Clean up clunky eval strings

- Remove unnecessary \ at EOL
  - This allows to omit ; too
- Remove unnecessary quotes for arguments of atf_set
- Don't expand $DEBUG in eval
  - We expect it's expanded on execution

Suggested by kre@

(ozaki-r)

Don't allow sav->lft_c to be NULL

lft_c of an sav that was created by SADB_GETSPI could be NULL.

(ozaki-r)

ISC DHCP 4.3.6 has released.

(taca)

Make the 'no results found' message sound less harsh.

Changes "try using better keywords" to "try using different keywords".

(jmcneill)

Kernel output format nit.

(simonb)

Use endian conversion macros (and some rearrange for readability).
No functional changes.

(isaki)

Reorg the virtio(4) config(9) code to be explicitly pci specific.

We'll later use this for a non-pci virtio(4) usecase.

ok martin

(cherry)

regen.

(msaitoh)

PCI device ID of Atom x5-Z8350's video is 0x22b0. ark.intel.com says 0x22b0 is
Intel HD Graphics 400.

(msaitoh)

Add test cases that there are SPs but no relevant SAs

(ozaki-r)

Comment out unused functions

(ozaki-r)

Don't use KEY_NEWSP for dummy SP entries

By the change KEY_NEWSP is now not called from softint anymore
and we can use kmem_zalloc with KM_SLEEP for KEY_NEWSP.

(ozaki-r)

Add test cases for setsockopt(IP_IPSEC_POLICY)

(ozaki-r)

Add missing IPsec policy checks to icmp6_rip6_input

icmp6_rip6_input is quite similar to rip6_input and the same checks exist
in rip6_input.

(ozaki-r)

Fix updating ipsec_used; turn on when SPs on sockets are added

(ozaki-r)

Fix updating ipsec_used

- key_update_used wasn't called in key_api_spddelete2 and key_api_spdflush
- key_update_used wasn't called if an SP had been added/deleted but
  a reply to userland failed

(ozaki-r)

Make IPsec SPD MP-safe

We use localcount(9), not psref(9), to make the sptree and secpolicy (SP)
entries MP-safe because SPs need to be referenced over opencrypto
processing that executes a callback in a different context.

SPs on sockets aren't managed by the sptree and can be destroyed in softint.
localcount_drain cannot be used in softint so we delay the destruction of
such SPs to a thread context. To do so, a list to manage such SPs is added
(key_socksplist) and key_timehandler_spd deletes dead SPs in the list.

For more details please read the locking notes in key.c.

Proposed on tech-kern@ and tech-net@

(ozaki-r)

Fix the previous fix that doesn't even compile.

(uwe)

tickets 162-169, 171, 173, 174

(snj)

Pull up following revision(s) (requested by mrg in ticket #174):
external/gpl3/binutils/usr.sbin/mdsetimage/bin_bfd.c: revision 1.2
sys/arch/evbmips/conf/CI20: revision 1.23
initialise some stack variables to zero so that later checking ensures
they are considered "unfound" instead of some random value.
part 1 in fixing the CI20 kernel build.  this one actually will cause
the build to fail, but it's outputting broken objects right now anyway.
part 2 coming up in the CI20 configuration itself.
--
don't set NEED_MDSETIMAGE when we don't enable the md_root.  move the
option in with the commented md_root options.
part 2 of the CI20 kernel build fix.  now evbmips builds completes
sanely for me.

(snj)

Pull up following revision(s) (requested by msaitoh in ticket #173):
sys/dev/pci/if_wmreg.h: 1.99-1.103
sys/dev/pci/if_wmvar.h: 1.34-1.37
sys/dev/pci/if_wm.c: 1.510-1.537
sys/dev/mii/ihphyreg.h: 1.2
sys/dev/mii/inbmphyreg.h: 1.10
sys/dev/mii/igphy.c: 1.27
Sync wm(4) up to if_wm.c rev. 1.537:
- Set WMREG_KABGTXD not in wm_init_locked() but in wm_reset().
- If a legacy interrupt is a spurious interrupt, don't print debug
  message.
- Don't print the Image Unique ID if an NVM is iNVM (i210 and I211).
- Fix a bug that a RAL was written at incorrect address when the index
  number is more than 16 on 82544 and newer.
- The layout of RAL on PCH* are different from others. Fix it.
- Flush every MTA write. Same as Linux.
- Move the location of calling wm_set_filter. Same as some other OSes.
- Flush writing WMREG_CTRL in wm_gmii_mediachange().
- Make new wm_phy_post_reset() and use this function at all location
  after resetting phy.
- Add I219 specific workaround for legacy interrupt. From OpenBSD.
- Move the location of calling wm_lplu_d0_disable().
- Disable D0 LPLU on 8257[12356], 82580, I350 and I21[01], too. Before
  this commit, above devices and non-PCIe devices accessed wrong
  register.
- Fix latency calculation in wm_platform_pm_pch_lpt().
- Set OBFF water mark and enable OBFF on PCH_LPT and newer.
- Fix a bug that wm_gmii_i82544_{read,write}reg() didn't take care of
  page select. PHY access from igphy() automatically did it, but
  accessing from wm(4) for workaround didn't work correctly. This
  change affects 8254[17], 8257[12] ICH8, ICH9 and ICH10.
- Call wm_kmrn_lock_loss_workaround_ich8lan() before any PHY access in
  wm_linkintr_gmii().
- Register access in wm_kmrn_lock_loss_workaround_ich8lan() now works
  correctly. Enable this function.
- IF TX is not required to flush, RX is also not required to flush in
  wm_flush_desc_rings(). Same as other OSes.
- Remove wrong semaphore access in wm_nvm_{read,write}_{ich8,spt} to
  prevent hangup. A semaphore is get/put in wm_nvm_{read,write}.
- Move some initialization stuff in wm_attach() before wm_reset().
  Some flags and callback function is required to set correctly before
  wm_reset() because wm_reset() and some helper functions refer them.
- Add wm_write_smbus_addr() to set SMBus address by software.
- Modify wm_gmii_hv_{read,write}reg_locked() to make them access
  HV_SMB_ADDR correctly.
- Configure the LCD with the extended configuration region in NVM if
  it's required. Tested with Thinkpad X220.
- 8257[12]: Don't directly access SPI but use EERD register.
- 82575-I354: If the size of SPI ROM >= 32K words, use direct SPI
  access instead of EERD register access.
- Add wm_nvm_eec_clock_raise() and wm_nvm_eec_clock_lower() and use
  them for Microwire/SPI bus control. Same as Linux and FreeBSD.
- Reduce timeout value for 80003 in wm_get_swfw_semaphore(). Same as
  Linux and FreeBSD.
- Change API of kmrn_{read,write}reg() and check the return value.
- Check whether it's required to use MDIC workaround for 80003 or not
  in wm_reset(). If the workaround isn't required, don't use the
  workaround code in wm_gmii_i80003_{read,write}reg.
- Add WM_F_WA_I210_CLSEM flag for a workaround. FreeBSD/Linux drivers
  say "In rare circumstances, the SW semaphore may already be held
  unintentionally." on I21[01]. PXE boot is one of the case.
- Qemu's e1000e emulation (82574L)'s SPI has only 64 words. I've never
  seen on real 82574 hardware with such small SPI ROM. Check
  NVM word size before accessing higher address words to prevent
  timeout.
- Check some wm_nvm_read()'s return vale.
- Use device_printf() instead of aprint_error_dev() for PHY read/write
  functions because those are used not only in device attach.
- Print internal driver flags when attaching.
- Add debug printf()s.
- Rename variables.
- Add comment, update comment and remove wrong comment.

(snj)

Pull up following revision(s) (requested by maya in ticket #171):
external/gpl3/gcc/dist/gcc/config/i386/i386.c: revision 1.13
Apply upstream patch:
Incorrect codegen from rdseed intrinsic use (CVE-2017-11671)
We should not expand call arguments in between flags reg setting and
flags reg using instructions, as it may expand with flags reg
clobbering insn (ADD in this case).
Attached patch moves expansion out of the link. Also, change
zero-extension to non-flags reg clobbering sequence in case we perform
zero-extension with and.
2017-03-25  Uros Bizjak  <ubizjak@gmail.com>

(snj)

Pull up following revision(s) (requested by skrll in ticket #169):
sys/dev/usb/if_ural.c: revision 1.52
Free the RX list if ural_alloc_rx_list fails part way through.
Reported by Ilja Van Sprundel.

(snj)

Pull up following revision(s) (requested by martin in ticket #168):
sys/kern/kern_malloc.c: revision 1.146
Avoid integer overflow in kern_malloc(). Reported by Ilja Van Sprundel.

(snj)

Pull up following revision(s) (requested by nat in ticket #167):
sys/dev/audio.c: revision 1.373
sys/dev/audio.c: revision 1.374
sys/dev/audio.c: revision 1.375
Refactoring of order of auto config combinations.  Most common formats
come first.
Do not exceed AUDIO_MAX_CHANNELS.
Addresses PR kern/52435.
--
Remove unused code from audio_set_params.
Addresses PR kern/52434.
--
Remove dead code from audio open.  Found by isaki@.
Better error handling incase memory for the streams' ringbuffers cannot be
allocated.
Addresses PR kern/52433.

(snj)

Pull up following revision(s) (requested by nat in ticket #166):
sys/dev/auconv.c: revision 1.30
sys/dev/audio.c: revision 1.372
sys/dev/ic/ac97.c: revision 1.97
sys/dev/pci/azalia_codec.c: revision 1.81
Mixer device bounds checking.
Analysis by Ilja van Sprundel.

(snj)

Pull up following revision(s) (requested by nat in ticket #165):
sys/dev/auconv.c: revision 1.29
sys/dev/auconv.h: revision 1.19
sys/dev/ic/am7930.c: revision 1.56
Add a null_filter to help with the audio autoconfig of pmax.
Tested by flxd@.

(snj)

Pull up following revision(s) (requested by maxv in ticket #164):
distrib/sets/lists/base/md.amd64: revision 1.269
distrib/sets/lists/debug/md.amd64: revision 1.97
sys/arch/amd64/conf/GENERIC: revision 1.460
sys/arch/amd64/conf/files.amd64: revision 1.89
sys/arch/i386/conf/GENERIC: revision 1.1157
sys/arch/i386/conf/files.i386: revision 1.379
sys/arch/i386/i386/i386_trap.S: revision 1.7-1.8
sys/arch/i386/include/frameasm.h: revision 1.16
sys/arch/x86/include/sysarch.h: revision 1.12
sys/arch/x86/x86/pmc.c: revision 1.8-1.10
sys/arch/x86/x86/sys_machdep.c: revision 1.36
sys/arch/xen/conf/files.compat: revision 1.26
sys/secmodel/suser/secmodel_suser.c: revision 1.43
sys/sys/kauth.h: revision 1.74
usr.bin/pmc/Makefile: revision 1.5
usr.bin/pmc/pmc.1: revision 1.12-1.13
usr.bin/pmc/pmc.c: revision 1.24-1.25
style
--
style
--
Disable interrupts for T_NMI (inline calltrap). Note that there's still a
way to evade the NMI mode here, if a segment register faults in
INTRFASTEXIT; but we don't care. I didn't test this change, but it seems
fine enough.
--
Make the PMC syscalls privileged.
--
Check argc, and add a message.
--
include opt_pmc.h
--
Build the pmc tool on amd64.
--
Properly handle overflows, and take them into account in userland.
--
Update.
--
Enable PMCs by default.
--
Sort sections. Fix macro usage.

(snj)

Pull up following revision(s) (requested by nonaka in ticket #163):
sys/external/bsd/drm2/pci/drm_pci.c: revision 1.18
PR/52409: Avoid panic at resume.

(snj)

Pull up following revision(s) (requested by maya in ticket #162):
sys/dev/pci/if_ral_pci.c: revision 1.24
sys/dev/ic/rt2860.c: revision 1.28
sys/dev/ic/rt2860var.h: revision 1.5
enable rt2860 power management code
adjust to fit netbsd:
make suspend,resume functions match desired pmf* prototypes
remove wakeup/activate wrapper functions
avoid jumping to NULL on resume by initializing if_stop
the problem machine has other issues on resume, so there might be further
issues, but it's an improvement over a jump to NULL.
tested by Riccardo Mottola

(snj)

another one down

(jdolecek)

fix logic bug in processing of finished commands - mask of active
commands can change during the loop as c_intr() callback can queue
new commands, so the interrupt routine should only mark as finished
those which were actually active before the loop started; otherwise
the code marked as finished commands which were just started, and
being executed by HBA, leading to all sorts of data corruption

while here mark the active mask volatile, as it is modified from
interrupt context

this fixes for good the random crashes, short reads, and fatal command
errors which I've been tracing down for past couple weeks

thanks to Jonathan (jakllsch@) for testing, and a script to easily
triggered the condition, and led to this bug being finally found and squashed

(jdolecek)

convert two pairs of MACHINE_ARCH into two single MACHINE_CPU checks.

(mrg)

adjust code to be closer to HEAD to it's easier to compare and find
regressions, undoing some changes which were actually not necessary

(jdolecek)

make atastart() schedule as many commands as possible, instead of always
only one; makes it able to pick up pace again after processing non-NCQ command

(jdolecek)

Fix linking for the optional llvm-ar tool.

(joerg)

move the drive recovery block to drive struct, it's inherently per-drive

(jdolecek)

Update LLVM entry.

(joerg)

Update LLVM build glue for 5.0RC1.

(joerg)

Mark files not present in 309604 as dead.

(joerg)

Support loading collation data from file.  Began with FreeBSD's
xlocale_collate, but had to change it somewhat to accommodate the
requirements of the Unicode Collation Algorithm (in particular,
there are maps from single-character collation elements to
multiple collation weight vectors, and multiple-to-multiple
mappings as well).

(perseant)

Inline storage size in one place to avoid depending on DCE for removing
the USE_FORT=yes warning.

(joerg)

Mark the compat_svr4 module obsolete.

hopefully fixes the build, this is a blind commit.

(maya)

Restore marking ./usr/include/i386/svr4_machdep.h obsolete
instead of removing

(so it can be removed by postinstall)

(maya)

mandoc-1.14.2 out.

(wiz)

Don't use the custom tokenizer when compiled with debugging on

Using the custom tokenizer means one cannot interactively query the database
through the SQLite shell, thus thwarting the purpose of the debug build option.

Thanks to leot@ for reporting it.

(While there change the debug macro from DEBUG to APROPOS_DEBUG)

(abhinav)

Wrong architecture for IBCS2 compat! it's stringly an x86 thing.

(maya)

Move arch/i386/i386/freebsd_* into compat/freebsd/. COMPAT_FREEBSD is
i386-specific.

(maxv)

Remove references to compat_freebsd when it is not supported.

(maxv)

Remove svr4_machdep.h right away, no one should include it.

(maxv)

Don't build the svr4 module on i386.

(maxv)

Don't include files.svr4 and files.svr4_32.

(maxv)

Remove commented vestiges of bygone compat options.

COMPAT_MACH, COMPAT_DARWIN, EXEC_MACHO, and COMPAT_PECOFF were all
removed ages ago.

(riastradh)

pthread__attr_init_private:
malloc+memset -> calloc. Also initialize all values to the proper
defaults.
This fixes the "rustc panic" discussed on pkgsrc-users.
OK: joerg

(martin)

From bjoern johannesson, in netbsd-general... Reset the mask to 0, used to
be 0xa. Is it wrong in the cfe? Should this better be done via a quirk?

(christos)

PR/52446: Devid Binderman: Fix error checking for wcstombs

(christos)

PR/52445: David Binderman: Fix wcstombs error checking

(christos)

Use bullet list.

(wiz)

Sort SEE ALSO.

(wiz)

Remove superfluous Pp.

(wiz)

New sentence, new line.

(wiz)

Mark files not present in 309604 as dead.

(joerg)

Enable GPCLK2 to Bluetooth.

(jmcneill)

mark ISR as mpsafe, detect when this is the console

(jmcneill)

Broadcom AUX UART doesn't seem to set LSR RXRDY bit when data is available
in the RX FIFO.

(jmcneill)

some more notes

(jdolecek)

modify code handling mismatch of nsegs in _bus_dmamem_alloc_range() to
a KASSERT() - plain return leaks memory, and this condition should
never trigger unless there is bug in uvm_pglistalloc(), so it seems
to be waste to check this

other ports usually simply do not check this, with exception of arm,
which does even full cleanup (checks it and calls uvm_pglistfree())

(jdolecek)

Use idt_vec_set instead.

(maxv)

Fix vax/* cross references

(ryoon)

Create a config.txt for RPI and set 'enable_uart=1' and 'force_turbo=0'.
This config gives us a stable core_freq clock that is used to drive the
SD HOST and AUX UART devices.

More information on config.txt settings for the UART can be found here:

https://www.raspberrypi.org/documentation/configuration/uart.md

(jmcneill)

Fix TCPCTL_NAMES, and remove TCPCTL_VARIABLES.

(maxv)

Disable all the compat options until COMPAT_10. NetBSD 1.0 was released on
October 26 1994; 23 years of compatibility is enough. Discussed with
christos quickly.

(maxv)

Remove references to COMPAT_OLDSOCK (itself removed years ago).

(maxv)

Update confused vs_set_params().
play and rec are identical but pfil and rfil are independent.
XXX I introduce VS_USE_PREC8 option for debugging purposes
    temporarily.  I'll remove it if the problem is solved.

(isaki)

Only remap sdhost/sdhci on boards with wifi.

(jmcneill)

Enable AUX UART for BT on btwifi boards

(jmcneill)

Add COM_TYPE_BCMAUXUART type for BCM2835 AUX UART.

(jmcneill)

rename unclear functions. pointed out by msaitoh@n.o.

(knakahara)

Support loading collation data from file.  Began with FreeBSD's
xlocale_collate, but had to change it somewhat to accommodate the
requirements of the Unicode Collation Algorithm (in particular,
there are maps from single-character collation elements to
multiple collation weight vectors, and multiple-to-multiple
mappings as well).

(perseant)

use pool_cache(9) instead of pool(9) as they can be called concurrently.

(knakahara)

use kmem_alloc KPI instead of malloc KPI.

(knakahara)

Make the compilation conditional on __STDC_ISO_10646__, so that applications
that need one of the East Asian locales but don't want __STDC_ISO_10646__
don't have to pay the penalty for it.

(perseant)

pack crypto_drivers variables to struct and add cacheline_aligned qualifier.

(knakahara)

refactor: remove glue macros for FreeBSD code.

(knakahara)

Set the correct com type

(jmcneill)

Add support for BCM2835 AUX UART.

(jmcneill)

Cleanup debug messages

(jmcneill)

Bump date for previous.

(wiz)

Compare return value of fputs(3) with EOF instead of 0.
This is POSIX-ly correct and fixes csplit(1) on non-NetBSD systems.

(cheusov)

According to POSIX return value of fputs(3) should be a non-netagive value if it succeeds

(cheusov)

undo debug code

(jdolecek)

The proper way to validate a condition that's expected to be true is
to assert it, not to bolt it into the program logic in a way that will
cause strange behavior if it accidentally isn't true at some point.

(dholland)

adjust error recovery to similar shape as ahcisata; also switch to using
siisata_reset_channel(), as that seems more reliably get the HBA to running
state again after some errors

besides this, change several places checking PORT_READY to use
siisata_reinit_port() and properly timeout - seen several cases
where it would just loop there indefinitely on some errors with no
way to get into ddb and no error message, due to the tight DELAY() loop

(jdolecek)

actually change the retry to also use NCQ - with one drive I see
very frequent fatal errors on siisata when switching often between
NCQ and non-NCQ I/O commands with chaos monkey (basically, always
the next NCQ command on a slot which had non-NCQ I/O throws fatal
error); they completely vanish when just using NCQ all the time

(jdolecek)

Explicitly configure pin pull-down/up settings for SDIO.

(jmcneill)

enable sdhost driver (ok skrll@)

(jmcneill)

Add driver for the bcm2835 internal SD controller. On boards with SDIO
Wi-Fi, the internal SD controller is used for the SD card slot and the
Arasan SDHCI controller is used for SDIO.

(jmcneill)

Remove references to COMPAT_IRIX - does not exist anymore.

I believe svr4_machdep.h should be removed when the option is not
implemented on the target architecture; and we should also remove the
associated md.* entries.

(maxv)

Remove lurking reference to TCP_COMPAT_42.

(maxv)

kmem_xyz(sizeof(struct foo)) --> kmem_xyz(sizeof(*foo))

No change to amd64 binary.

(riastradh)

Fix error branch in ttm_dma_tt_init to avoid double-free.

Should fix symptom of PR kern/52438, but who knows what underlying
problem causes us to reach the error branch in the first place.

(riastradh)

Disable svr4 and svr4_32 on sparc, sparc64 and amiga - the only places
where they were still enabled.

(maxv)

make it compile again.

(christos)

Initialize iospace_size instead of iospace (will be overwritten) to 64k.
Brings boot of emips on giano a bit closer, but now it's stuck (hangs).

(he)

Improve diagnostic code.
If the specified encoding is wrong, it outputs a sound like a buzzer :)
rather than panic.

(isaki)

Typo in debug message.

(isaki)

Audio now compiles with WARNS=5, so there is no need to specify WARNS=3.

(nat)

The pad module will now compile with WARNS=5.

(nat)

The audio module will now compile with WARNS=5.

(nat)

dts updated to 4.12.4

(jmcneill)

merge

(jmcneill)

Use cp -RL to follow symlinks

(jmcneill)

Standardize NAME section.

Remove unneeded troff macros.

XXX: this is still MLINKed as names that are not in Nm.

(wiz)

add macro for getting the slot from context register, just for reference for now

(jdolecek)

Clarify compile-time and run-time arithmetic safety assertions.

This is an experiment with a handful of macros for writing the
checks, most of which are compile-time:

MUL_OK(t, a, b)        Does a*b avoid overflow in type t?
ADD_OK(t, a, b)        Does a + b avoid overflow in type t?
TOOMANY(t, x, b, m)    Are there more than m b-element blocks in x in type t?
                        (I.e., does ceiling(x/b) > m?)

Addenda that might make sense but are not needed here:

MUL(t, a, b, &p)        Set p = a*b and return 0, or return ERANGE if overflow.
ADD(t, a, b, &s)        Set s = a+b and return 0, or return ERANGE if overflow.

Example:

uint32_t a = ..., b = ..., y = ..., z = ..., x, w;

        /* input validation */
        error = MUL(size_t, a, b, &x);
        if (error)
                fail;
        if (TOOMANY(uint32_t, x, BLKSIZ, MAX_NBLK))
                fail;
        y = HOWMANY(x, BLKSIZ);
        if (z > Z_MAX)
                fail;
        ...
        /* internal computation */
        __CTASSERT(MUL_OK(uint32_t, Z_MAX, MAX_NBLK));
        w = z*y;

Obvious shortcomings:

1. Nothing checks your ctassert matches your subsequent arithmetic.
  (Maybe we could have BOUNDED_MUL(t, x, xmax, y, ymax) with a
  ctassert inside.)

2. Nothing flows the bounds needed by the arithmetic you use back
  into candidate definitions of X_MAX/Y_MAX.

But at least the reviewer's job is only to make sure that (a) the
MUL_OK matches the *, and (b) the bounds in the assertion match the
bounds on the inputs -- in particular, the reviewer need not derive
the bounds from the context, only confirm they are supported by the
paths to it.

This is not meant to be a general-purpose proof assistant, or even a
special-purpose one like gfverif <http://gfverif.cryptojedi.org/>.
Rather, it is an experiment in adding a modicum of compile-time
verification with a simple C API change.

This also is not intended to serve as trapping arithmetic on
overflow.  The goal here is to enable writing the program with
explicit checks on input and compile-time annotations on computation
to gain confident that overflow won't happen in the computation.

(riastradh)

If i386/svr4_machdep.h is no longer to be installed in /usr/include
it should no longer be listed in the sets lists as installed.
Mark it obsolete (hopefully unbreak builds.)

(kre)

Remove TCP_COMPAT_42 from the config files. Pass 3.

(maxv)

whitespace

(jdolecek)

make compile without AHCI_DEBUG

(jdolecek)

make ahci_channel_recover() non-static, so that it's visible in backtrace,
and can set a separate breakpoint there

(jdolecek)

Remove unused.

(maxv)

fix AHCI_P_CMD_CCS_SHIFT - must shift only by 8, otherwise the result would
be always 0

(jdolecek)

do not do the drive reset on non-fatal recovery - spec explicitely says
says the READ LOG EXT should only be done when neither COMRESET nor software
reset was done, and indeed it returns junk in this case

use C_RECOVERY slot for drive reset, so that it now will always succeed
in getting a slot, in cases when it would be necessary

adjust code and comments on the recovery path to explain better what's going on

with this AHCI error recovery works with real hardware without timeouts again

(jdolecek)

reserve the highest slot for error recovery, and also have ata_channel
include space for the READ LOG EXT sector, so that it's not necessary
to allocate memory on the error handling path; now ata_read_log_ext_ncq()
will never fail due to resource shortage

(jdolecek)

actually count the REQUEUE as retry also, so that it will be retried
as non-NCQ, will not be subject to chaos monkey, and reported as fixed
once finished, too

(jdolecek)

Remove undocumented hack.

(maxv)

Remove TCP_COMPAT_42 from the config files. Pass 2.

(maxv)

Remove references to i386.

(maxv)

Unlink svr4_machdep.h.

(maxv)

Remove i386. By the way, it looks like several architectures are missing
here.

(maxv)

Remove svr4 from the config files.

(maxv)

Drop support for svr4 on i386. This feature is not maintained, not
reliable, and of a limited use case. Most svr4 applications got time to be
ported to linux, and we do have a functional, maintained linux emulation.

Reduces the number of entry points into the kernel, the number of
places that need special care (cpu context).

Note that compat_svr4 is still available on sparc.

(maxv)