Fix RB_KDB by calling the debugger entry points after pmap_bootstrap, etc

(skrll)

Bump size of scratchstr - some KASSERTMGS exceed 256 characters

(skrll)

Sprinle RCSID

(skrll)

Print device ID with 8bits again (revert part of rev. 1.27) because
the revision field also printed next.

(msaitoh)

New sentence, new line.

(wiz)

- Add 0x54-0x57.
- Add note about sdtemp(4).

(msaitoh)

Add ichsmb(4), spdmem(4) and sdtemp(4).

(msaitoh)

Add sdtemp(4)

(msaitoh)

Don't mutex_exit() if we didn't mutex_enter().

Pointed out by coypu. Thanks!

(pgoyette)

remove extra ifdefs. no functional changes.

ip_flow.c becomes build target only if GATEWAY kernel option is on.
So, "#ifdef GATEWAY" in ip_flow.c is not needed.

(knakahara)

Repair the inevitable tyops in previous.

(pgoyette)

start creating a table and say I will do binutils.

(christos)

Update a few more drivers for localcount(9)

(pgoyette)

If we're going to check for a NULL pointer, do the check before we
dereference it (to get the lock address).

(pgoyette)

Update supported masters and slaves.

(msaitoh)

Fix downmatch increment

(ozaki-r)

Move functions to make module related code at the bottom of this file.

(msaitoh)

Note MCP9804 now also supported by sdtemp(4) - thx msaitoh.

(rkujawa)

Typo

(pgoyette)

Fix conversion to device_lookup_acquire()

(pgoyette)

Fix botched edit.

(pgoyette)

Note sdtemp(4).

(msaitoh)

- Add new devices.
- s/DDR2 and DDR3 SDRAM DIMMs/DDR2 and newer DIMMs/

(msaitoh)

Reviewed and OK'd by pgoyette:
- Add supoprt for Atmel AT30TS00, AT30TSE004, Giantec GT30TS00, GT34TS02,
  Microchip MCP9804, MCP98244, IDT TS3000GB[02], TS3001GB2, TSE2004GB2,
  and On Semiconductor CAT34TS04. Taken from OpenBSD.
- Add IDT TSE2002GB2.
- Check the temperature resolution field in the capability register instead
  of the hard coded value in the match table. With this change, some devices'
  temperature resolution would be fixed.
- The resolution register is a vendor specific register.
- All of IDT devices have the resolution register.
- The address of the resolution register of Microchip MCP98244 is different
  from other Microchip devices.
- Show accuracy, range, resolution and timeout with aprint_normal().
- Show timeout with aprint_debug().

(msaitoh)

- Print device ID in 16bits instead of 8bits in aprint_debug().
- s/STTS2002/STTS2004/ for STTS2004.

(msaitoh)

More renaming of LOCALCOUNT_INITIALIZER --> DEVSW_MODULE_INIT

(pgoyette)

Rename LOCALCOUNT_INITIALIZER to DEVSW_MODULE_INIT.  This better describes
what we're doing, and why.

(pgoyette)

Simplify by using atomic_swap instead of mutex

Suggested by kefren@

(ozaki-r)

Set the cause register to zero after disabling interrupts now that spl0
doesn't do it.

My cobalt now boots (again again)

(skrll)

Fix some sloppy typos

(pgoyette)

Update delta to version 20160725 in light of the SA to bspatch at
https://www.freebsd.org/security/advisories/FreeBSD-SA-16:25.bspatch.asc

+ check for negative offsets provided in the binary patch file.
Bug reported (independently) by The Chromium Project and Lu Ting-Pin.

+ add a -V option to delta to print the version number of the
delta(1) utility to stderr, and then exit

(agc)

First cut at netbsd-8 branch objectives.

(martin)

Fix typo in local copy

(pgoyette)

Sync with HEAD

(pgoyette)

From Felix Deichmann:
While writing "slhci at tc" support, I noticed TC bus_space functions
for alpha don't work as they should (tc_bus_mem.c). Consideration of
dense vs. sparse space seems problematic, especially for widths < 4.

I found information mainly in

[1] https://web-docs.gsi.de/~kraemer/COLLECTION/DEC/d3syspmb.pdf
[2]
http://h20565.www2.hpe.com/hpsc/doc/public/display?docId=emr_na-c04623255

and a dirty (but tested) hack based on this is attached as an example.
It would be great if someone could have a look and do a cleaner version
preferably.

Especially the end of ([2] section 12.2.2) provides a summary of which
"instruction" to use for which data width in which space, and possible
side effects (unintentional double reads/writes) in dense space...

(christos)

When calling devsw_attach() we need to use the expected/official driver
name (as listed in the devsw_conv[] table) to get the expected device
majors.  Once rump initialization is finished (ie, it has created its
required device nodes), we need to detach the [bc]devsw so the module
initialization code doesn't get EEXIST.

(pgoyette)

IFQ_PURGE() must be called only after call if_initialize().
it lock ifp->if_snd.ifq_lock.

(nonaka)

Restore correct test for return value from aarpresolve().

(rjs)

Redo previous

(pgoyette)

xburst needs CPU_MIPS_D_CACHE_COHERENT

(macallan)

Update vnd for using localcount(9)

(pgoyette)

Remove one of the two SEE ALSO sections.

(wiz)

Remove lapic_id, lapic_ppr and lapic_isr. We need to be careful though: the
offset of lapic_tpr must not change, and the whole area must be exactly one
page.

(maxv)

Unused.

(maxv)

This needs to be page-aligned anyway.

(maxv)

When initializing the rump cgd component, use the correct driver name
(as found in the devsw_conv[] table).  This will get us the "official"
major numbers for the cgd device.

After creating the rump file-space nodes for /dev/cgd* we then need to
detach the [bc]devsw's because normal module initialization will do its
own attachment, and we don't want that to fail.

While here, since we're doing the devsw_attach() twice, share the
results from the first call rather than starting from scratch.

(pgoyette)

The L1 entry of the first page of the data segment is overwritten for the
LAPIC page, and set as RWX+PG_N. The LAPIC pa is fixed, and its va resides
in the data segment. Because of this error-prone design, the kernel image
map is not linear, and I first thought it was a bug (as I vaguely said in
PR/51148). Using large pages for the data segment is therefore wrong, since
the first page does not actually belong to the data segment (even if its va
is in the range). This bug is not triggered currently, since local_apic is
not large-page-aligned.

We will certainly have to allocate a va dynamically instead of using the
first page of data; but for now, disable large pages on the data segment,
and map the LAPIC as RW.

This is the last x86-specific RWX page.

(maxv)

Apply if_get/if_put to compat_ifioctl

(ozaki-r)

-Wno-stack-protector

(christos)

Eliminate the distinct path check, since the paths don't actually need to be
distinct and allowing this makes certain useful tasks possible, such as
fixing an unpopulated /dev while a tmpfs is mounted over it.
However, require the paths to be different, as mounting a path directly over
itself has the side effect of causing any other mount points within that path
to no longer be accessible, and is difficult to unmount when done on /.

(erh)

Make DAD of ARP/NDP MP-safe with coarse-grained locks

The change also prevents arp_dad_timer/nd6_dad_timer from running if
arp_dad_stop/nd6_dad_stop is called, which makes sure that callout_reset
won't be called during callout_halt.

(ozaki-r)

Add a comment to describe why we didn't convert one caller of
device_lokup_private().

(pgoyette)

Update a few drivers for localcount(9)

(pgoyette)

Use KASSERT for checking non-NULL of ifa->ifa_ifp

ifa->ifa_ifp should be always non-NULL, so doing the check only if
DIAGNOSTIC is ok.

(ozaki-r)

fix: unlock in reverse order

(knakahara)

Two fixes:

1) invesion of enable bits in splx (ipl_sr_map is disable mask)

2) Don't overwrite the cause register in spl0 - there might be pending
  softints.

The second helps with recent boot issues after several new workqueues
are created.  lwp_startup would call spl0 and lose the pending softints
status.

(skrll)

use bus_space_map()

(macallan)

The MSR EFER state is not saved and restored when sleeping on i386. On PAE,
the CPU crashes right after waking up, since it needs to access NOX-ed
pages, which are to be enabled in an MSR.

Fix this by properly saving and restoring the EFER MSR. It's a little
tricky since the wakeup code uses %edx, but rdmsr overwrites it. We just
save it in %esi.

Now, the CPU sleeps properly on PAE kernels.

(maxv)

Regen for readlink/readlinkat signatures.

(njoly)

Small cleanup. Adjust readlink/readlinkat signatures to reduce diffs with
native.

(njoly)

KNF, and reduce the diff between amd64 and i386.

(maxv)

Call device_release() in appropriate error paths.

In the module initialization code, make the bmajor/cmajor variables
global so they can be shared with the rump component initialization.

(pgoyette)

since we're now attached the [bc]devsw's, grabbing the assigned majors,
and then detaching (in anticipation of the driver module doing its own
attach), we need to make sure that the driver name matches what is
expected in devsw_attach().  In particular, the driver name is "cgd"
and not "/dev/cgd0"  :)

While here, we might as well record the major numbers from the first
call, and just reuse them later.  So make the module's variables global,
and reference them in the rump initialization code.

Yay - cgd now works in the localcount world, both as a kernel module
and as a rump component.

(pgoyette)

Add a device_acquire() for when we need to grab a reference and we
already have a pointer to the device.

(pgoyette)

Remove cgd_release() - it's not needed now that getcgd_softc() provides
access to the device_t

Restore original cgdattach() - seems to be needed after all.

Remove some debug printf's from GETCGD_SOFTC() macro.

(pgoyette)

Simplify, remove redundant code.

(pgoyette)

Lock the tlbinfo if it wasn't when doing a pmap_tlb_pai_check

(matt)

Fix typo in string.

This is a driver for the AS3722 PMIC, not the AS3822 LED driver.

(jakllsch)

Also check the compatibles table when the size is below zero.

Fixes an aprint_error() on Jetson TK1.

(jakllsch)

Use one more value from "locators.h", also make sure 'size' is initialized.

(jakllsch)

Use values from "locators.h" when constructing locators array to be sure
of correct index;

(jakllsch)

Print the IPv6 or IPv4 source addresses of packets with wrong hash, to
help debugging.

(is)

backout last change (wrong branch).

(is)

Log the IPv4/IPv6 source of incorrect hash packets, too. Needed for
meaningful debugging.

(is)

Workaround for PR 47013 by bouyer@. Only works for mixed IPv4/IPv6
environemnts, not for pure-IPv6 yet. A real fix is still needed.

(is)

Use correct function name - devsw_detach() vs cdevsw_detach()
----------------------------------------------------------------------

(pgoyette)

Always define MKNOD, even if we don't think we need it (turns out we do,
sometimes...)  Noticed by Michael Plass reported on current-users.

(kre)

The rump_cgd component needs to do an early cdevsw_attact() to figure
out the device major numbers being used.  But we then need to detach
the [bc]devsw's to allow them to get attached normally during module
initialization.

(pgoyette)

Restore original handling of ioctl() returns.  If the underlying disk's
ioctl() returns success, we call uvm_vnp_setsize().  Regardless of any
error from the ioctl() call we should return success.

(pgoyette)

Remove debug printfs.

For MODULE builds, define cgd_cd via CFDRIVER_DECL

(pgoyette)

Revert changes in revision 1.108.2.3

For pseudo-devices, config(1) doesn't provide a valid cfattach, so when
a modular driver calls config_pseudo_attach() it will fail.  This is not
an issue for built-in drivers.

(pgoyette)

Cause rc to abort if mounting any of the critical local filesystems fails
rather than just ignoring the error.

Don't bother attempting to clear the contents of /var/run if /var/run
does not exist.

In that case the mkdir of /var/run/lvm would have failed - correct that
by using mkdir -p (which as a side effect will ensure /var/run exists
and is available for later scripts to use if for some reason it did not exist.)

(kre)

Be consistent in the ordering of operations.

(pgoyette)

Fix a comment

(pgoyette)

Continue previous - finish what we started

(pgoyette)

Only retrieve the bdev/cdev from the xdevsw[] array once.

(pgoyette)

When mapping a dmamem page, make sure to trunc_pae the starting address

(matt)

Remove pmap_prealloc_lowmem_ptps on amd64. This function creates levels in
the page tree so that the first 2MB of virtual memory can be kentered in
L1.

Strictly speaking, the kernel should never kenter a virtual page below
VM_MIN_KERNEL_ADDRESS, because then it wouldn't be available in userland.
It used to need the first 2MB in order to map the CPU trampoline and the
initial VAs used by the bootstrap code. Now, the CPU trampoline VA is
allocated with uvm_km_alloc and the VAs used by the bootstrap code are
allocated with pmap_bootstrap_valloc, and in either case the resulting VA
is above VM_MIN_KERNEL_ADDRESS.

The low levels in the page tree are therefore unused. By removing this
function, we are making sure no one will be tempted to map an area below
VM_MIN_KERNEL_ADDRESS in kernel mode, and particularly, we are making sure
NULL cannot be kentered.

In short, there is no way to map NULL in kernel mode anymore.

(maxv)

Simplify pmap_alloc_level. It is designed to work only with normal_pdes and
PTP_LEVELS, so don't pass them as argument. While here, explain what we are
doing.

(maxv)

Space before tab and trailing whitespace fixes.

(jakllsch)

Unused.

(maxv)

Fix logic to avoid dereferencing NULL pointer

(pgoyette)

Style: indent label 'out:' one space from the margin.

(pgoyette)

Reduce KERNEL_LOCK thereby ifq_lock is used by default.

if_snd is always excluded by ifq_lock now. So, the KERNEL_LOCK in if_transmit()
which serializes packet output processing is not needed now.

(knakahara)

Toward NET_MPSAFE-on in future, if_snd uses if_snd->ifq_lock by default.

That can reduce confusing difference between NET_MPSAFE on and off.

(knakahara)

Remove debug

(pgoyette)

Call cgd_spawn() if the requested device doesn't exist, rather than if
the device exists but without any softc data.

(pgoyette)

Add ATA8-ACS Long Logical Sector Feature Set support to wd(4).

(jakllsch)

Fix up the error/exit path.

Return actual error, not blindly report success!

(pgoyette)

Pass correct arg to cgd_spawn()

(pgoyette)

Use correct prototype.

(pgoyette)

Make sure that whenever we're using the cgd device's softc, we maintain
a reference to the device so things won't get deleted out from under us!

(pgoyette)

Remove stray debugging info

(pgoyette)

In config_devfree(), free the 'struct localcount'

In device_lookup_private_acquire() we need to ensure that the caller has
access to the device_t so the reference that we're acquiring can later
be device_release()d.  So we must require that the device has non-NULL
private data where the pointer back to the device_t can be stored (ie,
in xxx->sc_dev).

(pgoyette)

Return the actual error code, rather than blind success.

(pgoyette)

Don't include <mips/locore.h> in <machine/intr.h>, introduces circular
dependencies; instead include it in the 4 driver files that need it,
and reorder it in machdep.c

(christos)

Call wd_params_to_properties() after softc is sufficently
initialized.

(jakllsch)

Remove unused 'params' argument of local function wd_params_to_properties()

(jakllsch)

add NOPIE

(christos)

replace variable stack declaration with a large enough one and KASSERT.

(christos)

Don't do variable stack allocations for systems with non-const PAGE_SIZE;
instead assume that the smallest pagesize is 1024.

(christos)

remove .MIPS.abiflags to avoid objcopy creating a 0x20000000 large boot file.

(christos)

make this compile

(christos)

make this compile.

(christos)

more stack protector

(christos)

According to Opengroup online specifications[1], _POSIX_CPUTIME and
_POSIX_THREAD_CPUTIME options should be present in unistd.h not
limits.h. Move them with other options.

[1] http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/unistd.h.html

(njoly)

Ticket #1392

(martin)

Pull up following revision(s) (requested by shm in ticket #1392):
libexec/mail.local/mail.local.c: revision 1.28
Fix error checks in open(2) calls.

(martin)

Ticket #1392

(martin)

fix an error patch to call {b,c}devsw_release()

(pgoyette)

Pull up following revision(s) (requested by shm in ticket #1392):
libexec/mail.local/mail.local.c: revision 1.28
Fix error checks in open(2) calls.

(martin)

Ticket #1392

(martin)

Pull up following revision(s) (requested by shm in ticket #1392):
libexec/mail.local/mail.local.c: revision 1.28
Fix error checks in open(2) calls.

(martin)

Ticket #1207

(martin)

Pull up following revision(s) (requested by shm in ticket #1207):
libexec/mail.local/mail.local.c: revision 1.28
Fix error checks in open(2) calls.

(martin)

Ticket #1207

(martin)

Pull up following revision(s) (requested by shm in ticket #1207):
libexec/mail.local/mail.local.c: revision 1.28
Fix error checks in open(2) calls.

(martin)

Fix error checks in open(2) calls.

(shm)

Add -lz to makefile to fix the build.

(christos)

Fix typo in comment

(skrll)

Add missing semicolon

(pgoyette)

Call bdevsw_release() in the normal-exit path.

(pgoyette)

Actually save the bdev value when it is retrieved, so we can use it
later in a call to bdevsw_release().

(pgoyette)

simplify for readability

(pgoyette)

Attach atphy instead of ukphy.

(kiyohara)

Update processor families from the latest Intel SDM:
- 06_4FH: Add Xeon E7 v4 and Core i7-69xx Extreme Edition
- 06_57H: Xeon Phi [357]200

(msaitoh)

Remove a stray call to mutex_exit() that is left over from a previous
incarnation of this code.

(pgoyette)

+ PTO    parent-teacher organization
(also PTA)

(dholland)

Make complex RTM_CHANGE code understandable

Tests for route change added recently would reduce the possibility of
regressions.

Reviewed by ryo@

(ozaki-r)

Add some tests for route change

(ozaki-r)

Remove extra grep

(ozaki-r)

Make the test name self-descriptive

(ozaki-r)

don't die printing a name we don't have

(christos)

Partial adaptation to the new {b,c}devsw locking.  Still needs some work.
(See XXX comments in code)

(pgoyette)

Adapt the machine/arch dependent code to the new {b,c}devsw reference
counting.

XXX Most of these will require testing by someone other than myself, as
I have a limited selection of hardware!

(pgoyette)

Adapt machine-independant code to the new {b,c}devsw reference-counting
(using localcount(9)).  All callers of {b,c}devsw_lookup() now call
{b,c}devsw_lookup_acquire() which retains a reference on the 'struct
{b,c}devsw'.  This reference must be released by the caller once it is
finished with the structure's content (or other data that would disappear
if the 'struct {b,c}devsw' were to disappear).

(pgoyette)

put ctf lint library back. need to investigate more.

(christos)

PR/51353: Patrick Mackey: mld6query  "-r" option is not recognised
XXX: pullup 7

(christos)

use bus_space_map() instead of fudging our own bus_space_handle_t

(macallan)

unnamed unions need special treatment since they need braced initializers
for old style initializations.

(christos)

CID 976523: add FALLTHROUGH

(shm)

cplusplus does not like complex named initializers...

(christos)

use named initializers

(christos)

accept attributes in param decls

(christos)

fix tpyo, now n32 kernels have a chance to work again

(macallan)

handle "extern __attribute__((__gnu_inline__)) __inline"

(christos)

add uvm_km_protect()

(christos)

This comment is wrong. In fact, we are in low physical memory, but in high
virtual memory, and only the latter matters. I'm not exactly sure why, but
it appears that the kernel modules must be placed above the kernel image.
Just make this comment more ambiguous, in case the next passer-by gets
inspired.

(maxv)

Split the data+bss+rodata segment in two data+bss and rodata segments. The
latter is made read-only.

(maxv)

Change the protection of the kernel modules segments once we are done
relocating them. The text is allocated as RWX, and then mprotected to RW.

There is a bug that prevents us from doing RW->RX on amd64 and perhaps
sparc64. On x86, the pmap waits for the page to fault before granting it
the X permission. But in the trap handler, such a page is considered as
belonging to kernel_map, while it actually belongs to module_map. The
kernel then finds out the page is not present in kernel_map, and panics.
In all cases, module_map is non pageable, so even if the trap were handled
properly, it still wouldn't work.

Therefore, there is a small window in which the segment is RWX. But that's
fine enough, for now.

(maxv)

Introduce uvm_km_protect.

(maxv)

There is a huge bug in the way a uvm_map_protect is processed on x86.

When mprotecting a page, the kernel updates the uvm protection associated
with the page, and then gives control to the x86 pmap which splits the
procedure in two: if we are restricting the permissions it updates the page
tree right away, and if we are increasing the permissions it just waits for
the page to fault.

In the first case, it forgets to take care of the X permission. Which means
that if we allocate an executable page, it is impossible to remove the X
permission on it, this being true regardless of whether the mprotect call
comes from the kernel or from userland. It is not possible to make sure the
page is non executable either, since the only holder of the permission
information is uvm, and no track is kept at the pmap level of the actual
permissions enforced. In short, the kernel believes the page is non
executable, while the cpu knows it is.

Fix this by properly taking care of the !VM_PROT_EXECUTE case. Since the
bit manipulation is a little tricky we use two vars: bit_rem (remove) and
bit_put.

(maxv)

mark tag_signal as noreturn.

(christos)

Get rid of extra ifafree

It was wrongly imported from FreeBSD.

(ozaki-r)

Apply pserialize to some iterations of IP address lists

(ozaki-r)

Improved comment.  No code change.

(pgoyette)