summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2020-03-11Make sure hdrlen is initialized.tobhe
ok stsp@
2020-03-11The NXP-provided device trees used to make use of the assigned-clocksPatrick Wildt
properties to make sure that the PCIe clocks are configured correctly. While upstreaming these into Linux mainline, those have been removed. Instead the clock driver has the responsibility of setting these up on attach. For us we'll just hardcode the parent and enable the setting when we turn on the PCIe clocks. ok kettenis@
2020-03-11Power domains on i.MX8MQ can reference other power domains, whichPatrick Wildt
need to be enabled first. Notable users are the PCIe domains. ok kettenis@
2020-03-11Allow alpha boot(8) to read from an ffs2 filesystem and adapt itsOtto Moerbeek
custom installboot to deal with ffs2. While there, fix the partition read code to deal with offsets > 2G; ok deraadt@
2020-03-11typo; from bryan stensonJason McIntyre
2020-03-11Take a swing at blocking Load-Value-Injection attacks against thePhilip Guenther
kernel by using lfence in place of stac/clac on pre-SMAP CPUs. To quote from https://software.intel.com/security-software-guidance/insights/deep-dive-load-value-injection "If the OS makes use of Supervisor Mode Access Prevention (SMAP) on processors with SMAP enabled, then LVI on kernel load from user pages will be mitigated. This is because the CLAC and STAC instructions have LFENCE semantics on processors affected by LVI, and this serves as a speculation fence around kernel loads from user pages." ok deraadt@
2020-03-11properly limit indexing into the aggr_periodic_times array.David Gwynne
coverity CID 1486819 pointed out by and ok tobhe@
2020-03-11Make traphandler_parse also set vbinds for SNMPv1 traps.Martijn van Duren
This also prevents the access of an initialized pointer in traphandler_fork_handler as found by jan@. OK jan@
2020-03-10Variables 'nowait' amd 'flags' should have died in 2008 with r1.36'sKenneth R Westerback
removal of internal queuing of scsi_xfer's. Put them out of their misery. Nuke an obsolete comment in passing. CID 1453357.
2020-03-10The return value of rt_ifa_purge() is ignored, so stopKenneth R Westerback
returning a (possibly uninitialized) value. CID 1483466. ok millert@
2020-03-10regenanton
2020-03-10Unlock fcntl(2).anton
ok visa@
2020-03-10Relookup policy based on received cryptographic parameter proposal.tobhe
The IKEv2 responder does not know which policy is negotiated until the ID payload is received in the IKE_AUTH exchange. iked therefore chooses a default policy until the final policy is selected. This change adds a policy relookup during the IKE_SA_INIT. If the received proposal is not compatible with the default policy we switch to the next highest ranked policy that is compatible with the received proposal. ok kn@ markus@
2020-03-10Use ctx->hs->secrets rather than the S3I(s) version.Joel Sing
ok inoguchi@ tb@
2020-03-10Update to follow handshake enum removal.Joel Sing
2020-03-10Remove some unnecessary handshake enums/functions.Joel Sing
Both session tickets and key updates are post-handshake handshake messages, which were originally included in the handshake code. ok inoguchi@ tb@
2020-03-10Add a return value check to tls13_buffer_extend().Joel Sing
In the unlikely event that the return value from the read callback is larger than the number of bytes we asked for, we can end up incrementing buf->len beyond capacity. Check the return value from the read callback to prevent this. ok inoguchi@ tb@
2020-03-10Remove the enc function pointers.Joel Sing
The enc function pointers do not serve any purpose these days - remove a layer of indirection and call dtls1_enc()/tls1_enc() directly. ok inoguchi@ tb@
2020-03-10Coverity points out (12 + MAXMPS) can never be less thanKenneth R Westerback
12. PAGE_SHIFT is 12 (a.k.a. PAGE_SIZE == 4096) or more on OpenBSD architectures. So remove some dead code by unconditionally setting sc_mps (memory page size) to 1 << PAGE_SHIFT. CID 1491655. No functional change.
2020-03-10Narrow the visibility of some functions and variables local to output.cJeremie Courreges-Anglas
Also make the outputs table const. Based on a suggestion from claudio@, ok deraadt@ claudio@
2020-03-10Ooops. Need both files.Kenneth R Westerback
sgl (scatter gather list) and prpl (physical region page list) are two different things in the NVMe world. Only the latter is currently implemented in nvme(4) so rename sc_max_sgl to sc_max_prpl to reflect this. No functional change.
2020-03-10sgl (scatter gather list) and prpl (physical region page list) are twoKenneth R Westerback
different things in the NVMe world. Only the latter is currently implemented in nvme(4) so rename sc_max_sgl to sc_max_prpl to reflect this. No functional change.
2020-03-10Import openssl-1.1.1d test data to base64test.cKinichiro Inoguchi
ok bcook@ tb@
2020-03-10Modify regress base64test.cKinichiro Inoguchi
- Don't remove multi line CR/LF from bt->out when NL mode base64_encoding_test removes CR/LF from bt->out to compare with the encoding result. This is fine with NO NL mode, but it goes wrong with NL mode if encoding result is larger than 64 and multi line, like below. "eHh4eHh4eHh4eHh4eHh4eHh4eHh4eHh4eHh4eHh4eHh4eHh4eHh4eHh4eHh4eHh4\neHh4eHh4eHh4eHh4\n" - Use memcpy instead of asprintf to avoid lost '\0' at the end of data This test data loses trailing '\0' if using asprintf. "\x61\x47\x56\x73\x62\x47\x38\x3d\x0a\x00" - Print original data if decoding result comparison fails This change is not for importing test data, but I just notice. It prints bt->out if fail to memcmp bt->in with decoding result. ok bcook@ tb@
2020-03-10syncStuart Henderson
2020-03-10add 6.8 fw pubkeyStuart Henderson
2020-03-10Make sure 'pooladdr' is zero initialized.tobhe
ok markus@
2020-03-10Fix memory leak of 'cr' if allocation of 'cr->data' fails.tobhe
ok markus@
2020-03-10Make sure ikev2_next_payload() is only called when there is a previoustobhe
payload. Not doing so may result in a nullptr dereference. ok gerhard@
2020-03-10Return when SA lookup fails in ikev2_init_recv().tobhe
ok patrick@ markus@
2020-03-10Delete dead code in ikev2_msg_valid_ike_sa().tobhe
ok markus@
2020-03-10Make sure 'e' is NULL initialized to prevent nullptr dereference intobhe
ibuf_release(). ok gerhard@
2020-03-10Properly exit loop at end of hooks TAILQ.tobhe
Feedback from and ok dlg@ ok kn@ todd@
2020-03-10Make sure return value 'error' is initialized to '0'.tobhe
ok dlg@ deraadt@
2020-03-10Remove unused argument to sr_error.tobhe
ok kn@ deraadt@
2020-03-10Access 4-byte address field with _4btol() not _8btol(), fixingKenneth R Westerback
detecton of a 0 value. CID 1488899
2020-03-10syncKevin Lo
2020-03-10Match on Lenovo OneLink Plus Dock Ethernet.Kevin Lo
Tested by David Demelier
2020-03-09Ensure that we properly flush, close and rename temporary output filesJeremie Courreges-Anglas
Makes sure we don't feed an incomplete/garbage file to consumers. Input and ok claudio@ deraadt@
2020-03-09logx->warn, we want to know why operations on output files failedJeremie Courreges-Anglas
ok claudio@ deraadt@
2020-03-09Return EINVAL for KERN_PROC if the size parameter is 0.Todd C. Miller
Prevents a panic due to a NULL dereference; Coverity CID 1452899. Based on a diff from mpi@, OK deraadt@ kettenis@
2020-03-09Avoid a tight CPU loop when no unlocked worklist items can be processed.Todd C. Miller
If process_worklist_item() is unable to process locked vnodes, num_on_worklist will still be non-zero, preventing the loop in softdep_process_worklist() from exiting. This can result in a kernel hang. To fix this, process_worklist_item() now returns non-zero if it was able to process a worklist item (regardless of whether it matched the mountpoint) and takes a pointer to matchcnt as a function argument. We now break out of the loop in softdep_process_worklist() if process_worklist_item() is unable to make progress. OK beck@ bluhm@
2020-03-09Use TAILQ_FOREACH_SAFE instead of hand rolled loops.tobhe
From Wataru <wataash at wataash dot com> ok patrick@
2020-03-09cstyle in illumos noticed some weird syntax, which this fixes.David Gwynne
some errx lines in if statements were terminated with commas, which caused the following statement to be considered part of the error handling. while it is bad style, ingo points out it was also a bug which which caused some tests in the code to be skipped. this reminds me of a haiku that Chris Pascoe (cpascoe@) had behind his desk: Also, that comma Should be a semi-colon. Cherry blossoms fall. this was found by Robert Mustacchi when porting the tests to illumos. ok schwarze@ stsp@ thank you robert.
2020-03-09Recommit, now that we found out how to fix the BIOS related issues: goOtto Moerbeek
back to a 4 byte add instruction. We do not know *why* though, so if somebody likes a challenge... Lots of help from semarie@ who has a few systems showing the issue. ok deraadt@
2020-03-09Return error value when sending "sensor reading" is failed. This fixesYASUOKA Masahiko
"ipmi0: sendcmd fails" errors when there is a sensor which is enumurated but reading it is failed. ok mpi
2020-03-09Update the console device always when attaching the real deviceYASUOKA Masahiko
driver. The information by the driver is supposed more reliable than the information which was set up earlier. ok patrick
2020-03-08Rewrite qp_encoded_write.Martijn van Duren
Among the fixes are: - Don't count a leading dot, since it's part of smtp and not the message. - Let qp_encoded_write handle line wrappings, since it can do a better job at the bookkeeping than the caller - Make sure a soft linebreak that starts with a dot is escaped for the SMTP-layer - Only encode a space character if it's followed by a hard linebreak, since this is the only case mandated by RFC5321 and it gives a more readable output and saves a few additional bytes. - Fix that we actually limit the encoded linelength to 76 characters as specified in RFC5321. Feedback and OK eric@
2020-03-08Rework ucom(4) a bit. The logic of RTS follows DTR should only be doneClaudio Jeker
when the port is initialized (with a tcsetattr() call) but not on every changed of DTR (especially those caused by other tty ioctls). Fixes firmware upload on various microcontroller boards that use DTR and RTS as signaling lines to reset the device and enter the bootloader. Tested by tracey@, OK deraadt@
2020-03-08fs_name[] must match file_system[], adapt for ufs2 support.Landry Breuil
fixes pxeboot - committing on behalf of otto@ who cant right now.
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-01 11:15:47 +0000