The idea here is to avoid a memory access and conditional branch per
probe site. Instead, the probe is represented by an "unreachable"
unconditional function call. asm goto is used to store the address of
the probe site (represented by a no-op sled) and the address of the
function call into a tracepoint record. Each SDT probe carries a list
of tracepoints.
When the probe is enabled, the no-op sled corresponding to each
tracepoint is overwritten with a jmp to the corresponding label. The
implementation uses smp_rendezvous() to park all other CPUs while the
instruction is being overwritten, as this can't be done atomically in
general. The compiler moves argument marshalling code and the
sdt_probe() function call out-of-line, i.e., to the end of the function.
Per gallatin@ in D43504, this approach has less overhead when probes are
disabled. To make the implementation a bit simpler, I removed support
for probes with 7 arguments; nothing makes use of this except a
regression test case. It could be re-added later if need be.
The approach taken in this patch enables some more improvements:
1. We can now automatically fill out the "function" field of SDT probe
names. The SDT macros let the programmer specify the function and
module names, but this is really a bug and shouldn't have been
allowed. The intent was to be able to have the same probe in
multiple functions and to let the user restrict which probes actually
get enabled by specifying a function name or glob.
2. We can avoid branching on SDT_PROBES_ENABLED() by adding the ability
to include blocks of code in the out-of-line path. For example:
if (SDT_PROBES_ENABLED()) {
int reason = CLD_EXITED;
if (WCOREDUMP(signo))
reason = CLD_DUMPED;
else if (WIFSIGNALED(signo))
reason = CLD_KILLED;
SDT_PROBE1(proc, , , exit, reason);
}
could be written
SDT_PROBE1_EXT(proc, , , exit, reason,
int reason;
reason = CLD_EXITED;
if (WCOREDUMP(signo))
reason = CLD_DUMPED;
else if (WIFSIGNALED(signo))
reason = CLD_KILLED;
);
In the future I would like to use this mechanism more generally, e.g.,
to remove branches and marshalling code used by hwpmc, and generally to
make it easier to add new tracepoint consumers without having to add
more conditional branches to hot code paths.
Reviewed by: Domagoj Stolfa, avg
MFC after: 2 months
Differential Revision: https://reviews.freebsd.org/D44483
This way we can avoid making assumptions about the SDT probe
implementation. No functional change intended.
This was submitted upstream as https://github.com/openzfs/zfs/pull/16284
MFC after: 1 week
Match the "U" in RFSPAWN's value (1U<<31) (and "u" for completeness).
Reject kernel-only RFPPWAIT to avoid a duplicate entry.
Reviewed by: imp, markj
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D45597
While most "cloudware" (AWS, Azure, GCP, Vagrant) images get uploaded
to their respective clouds, we have two images -- BASIC-CI and
BASIC-CLOUDINIT -- which are published via the FTP mirrors. We have
been handling these using a manual and error-prone process of copying
them from the internal build directories into the FTP staging tree,
but they should be handled using the "make install ftp-stage"
mechanism which is used for all the other published images.
This commit 1. Adds a new 'make cloud-install' target which is run as
part of 'make install' and installs BASIC-CI and BASIC-CLOUDINIT images,
and 2. Extends 'make ftp-stage' in Makefile.mirrors to copy those bits
into the FTP staging area.
As a side benefit, this will also make BASIC-CI and BASIC-CLOUDINIT
images available for arm64; until now they have only been available
for amd64.
MFC after: 1 week
This is derived from swills@ fork of the Juniper virtfs with many
changes by me including bug fixes, style improvements, clearer layering
and more consistent logging. The filesystem is renamed to p9fs to better
reflect its function and to prevent possible future confusion with
virtio-fs.
Several updates and fixes from Juniper have been integrated into this
version by Val Packett and these contributions along with the original
Juniper authors are credited below.
To use this with bhyve, add 'virtio_p9fs_load=YES' to loader.conf. The
bhyve virtio-9p device allows access from the guest to files on the host
by mapping a 'sharename' to a host path. It is possible to use p9fs as a
root filesystem by adding this to /boot/loader.conf:
vfs.root.mountfrom="p9fs:sharename"
for non-root filesystems add something like this to /etc/fstab:
sharename /mnt p9fs rw 0 0
In both examples, substitute the share name used on the bhyve command
line.
The 9P filesystem protocol relies on stateful file opens which map
protocol-level FIDs to host file descriptors. The FreeBSD vnode
interface doesn't really support this and we use heuristics to guess the
right FID to use for file operations. This can be confused by privilege
lowering and does not guarantee that the FID created for a given file
open is always used for file operations, even if the calling process is
using the file descriptor from the original open call. Improving this
would involve changes to the vnode interface which is out-of-scope for
this import.
Differential Revision: https://reviews.freebsd.org/D41844
Reviewed by: kib, emaste, dch
MFC after: 3 months
Co-authored-by: Val Packett <val@packett.cool>
Co-authored-by: Ka Ho Ng <kahon@juniper.net>
Co-authored-by: joyu <joyul@juniper.net>
Co-authored-by: Kumara Babu Narayanaswamy <bkumara@juniper.net>
The targ example program doesn't compile with current clang, and
probably hasn't for multiple releases. Fix the build. I don't have the
right hardware to test it, though.
MFC after: 2 weeks
Sponsored by: Axcient
It was mostly removed from the man page in
9c887a4f86, but left in the command.
Fully remove it from the man page, and warn if anybody uses it. Remove
it entirely for FreeBSD 16.
PR: 279546
MFC after: 2 weeks
Reviewed by: mav
Sponsored by: Axcient
Differential Revision: https://reviews.freebsd.org/D45522
Since we e07f917850 ("sound: Separate implementations for
SNDCTL_AUDIOINFO[_EX] and SNDCTL_ENGINEINFO") support more than
mono/stereo.
Sponsored by: The FreeBSD Foundation
MFC after: 1 day
Reviewed by: dev_submerge.ch, markj, emaste
Differential Revision: https://reviews.freebsd.org/D45605
Although the docs advise against using them, it doesn't hurt to fill
them out correctly.
Sponsored by: The FreeBSD Foundation
MFC after: 1 day
Reviewed by: dev_submerge.ch, markj
Differential Revision: https://reviews.freebsd.org/D45604
In commit d8237b9555, as part of the deprecation of IPv4 address
classes, Mike Karels introduced a warning in ifconfig for setting an
address without a width or mask. The commit message says "This will
hopefully be an error in the future."
As the warning has been included in releases from FreeBSD 13.1 on, and
sufficient time has passed, turn this into an error.
Reviewed by: allanjude, olce, grembo, philip, gordon
Relnotes: Yes
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D45585
Make sure that the transmit traffic is tagged correctly or else the
firmware will refuse to transmit and will report an ACL violation.
On receive the hardware will make sure that tagged traffic is delivered
to the appropriate VM. The driver only asserts that the VLAN id that
was extracted from the wire traffic matches the VF's configuration.
All this works when associating a specific VLAN id with a VF. The
'trunk' setting likely needs more work.
MFC after: 1 week
Sponsored by: Chelsio Communications
I am unsure if AArch64 Windows systems keep UTC or local time in the RTC
by default, so keep tzsetup consistent across architectures for now.
This reverts commit aa6fb1d277.
Reported by: Mark Millard
Allow iovctl to create VFs that are restricted to specific VLAN IDs.
Reviewed by: kib, np
MFC after: 2 weeks
Sponsored by: Orange Business Services
Differential Revision: https://reviews.freebsd.org/D45402
hw.cxgbe.doorbells_allowed="0xf"
The adapter's doorbells bitmap is clipped to the value specified in the
tunable, which is meant for debug and workarounds only. There is no
change in default behavior.
MFC after: 1 week
Sponsored by: Chelsio Communications
It is pointless to attempt an operation that is not permitted. It spams
the firmware devlog with "insufficient caps" errors that distract from
real errors.
78 2463625358 ERR CORE insufficient caps to process mailbox cmd: pfn 0x0 vfn 0x1; r_caps 0x86 wx_caps 0x82 required r_caps 0x81 w_caps 0x5
MFC after: 1 week
Sponsored by: Chelsio Communications
Use the name as-is but perform cxgbe specific ioctls on the device to
make sure that it is a Terminator device nexus. Determine the chip
type, pf/vf, etc. from the device registers rather than the nexus name.
This allows cxgbetool to work with the VF driver.
MFC after: 1 week
Sponsored by: Chelsio Communications
This is a scheme to avoid taking the bufobj lock and doing a second
lookup in the case where in getblk we do an unlocked lookup and find no
buf. Was there really no buf, or were we in the middle of a reassignbuf
race? By tracking any use of reassignbuf with a flag, we can know if
there can't have been a race because there has been no reassignbuf.
Because this scheme is spoiled on the first use of reassignbuf, it is
mostly only beneficial for cases where a certain vnode is never expected
to use dirty bufs at all.
Reviewed by: kib
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D45571
Note that it is ok to use device_get_desc() as one of the format string
parameters because it is set using device_set_desc() (not
device_set_desc_copy()) and so won't be freed when the description is
updated.
No functional change intended.
MFC after: 1 week
