Add ktrace struct tracepoints for siginfo_t to the kernel side of
waitid(2) and __thrsigdivert(2) and teach kdump(1) to handle them.
Also report more from the siginfo_t inside PSIG tracepoints.

ok mpi@

Support the sendmmsg and recvmmsg system calls.
Input guenther@
OK bluhm@

Show scope id when set; ok florian@

Be more careful when dumping cmsghdr struct. In the SCM_RIGHTS case an
extra check for a truncated cmsghdr needs to be done since the embeded
lenght may be longer than the supplied buffer (MSG_CTRUNC case).
OK deraadt@

Peldge and unveil can have struct data size 0. While there, fix "fds"
handler; ok guenther@

trace struct flock; ok visa@

Unveiling unveil(2).
This brings unveil into the tree, disabled by default - Currently
this will return EPERM on all attempts to use it until we are
fully certain it is ready for people to start using, but this
now allows for others to do more tweaking and experimentation.

Still needs to send the unveil's across forks and execs before
fully enabling.

Many thanks to robert@ and deraadt@ for extensive testing.
ok deraadt@

pledge()'s 2nd argument becomes char *execpromises, which becomes the
pledge for a new execve image immediately upon start. Also introduces
"error" which makes violations return -1 ENOSYS instead of killing the
program ("error" may not be handed to a setuid/setgid program, which
may be missing/ignoring syscall return values and would continue with
inconsistant state)
Discussion with many
florian has used this to improve the strictness of a daemon

Add ktracing of the fds returned by pipe() and socketpair()

ok deraadt@

Pull in <sys/select.h> for fd_set

ok deraadt@

ktrace support for pollfd[] arrays
ok guenther

add a hack to avoid printing the long string format for times in the 70s.
this is usually a result of a timestamp which hasn't been identified as
"relative" and decoding them to precise dates long in the past is a
distraction. (eagerly awaiting a cleaner, better diff to do the same.)

On hppa, function pointer comparison can require dereferencing them.
kdump can't do that for a sigaction sa_handler pointer from the trace,
so cast to void* to suppress it.

ok deraadt@

Display NAMI records and AF_UNIX socket paths with vis, using
VIS_CSTYLE | VIS_DQ | VIS_TAB | VIS_NL; add the latter three flags
to the existing vis encoding of exec argv/environ and pledge
requests/paths.
Delete local variables left unused when showbuf() and showbufc() were split

ok otto@ millert@

Make the fd_set* passed to FD_ISSET() non-const: the non-macro version is
stricter (per POSIX, dang it)

Display struct event more like other structs: show all members and use cpp
define names. Use the mksubr infrastructure for EVFILT_*,
EV_*, and (with some kludging) NOTE_* values. If EV_ERROR is set,
include the errno string for the data member.

Also, do MSG_* parsing on the msg_flags member of struct msghdr.

ok millert@

print more info about kevent

very basic kevent printing

Use offsetof() instead of adding the sizes of the preceeding struct members

ok millert@

Change all tame callers to namechange to pledge(2).

Add ktracing of tame()'s arguments' values

"every tool helps" deraadt@

Add ktracing of structs iovec, msghdr, and cmsghdr for {,p}{read,write}v(),
sendmsg(), and recvmsg(). For cmsghdr, the len, level, and type are always
shown, and for SOL_SOCKET,SCM_RIGHTS the fd numbers being passed are shown.

ok millert@ deraadt@

Don't display formatted time if localtime() fails.
Avoids a crash in strftime() found with the afl fuzzer.
ok guenther@

Eliminate the -r option and always do sysctl OID, username, groupname,
and ctime presentation, but combined with the numeric form ala 0<"root">.
Do username and groupname presentation on syscall arguments and retvals.

ok millert@ otto@

Be more POSIXy by using blksize_t (a.k.a. int32) for st_blksize, rather than the current uint32_t.

ok guenther@ deraadt@

Add dumping of struct dqblk done by quotactl(2)

ok millert@

remove (commented) bluetooth reference

Negative time{spec,val}s with non-zero subsecond parts require extra
handling. E.g., a timeval of { -1, 1 } is -0.999999, as the subsecond
part is always positive.

Also, update the list of sockaddr_* types that we might add support for

exit1() needs to do a final aggregation of the thread's [us]ticks
and runtime to the process totals. Also, add ktracing of struct
rusage in wait4() and getrusage().

problem pointed out by tedu@
ok deraadt@

Recognize the UTIME_OMIT and UTIME_NOW values when printing timespecs

ok otto@

Split out from kdump.c the ktrstruct.c bits into ktrstruct.c
Reduce the #includes to take advantage of that.

ok millert@ otto@

Support the sendmmsg and recvmmsg system calls.
Input guenther@
OK bluhm@

Show scope id when set; ok florian@

Be more careful when dumping cmsghdr struct. In the SCM_RIGHTS case an
extra check for a truncated cmsghdr needs to be done since the embeded
lenght may be longer than the supplied buffer (MSG_CTRUNC case).
OK deraadt@

Peldge and unveil can have struct data size 0. While there, fix "fds"
handler; ok guenther@

trace struct flock; ok visa@

Unveiling unveil(2).
This brings unveil into the tree, disabled by default - Currently
this will return EPERM on all attempts to use it until we are
fully certain it is ready for people to start using, but this
now allows for others to do more tweaking and experimentation.

Still needs to send the unveil's across forks and execs before
fully enabling.

Many thanks to robert@ and deraadt@ for extensive testing.
ok deraadt@

pledge()'s 2nd argument becomes char *execpromises, which becomes the
pledge for a new execve image immediately upon start. Also introduces
"error" which makes violations return -1 ENOSYS instead of killing the
program ("error" may not be handed to a setuid/setgid program, which
may be missing/ignoring syscall return values and would continue with
inconsistant state)
Discussion with many
florian has used this to improve the strictness of a daemon

Add ktracing of the fds returned by pipe() and socketpair()

ok deraadt@

Pull in <sys/select.h> for fd_set

ok deraadt@

ktrace support for pollfd[] arrays
ok guenther

add a hack to avoid printing the long string format for times in the 70s.
this is usually a result of a timestamp which hasn't been identified as
"relative" and decoding them to precise dates long in the past is a
distraction. (eagerly awaiting a cleaner, better diff to do the same.)

On hppa, function pointer comparison can require dereferencing them.
kdump can't do that for a sigaction sa_handler pointer from the trace,
so cast to void* to suppress it.

ok deraadt@

Display NAMI records and AF_UNIX socket paths with vis, using
VIS_CSTYLE | VIS_DQ | VIS_TAB | VIS_NL; add the latter three flags
to the existing vis encoding of exec argv/environ and pledge
requests/paths.
Delete local variables left unused when showbuf() and showbufc() were split

ok otto@ millert@

Make the fd_set* passed to FD_ISSET() non-const: the non-macro version is
stricter (per POSIX, dang it)

Display struct event more like other structs: show all members and use cpp
define names. Use the mksubr infrastructure for EVFILT_*,
EV_*, and (with some kludging) NOTE_* values. If EV_ERROR is set,
include the errno string for the data member.

Also, do MSG_* parsing on the msg_flags member of struct msghdr.

ok millert@

print more info about kevent

very basic kevent printing

Use offsetof() instead of adding the sizes of the preceeding struct members

ok millert@

Change all tame callers to namechange to pledge(2).

Add ktracing of tame()'s arguments' values

"every tool helps" deraadt@

Add ktracing of structs iovec, msghdr, and cmsghdr for {,p}{read,write}v(),
sendmsg(), and recvmsg(). For cmsghdr, the len, level, and type are always
shown, and for SOL_SOCKET,SCM_RIGHTS the fd numbers being passed are shown.

ok millert@ deraadt@

Don't display formatted time if localtime() fails.
Avoids a crash in strftime() found with the afl fuzzer.
ok guenther@

Eliminate the -r option and always do sysctl OID, username, groupname,
and ctime presentation, but combined with the numeric form ala 0<"root">.
Do username and groupname presentation on syscall arguments and retvals.

ok millert@ otto@

Be more POSIXy by using blksize_t (a.k.a. int32) for st_blksize, rather than the current uint32_t.

ok guenther@ deraadt@

Add dumping of struct dqblk done by quotactl(2)

ok millert@

remove (commented) bluetooth reference

Negative time{spec,val}s with non-zero subsecond parts require extra
handling. E.g., a timeval of { -1, 1 } is -0.999999, as the subsecond
part is always positive.

Also, update the list of sockaddr_* types that we might add support for

exit1() needs to do a final aggregation of the thread's [us]ticks
and runtime to the process totals. Also, add ktracing of struct
rusage in wait4() and getrusage().

problem pointed out by tedu@
ok deraadt@

Recognize the UTIME_OMIT and UTIME_NOW values when printing timespecs

ok otto@

Split out from kdump.c the ktrstruct.c bits into ktrstruct.c
Reduce the #includes to take advantage of that.

ok millert@ otto@

Show scope id when set; ok florian@

Be more careful when dumping cmsghdr struct. In the SCM_RIGHTS case an
extra check for a truncated cmsghdr needs to be done since the embeded
lenght may be longer than the supplied buffer (MSG_CTRUNC case).
OK deraadt@

Peldge and unveil can have struct data size 0. While there, fix "fds"
handler; ok guenther@

trace struct flock; ok visa@

Unveiling unveil(2).
This brings unveil into the tree, disabled by default - Currently
this will return EPERM on all attempts to use it until we are
fully certain it is ready for people to start using, but this
now allows for others to do more tweaking and experimentation.

Still needs to send the unveil's across forks and execs before
fully enabling.

Many thanks to robert@ and deraadt@ for extensive testing.
ok deraadt@

pledge()'s 2nd argument becomes char *execpromises, which becomes the
pledge for a new execve image immediately upon start. Also introduces
"error" which makes violations return -1 ENOSYS instead of killing the
program ("error" may not be handed to a setuid/setgid program, which
may be missing/ignoring syscall return values and would continue with
inconsistant state)
Discussion with many
florian has used this to improve the strictness of a daemon

Add ktracing of the fds returned by pipe() and socketpair()

ok deraadt@

Pull in <sys/select.h> for fd_set

ok deraadt@

ktrace support for pollfd[] arrays
ok guenther

add a hack to avoid printing the long string format for times in the 70s.
this is usually a result of a timestamp which hasn't been identified as
"relative" and decoding them to precise dates long in the past is a
distraction. (eagerly awaiting a cleaner, better diff to do the same.)

On hppa, function pointer comparison can require dereferencing them.
kdump can't do that for a sigaction sa_handler pointer from the trace,
so cast to void* to suppress it.

ok deraadt@

Display NAMI records and AF_UNIX socket paths with vis, using
VIS_CSTYLE | VIS_DQ | VIS_TAB | VIS_NL; add the latter three flags
to the existing vis encoding of exec argv/environ and pledge
requests/paths.
Delete local variables left unused when showbuf() and showbufc() were split

ok otto@ millert@

Make the fd_set* passed to FD_ISSET() non-const: the non-macro version is
stricter (per POSIX, dang it)

Display struct event more like other structs: show all members and use cpp
define names. Use the mksubr infrastructure for EVFILT_*,
EV_*, and (with some kludging) NOTE_* values. If EV_ERROR is set,
include the errno string for the data member.

Also, do MSG_* parsing on the msg_flags member of struct msghdr.

ok millert@

print more info about kevent

very basic kevent printing

Use offsetof() instead of adding the sizes of the preceeding struct members

ok millert@

Change all tame callers to namechange to pledge(2).

Add ktracing of tame()'s arguments' values

"every tool helps" deraadt@

Add ktracing of structs iovec, msghdr, and cmsghdr for {,p}{read,write}v(),
sendmsg(), and recvmsg(). For cmsghdr, the len, level, and type are always
shown, and for SOL_SOCKET,SCM_RIGHTS the fd numbers being passed are shown.

ok millert@ deraadt@

Don't display formatted time if localtime() fails.
Avoids a crash in strftime() found with the afl fuzzer.
ok guenther@

Eliminate the -r option and always do sysctl OID, username, groupname,
and ctime presentation, but combined with the numeric form ala 0<"root">.
Do username and groupname presentation on syscall arguments and retvals.

ok millert@ otto@

Be more POSIXy by using blksize_t (a.k.a. int32) for st_blksize, rather than the current uint32_t.

ok guenther@ deraadt@

Add dumping of struct dqblk done by quotactl(2)

ok millert@

remove (commented) bluetooth reference

Negative time{spec,val}s with non-zero subsecond parts require extra
handling. E.g., a timeval of { -1, 1 } is -0.999999, as the subsecond
part is always positive.

Also, update the list of sockaddr_* types that we might add support for

exit1() needs to do a final aggregation of the thread's [us]ticks
and runtime to the process totals. Also, add ktracing of struct
rusage in wait4() and getrusage().

problem pointed out by tedu@
ok deraadt@

Recognize the UTIME_OMIT and UTIME_NOW values when printing timespecs

ok otto@

Split out from kdump.c the ktrstruct.c bits into ktrstruct.c
Reduce the #includes to take advantage of that.

ok millert@ otto@

Be more careful when dumping cmsghdr struct. In the SCM_RIGHTS case an
extra check for a truncated cmsghdr needs to be done since the embeded
lenght may be longer than the supplied buffer (MSG_CTRUNC case).
OK deraadt@

Peldge and unveil can have struct data size 0. While there, fix "fds"
handler; ok guenther@

trace struct flock; ok visa@

Unveiling unveil(2).
This brings unveil into the tree, disabled by default - Currently
this will return EPERM on all attempts to use it until we are
fully certain it is ready for people to start using, but this
now allows for others to do more tweaking and experimentation.

Still needs to send the unveil's across forks and execs before
fully enabling.

Many thanks to robert@ and deraadt@ for extensive testing.
ok deraadt@

pledge()'s 2nd argument becomes char *execpromises, which becomes the
pledge for a new execve image immediately upon start. Also introduces
"error" which makes violations return -1 ENOSYS instead of killing the
program ("error" may not be handed to a setuid/setgid program, which
may be missing/ignoring syscall return values and would continue with
inconsistant state)
Discussion with many
florian has used this to improve the strictness of a daemon

Add ktracing of the fds returned by pipe() and socketpair()

ok deraadt@

Pull in <sys/select.h> for fd_set

ok deraadt@

ktrace support for pollfd[] arrays
ok guenther

add a hack to avoid printing the long string format for times in the 70s.
this is usually a result of a timestamp which hasn't been identified as
"relative" and decoding them to precise dates long in the past is a
distraction. (eagerly awaiting a cleaner, better diff to do the same.)

On hppa, function pointer comparison can require dereferencing them.
kdump can't do that for a sigaction sa_handler pointer from the trace,
so cast to void* to suppress it.

ok deraadt@

Display NAMI records and AF_UNIX socket paths with vis, using
VIS_CSTYLE | VIS_DQ | VIS_TAB | VIS_NL; add the latter three flags
to the existing vis encoding of exec argv/environ and pledge
requests/paths.
Delete local variables left unused when showbuf() and showbufc() were split

ok otto@ millert@

Make the fd_set* passed to FD_ISSET() non-const: the non-macro version is
stricter (per POSIX, dang it)

Display struct event more like other structs: show all members and use cpp
define names. Use the mksubr infrastructure for EVFILT_*,
EV_*, and (with some kludging) NOTE_* values. If EV_ERROR is set,
include the errno string for the data member.

Also, do MSG_* parsing on the msg_flags member of struct msghdr.

ok millert@

print more info about kevent

very basic kevent printing

Use offsetof() instead of adding the sizes of the preceeding struct members

ok millert@

Change all tame callers to namechange to pledge(2).

Add ktracing of tame()'s arguments' values

"every tool helps" deraadt@

Add ktracing of structs iovec, msghdr, and cmsghdr for {,p}{read,write}v(),
sendmsg(), and recvmsg(). For cmsghdr, the len, level, and type are always
shown, and for SOL_SOCKET,SCM_RIGHTS the fd numbers being passed are shown.

ok millert@ deraadt@

Don't display formatted time if localtime() fails.
Avoids a crash in strftime() found with the afl fuzzer.
ok guenther@

Eliminate the -r option and always do sysctl OID, username, groupname,
and ctime presentation, but combined with the numeric form ala 0<"root">.
Do username and groupname presentation on syscall arguments and retvals.

ok millert@ otto@

Be more POSIXy by using blksize_t (a.k.a. int32) for st_blksize, rather than the current uint32_t.

ok guenther@ deraadt@

Add dumping of struct dqblk done by quotactl(2)

ok millert@

remove (commented) bluetooth reference

Negative time{spec,val}s with non-zero subsecond parts require extra
handling. E.g., a timeval of { -1, 1 } is -0.999999, as the subsecond
part is always positive.

Also, update the list of sockaddr_* types that we might add support for

exit1() needs to do a final aggregation of the thread's [us]ticks
and runtime to the process totals. Also, add ktracing of struct
rusage in wait4() and getrusage().

problem pointed out by tedu@
ok deraadt@

Recognize the UTIME_OMIT and UTIME_NOW values when printing timespecs

ok otto@

Split out from kdump.c the ktrstruct.c bits into ktrstruct.c
Reduce the #includes to take advantage of that.

ok millert@ otto@

Peldge and unveil can have struct data size 0. While there, fix "fds"
handler; ok guenther@

trace struct flock; ok visa@

Unveiling unveil(2).
This brings unveil into the tree, disabled by default - Currently
this will return EPERM on all attempts to use it until we are
fully certain it is ready for people to start using, but this
now allows for others to do more tweaking and experimentation.

Still needs to send the unveil's across forks and execs before
fully enabling.

Many thanks to robert@ and deraadt@ for extensive testing.
ok deraadt@

pledge()'s 2nd argument becomes char *execpromises, which becomes the
pledge for a new execve image immediately upon start. Also introduces
"error" which makes violations return -1 ENOSYS instead of killing the
program ("error" may not be handed to a setuid/setgid program, which
may be missing/ignoring syscall return values and would continue with
inconsistant state)
Discussion with many
florian has used this to improve the strictness of a daemon

Add ktracing of the fds returned by pipe() and socketpair()

ok deraadt@

Pull in <sys/select.h> for fd_set

ok deraadt@

ktrace support for pollfd[] arrays
ok guenther

add a hack to avoid printing the long string format for times in the 70s.
this is usually a result of a timestamp which hasn't been identified as
"relative" and decoding them to precise dates long in the past is a
distraction. (eagerly awaiting a cleaner, better diff to do the same.)

On hppa, function pointer comparison can require dereferencing them.
kdump can't do that for a sigaction sa_handler pointer from the trace,
so cast to void* to suppress it.

ok deraadt@

Display NAMI records and AF_UNIX socket paths with vis, using
VIS_CSTYLE | VIS_DQ | VIS_TAB | VIS_NL; add the latter three flags
to the existing vis encoding of exec argv/environ and pledge
requests/paths.
Delete local variables left unused when showbuf() and showbufc() were split

ok otto@ millert@

Make the fd_set* passed to FD_ISSET() non-const: the non-macro version is
stricter (per POSIX, dang it)

Display struct event more like other structs: show all members and use cpp
define names. Use the mksubr infrastructure for EVFILT_*,
EV_*, and (with some kludging) NOTE_* values. If EV_ERROR is set,
include the errno string for the data member.

Also, do MSG_* parsing on the msg_flags member of struct msghdr.

ok millert@

print more info about kevent

very basic kevent printing

Use offsetof() instead of adding the sizes of the preceeding struct members

ok millert@

Change all tame callers to namechange to pledge(2).

Add ktracing of tame()'s arguments' values

"every tool helps" deraadt@

Add ktracing of structs iovec, msghdr, and cmsghdr for {,p}{read,write}v(),
sendmsg(), and recvmsg(). For cmsghdr, the len, level, and type are always
shown, and for SOL_SOCKET,SCM_RIGHTS the fd numbers being passed are shown.

ok millert@ deraadt@

Don't display formatted time if localtime() fails.
Avoids a crash in strftime() found with the afl fuzzer.
ok guenther@

Eliminate the -r option and always do sysctl OID, username, groupname,
and ctime presentation, but combined with the numeric form ala 0<"root">.
Do username and groupname presentation on syscall arguments and retvals.

ok millert@ otto@

Be more POSIXy by using blksize_t (a.k.a. int32) for st_blksize, rather than the current uint32_t.

ok guenther@ deraadt@

Add dumping of struct dqblk done by quotactl(2)

ok millert@

remove (commented) bluetooth reference

Negative time{spec,val}s with non-zero subsecond parts require extra
handling. E.g., a timeval of { -1, 1 } is -0.999999, as the subsecond
part is always positive.

Also, update the list of sockaddr_* types that we might add support for

exit1() needs to do a final aggregation of the thread's [us]ticks
and runtime to the process totals. Also, add ktracing of struct
rusage in wait4() and getrusage().

problem pointed out by tedu@
ok deraadt@

Recognize the UTIME_OMIT and UTIME_NOW values when printing timespecs

ok otto@

Split out from kdump.c the ktrstruct.c bits into ktrstruct.c
Reduce the #includes to take advantage of that.

ok millert@ otto@

Unveiling unveil(2).
This brings unveil into the tree, disabled by default - Currently
this will return EPERM on all attempts to use it until we are
fully certain it is ready for people to start using, but this
now allows for others to do more tweaking and experimentation.

Still needs to send the unveil's across forks and execs before
fully enabling.

Many thanks to robert@ and deraadt@ for extensive testing.
ok deraadt@

pledge()'s 2nd argument becomes char *execpromises, which becomes the
pledge for a new execve image immediately upon start. Also introduces
"error" which makes violations return -1 ENOSYS instead of killing the
program ("error" may not be handed to a setuid/setgid program, which
may be missing/ignoring syscall return values and would continue with
inconsistant state)
Discussion with many
florian has used this to improve the strictness of a daemon

Add ktracing of the fds returned by pipe() and socketpair()

ok deraadt@

Pull in <sys/select.h> for fd_set

ok deraadt@

ktrace support for pollfd[] arrays
ok guenther

add a hack to avoid printing the long string format for times in the 70s.
this is usually a result of a timestamp which hasn't been identified as
"relative" and decoding them to precise dates long in the past is a
distraction. (eagerly awaiting a cleaner, better diff to do the same.)

On hppa, function pointer comparison can require dereferencing them.
kdump can't do that for a sigaction sa_handler pointer from the trace,
so cast to void* to suppress it.

ok deraadt@

Display NAMI records and AF_UNIX socket paths with vis, using
VIS_CSTYLE | VIS_DQ | VIS_TAB | VIS_NL; add the latter three flags
to the existing vis encoding of exec argv/environ and pledge
requests/paths.
Delete local variables left unused when showbuf() and showbufc() were split

ok otto@ millert@

Make the fd_set* passed to FD_ISSET() non-const: the non-macro version is
stricter (per POSIX, dang it)

Display struct event more like other structs: show all members and use cpp
define names. Use the mksubr infrastructure for EVFILT_*,
EV_*, and (with some kludging) NOTE_* values. If EV_ERROR is set,
include the errno string for the data member.

Also, do MSG_* parsing on the msg_flags member of struct msghdr.

ok millert@

print more info about kevent

very basic kevent printing

Use offsetof() instead of adding the sizes of the preceeding struct members

ok millert@

Change all tame callers to namechange to pledge(2).

Add ktracing of tame()'s arguments' values

"every tool helps" deraadt@

Add ktracing of structs iovec, msghdr, and cmsghdr for {,p}{read,write}v(),
sendmsg(), and recvmsg(). For cmsghdr, the len, level, and type are always
shown, and for SOL_SOCKET,SCM_RIGHTS the fd numbers being passed are shown.

ok millert@ deraadt@

Don't display formatted time if localtime() fails.
Avoids a crash in strftime() found with the afl fuzzer.
ok guenther@

Eliminate the -r option and always do sysctl OID, username, groupname,
and ctime presentation, but combined with the numeric form ala 0<"root">.
Do username and groupname presentation on syscall arguments and retvals.

ok millert@ otto@

Be more POSIXy by using blksize_t (a.k.a. int32) for st_blksize, rather than the current uint32_t.

ok guenther@ deraadt@

Add dumping of struct dqblk done by quotactl(2)

ok millert@

remove (commented) bluetooth reference

Negative time{spec,val}s with non-zero subsecond parts require extra
handling. E.g., a timeval of { -1, 1 } is -0.999999, as the subsecond
part is always positive.

Also, update the list of sockaddr_* types that we might add support for

exit1() needs to do a final aggregation of the thread's [us]ticks
and runtime to the process totals. Also, add ktracing of struct
rusage in wait4() and getrusage().

problem pointed out by tedu@
ok deraadt@

Recognize the UTIME_OMIT and UTIME_NOW values when printing timespecs

ok otto@

Split out from kdump.c the ktrstruct.c bits into ktrstruct.c
Reduce the #includes to take advantage of that.

ok millert@ otto@

pledge()'s 2nd argument becomes char *execpromises, which becomes the
pledge for a new execve image immediately upon start. Also introduces
"error" which makes violations return -1 ENOSYS instead of killing the
program ("error" may not be handed to a setuid/setgid program, which
may be missing/ignoring syscall return values and would continue with
inconsistant state)
Discussion with many
florian has used this to improve the strictness of a daemon

Add ktracing of the fds returned by pipe() and socketpair()

ok deraadt@

Pull in <sys/select.h> for fd_set

ok deraadt@

ktrace support for pollfd[] arrays
ok guenther

add a hack to avoid printing the long string format for times in the 70s.
this is usually a result of a timestamp which hasn't been identified as
"relative" and decoding them to precise dates long in the past is a
distraction. (eagerly awaiting a cleaner, better diff to do the same.)

On hppa, function pointer comparison can require dereferencing them.
kdump can't do that for a sigaction sa_handler pointer from the trace,
so cast to void* to suppress it.

ok deraadt@

Display NAMI records and AF_UNIX socket paths with vis, using
VIS_CSTYLE | VIS_DQ | VIS_TAB | VIS_NL; add the latter three flags
to the existing vis encoding of exec argv/environ and pledge
requests/paths.
Delete local variables left unused when showbuf() and showbufc() were split

ok otto@ millert@

Make the fd_set* passed to FD_ISSET() non-const: the non-macro version is
stricter (per POSIX, dang it)

Display struct event more like other structs: show all members and use cpp
define names. Use the mksubr infrastructure for EVFILT_*,
EV_*, and (with some kludging) NOTE_* values. If EV_ERROR is set,
include the errno string for the data member.

Also, do MSG_* parsing on the msg_flags member of struct msghdr.

ok millert@

print more info about kevent

very basic kevent printing

Use offsetof() instead of adding the sizes of the preceeding struct members

ok millert@

Change all tame callers to namechange to pledge(2).

Add ktracing of tame()'s arguments' values

"every tool helps" deraadt@

Add ktracing of structs iovec, msghdr, and cmsghdr for {,p}{read,write}v(),
sendmsg(), and recvmsg(). For cmsghdr, the len, level, and type are always
shown, and for SOL_SOCKET,SCM_RIGHTS the fd numbers being passed are shown.

ok millert@ deraadt@

Don't display formatted time if localtime() fails.
Avoids a crash in strftime() found with the afl fuzzer.
ok guenther@

Eliminate the -r option and always do sysctl OID, username, groupname,
and ctime presentation, but combined with the numeric form ala 0<"root">.
Do username and groupname presentation on syscall arguments and retvals.

ok millert@ otto@

Be more POSIXy by using blksize_t (a.k.a. int32) for st_blksize, rather than the current uint32_t.

ok guenther@ deraadt@

Add dumping of struct dqblk done by quotactl(2)

ok millert@

remove (commented) bluetooth reference

Negative time{spec,val}s with non-zero subsecond parts require extra
handling. E.g., a timeval of { -1, 1 } is -0.999999, as the subsecond
part is always positive.

Also, update the list of sockaddr_* types that we might add support for

exit1() needs to do a final aggregation of the thread's [us]ticks
and runtime to the process totals. Also, add ktracing of struct
rusage in wait4() and getrusage().

problem pointed out by tedu@
ok deraadt@

Recognize the UTIME_OMIT and UTIME_NOW values when printing timespecs

ok otto@

Split out from kdump.c the ktrstruct.c bits into ktrstruct.c
Reduce the #includes to take advantage of that.

ok millert@ otto@