Newer versions of `shellcheck` and `checkbashism` finds more than
previous, so fix those.
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Closes#18000
The `type` command is an optional feature in POSIX, so shouldn't be
used.
Instead, use `command -v`, which commit
e865e7809e
did, but it missed this file.
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Closes#18000
There's no real documenation (which should probably be written!),
so instead document the code the best we can on what's going and
with the mounting of file systems to make future updates easier.
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Closes#18000
More code standard changes, where if/then is on different lines.
To have it on the same, or on different lines, can be argued, but
we need to pick one, and try not to mix how to do things.
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Closes#18000
The `ZFS_INITRD_ADDITIONAL_DATASETS` variable is used in the initrd
script to boot additional OS file systems besides the root file system.
But it wasn't included as an example in the config files.
The `ZFS_POOL_EXCEPTIONS` *was* included in the example defaults file,
but it was not exported, so not available in the initrd.
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Closes#18000
The file `/etc/default/zfs` is already sourced by the `/etc/zfs/zfs-functions`,
so no need to source it again.
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Closes#18000
When a pool is degraded, or needs special action, the `zpool import`
(without pool to import) line will report:
```
pool: rpool
id: 01234567890123456789
state: ONLINE
action: The pool can be imported using its name or numeric identifier.
config:
[..]
```
If the import with the pool name fails, it is supposed to try importing
using the pool ID.
However, the script is also getting the `action` line (and probably `scrub:`
if/when that's available):
pool; The pool can be imported using its name or numeric identifier.;config:;
which causes issues on consequent import attempts.
Cleanup the information by rewriting the `sed` command line.
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Closes#18000
It's considered good practice to:
1) Wrap the variable name in `{}`.
As in `${variable}` instead of `$variable`.
2) Put variables in `"`.
Also some minor error message tuning.
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Closes#18000
In a previous commit (e865e7809e), the
`local` keyword was removed in functions because of bashism.
Removing bashisms is correct, however this could cause variable overwrites,
since several functions use the same variable name.
So this commit make function variables unique in the (now) global name
space.
The problem from the original bug report (see #17963) could not be duplicated,
but it is still sane to make sure that variables stay unique.
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Closes#18000
This is now consistently passing with 100+ consecutive runs.
Signed-off-by: Siva Mahadevan <me@svmhdvn.name>
PR: 244170
MFC after: 3 days
Sponsored by: The FreeBSD Foundation
When a module the environment must be explicitly fetched.
Fixes: d9788eabff
PR: 291548
Noted by: markj
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D54242
MFC after: 3 days
When copying ipfs data from user space, don't just check that the payload
length is nonzero, but also that it does not exceed the size of the stack
buffer we're copying it into.
While we're at it, use a union to create a buffer of the exact size we
need instead of guessing that 2048 will be enough (and not too much).
Finally, check the size of the payload once it gets to where it's used.
MFC after: 3 days
Reported by: Ilja Van Sprundel <ivansprundel@ioactive.com>
Reviewed by: cy
Differential Revision: https://reviews.freebsd.org/D54194
Copy the text change from the shell makeman to makeman.lua.
Fixes: dd8c666d8b ("src.sys.mk: Support src.conf in SRCTOP")
Sponsored by: The FreeBSD Foundation
Add PNP info so the module can be matched by devmatch(8) and automatically
loaded.
Reviewed by: adrian
Approved by: adrian
Differential Revision: https://reviews.freebsd.org/D54254
This testcase passes consistently (in 100+ runs) now.
Signed-off-by: Siva Mahadevan <me@svmhdvn.name>
PR: 244163, 251726
MFC after: 3 days
Sponsored by: The FreeBSD Foundation
This test now consistently passes (300+ consecutive runs).
Signed-off-by: Siva Mahadevan <me@svmhdvn.name>
PR: 244172
Sponsored by: The FreeBSD Foundation
When building with WITHOUT_TESTS this result in a FreeBSD-atf-dev
package with only this directory and a dependency on FreeBSD-atf which
doesn't exists.
Reviewed by: ivy
Differential Revision: https://reviews.freebsd.org/D54236
Fixes: 436618a427 ("etc/mtree: Add package tags for /usr/include")
Sponsored by: Beckhoff Automation GMbH & Co. KG
When building with WITHOUT_BSNMP this result in a FreeBSD-bsnmp-dev
package with only this directory and a dependency on FreeBSD-bsnmp which
doesn't exists.
Reviewed by: ivy
Differential Revision: https://reviews.freebsd.org/D54235
Fixes: 436618a427 ("etc/mtree: Add package tags for /usr/include")
Sponsored by: Beckhoff Automation GMbH & Co. KG
Upstream fixed a couple of bugs:
1. Only attempt to restore the blocking rules if the database file
exists. Otherwise, when the service starts for the first time, it
fails (PR 258411).
2. Revert a commit that removed a call to close(bi->bi_fd), preventing
the descriptor from being deleted.
PR: 258411
PR: 291680
MFC after: 1 week
Drop not needed cast.
Group sigchld state check as single KASSERT condition.
Remove useless comment.
Reviewed by: des, olce
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D54234
While here, replace loop copying the MAC address with memcpy() for
better readability.
Reviewed by: markj
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D54177
Add vendor import instructions for blocklist.
It includes a "freebsd-changes.sh" script that takes care of adapting
paths and functions into FreeBSD.
Reviewed by: emaste
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D49510
So it isn't matched by devmatch(8) and automatically loaded. The PNP
info will be readded once the USB4 driver is more complete.
PR: 290827
Reported by: fuz, Marco Siedentopf <siedentm@me.com>
Fixes: 2ed9833791 (thunderbolt: Import USB4 code)
Sponsored by: The FreeBSD Foundation
The 'Setup QEMU' CI step updates and installs all packages necessary to
startup QEMU. Typically the step takes a little over a minute, but
we've seen cases where it can take legitimately take more than 45min
minutes. Change the timeout to 60 minutes.
In addition, change the 'Install dependencies' timeout to 60min since
we've also seen timeouts there.
Lastly, remove all timeouts from the zfs-qemu-packages workflow.
We do this so that we can always build packages from a branch, even if
the time it takes to do a CI step changes over time. It's ok to
eliminate the timeouts from the zfs-qemu-packages completely since that
workflow is only run manually.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tony Hutter <hutter2@llnl.gov>
Closes#18056
The first byte of the entry after compression is used for algorithm
and byte order flag. We should decrement when calling compression/
decompression algorithm.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <alexander.motin@TrueNAS.com>
Closes#18055
Unlike other ZAP consumers due to compression DDT does not know
how big entry it is reading from ZAP. Due to this it called
zap_length_uint64_by_dnode() and zap_lookup_uint64_by_dnode(),
each of which does full ZAP entry lookup.
Introduction of the combined ZAP method dramatically reduces the
CPU overhead and locks contention at DBUF layer.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <alexander.motin@TrueNAS.com>
Closes#18048
Dynamically allocate bpf tap points for every rule that has "log".
The name is "ipfw%u", where %u is substituted to the rule number.
The default catch all "ipfw0" tap still exists for compatibility
and it will catch packets in case if there are no bpf listeners
on a per-rule tap.
Reviewed by: ae
Differential Revision: https://reviews.freebsd.org/D53877
This should provide people a chance to remove ipfw0 and ipfwlog0 from
cloned_interfaces in their rc.conf during FreeBSD 16.x lifetime.
Differential Revision: https://reviews.freebsd.org/D53876
Imagine that bpf(9) tapping can happen at any point in the network stack,
not necessarily at interface transmit or receive. To achieve that we need
a thin layer of abstraction defined by struct bif_methods, that defines
how generic bpf layer works with a tap point of this kind.
Implement ifnet(9) specific methods in a separate file bpf_ifnet.c. At
this point there is 100% compatibility for all existing interfaces, there
is no KPI change, yet. The legacy attaching KPI is layered over new ifnet
agnostic KPI. The new KPI may change though, as we can implement multiple
DLTs per single tap point in a prettier fashion.
The new abstraction layer allows us to move all the 802.11 radio injection
hacks out of bpf.c into ieee80211_radiotap.c, so do that immediately as a
good proof of concept.
Reviewed by: bz
Differential Revision: https://reviews.freebsd.org/D53872
As was previously done for BRT, avoid holding/releasing DDT ZAP
dnodes for every access. Instead hold the dnodes during all their
life time, never releasing.
While at this, add _by_dnode() interfaces for zap_length_uint64()
and zap_count(), actively used by DDT code.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <alexander.motin@TrueNAS.com>
Closes#18047
These two files are broken due to Linux 6.5 DTS import.
Both of these boards have support in Linux DTS tree,
please use these DTS instead.
Removed and not fixed because of commit 949efdaa1d
Approved by: br, manu (mentor)
Differential revision: https://reviews.freebsd.org/D54216
Gleb Smirnoff