summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Make VOP_BMAP return a struct bufobj for the underlying storage devicephk2004-11-1515-51/+53
| | | | | | | | | instead of a vnode for it. The vnode_pager does not and should not have any interest in what the filesystem uses for backend. (vfs_cluster doesn't use the backing store argument.)
* Add pbgetbo()/pbrelbo() lighter weight versions of pbgetvp()/pbrelvp().phk2004-11-152-0/+44
|
* Don't manually set b_bufobj, pbgetvp() does this for us.phk2004-11-151-1/+0
|
* More kasserts.phk2004-11-151-1/+6
|
* style polishing.phk2004-11-151-7/+3
|
* Move pbgetvp() and pbrelvp() to vm_pager.c with the rest of the pbuf stuff.phk2004-11-152-46/+44
|
* expect the caller to have called pbrelvp() if necessary.phk2004-11-151-3/+0
|
* Explicitly call pbrelvp()phk2004-11-152-0/+3
|
* Order of arguments after "optional" is not without significance:phk2004-11-151-4/+4
| | | | | | the first field must be the name of the enabling device/option, otherwise config doesn't know there is an option/device of that name.
* Oops, all my test boxes have NO_KERBEROS set, so I didn't hit the kerberos5peter2004-11-151-0/+10
| | | | | build tools problem. I'd missed the kerberos5/tools stuff entirely. Add the missing bits.
* After discussions with Nate, repo copy the acpi assist drivers fromimp2004-11-1511-1886/+17
| | | | | | | | | | i386 to dev/acpi_support. In theory, these devices could be found other than in i386 machines only as amd64 becomes more popular. These drivers don't appear to do anything i386 specific, so move them to dev/acpi_support. Move config lines to files so that those architectures that don't support kernel modules can build them into the kernel. At the same time, rename acpi_snc to acpi_sony to follow the lead of all the other specialty devices.
* Merge 1.3 from acpi_snc.c: don't use return_VALUE() macroimp2004-11-151-24/+22
| | | | Improve style(9) compliance.
* Add comment about why amd64 and ia64 don't build acpi modules.imp2004-11-151-2/+2
|
* Apply a bandaid to avoid hangs on Ultra 2 machines. The second Z8530marcel2004-11-151-1/+1
| | | | | | connects to the keyboard and mouse and needs some special treatment. Until this is fully understood, implemented and tested, simply avoid probing the second Z8530. This is also what the zs(4) driver does.
* Since I'm not using the public API for writingkientzle2004-11-151-6/+8
| | | | | | | the the pax attributes, I shouldn't try using the public API for finishing out the attribute entry, either. This also removes some old dubious state manipulations.
* Pax extended headers were always failingkientzle2004-11-151-1/+1
| | | | | | | | | because the code was using the external API (archive_write_data) and assuming internal error-return conventions. Use the internal API for writing data. Thanks to: Joe Marcus Clarke
* Be slightly more paranoid about using the divisor in a division andmarcel2004-11-151-2/+6
| | | | the calculated baudrate. Neither should be 0.
* Remove the BR tag. When the machine doesn't have the DIG64 HCDPmarcel2004-11-141-1/+1
| | | | | | | | table with console settings, we now only need to know at which address the UART lives. Leaving the baudrate unspecified results in us using the baudrate at which the UART operates. This removes one parameter that can interfere with a successful installation out of the box.
* Implement UART_IOCTL_BAUD. Consequently, when the baudrate was unsetmarcel2004-11-141-1/+11
| | | | for the console, we emit the actual baudrate during bus enumeration.
* Add UART_IOCTL_BAUD to allow us to query the hardware about themarcel2004-11-142-0/+6
| | | | | | | | | current baudrate setting. Use this ioctl() when we don't know the baudrate of the sysdev (as represented by a 0 value). When the ioctl() fails, e.g. when the backend hasn't implemented it or the hardware doesn't provide the means to determine its current baudrate setting, we invalidate the baudrate setting by setting it to -1. None of the backends currently implement the new ioctl().
* So do it like we do in usr.bin/tip/tip/Makefile. ;)ru2004-11-141-0/+2
|
* Make sure the baudrate specified with the BR tag is somewhat sane.marcel2004-11-141-0/+20
| | | | | | | | A baudrate we consider insane is silently replaced with 0. When the baudrate is 0, we will not try to program the hardware. Instead we leave the communication speed unaltered, maximizing the chance to have a working console. Obviously this means we allow specifying a 0 baudrate for exactly that purpose.
* Revert previous commit. As ru explains:schweikh2004-11-141-1/+1
| | | | | | | | | | | In the old world (as the surrounding comment in makefile says), there was the /usr/libexec/ld-elf.so.1 binary which is now a symlink to /libexec/ld-elf.so.1. To symlink, we need to make sure that the _target_ (and the target is /usr/libexec/ld-elf.so.1) doesn't have "schg" flag set. A real solution is to protect the chflags call only if target exists, like we do in usr.bin/tip/tip/Makefile. Requested by: ru
* Make it clear that the acpi modules can only be compiled on i386.scottl2004-11-141-0/+8
|
* s/return_VALUE/return / to fix build breakage when ACPI_DEBUG istruckman2004-11-141-3/+3
| | | | defined.
* Further refine the if_em vlan fix in if_em.c:1.53:rwatson2004-11-141-4/+19
| | | | | | | | | | | | | - Because em_encap() can now fail in a way that leaves us without an mbuf chain, potentially set *m_headp to NULL if that happens, so that the caller can do the right thing. This case can occur when we try to prepend the vlan header mbuf but can't allocate additional memory. - Modify the caller of em_encap() to detect a NULL m_head and not try to queue the mbuf if that happens. - When em_encap() fails, make sure to call bus_dmamap_destroy() to clean up.
* Commit more debugging output. This is a little bit of using a large hammer,obrien2004-11-142-0/+20
| | | | | | | | | but sk(4) is so prevalent on AMD64 motherboards we need to reduce the number of round trips in the mailing lists trying to get sufficient information to make sure we've got a handle on all the problems and are working towards making sk(4) solid. Submitted by: bz
* Unify the ci/co variables now that the the tty drivers now use the sameobrien2004-11-141-40/+35
| | | | character for both.
* Catch up with PHK's sio(4) rework [sys/dev/sio/sio.c rev. 1.456].obrien2004-11-141-23/+23
|
* Currently if the user specifies -e and procfs is not mounted on /proc,csjp2004-11-141-0/+25
| | | | | | | | | | | printing of the process environment will fail silently. -define a function which will check to see if procfs is mounted on /proc -Implement this test if the user specified -e -If procfs is not mounted on /proc and -e was specified, print a warning. informing the user that procfs(5) is required. Reviewed by: wes, rwatson
* Fix a reference from pool(9) -> zone(9), but keep on talking about "memorymlaier2004-11-141-1/+1
| | | | | | pools" as that is what UMA provides. Submitted by: Jay <jay NO meangrape SPAM com>
* Move the bit of the syncer which deals with vnodes into a separatephk2004-11-141-36/+46
| | | | function.
* Rearrange memory management for ioctl arguments to use stronger checksphk2004-11-141-25/+16
| | | | for illegal values and don't store them on the stack any more.
* Avoid an (ignored) error by invoking chflags on the link target, not theschweikh2004-11-141-1/+1
| | | | | | | | symlink. PR: kern/73016 Submitted by: John E. Hein <jhein@timing.com> MFC after: 1 week
* style polish.phk2004-11-141-8/+6
|
* Move #define up.phk2004-11-141-5/+5
|
* tr(1) attempts to convert \n[n][n] sequences into octal digits, but doesn'tjkh2004-11-141-1/+1
| | | | | | | | check to see that a given digit is actually an octal digit. This leads to unusual consequences if passed in values like \9. Reported by: Joseph Davison (OpenDarwin project) MFC after: 1 week
* Use PORTSDIR, falling back to /usr/portsimp2004-11-141-1/+1
|
* Remove stub libxpg4. All functionality was merged into libc a long time ago.tjr2004-11-133-19/+1
|
* Kill ACPI_FUNCTION_TRACE. It is killing the tinderbox build, and itimp2004-11-132-20/+12
| | | | | | | | isn't worth adding to the modules lists that we have to hard code for this to work. Since we print PID right away, we have a trace point already. Minor knf while I'm here.
* Eliminate vop_revoke() function now that devfs_revoke() does the entire job.phk2004-11-132-43/+0
|
* Integrate most of vop_revoke() into devfs_revoke() where it belongs.phk2004-11-131-1/+24
|
* Add the devfs_fp_check() function which helps us get from a struct filephk2004-11-131-70/+34
| | | | | | | | to a cdev and a devsw, doing all the relevant checks along the way. Add the check to see if fp->f_vnode->v_rdev differs from our cached fp->f_data copy of our cdev. If it does the device was revoked and we return ENXIO.
* The default VOP_REVOKE() should be vop_panic() as we should neverphk2004-11-131-1/+1
| | | | get here in the first place.
* VOP_REVOKE() is only ever for VCHR vnodes, so unionfs does notphk2004-11-131-20/+0
| | | | need a vop_revoke() method.
* Add knob NO_NIS (fka NO_YP_LIBC) and make world compileable when set.bz2004-11-1325-32/+125
| | | | | | | | | | | If turned on no NIS support and related programs will be built. Lost parts rediscovered by: Danny Braniss <danny at cs.huji.ac.il> PR: bin/68303 No objections: des, gshapiro, nectar Reviewed by: ru Approved by: rwatson (mentor) MFC after: 2 weeks
* Set ramsize depending on a value from eprom instead of usingbz2004-11-132-6/+18
| | | | | | | | | | hardcoded 128k for Yukon devices. 88E8001 only has 64k of on-chip RAM[1]. [1] http://www.marvell.com/products/pcconn/yukon/Yukon_88E8001_10_073103_final.pdf Tested by: amd64, current Approved by: rwatson (mentor) MFC after: 1 week
* fifos doesn't need a vop_lookup, the default will do fine.phk2004-11-131-19/+0
|
* Sync usage() and SYNOPSIS.ru2004-11-132-2/+3
|
* Polish two functions a bit so that it is easier to wrap them inphk2004-11-131-21/+33
| | | | locks if/when we need that.
OpenPOWER on IntegriCloud