate struct device everywhere. It was bogus.
o Initialize softc for pccard device so we have valid pointers to
ourselves.
o Implement routine to find the pcic ivar for a child device of the pccard so
we can use it to talk to the pcic hardware.
o Lots of minor tiding up.
This version now panics when we try to read the CIS. The next batch
of work to make this work is what was outlined in my posting to mobile
about resource allocation and such.
