Correct function name.
freem is used to free memory allocated by allocm.

(isaki)

Sync prototypes with reality.  These have been changed in 2011.

(isaki)

make

options NO_PCI_MSI_MSIX

work again for arch/x86/

(cherry)

protect from redefinition

(christos)

protect linux sized types from redefinition

(christos)

Fix MSIX allocation in pci_intr_alloc

(skrll)

since NVMe 1.2 (which defined the ONCS bit), the bit (4) is whether controller
supports 'Save' field in Set Features, and 'Select' field for Get Features,
not the support for the 'Set Features' command itself

(jdolecek)

rename NVME_ID_CTRLR_ONCS_SAVE back to NVME_ID_CTRLR_ONCS_SET_FEATURES;
it's actually used in nvmectl sources too, and while 'Save' is more
correct it's not complete either, so better to leseen churn -
according to spec it's support for 'Save' for Set Features, and
support for 'Select' for Get Features

(jdolecek)

support DIOCSCACHE + DKCACHE_WRITE if volatile write cache is present

fix the Get Features call for DIOCGCACHE to actually retrieve the current
value properly

(jdolecek)

Fix kernel info leak, 4 bytes of padding in struct _ksiginfo. Maybe we
should just set _pad to zero on LP64?

+ Possible info leak: [len=40, leaked=4]
| #0 0xffffffff80baf397 in kleak_copyout
| #1 0xffffffff80bda817 in sigtimedwait1
| #2 0xffffffff80bdab95 in sys_____sigtimedwait50
| #3 0xffffffff80259c42 in syscall

(maxv)

fix pasto (MSI -> MSIX)

(jdolecek)

allow MSI-X even with just one interrupt, on some systems MSI/INTx might
not be available (or working, for that matter)

(jdolecek)

simplify the interrupt allocation - it's not necessary to do the explicit
fallbacks, pci_intr_alloc() does this already internally

(jdolecek)

Fix typos

(skrll)

now that Jared fixed ena(4) to work, add manpage for it; adapted from FreeBSD

(jdolecek)

disestablish the interrupt on failure in nvme_q_create()

(jdolecek)

Rename the internal function "makename" to "makeword" to better reflect
what it actually does (makearg would have been an alternative).
While here, notice the one remaining place where it should have been
used, but was left open coded, and consume that one.

NFCI.

(kre)

Correct the documentation of cdbr_open_mem(3)

Fix function prototype in the man-page.
Add link cdbr_open_mem(3) -> cdbr(3).

(kamil)

Fix compile error.

(msaitoh)

Use DVF_ATTACH_INPROGRESS.

(msaitoh)

Add new dv_flags value DVF_ATTACH_INPROGRESS. Currenty, this flags is used
only for checking the registration of pmf.

This flag should be set when an attach function(ca_attach) doesn't complete
the whole attach work when the function returned. config_interrupts() set it.
It's also required for device drivers who do a part of the attach work in their
own kthread to set it.

(msaitoh)

Save control registers in PCI-X, PCIe, MSI and MSI-X capability area when
suspend and restore them when resume. For PCIe cababilities register, it's
required to check the existence of each register to not to write the next area.

This chagnge fixes a stability of suspend/resume.

(msaitoh)

NFC.  Need a grain of const

(kre)

NFC.  KNF.  return (x) -> return x

(kre)

Don't abuse config_interrupts()'s first argument. Use kthread instead of
config_interrupt(). OK'd by jmcneill and macallan.

(msaitoh)

ena(4): Add support for Amazon.com Elastic Network Adapter (ENA).

Many people contributed to this so I set the developer to "netbsd", please
change if there is a better way to do this.

(jmcneill)

It is not only the EXIT trap we need to check for when deciding no
fork is required, but any trap  (dumb mistake...)

XXX - include in 48875 pullup to -8

(kre)

Add support for configuring Amazon.com EC2 SSH keys and hostnames. While
here, only set wscons=YES if a wsdisplay0 device is present.

(jmcneill)

allow MSI-X if supported by the hardware, discussed with Manuel

(jdolecek)

simplify intr establish code - rely on pci_intr_alloc() to return
interrupt types which are possible for pci_intr_establish(); remove
fallbacks to retry with MSI explicitly disabled

bge(4) specifically needs to disable MSI on some boards which are
known to have broken MSI support, so this can't use pci_intr_alloc()
with just NULL counts

discussed on tech-kern@, and specifically bge(4) with Manuel

https://mail-index.netbsd.org/tech-kern/2018/11/27/msg024240.html

(jdolecek)

simplify intr establish code - rely on pci_intr_alloc() to allow
also MSI-X, and to return interrupt types which are possible for
pci_intr_establish(); remove fallbacks to retry with MSI/MSI-X
explicitly disabled

discussed on tech-kern@

https://mail-index.netbsd.org/tech-kern/2018/11/27/msg024240.html

(jdolecek)

Add ena at pci

(jmcneill)

Add puc at pci and com at puc

(jmcneill)

Add support for polling com devices when no interrupt is available.

(jmcneill)

Add support for com(4) without an interrupt by setting the COM_HW_POLL flag

(jmcneill)

The SHOULDYIELD flag doesn't indicate that other LWPs could run but only
that the current LWP was seen on two consecutive scheduler intervals.

There are currently at least 3 cases for calling preempt().
- always call preempt()
- check the SHOULDYIELD flag
- check the real ci_want_resched

So the forced check for SHOULDYIELD changed the scheduler timing. Revert
it for now.

(mlelstv)

Get this driver into a functional state.

(jmcneill)

workqueue and callout fixes

(jmcneill)

Ticket #1661

(martin)

Pull up following revision(s) (requested by maxv in ticket #1661):

sys/kern/kern_sig.c: revision 1.350

Fix kernel info leak, 4 bytes of padding at the end of struct sigaction.

        + Possible info leak: [len=32, leaked=4]
        | #0 0xffffffff80baf327 in kleak_copyout
        | #1 0xffffffff80bd9ca8 in sys___sigaction_sigtramp
        | #2 0xffffffff80259c42 in syscall

(martin)

Ticket #1661

(martin)

Pull up following revision(s) (requested by maxv in ticket #1661):

sys/kern/kern_sig.c: revision 1.350

Fix kernel info leak, 4 bytes of padding at the end of struct sigaction.

        + Possible info leak: [len=32, leaked=4]
        | #0 0xffffffff80baf327 in kleak_copyout
        | #1 0xffffffff80bd9ca8 in sys___sigaction_sigtramp
        | #2 0xffffffff80259c42 in syscall

(martin)

Ticket #1661

(martin)

Pull up following revision(s) (requested by maxv in ticket #1661):

sys/kern/kern_sig.c: revision 1.350

Fix kernel info leak, 4 bytes of padding at the end of struct sigaction.

        + Possible info leak: [len=32, leaked=4]
        | #0 0xffffffff80baf327 in kleak_copyout
        | #1 0xffffffff80bd9ca8 in sys___sigaction_sigtramp
        | #2 0xffffffff80259c42 in syscall

(martin)

Ticket #1114, ammend ticket #1110 for additional changes pulled up.

(martin)

Pull up following revision(s) (requested by maxv in ticket #1114):

sys/kern/kern_exit.c: revision 1.273

Fix info leak. There is one branch where 'status' is not initialized at
all.

        + Possible info leak: [len=4, leaked=4]
        | #0 0xffffffff80baf397 in kleak_copyout
        | #1 0xffffffff80b56d0c in sys_wait6
        | #2 0xffffffff80259c42 in syscall

(martin)

Additionally pull up following revision(s) (requested by maxv in ticket #1110):

sys/compat/linux/common/linux_misc_notalpha.c: revision 1.110
sys/kern/kern_time.c: revision 1.193

Improve my kern_time.c::rev1.192, systematically clear the buffers we get
from 'ptimer_pool' to prevent more leaks.

(martin)

Add new PCIE_HAS_LINKREGS(pcie_devtype) and use it. No functional change.

(msaitoh)

Zfs_write() may hide write errors if uiomove() succeeds but a
further dmu_tx_assign() fails because over quota or no space.

Use the emulated uiocopy() and uioskip() like illumos does.

Fix the uiocopy() emulation to not clobber the iovecs.

(hannken)

Skip zfs, it does not GOP_ALLOC.

PR kern/47656 test zfs_fillfs.

(hannken)

Add new macro PCIE_HAS_ROOTREGS(pcie_devtype) and use it. No functional
change.

(msaitoh)

The downstream port of PCIe switch is not a root port, so don't print
root port related register. For example, Intel 63xxESB controller's
downstream port device was printed by pcictl(8) with this bug:

-----------------------------
PCI configuration registers:
  Common header:
    0x00: 0x35108086 0x00100147 0x06040001 0x00010010

    Vendor Name: Intel (0x8086)
    Device Name: 63xxESB PCI Express Downstream Port #1 (0x3510)
(snip)
  Capability register at 0x44      <=============
    type: 0x10 (PCI Express)
  Capability register at 0x60      <============= 0x60 - 0x44 = 0x1c
    type: 0x05 (MSI)
  Capability register at 0x70
    type: 0x01 (Power Management)
  Capability register at 0x80
    type: 0x0d (Subsystem vendor ID)
(snip)
  PCI Message Signaled Interrupt
    Message Control register: 0x0080
      MSI Enabled: off
      Multiple Message Capable: no (1 vector)
      Multiple Message Enabled: off (1 vector)
      64 Bit Address Capable: on
      Per-Vector Masking Capable: off
      Extended Message Data Capable: off
      Extended Message Data Enable: off
    Message Address (lower) register: 0x00000000
    Message Address (upper) register: 0x00000000
    Message Data register: 0x0000
(snip)
  PCI Express Capabilities Register
(snip)
    Root Control Register: 0x7005 <=== 0x7005 is the first two byte
      SERR on Correctable Error Enable: on  of the MSI capability structure
      SERR on Non-Fatal Error Enable: off
      SERR on Fatal Error Enable: on
      PME Interrupt Enable: off
      CRS Software Visibility Enable: off
    Root Capability Register: 0x0080
      CRS Software Visibility: off
    Root Status Register: 0x00000000
      PME Requester ID: 0x0000
      PME was asserted: off
      another PME is pending: off

  Device-dependent header:
    0x40: 0x00c00000 0x00616010 0x00000001 0x00005026  |  |<- PCIe      |
    0x50: 0x0203f441 0x10010020 0x00000000 0x004803c0  |              ->|
    0x60: 0x00807005 0x00000000 0x00000000 0x00000000  |<-    MSI    ->|
    0x70: 0xc8028001 0x00000100 0x00000000 0x00000000  |<-    PM      ->|
    0x80: 0x0000000d 0x00000000 0x00000000 0x00000000  |<- subsystem ID  |
--------------------------------------

(msaitoh)

Use pmf_device_{register,deregister}().

(msaitoh)

Remove extra whitespaces.

(msaitoh)

Add script for assisting with debugging curses tests.

(blymn)

curses: Fix inverted logic when drawing LR corner with auto margins

Thanks to M. Levinson.

(roy)

Add base-256 test (Micha Gorny)

(christos)

Add base-256 decoding support (Micha Gorny)

(christos)

Add Amazon.com EC2 virtual 16650-compatible PCI serial device

(jmcneill)

regen

(jmcneill)

Add Amazon.com UART, NVMe, and ENA PCI IDs.

(jmcneill)

rasops_do_cursor():
- simplify & sanitize the unaligned case
- use only 32bit accesses
... no longer crash with odd sized fonts in 8bit

(macallan)

Add hw.wsdisplay.multicons sysctl that can be used to disable mirroring wsdisplay output to serial port.

(jmcneill)

sun4v: fix incorrect calculation of phys addr for the mdesc - now the virtual-devices can (again) be detected (worked previously...)

(palle)

use %u for u_int, rather than %d; corrects printing of frequencies at or above 2**31 Hz

(jakllsch)

Rewrite the gpa map/unmap functions. Dig holes in the mapped areas when
there is an overlap. Close to what Qemu expects.

(maxv)

Correct abuses of aprint_error_dev(9).

(jakllsch)

Improve my kern_time.c::rev1.192, systematically clear the buffers we get
from 'ptimer_pool' to prevent more leaks.

(maxv)

new openssl versions out

(wiz)

Tickets #1112 and #1113

(martin)

Pull up following revision(s) (requested by maxv in ticket #1113):

sys/kern/kern_sig.c: revision 1.350

Fix kernel info leak, 4 bytes of padding at the end of struct sigaction.

        + Possible info leak: [len=32, leaked=4]
        | #0 0xffffffff80baf327 in kleak_copyout
        | #1 0xffffffff80bd9ca8 in sys___sigaction_sigtramp
        | #2 0xffffffff80259c42 in syscall

(martin)

Pull up following revision(s) (requested by maxv in ticket #1112):

sys/kern/sys_ptrace_common.c: revision 1.46

Fix stack info leak.

        + Possible info leak: [len=136, leaked=92]
        | #0 0xffffffff80baf397 in kleak_copyout
        | #1 0xffffffff80bd4155 in ptrace_copyout_siginfo
        | #2 0xffffffff80bd5348 in do_ptrace
        | #3 0xffffffff80bd40fe in sys_ptrace
        | #4 0xffffffff80259c42 in syscall

(martin)

Fix info leak. There is one branch where 'status' is not initialized at
all.

+ Possible info leak: [len=4, leaked=4]
| #0 0xffffffff80baf397 in kleak_copyout
| #1 0xffffffff80b56d0c in sys_wait6
| #2 0xffffffff80259c42 in syscall

(maxv)

Fix stack info leak.

+ Possible info leak: [len=136, leaked=92]
| #0 0xffffffff80baf397 in kleak_copyout
| #1 0xffffffff80bd4155 in ptrace_copyout_siginfo
| #2 0xffffffff80bd5348 in do_ptrace
| #3 0xffffffff80bd40fe in sys_ptrace
| #4 0xffffffff80259c42 in syscall

(maxv)

Add nvme at pci

(jmcneill)

Fix kernel info leak, 4 bytes of padding at the end of struct sigaction.

+ Possible info leak: [len=32, leaked=4]
| #0 0xffffffff80baf327 in kleak_copyout
| #1 0xffffffff80bd9ca8 in sys___sigaction_sigtramp
| #2 0xffffffff80259c42 in syscall

(maxv)

Don't validate the source address of forwarding IPv6 packets (same as IPv4)

(ozaki-r)

Don't run DAD on link-up if it's explicitly disabled

(ozaki-r)

Introduce and use ip_dad_enabled() and ip6_dad_enabled() functions

(ozaki-r)

Ticket #1660

(martin)

Pull up following revision(s) (requested by maxv in ticket #1660):

sys/kern/kern_time.c: revision 1.192

Fix kernel info leak.

        + Possible info leak: [len=32, leaked=16]
        | #0 0xffffffff80baf3a7 in kleak_copyout
        | #1 0xffffffff80b940f8 in sys___timer_settime50
        | #2 0xffffffff80259c42 in syscall

(martin)

Ticket #1660

(martin)

Pull up following revision(s) (requested by maxv in ticket #1660):

sys/kern/kern_time.c: revision 1.192

Fix kernel info leak.

        + Possible info leak: [len=32, leaked=16]
        | #0 0xffffffff80baf3a7 in kleak_copyout
        | #1 0xffffffff80b940f8 in sys___timer_settime50
        | #2 0xffffffff80259c42 in syscall

(martin)

Ticket #1660

(martin)

Pull up following revision(s) (requested by maxv in ticket #1660):

sys/kern/kern_time.c: revision 1.192

Fix kernel info leak.

        + Possible info leak: [len=32, leaked=16]
        | #0 0xffffffff80baf3a7 in kleak_copyout
        | #1 0xffffffff80b940f8 in sys___timer_settime50
        | #2 0xffffffff80259c42 in syscall

(martin)

Tickets #1110 and #1111

(martin)

Pull up following revision(s) (requested by maxv in ticket #1111):

sys/arch/amd64/amd64/machdep.c: revision 1.321

Fix stack info leak. There is a big padding in struct sigframe_siginfo.

[  224.006287] kleak: Possible leak in copyout: [len=920, leaked=92]
[  224.016977] #0 0xffffffff80224d0a in kleak_note <netbsd>
[  224.026268] #1 0xffffffff80224d8a in kleak_copyout <netbsd>
[  224.026268] #2 0xffffffff802224b5 in sendsig_siginfo <netbsd>
[  224.036261] #3 0xffffffff80b51564 in sendsig <netbsd>
[  224.046475] #4 0xffffffff80b51282 in postsig <netbsd>
[  224.046475] #5 0xffffffff80b2fc5d in lwp_userret <netbsd>
[  224.056273] #6 0xffffffff8025a951 in mi_userret <netbsd>
[  224.066277] #7 0xffffffff8025ab89 in syscall <netbsd>

(martin)

Pull up following revision(s) (requested by maxv in ticket #1110):

sys/kern/kern_time.c: revision 1.192

Fix kernel info leak.

        + Possible info leak: [len=32, leaked=16]
        | #0 0xffffffff80baf3a7 in kleak_copyout
        | #1 0xffffffff80b940f8 in sys___timer_settime50
        | #2 0xffffffff80259c42 in syscall

(martin)

Better log entry for ticket #1656

(martin)

Fix off by one <tsahara at iij>

(christos)

Allow non-power of 2 counts, and support alloc/release/alloc patterns for a device as long as the ITT size is sufficient

(jmcneill)

Force a matching com@puc by seg/bus/dev/func to be the console device if specified in SPCR

(jmcneill)

Add support for a "force_console" property, where when set, reuses the console dev configuration

(jmcneill)

Add calls to if_attach and if_deferred_start_init

(jmcneill)

Ticket #1659

(martin)

Pull up following revision(s) (requested by mrg in ticket #1659):

libexec/httpd/main.c: revision 1.22
libexec/httpd/CHANGES: revision 1.29
libexec/httpd/cgi-bozo.c: revision 1.45
libexec/httpd/bozohttpd.h: revision 1.57
libexec/httpd/CHANGES: revision 1.30
libexec/httpd/bozohttpd.c: revision 1.97
libexec/httpd/bozohttpd.c: revision 1.98
libexec/httpd/bozohttpd.c: revision 1.99

one semicolon is usually enough.

-

appease lint

- add FALLTHROUGH comment
- one return is usually enough.

-

avoid c99ism.

-

fix -X option parsing.  noted by Rajeev V. Pillai.

-

add option fixes here.

-

normalise some messages.

(martin)

Ticket #1659

(martin)

Pull up following revision(s) (requested by mrg in ticket #1659):

libexec/httpd/main.c: revision 1.22
libexec/httpd/CHANGES: revision 1.29
libexec/httpd/cgi-bozo.c: revision 1.45
libexec/httpd/bozohttpd.h: revision 1.57
libexec/httpd/CHANGES: revision 1.30
libexec/httpd/bozohttpd.c: revision 1.97
libexec/httpd/bozohttpd.c: revision 1.98
libexec/httpd/bozohttpd.c: revision 1.99

one semicolon is usually enough.

-

appease lint

- add FALLTHROUGH comment
- one return is usually enough.

-

avoid c99ism.

-

fix -X option parsing.  noted by Rajeev V. Pillai.

-

add option fixes here.

-

normalise some messages.

(martin)

Ticket #1659

(martin)

Pull up following revision(s) (requested by mrg in ticket #1659):

libexec/httpd/main.c: revision 1.22
libexec/httpd/CHANGES: revision 1.29
libexec/httpd/cgi-bozo.c: revision 1.45
libexec/httpd/bozohttpd.h: revision 1.57
libexec/httpd/CHANGES: revision 1.30
libexec/httpd/bozohttpd.c: revision 1.97
libexec/httpd/bozohttpd.c: revision 1.98
libexec/httpd/bozohttpd.c: revision 1.99

one semicolon is usually enough.

-

appease lint

- add FALLTHROUGH comment
- one return is usually enough.

-

avoid c99ism.

-

fix -X option parsing.  noted by Rajeev V. Pillai.

-

add option fixes here.

-

normalise some messages.

(martin)

Ticket #1109

(martin)

Pull up following revision(s) (requested by mrg in ticket #1109):

libexec/httpd/main.c: revision 1.22
libexec/httpd/CHANGES: revision 1.29
libexec/httpd/cgi-bozo.c: revision 1.45
libexec/httpd/bozohttpd.h: revision 1.57
libexec/httpd/CHANGES: revision 1.30
libexec/httpd/bozohttpd.c: revision 1.97
libexec/httpd/bozohttpd.c: revision 1.98
libexec/httpd/bozohttpd.c: revision 1.99

one semicolon is usually enough.

-

appease lint

- add FALLTHROUGH comment
- one return is usually enough.

-

avoid c99ism.

-

fix -X option parsing.  noted by Rajeev V. Pillai.

-

add option fixes here.

-

normalise some messages.

(martin)

Move counting involuntary switches into mi_switch. preempt() passes that
information by setting a new LWP flag.

While here, don't even try to switch when the scheduler has no other LWP
to run. This check is currently spread over all callers of preempt()
and will be removed there.

ok mrg@.

(mlelstv)

Ticket #1658

(martin)

Pull up following revision(s) (requested by maxv in ticket #1658):

sys/kern/kern_exec.c: revision 1.462

Fix stack info leak. There are 2x4 bytes of padding in struct ps_strings.

[  223.896199] kleak: Possible leak in copyout: [len=32, leaked=8]
[  223.906430] #0 0xffffffff80224d0a in kleak_note <netbsd>
[  223.906430] #1 0xffffffff80224d8a in kleak_copyout <netbsd>
[  223.918363] #2 0xffffffff80b1e26c in copyoutpsstrs <netbsd>
[  223.926560] #3 0xffffffff80b1e331 in copyoutargs <netbsd>
[  223.936216] #4 0xffffffff80b21768 in execve_runproc <netbsd>
[  223.946225] #5 0xffffffff80b21cc9 in execve1 <netbsd>
[  223.946225] #6 0xffffffff8025a89c in sy_call <netbsd>
[  223.956225] #7 0xffffffff8025aace in sy_invoke <netbsd>
[  223.966232] #8 0xffffffff8025ab54 in syscall <netbsd>

(martin)

Ticket #1658

(martin)

Pull up following revision(s) (requested by maxv in ticket #1658):

sys/kern/kern_exec.c: revision 1.462

Fix stack info leak. There are 2x4 bytes of padding in struct ps_strings.

[  223.896199] kleak: Possible leak in copyout: [len=32, leaked=8]
[  223.906430] #0 0xffffffff80224d0a in kleak_note <netbsd>
[  223.906430] #1 0xffffffff80224d8a in kleak_copyout <netbsd>
[  223.918363] #2 0xffffffff80b1e26c in copyoutpsstrs <netbsd>
[  223.926560] #3 0xffffffff80b1e331 in copyoutargs <netbsd>
[  223.936216] #4 0xffffffff80b21768 in execve_runproc <netbsd>
[  223.946225] #5 0xffffffff80b21cc9 in execve1 <netbsd>
[  223.946225] #6 0xffffffff8025a89c in sy_call <netbsd>
[  223.956225] #7 0xffffffff8025aace in sy_invoke <netbsd>
[  223.966232] #8 0xffffffff8025ab54 in syscall <netbsd>

(martin)

Ticket #1658

(martin)

Pull up following revision(s) (requested by maxv in ticket #1658):

sys/kern/kern_exec.c: revision 1.462

Fix stack info leak. There are 2x4 bytes of padding in struct ps_strings.

[  223.896199] kleak: Possible leak in copyout: [len=32, leaked=8]
[  223.906430] #0 0xffffffff80224d0a in kleak_note <netbsd>
[  223.906430] #1 0xffffffff80224d8a in kleak_copyout <netbsd>
[  223.918363] #2 0xffffffff80b1e26c in copyoutpsstrs <netbsd>
[  223.926560] #3 0xffffffff80b1e331 in copyoutargs <netbsd>
[  223.936216] #4 0xffffffff80b21768 in execve_runproc <netbsd>
[  223.946225] #5 0xffffffff80b21cc9 in execve1 <netbsd>
[  223.946225] #6 0xffffffff8025a89c in sy_call <netbsd>
[  223.956225] #7 0xffffffff8025aace in sy_invoke <netbsd>
[  223.966232] #8 0xffffffff8025ab54 in syscall <netbsd>

(martin)

Revert previous for a better fix.

(mlelstv)

Avoid kpause while cold

(jmcneill)

Comment out implementation specific registers to avoid illegal instruction trap on ThunderX

(ryo)

Our softc is the private data of self, not parent. Spotted by mlelstv@

(jmcneill)

Use correct PCI BAR offsets

(jmcneill)

whitespace fix, NFC

(jdolecek)

Ticket #1657

(martin)

Pull up following revision(s) (requested by maxv in ticket #1657):

sys/net/rtsock.c: revision 1.244 (adapted)

Fix kernel info leak. There are 2 bytes of padding in struct if_msghdr.
[  944.607323] kleak: Possible leak in copyout: [len=176, leaked=2]
[  944.617335] #0 0xffffffff80b7c44a in kleak_note <netbsd>
[  944.627332] #1 0xffffffff80b7c4ca in kleak_copyout <netbsd>
[  944.627332] #2 0xffffffff80c91698 in sysctl_iflist_if <netbsd>
[  944.637336] #3 0xffffffff80c91d3c in sysctl_iflist <netbsd>
[  944.647343] #4 0xffffffff80c93855 in sysctl_rtable <netbsd>
[  944.647343] #5 0xffffffff80b5b328 in sysctl_dispatch <netbsd>
[  944.657346] #6 0xffffffff80b5b62e in sys___sysctl <netbsd>
[  944.667354] #7 0xffffffff8025ab3c in sy_call <netbsd>
[  944.667354] #8 0xffffffff8025ad6e in sy_invoke <netbsd>
[  944.677365] #9 0xffffffff8025adf4 in syscall <netbsd>

(martin)

Ticket #1657

(martin)

Pull up following revision(s) (requested by maxv in ticket #1657):

sys/net/rtsock.c: revision 1.244 (adapted)

Fix kernel info leak. There are 2 bytes of padding in struct if_msghdr.
[  944.607323] kleak: Possible leak in copyout: [len=176, leaked=2]
[  944.617335] #0 0xffffffff80b7c44a in kleak_note <netbsd>
[  944.627332] #1 0xffffffff80b7c4ca in kleak_copyout <netbsd>
[  944.627332] #2 0xffffffff80c91698 in sysctl_iflist_if <netbsd>
[  944.637336] #3 0xffffffff80c91d3c in sysctl_iflist <netbsd>
[  944.647343] #4 0xffffffff80c93855 in sysctl_rtable <netbsd>
[  944.647343] #5 0xffffffff80b5b328 in sysctl_dispatch <netbsd>
[  944.657346] #6 0xffffffff80b5b62e in sys___sysctl <netbsd>
[  944.667354] #7 0xffffffff8025ab3c in sy_call <netbsd>
[  944.667354] #8 0xffffffff8025ad6e in sy_invoke <netbsd>
[  944.677365] #9 0xffffffff8025adf4 in syscall <netbsd>

(martin)

Tickets #1656 and #1657

(martin)

Pull up following revision(s) (requested by maxv in ticket #1657):

sys/net/rtsock.c: revision 1.244 (adapted)

Fix kernel info leak. There are 2 bytes of padding in struct if_msghdr.
[  944.607323] kleak: Possible leak in copyout: [len=176, leaked=2]
[  944.617335] #0 0xffffffff80b7c44a in kleak_note <netbsd>
[  944.627332] #1 0xffffffff80b7c4ca in kleak_copyout <netbsd>
[  944.627332] #2 0xffffffff80c91698 in sysctl_iflist_if <netbsd>
[  944.637336] #3 0xffffffff80c91d3c in sysctl_iflist <netbsd>
[  944.647343] #4 0xffffffff80c93855 in sysctl_rtable <netbsd>
[  944.647343] #5 0xffffffff80b5b328 in sysctl_dispatch <netbsd>
[  944.657346] #6 0xffffffff80b5b62e in sys___sysctl <netbsd>
[  944.667354] #7 0xffffffff8025ab3c in sy_call <netbsd>
[  944.667354] #8 0xffffffff8025ad6e in sy_invoke <netbsd>
[  944.677365] #9 0xffffffff8025adf4 in syscall <netbsd>

(martin)

Pull up following revision(s) (requested by msaitoh in ticket #1656):

sys/dev/pci/if_wm.c: revision 1.599

- Fix a bug that i82578 PHY can't detect correctly.
- Print PHY OUI and model number if a PHY's can't be identified.

XXX pullup-[78]

(martin)

Fix kernel info leak.

+ Possible info leak: [len=32, leaked=16]
| #0 0xffffffff80baf3a7 in kleak_copyout
| #1 0xffffffff80b940f8 in sys___timer_settime50
| #2 0xffffffff80259c42 in syscall

(maxv)

Adapt ena_rx_checksum() to NetBSD.
It wasn't ported to the NetBSD conventions of indicating hardware checkum
status.

Compile tested only.

(bad)

Redo the access check for setting va_flags in zfs_netbsd_setattr().

Use user flag UF_NODUMP instead of UF_IMMUTABLE for the test as it
is the only user flag supported by all tested file systems.

PR kern/47656 test zfs_flags.

(hannken)

Add missing access check for REMOVE into zfs_netbsd_lookup().

PR kern/47656 test zfs_dirperms.

(hannken)

Add missing access check for setting va_Xtime into zfs_netbsd_setattr().

PR kern/47656 test zfs_times.

(hannken)

We expect VOP_ACCESS() to return EACCESS as general error.

Change zfs_netbsd_access() to translate the common EPERM to EACCES.

(hannken)

Don't try to release a NULL vnode in zfs_netbsd_rename().

(hannken)

Remove an early test for "source and target are equal" from zfs_rename()
that broke BSD semantics.

(hannken)

Add missing sa_buf_rele() into zfs_zget_cleaner().

(hannken)

Always unbusy pages in zfs_putapage() after the data has been written
into the DMU.  Running fsx no longer hangs the kernel.

(hannken)

Fix statistics in case mi_switch didn't actually switch LWPs.

(mlelstv)

support boot option "-1" to disable multiprocessor boot, and "-z" to set AB_SILENT flag.

(ryo)

* Increase minimum finger width to prevent entering scroll mode erroneously
* Attempt to clarify what the sysctl variables for finger scroll do
* Add hysteresis to validity check so changing it does not get rejected

Thanks to Martin Husemann and Michael van Elst for reporting the issues.

(blymn)

The register offset of the mask and pending register is depend on the 64bit
address capable bit, so fix the definition of PCI MSI vector mask and pending
register. This problem was not a real bug because PCI_MSI{MASK,PENDING} were
not used from anywhere.

(msaitoh)

Fix comment. No functional change.

(msaitoh)

don't pass illegal cpu index to cpu_lookup(). it may cause KASSERT.

(ryo)

don't exec 32bit binary on the cpu that has no aarch32.

(ryo)

Define MD4_DIGEST_STRING_LENGTH in <md4.h>

This is a kind of a symbol existing in other implementations and included
in headers for other hash APIs (like MD5).

(kamil)

Use better description for ticket 1108.

(snj)

Replace SPCR_INTERFACE_TYPE_* defines with ACPI_DBG2_* from acpica. Suggested by msaitoh@

(jmcneill)

Divide GIG by 512 (DEV_BLKSIZE) so we don't overflow 32 bit u_long
when multiplying by 128 (in 32 bits, 128 * GIG == 0)

Should fix i386 build )and other 32 bit ports).

An alternative would be to use ULL rather than UL for the numbers,
but that would mean needlessly doing 64 bit arith for data that is
resticted to 32 bit results anyway.

(kre)

actually allow pci_intr_alloc() with NULL count with MSI-X

(jdolecek)

document pci_intr_alloc() with NULL now also trying MSI-X

note aarch64 support

(jdolecek)

make pci_intr_alloc() try also MSI-X by default (with NULL count);
there are boards/emulators which only have MSI-X and no MSI, and
so far there is no evidence there are devices which support both
and don't work in MSI-X mode

this change is supposed to reduce amount of needed cut&paste code in drivers

discussed briefly with jmcneill@

(jdolecek)

More TSO4 fixes, from the freebsd driver:
- the chip doens't want the lenght of options, but the complete lenght of
  ip headers (ip + tcp + options). Fix this for the BGE_IS_5717_PLUS()
  and BGE_IS_5705_PLUS() cases; FreeBSD doens't cover the last case so
  leave it as is for now. This fixes checksum failures for heavy transfers.
- It looks like the transmit engine hangs if the TCP segment crosses a 4GB
  boundary. FreeBSD fixes it by mapping everything below 4GB; instead
  try detect when this happens and do the bounce only when needed.

With these fixes I could transfers 3GB images over ftp at gigabit speed
(112MB/s with wget) without problems. Tested on a
bge0 at pci4 dev 0 function 0: Broadcom BCM5720 Gigabit Ethernet
bge0: APE firmware NCSI 1.4.22.0
bge0: interrupting at msi1 vec 0
bge0: HW config 002b1194, 00006014, 0002aa38, 00000000 0000000c
bge0: ASIC BCM5720 A0 (0x5720000), Ethernet address d0:94:66:8b:9c:18
bge0: setting short Tx thresholds
brgphy0 at bge0 phy 1: BCM5720C 1000BASE-T media interface, rev. 0

(bouyer)

Add support for SPCR 16550 and 16450 interface types

(jmcneill)

Cosmetics: consistently use MEG and GIG defines for size calculations

(martin)

Add tprof

(msaitoh)

Fix widespread leak in the sendsig_siginfo() functions. sigframe_siginfo
has padding, so zero it out properly. While here I'm also zeroing out some
other things in several ports, for safety. Same problem in netbsd32, so
fix that too.

I can't compile-test on each architecture, but there should be no
breakage (tm).

Overall this fixes at least 14 info leaks. Prompted by the discovery by
KLEAK of a leak in amd64's sendsig_siginfo.

(maxv)

More macros, less whitespace.

(wiz)

Remove superfluous Ns.

(wiz)

Add two new test cases to add more tests for $@ expansions
(and a few more subtests in an existing test case).

The two new test cases currently fail, because of issues with
expanding "${1+$@}" which will (hopefully) be fixed soon.
(This looks to have been broken sometime during 2013 ... then I
made it even worse with some of the parser changes a while ago,
though the end result is that it appears less broken than it
really is.)

(kre)

Add a new test case to test <> (open for input & output),
includion a test to make sure that the file doesn't get truncated.

Add new subtests to the "incorrect redirections" test case, to
validate correct behaviour of the shell when redirections fail in
various scenarios, including when the redirect is the whole command.

More along these lines are really needed, but this is better than nothing.

All the added tests pass on the /bin/sh currently in netbsd HEAD.

(kre)

Remove the rest of my local change that somehow snuck in to an
earlier commit.

(pgoyette)

Remove obsolete ksem entry to prevent build break.

(msaitoh)

Fix link sha1.3 <- SHA1File.3

This SHA1File used to link to sha2(3).

(kamil)

Document SHA1FileChunk(3) in sha1(3)

Description taken from OpenBSD.

(kamil)

Include the compat_netbsd32* modules for sparc64

Their absence was noticed by martin@ when the recent libnv-related
build failure on amd64 did not affect sparc64!

(pgoyette)

tprof: Add goldmont plus support. Tested by msaitoh@n.o, thanks.

(knakahara)

add AHCI_PCI_QUIRK_FORCE for apu2 to be recognized; patch by Mike Pumford
on current-users@

(jdolecek)

Fix typo: O_ALTIO -> O_ALT_IO

Noted by @jbeich via GitHub.

(kamil)

Add ThunderX IDs

(ryo)

Now that lib/csu no longer uses the object handle, drop the magic and
version field from the internal object. It is kept in the legacy version
of the main object.

(joerg)