Document import of terminfo

(roy)

Remove termcap as we now use terminfo.

OK: core@, jdc@

(roy)

s/u_int/uint/

(msaitoh)

- Make the initialize sequence close to the document.
- Fix ASF heartbeat sending bug (FreeBSD rev. 1.271 (r202821))
- Use new handshake command for BCM5750 or new controllers (FreeBSD rev. 1.272
  (r202822))

(msaitoh)

Userland now builds and uses terminfo instead of termcap.

OK: core@, jdc@

(roy)

Import my terminfo implementation.
This uses the ncurses terminal definitions.

OK: core@, jdc@

(roy)

pkg_install-20100203:
- Restore PKG_PREFIX in pkg_delete (PR 42731)
- Ensure that the current pkg_install version is at least as new as
  the version used to build the package

Status:

Vendor Tag: PKGSRC
Release Tags: pkg_install-20100203

(joerg)

Enable all devices which can be compiled.
(Tested, or even working, are only few of them.)

(phx)

A few assertions & comments.

(uebayasi)

Ensure reorder protection for amigappc in all bus space functions.
The ne(4) driver (XSurf2), which is the only bus_space device I got,
works now.

(phx)

Add missing parentheses. Found by cppcheck, reported by
Henning Petersen in PR 42728.

(wiz)

Remove extra parenthesis. Found by cppcheck, reported by
Henning Petersen in PR 42732.

(wiz)

Remove extra parenthesis. Found by cppcheck, reported by
Henning Petersen in PR 42730.

(wiz)

Remove extra parenthesis. Found by cppcheck, reported by
Henning Petersen in PR 42729.

(wiz)

No need to compile strlcpy.c and strlcat.c here.

They are in libnbcompat.a, not in src/lib/libc/string
(moved into src/common/lib/libc/string), and
strlcpy.c and strlcat.c derived from crypto/dist/heimdal/lib/roken
won't include the target functions if nbtool_config.h defines
HAVE_STRLCPY and HAVE_STRLCAT.

(tsutsui)

uvm_fault_lower_generic_io: Reduce diff from uvm_loanuobj().

(uebayasi)

Remove trailing whitespace. Use Fn when referencing functions. Sort sections.
New sentence, new line.

(wiz)

uvm_fault_lower_generic_io: One missing mutex_exit(vmobjlock).  Found while
comparing this function with uvm_loanuobj().  (Part of) these should be
merged.

(uebayasi)

Plug liblzf into build.  Liblzf is a small (3504 byte shared library on i386),
very very fast, 2-clause BSD-licensed compressor.  We provide the LZF_STATE
version of the API, which is not the default for generic liblzf.

(tls)

Adjust liblzf to offer the LZF_STATE_ARG API, which is better for small-stack
environments (think bootblocks etc) and no worse elsewise (you can always
just pass the same state argument...) and have correct prototypes when doing
so.

(tls)

Import liblzf version 3.5 (2-clause BSD licensed by Marc Lehmann)

Status:

Vendor Tag: schmorpforge
Release Tags: liblzf-3-5

(tls)

Tickets 1294 and 1295.

(snj)

Apply patch (requested by joerg in ticket #1295):
Update pkg_install to 20100130.
pkg_install-20091115:
Completely ignore @src in pkg_create. Silently ignore the -L option.
The combination of -I and -p are used by pkgsrc for the same result.

Do not overwrite a string with itself using snprintf. This breaks
setting the pkgdb directory internally on Linux. Explicitly check
if the string is the same and otherwise just use xstrdup.

Add support to query arbitrary variables with pkg_admin config-var.

pkg_install-20100130:
- pkg_add -U to rplace an installed version
- refactored man pages
- PKG_DBDIR / PKG_REFCOUNT_DBDIR as pkg_install.conf options
- synced license list
- use connection cache from libfetch

(snj)

Apply patch (requested by joerg in ticket #1294):
Update libfetch to 2.30.
Changes:
libfetch-2.25:
- address a number of lint warnings
- fix strict-alignment issues for GCC 4.4
- fix a bug in the line reading optimisation
- do not reuse a FTP connection if there is an active transfer on it

libfetch-2.26:
- Add support to aggressively cache directory listenings, useful for
  HTTP
- Avoid leaking memory in error cases. From Xavier from Arch Linux.

libfetch-2.30:
- Revamped connection cache, allowing more than one active session
- HTTP keep-alive support

(snj)

Introduce the uhub_ubermatch variable.  Setting it to 1 makes sure
than a uhub attaches even at higher priority than ugen when
ugen_override is set to 1.  This allows to probe the whole USB bus
and attach its functions with ugen.

A better infrastructure to control this would be nice ....

(pooka)

We need to provide eclockfreq, just for the aucc driver. The amigappc uses
mftb timecounters.

(phx)

Typo in comment.

(skrll)

Add my name, as specified in the welcome message.

(yhardy)

Make it compile with all devices from amiga/dev.

(phx)

Rewrote the "amiga" PIC from scratch. Interrupts are based on IPL 1-6 now
and no longer on IRQ 0-13.
Fixed interrupt levels in amigappc_install_handlers() (IPL_SOFTxxx was very
bad!).
Compilation with all (most?) amiga devices was tested and fixed.
Some smaller fixes.
Status:
  With Cybervision64 and A3000 SBIC SCSI (with DMA disabled) the system
  reaches multiuser!
Lots of problems left...

(phx)

IOBlix files were missing.

(phx)

According to TOG:
- asctime{,_r}, ctime{,_r} may return NULL; document that, and avoid coredumps.
- gmtime{,_r}, localtime{,_r} may return NULL and set EOVERFLOW, document and
  set errno.
- when mktime returns (time_t)-1, make it set EOVERFLOW and document it.

XXX: Should be pulled up to 5.x

(christos)

Fix compilation warnings and errors with amigappc.

(phx)

uobj->pgops->pgo_get doing PGO_SYNCIO returns a uobjpage whose uobj backpointer
refers to another "uobj" used to call pgo_get.  Revert the wrong assertion
I made.  My bad.

(This and pgo_get's possible ERESTART return value check is the only 2 behavioral
changes I made.)

Reported by drochner@, thanks.

(uebayasi)

Don't pass an unnecessary reference to uvm_loanbreak_anon().

Requested by rmind@.

(uebayasi)

give up some silly macros which were only used ~once, and reduce
dependency on "locators.h"
makes the code more LKM frienly

(drochner)

The structure returned by USB_DEVICEINFO has the vendor/device strings
UTF-8 encoded now. We can't simply print this to a terminal, so
convert it to the current codeset first.

(drochner)

add hooks to allow terminal emulations to be installed by LKMs
(these are not available in early bootstrap, so this is not an
option for the system's default emulation)

(drochner)

Missing printf in sys/compat/linux/arch/powerpc/linux_machdep.c
found by cppcheck and reported by Henning Petersen in PR 42720.

(wiz)

Missing 'if defined COMPAT13 or COMPAT50' in uvm_swap.c found by cppcheck
and reported by Henning Petersen in PR 42721.

(wiz)

Fix incomplete line (deleto?), found by cppcheck and reported by
Henning Petersen in PR 42719.

(wiz)

Add missing parentheses, found by cppcheck and reported by
Henning Petersen in PR 42718.

(wiz)

Make ./build release work with new amigappc port.

(phx)

Include majors.amigappc instead of majors.amiga.

(phx)

Include "ctype.h" in the central place which deals with building the
kernel debugger as a userland program.

(tron)

Be consistent to decide if PMAP_WIRED or not.

(uebayasi)

Move A->K loan break code to uvm_loan.c.

(uebayasi)

Indent.

(uebayasi)

uvm_fault: Split "neighbor" fault and loan handling into functions.

(uebayasi)

- split sp_tlb_flush_pte() and switchtoctx() into sp_tlb_flush_pte_us()/
  sp_tlb_flush_pte_usiii() and switchtoctx_us()/switchtoctx_usiii() and
  implement the latter while i'm here.  it works ... sometimes i think,
  but also sometimes panics/hangs.
- fix a comment in sparc64_ipi_flush_pte_usiii()

(mrg)

pud requires putter like it always has.
deliberately do NOT bump date.

(pooka)

for now, reduce a few #ifdef SPITFIRE cases by introducing a new
ASI_DCACHE_TAG_OR_INV that is either ASI_DCACHE_{TAG,_INVALIDATE}.

increase the common code in cache_flush_phys().

(mrg)

Sort struct uvm_faultctx members for better alignment.

(uebayasi)

Use uppercase for acronyms on output: DRAM, SRAM

(hubertf)

Ticket #1378

(bouyer)

Pull up following revision(s) (requested by taca in ticket #1378):
crypto/dist/openssl/crypto/comp/c_zlib.c: patch
Fix to deal with CVE-2009-4355 from OpenSSL's repositroy.

(bouyer)

Ticket #1378

(bouyer)

Pull up following revision(s) (requested by taca in ticket #1378):
crypto/dist/openssl/crypto/comp/c_zlib.c: patch
Fix to deal with CVE-2009-4355 from OpenSSL's repositroy.

(bouyer)

Switch SSP init output to aprint_debug() instead of aprint_normal()
under DIAGNOSTIC ifdefs.

(njoly)

Indent.

(uebayasi)

More split.

(uebayasi)

Aprintify.

(njoly)

Do not print ac97 link rate message unless verbose boot is requested.

(njoly)

Fix build without DIAGNOSTIC.

(uebayasi)

uvm_fault: Clarify when to wire what.

(uebayasi)

Include "ctype.h" if we are not building a kernel to fix the build
of crash(8).

(tron)

Fix incorrect notify handler type (ACPI_SYSTEM_NOTIFY -> ACPI_DEVICE_NOTIFY).

(jruoho)

uvm_fault_upper_lookup: This is totally my personal preference, but can't help
adding one goto to reduce one indent.

(uebayasi)

uvm_fault:
- Lower fault routines don't care the vm_anon array found in upper lookup.
  Don't pass the pointer down.
- The flag "shadowed" is known when we lookup upper layer.  Don't need to
  keep in the fault context struct.

(uebayasi)

Indent.

(uebayasi)

Rewrite uvm_fault() loop using while () than goto.

(uebayasi)

Split uvm_fault() into 2 more functions, uvm_fault_check() and
uvm_fault_upper_lookup().  Omit unnecessary arguments passed around.

(uebayasi)

- sp_tlb_flush_all() becomes sp_tlb_flush_all_us() and sp_tlb_flush_all_usiii()
- sparc64_ipi_flush_pte() becomes sparc64_ipi_flush_pte_us() and
  sparc64_ipi_flush_pte_usiii()
- add some commented out code to disable interrupts and raise the traplevel
  in sparc64_ipi_flush_pte_usiii()
- cache_flush_phys() was missing a little of code in the cheetah case

(mrg)

uvm_fault: Pack variables shared during fault / re-fault into a struct named
uvm_faultctx.  Unfortunately ~all of those values are overriden in various
ways.  Constification doesn't help much...

(uebayasi)

Move init of lwp0 uarea earlier so that badaddr has a stack and pcb.

(matt)

Allow port-specific code to init lwp0.l_addr early.  (pmax needs it so it
call badaddr).

(matt)

for USIII, use the new "tlb demap all" operation in sp_tlb_flush_all()

(mrg)

Update to reflect mips tree changes.

(matt)

ERESTART is already negative.  Give up negating error values to not override
the original values.  Pointed out by rmind@, thanks.

In the lower fault case, if (*pgo_get)() can return ERESTART and we should
re-fault for that remains a question.  The original code just returned the
error, so keep that behaviour for now.  In case (*pgo_get)() really returns
ERESTART, pass EIO to tell the uvm_fault caller that (*pgo_get)() failed.

(As far as I grep callers don't check if the return value is ERESTART or not.
So assuming (*pgo_get)() never returns ERESTART should be a safe bet.)

(uebayasi)

Fix the bug that unaligned access occurs on amd64. It also fixes the bug
that error bits aren't cleard because these bit are W2C (in other words W1C).

Reported by Michael van Elst.

(msaitoh)

add some macros to check cpu type:
GETVER_CPU_IMPL()
CPU_IS_JALAPENO()
CPU_IS_USIII_UP()
and use them in a couple places to choose the right method/index

(mrg)

fix fallout from frame/trapframe merger.

(matt)

fix fallout from frame/trapframe merger.

(matt)

fix fallout from frame/trapframe merger.

(matt)

fix fallout from frame/trapframe merger.

(matt)

Merge frame into trapframe.  While this costs a bit more stack space on
kernel exceptions, the resulting simplifications are worth it.  This is
a step to fast softints and kernel preemption.

trapframe now includes a struct reg instead of a separate array of registers.

(matt)

remove unused icache_flush_page() implementation.

(mrg)

- add a PDB_CTX_FLUSHALL debug type to pmap.c, and also log the cpu_number()
  in a bunch more cases

- make sparc64_ipi_halt_thiscpu() and sparc64_ipi_pause_thiscpu() return void,
  their callers never checked anyway.

- remove prototypes for sparc64_ipi_flush_ctx() and sparc64_ipi_flush_all(),
  there are no such functions

(mrg)

Make it proper KNF as it should have been done in my previous commit.

(ahoka)

Remove duplicate sys/sysctl.h inclusion and put includes in alpabetic
order.

(ahoka)

Fix an outdated comment, noticed by Paul Goyette.

(martin)

New modifiers for the "write" command: /BHL. They act like /bhl but do not
read the old value from memory before writing.

(phx)

Add a convenience function to find the maximum value currently reported
by a set (selected via a passed callback predicate) of sensors.
This provides an easy way to query the current temperature of a thermal
zone, for example, from within the kernel - assuming the caller knows
the topology.

(martin)

Bump date for SEE ALSO change.

(wiz)

Do not advertise deprecated APIs in an important introductory manual page.

(jruoho)

Import dhcpcd-5.1.5

(roy)

sync

(roy)

Import dhcpcd-5.1.5 with the following changes:
* -H, --xidhwaddr uses the last 4 bytes of the hardware address as
  the DHCP xid
* -J, --broadcast tells the DHCP server to broadcast replies
* no longer warn about InfiniBand being an unsupported family
* infinite lease times are valid when reading old leases
* discard address correctly if lease is invalid

Status:

Vendor Tag: roy
Release Tags: dhcpcd-5-1-5

(roy)

Verify the object type before using variables from it.
In addition, use CTASSERT(9) instead of KASSERT(9).

(jruoho)

Fixed for amigappc by executing eieio after writing to a vga register.

(phx)

Add a detachment routine.

(jruoho)

Also remove the ACPI notify handler upon detach.

(jruoho)

Add the missing newline to a verbose autoconf message.

(dyoung)

Add a detachment routine. While here, clarify the initialization of sensors.

(jruoho)

Ax uvm_fault_internal() & break it into functions.  "Upper" fault and "lower"
fault routines are separated now.

(uebayasi)

Skip handling of APPLEUFS_LABEL if it is smaller than a device block.
In particular:

- newfs will not try to erase the label
- fsck_ffs will not try to validate the label

This lets newfs and fsck work on 2048-byte-per-sector media.

Does Apple UFS support such media and how?

(mlelstv)

Properly register blocksize with disk(9) framework.

(mlelstv)

Include newly-created subr_device.c and remove few special case
device accessor copypastes.  This makes it possible to link static
binaries which use -lrumpdev.

(pooka)

Device accessors are only marginally related to autoconf, so put them
into subr_device.c instead of having them in subr_autoconf.c.

Since none of the copyrights in subr_autoconf.c really match the
history of device accessors, I took the liberty of slapping (c)
2006 TNF onto subr_device.c.

(pooka)

Use full month name in Dd.

(wiz)

Bump date for new debug section.
New sentence, new line.
Remove Pp after section begin.

(wiz)

Fix two typos to make documentation match the code:
* the CIOCFKEY ioctl is really called CIOCKEY
* CRK_MOD_EX => CRK_MOD_EXP

Bump date.

(hubertf)

Add missing "break" for CRYPTO_CAST_CBC, and some assorted comment fixes.
openssl(1) checks for CAST (and others) on ~every startup.

(hubertf)

filesystems now use getdisksize() which is in librumpdev_disk.

(mlelstv)

If RUMP_BLKSECTSHIFT is set in the environment, use that as
device sector size instead of DEV_BSHIFT.

(pooka)

Remove ACPICA_PEDANTIC.

(jruoho)

Remove ACPICA_PEDANTIC. Appears to be no longer in use.

(jruoho)

Miscellaneous small changes:

        * Use -compact to make the page a little shorter.
        * Capitalize the title in the sysctl-section.
        * Use .Ic for the sysctl variables.
        * Remove ACPI_PEDANTIC; no such option any more.

(jruoho)

1 CTASSERT(foo) is enough for anyone.

(skrll)

Release buffer in case a receive failed.

(mlelstv)

Add a short introduction to ACPI debugging.

(jruoho)

Add dynamic debug options for ACPI_DEBUG kernels.

ok jmcneill@, pooka@, pgoyette@

(jruoho)

No more ACPI_ACAD_DEBUG.

(jruoho)

Simplify and clean:

        * Semantics.
        * Remove ACPI_ACAD_DEBUG.
        * Reduce the amount of error reporting.
        * Remove the internal bookkeeping flags and variables;
          instead operate only with a single state variable.
        * Trust that the notify handler is doing its job;
          remove the refresh routine from the sensor.
        * Queue a status check upon resume; we want to know
          if the status has changed when in suspend (and it
          is unclear if a notify is sent to us).

(jruoho)

Fix block shift to work with different device block sizes.

Unlike other filesystems this has some side issues because
the shift values are stored in the superblock and because
userland utitlies share the same fsbtodb macros.

-> the kernel now ignores the value stored in the superblock.
-> the macro adaption is only done for defined(_KERNEL) code.

(mlelstv)

Replace individual queries for partition information with
new helper function.

(mlelstv)

Fix block shift to work with different device block sizes.

(mlelstv)

Replace individual queries for partition information with
new helper function.

(mlelstv)

Replace individual queries for partition information with
new helper function.

(mlelstv)

This is using device_t, so it needs to include <sys/device.h>.

(martin)

uvm_fault_internal:

Move local variables around to isolate contexts.  Note that remaining variables
are global in that function, and some hold state across re-fault.

Slilently clean-up the "eoff" mess.

(Superfluous braces will go once things settle down.)

(uebayasi)

Indent.

(uebayasi)

uvm_fault_internal: In lower fault handling case, put another goto to clarify
that we don't care lower neighboring pages for the zero-fill object.

(uebayasi)

uvm_fault_internal: Skip another long code segment (lower "neighbor" fault)
by a goto.

(uebayasi)

Add detachment routine.

(jruoho)

uvm_fault_internal: Put a goto label "Case1" as well as "Case2".  Clarify
that if the faulting page is shadowed, we don't care the lower layer at all.

(uebayasi)

Allocate the sensor structures dynamically.

(jruoho)

Add detachment routines.

(jruoho)

uncommit part which wasn't supposed to get committed yet

(pooka)

Add an example on how to use sysmon watchdogs in rump.

(pooka)

Add rump sysmon component to devlists.

(pooka)

Add sysmon rump component.

(pooka)

print newline after timer expiry message

(pooka)

include kern_hook.c

(pooka)

Pass root device as a parameter to domountroothook().

(pooka)

Correct previous; fix a miscalculation of offset-into-entry in MADV_SEQUENTIAL
case.  Pointed out by pooka@.

(uebayasi)

Place *hook implementations in kern_hook.c instead of them floating
around in the kern_subr.c gruel.  Arrrrr.

(pooka)

Some might argue that it is benefi_c_ial to spell words correctly.

(snj)

Use proper static initializers for *hooklist (currently they happened
to work accidentally anyway since the initializer is 0).

(pooka)

Replace more printfs with aprint_normal / aprint_verbose
Makes "boot -z" go mostly silent for me.

(hubertf)

Make (no pun intended) it possible to perform component symbol
protection from a shell prompt with the command line:

    make -f /sys/rump/Makefile RUMP_SYMREN=module.kobj

This is benefitial especially on i386, where rump ABI matches the
kernel ABI (sans the renaming isolation).  In the above example
module.kobj can then be loaded in a rump kernel either statically
at link-time simply by giving module.kobj as an object to the
linker, or dynamically at runtime by calling rump_sys_modctl(MODCTL_LOAD).

(pooka)

Change MIPS_CURLWP from s7 to t8.  In a MALTA64 kernel, s6 is used 9155 times
which means the compiler could really use s7 is was free to do so.  The least
used temporary was t8 (288 times).  Once the kernel was switched to use t8 for
MIPS_CURLWP, s7 was used 7524 times.

Additionally a MALTA32 kernel shrunk by 6205 instructions (24820 bytes) or
about 1% of its text size.

[For some reason, pre-change t1 was never used and post change t2 was never
used.  Not sure why.]

(matt)