]> git.ipfire.org Git - thirdparty/freeradius-server.git/log
thirdparty/freeradius-server.git
3 months agoremove unnecessary code
Alan T. DeKok [Sun, 15 Feb 2026 20:26:34 +0000 (15:26 -0500)] 
remove unnecessary code

3 months agouse CMP to avoid manual calculations
Alan T. DeKok [Sun, 15 Feb 2026 19:57:46 +0000 (14:57 -0500)] 
use CMP to avoid manual calculations

3 months agovarious bug fixes for tmpl_tokenize.c
Alan T. DeKok [Sun, 15 Feb 2026 19:52:30 +0000 (14:52 -0500)] 
various bug fixes for tmpl_tokenize.c

3 months agodon't mash string length
Alan T. DeKok [Sun, 15 Feb 2026 19:48:58 +0000 (14:48 -0500)] 
don't mash string length

3 months agoex->trigger_prefix needs to be talloc'd
Alan T. DeKok [Sun, 15 Feb 2026 19:46:55 +0000 (14:46 -0500)] 
ex->trigger_prefix needs to be talloc'd

3 months agodelay free until after we're done using pointer
Alan T. DeKok [Sun, 15 Feb 2026 19:44:19 +0000 (14:44 -0500)] 
delay free until after we're done using pointer

3 months agomove bracket for typo
Alan T. DeKok [Sun, 15 Feb 2026 19:43:27 +0000 (14:43 -0500)] 
move bracket for typo

3 months agovarious bug fixes for virtual_servers.c
Alan T. DeKok [Sun, 15 Feb 2026 19:42:13 +0000 (14:42 -0500)] 
various bug fixes for virtual_servers.c

3 months agopassword_normify() returns NULL on failure, not on OOM
Alan T. DeKok [Sun, 15 Feb 2026 19:34:09 +0000 (14:34 -0500)] 
password_normify() returns NULL on failure, not on OOM

3 months agoinstantiate / free all libraries, even if one is already done
Alan T. DeKok [Sun, 15 Feb 2026 19:32:13 +0000 (14:32 -0500)] 
instantiate  / free all libraries, even if one is already done

3 months agofix typos
Alan T. DeKok [Sun, 15 Feb 2026 19:31:20 +0000 (14:31 -0500)] 
fix typos

3 months agocheck correct variable
Alan T. DeKok [Sun, 15 Feb 2026 19:30:29 +0000 (14:30 -0500)] 
check correct variable

3 months agofix various bugs in module.c
Alan T. DeKok [Sun, 15 Feb 2026 19:29:23 +0000 (14:29 -0500)] 
fix various bugs in module.c

3 months agodon't use strcpy()
Alan T. DeKok [Sun, 15 Feb 2026 19:24:26 +0000 (14:24 -0500)] 
don't use strcpy()

3 months agoescape strings properly, and use heap instead of stack
Alan T. DeKok [Sun, 15 Feb 2026 19:20:00 +0000 (14:20 -0500)] 
escape strings properly, and use heap instead of stack

3 months agoreturn on error
Alan T. DeKok [Sun, 15 Feb 2026 19:07:26 +0000 (14:07 -0500)] 
return on error

3 months agocopy the entire environment
Alan T. DeKok [Sun, 15 Feb 2026 19:07:09 +0000 (14:07 -0500)] 
copy the entire environment

3 months agoactually do mprotect()
Alan T. DeKok [Sun, 15 Feb 2026 19:06:37 +0000 (14:06 -0500)] 
actually do mprotect()

3 months agotweaks to keep analyzer happy
Alan T. DeKok [Sun, 15 Feb 2026 17:26:17 +0000 (12:26 -0500)] 
tweaks to keep analyzer happy

3 months agovarious bug fixes for xlat_eval.c
Alan T. DeKok [Sun, 15 Feb 2026 17:17:30 +0000 (12:17 -0500)] 
various bug fixes for xlat_eval.c

3 months agovarious bug fixes for edit.c
Alan T. DeKok [Sun, 15 Feb 2026 17:16:54 +0000 (12:16 -0500)] 
various bug fixes for edit.c

3 months agotypo, use subcs and not parent cs
Alan T. DeKok [Sun, 15 Feb 2026 16:54:27 +0000 (11:54 -0500)] 
typo, use subcs and not parent cs

3 months agovarious bug fixes in subrequest.c
Alan T. DeKok [Sun, 15 Feb 2026 16:33:06 +0000 (11:33 -0500)] 
various bug fixes in subrequest.c

free child for completeness

if we fail to start the child, then we had pushed a frame in the
parent which is now invalid.  So we mark that frame as unwinding
which does little more than reduce maximum memory pressure, as it
was parented from "request"

3 months agotypos
Alan T. DeKok [Sun, 15 Feb 2026 16:32:12 +0000 (11:32 -0500)] 
typos

3 months agoold "foreach" syntax is an error
Alan T. DeKok [Sun, 15 Feb 2026 14:32:41 +0000 (09:32 -0500)] 
old "foreach" syntax is an error

3 months agovarious bug fixes in xlat_builtin.c
Alan T. DeKok [Sun, 15 Feb 2026 14:05:46 +0000 (09:05 -0500)] 
various bug fixes in xlat_builtin.c

3 months agovarious bug fixes in compile.c
Alan T. DeKok [Sun, 15 Feb 2026 14:05:29 +0000 (09:05 -0500)] 
various bug fixes in compile.c

3 months agopass correct thread data to retry callback
Alan T. DeKok [Sun, 15 Feb 2026 13:52:29 +0000 (08:52 -0500)] 
pass correct thread data to retry callback

3 months agoincrement callers only after success, and update ua on failure
Alan T. DeKok [Sun, 15 Feb 2026 13:50:12 +0000 (08:50 -0500)] 
increment callers only after success, and update ua on failure

3 months agovarious bug fixes in parallel.c
Alan T. DeKok [Sun, 15 Feb 2026 13:48:29 +0000 (08:48 -0500)] 
various bug fixes in parallel.c

cr->request may be NULL if a child was not initialized
i.e. if it's cancelled before it has a chance to run

on error initializing parallel, also clean up the child which caused
the error

3 months agopass correct context to timer callback
Alan T. DeKok [Sun, 15 Feb 2026 13:48:15 +0000 (08:48 -0500)] 
pass correct context to timer callback

3 months agoadd test harness for fifo, hash, inet, retry, and trie
Alan T. DeKok [Sun, 15 Feb 2026 13:25:52 +0000 (08:25 -0500)] 
add test harness for fifo, hash, inet, retry, and trie

3 months agochange struct to quiet compiler warnings
Alan T. DeKok [Sun, 15 Feb 2026 13:30:27 +0000 (08:30 -0500)] 
change struct to quiet compiler warnings

3 months agofix RT calculations
Alan T. DeKok [Sun, 15 Feb 2026 13:02:56 +0000 (08:02 -0500)] 
fix RT calculations

3 months agoset inlen if it's -1 on input
Alan T. DeKok [Sat, 14 Feb 2026 14:26:31 +0000 (09:26 -0500)] 
set inlen if it's -1 on input

3 months agoreorder operations for "update" to match "hash"
Alan T. DeKok [Sat, 14 Feb 2026 14:24:18 +0000 (09:24 -0500)] 
reorder operations for "update" to match "hash"

3 months agoScheduled fuzzing: Update src/tests/fuzzer-corpus/dhcpv6.tar
github-actions[bot] [Sun, 15 Feb 2026 05:10:49 +0000 (05:10 +0000)] 
Scheduled fuzzing: Update src/tests/fuzzer-corpus/dhcpv6.tar

3 months agoScheduled fuzzing: Update src/tests/fuzzer-corpus/radius.tar
github-actions[bot] [Sun, 15 Feb 2026 05:10:43 +0000 (05:10 +0000)] 
Scheduled fuzzing: Update src/tests/fuzzer-corpus/radius.tar

3 months agoScheduled fuzzing: Update src/tests/fuzzer-corpus/bfd.tar
github-actions[bot] [Sun, 15 Feb 2026 05:10:02 +0000 (05:10 +0000)] 
Scheduled fuzzing: Update src/tests/fuzzer-corpus/bfd.tar

3 months agoScheduled fuzzing: Update src/tests/fuzzer-corpus/dns.tar
github-actions[bot] [Sun, 15 Feb 2026 05:09:13 +0000 (05:09 +0000)] 
Scheduled fuzzing: Update src/tests/fuzzer-corpus/dns.tar

3 months agoScheduled fuzzing: Update src/tests/fuzzer-corpus/dhcpv4.tar
github-actions[bot] [Sun, 15 Feb 2026 05:09:08 +0000 (05:09 +0000)] 
Scheduled fuzzing: Update src/tests/fuzzer-corpus/dhcpv4.tar

3 months agovarious bug fixes for dict_tokenize.c
Alan T. DeKok [Sat, 14 Feb 2026 13:58:48 +0000 (08:58 -0500)] 
various bug fixes for dict_tokenize.c

3 months agovarious bug fixes for dict_util.c
Alan T. DeKok [Sat, 14 Feb 2026 13:58:24 +0000 (08:58 -0500)] 
various bug fixes for dict_util.c

3 months agoset fixup flag for aliases, and clear the correct flag
Alan T. DeKok [Sat, 14 Feb 2026 13:57:52 +0000 (08:57 -0500)] 
set fixup flag for aliases, and clear the correct flag

3 months agofix various bugs in sbuff.c
Alan T. DeKok [Sat, 14 Feb 2026 13:44:32 +0000 (08:44 -0500)] 
fix various bugs in sbuff.c

3 months agoupdate code / docs to match expected usage
Alan T. DeKok [Sat, 14 Feb 2026 13:38:56 +0000 (08:38 -0500)] 
update code / docs to match expected usage

3 months agouse sockaddr_storage
Alan T. DeKok [Sat, 14 Feb 2026 13:37:18 +0000 (08:37 -0500)] 
use sockaddr_storage

3 months agocheck for allocation failure
Alan T. DeKok [Sat, 14 Feb 2026 13:36:37 +0000 (08:36 -0500)] 
check for allocation failure

3 months agovarious bug fixes in trie.c
Alan T. DeKok [Sat, 14 Feb 2026 13:33:51 +0000 (08:33 -0500)] 
various bug fixes in trie.c

3 months agodon't close already closed socket
Alan T. DeKok [Sat, 14 Feb 2026 13:19:46 +0000 (08:19 -0500)] 
don't close already closed socket

3 months agoupdate code to match docs, and add assertions
Alan T. DeKok [Sat, 14 Feb 2026 13:17:22 +0000 (08:17 -0500)] 
update code to match docs, and add assertions

3 months agofix various bugs in inet.c
Alan T. DeKok [Sat, 14 Feb 2026 03:14:45 +0000 (22:14 -0500)] 
fix various bugs in inet.c

3 months agodon't double-close sockfd
Alan T. DeKok [Sat, 14 Feb 2026 03:11:46 +0000 (22:11 -0500)] 
don't double-close sockfd

3 months agoupdate doxygen to match code
Alan T. DeKok [Sat, 14 Feb 2026 03:09:56 +0000 (22:09 -0500)] 
update doxygen to match code

3 months agoshrink the heap after rebalancing it, not before.
Alan T. DeKok [Sat, 14 Feb 2026 03:07:01 +0000 (22:07 -0500)] 
shrink the heap after rebalancing it, not before.

shrink to 1/2 size if the heap is less than 1/3 full.  Which
leaves some room to insert new entries before we have to grow the
heap again.

3 months agoavoid overflow by just looking at actual allocated size
Alan T. DeKok [Sat, 14 Feb 2026 02:59:26 +0000 (21:59 -0500)] 
avoid overflow by just looking at actual allocated size

3 months agofix various bugs in event.c
Alan T. DeKok [Sat, 14 Feb 2026 02:53:29 +0000 (21:53 -0500)] 
fix various bugs in event.c

3 months agofix various bugs in calc.c
Alan T. DeKok [Sat, 14 Feb 2026 02:51:31 +0000 (21:51 -0500)] 
fix various bugs in calc.c

3 months agofix return code and don't free ourselves in a destructor
Alan T. DeKok [Sat, 14 Feb 2026 02:33:29 +0000 (21:33 -0500)] 
fix return code and don't free ourselves in a destructor

3 months agofix various bugs in value.c
Alan T. DeKok [Sat, 14 Feb 2026 02:32:37 +0000 (21:32 -0500)] 
fix various bugs in value.c

3 months agofix various bugs in the pair code.
Alan T. DeKok [Sat, 14 Feb 2026 02:10:57 +0000 (21:10 -0500)] 
fix various bugs in the pair code.

3 months agofix various bugs in the cbor code
Alan T. DeKok [Sat, 14 Feb 2026 02:10:40 +0000 (21:10 -0500)] 
fix various bugs in the cbor code

3 months agoMinor typo fix.
Marc-Andre Casavant [Fri, 13 Feb 2026 21:33:33 +0000 (16:33 -0500)] 
Minor typo fix.

3 months agoremove redundant code and do tweaks
Alan T. DeKok [Fri, 13 Feb 2026 19:50:15 +0000 (14:50 -0500)] 
remove redundant code and do tweaks

3 months agoadd SRV record
Alan T. DeKok [Fri, 13 Feb 2026 19:44:46 +0000 (14:44 -0500)] 
add SRV record

3 months agocheck for truncated packets and give an error.
Alan T. DeKok [Fri, 13 Feb 2026 19:33:12 +0000 (14:33 -0500)] 
check for truncated packets and give an error.

Truncation means that the packt is well-formed, but that it's not
containing all of the available data.  We then have to re-try the
query over TCP.

For now, this means we don't process the packet.

3 months agofix typos
Alan T. DeKok [Fri, 13 Feb 2026 19:28:58 +0000 (14:28 -0500)] 
fix typos

3 months agoreturn correct error offset for question section
Alan T. DeKok [Fri, 13 Feb 2026 19:25:43 +0000 (14:25 -0500)] 
return correct error offset for question section

3 months agostop when we reach the end of the input
Alan T. DeKok [Fri, 13 Feb 2026 18:59:14 +0000 (13:59 -0500)] 
stop when we reach the end of the input

3 months agonote todo for RFC 2136
Alan T. DeKok [Fri, 13 Feb 2026 18:56:15 +0000 (13:56 -0500)] 
note todo for RFC 2136

3 months agoupdate test harness function to catch error case
Alan T. DeKok [Fri, 13 Feb 2026 17:57:06 +0000 (12:57 -0500)] 
update test harness function to catch error case

3 months agoa DNS label can be zero length.
Alan T. DeKok [Fri, 13 Feb 2026 17:52:58 +0000 (12:52 -0500)] 
a DNS label can be zero length.

but a zero-length label can't then be in an array.

update the API for the encoding function, as none of the callers
looked for "room needed", and the callers did need to see a zero
return as an allowed, non-error return value.

3 months agoensure that "counter" is updated
Alan T. DeKok [Fri, 13 Feb 2026 17:46:20 +0000 (12:46 -0500)] 
ensure that "counter" is updated

3 months agoit doesn't make sense to encode no data for an array member
Alan T. DeKok [Fri, 13 Feb 2026 17:39:41 +0000 (12:39 -0500)] 
it doesn't make sense to encode no data for an array member

3 months agoUPDATE packets use the count fields for different purposes
Alan T. DeKok [Fri, 13 Feb 2026 17:37:18 +0000 (12:37 -0500)] 
UPDATE packets use the count fields for different purposes

3 months agoswap to use correct error codes
Alan T. DeKok [Fri, 13 Feb 2026 17:29:11 +0000 (12:29 -0500)] 
swap to use correct error codes

3 months agofix typo. MX preference is 16 bit.
Alan T. DeKok [Fri, 13 Feb 2026 17:07:09 +0000 (12:07 -0500)] 
fix typo.  MX preference is 16 bit.

3 months agotypos
Alan T. DeKok [Thu, 12 Feb 2026 16:47:24 +0000 (11:47 -0500)] 
typos

3 months agoMaximum-Msg-Size is uint16, not uint32
Alan T. DeKok [Thu, 12 Feb 2026 16:42:04 +0000 (11:42 -0500)] 
Maximum-Msg-Size is uint16, not uint32

3 months agono need to print this as hex
Alan T. DeKok [Thu, 12 Feb 2026 16:41:55 +0000 (11:41 -0500)] 
no need to print this as hex

3 months agoaccept only ethernet
Alan T. DeKok [Thu, 12 Feb 2026 16:28:38 +0000 (11:28 -0500)] 
accept only ethernet

3 months agoremove unused code
Alan T. DeKok [Thu, 12 Feb 2026 16:27:45 +0000 (11:27 -0500)] 
remove unused code

3 months agoallow up to FR_DHCP_CODE_MAX
Alan T. DeKok [Thu, 12 Feb 2026 16:24:40 +0000 (11:24 -0500)] 
allow up to FR_DHCP_CODE_MAX

3 months agoadd assert that fr_dhcpv4_ok() is called before printing
Alan T. DeKok [Thu, 12 Feb 2026 14:58:13 +0000 (09:58 -0500)] 
add assert that fr_dhcpv4_ok() is called before printing

and check for end of options before printing length

3 months agobe more aggressive about cleaning up on error
Alan T. DeKok [Thu, 12 Feb 2026 14:27:45 +0000 (09:27 -0500)] 
be more aggressive about cleaning up on error

there's already an "error" target which does full cleanups, so
use that.

All of the memory was parented from "ctx", so there were no actual
leaks, but it's good practice to be nice.

3 months agoclean up data on memory allocation failure
Alan T. DeKok [Thu, 12 Feb 2026 14:25:01 +0000 (09:25 -0500)] 
clean up data on memory allocation failure

3 months agoaccount for UDP header size, too
Alan T. DeKok [Thu, 12 Feb 2026 14:20:32 +0000 (09:20 -0500)] 
account for UDP header size, too

3 months agofix option field decoding
Alan T. DeKok [Thu, 12 Feb 2026 14:16:00 +0000 (09:16 -0500)] 
fix option field decoding

use macros instead of hard-coded values.

and update tests to match

3 months agofix bounds check
Alan T. DeKok [Thu, 12 Feb 2026 14:13:27 +0000 (09:13 -0500)] 
fix bounds check

3 months agofix typo
Alan T. DeKok [Thu, 12 Feb 2026 14:12:56 +0000 (09:12 -0500)] 
fix typo

3 months agoRe-enable message_set_test
Nick Porter [Thu, 12 Feb 2026 14:45:45 +0000 (14:45 +0000)] 
Re-enable message_set_test

3 months agoAdd test of messages_set where initial ring buffer is smaller than request size
Nick Porter [Thu, 12 Feb 2026 14:44:28 +0000 (14:44 +0000)] 
Add test of messages_set where initial ring buffer is smaller than request size

3 months agoUse fr_exit_now in test in place of fr_assert
Nick Porter [Thu, 12 Feb 2026 14:43:47 +0000 (14:43 +0000)] 
Use fr_exit_now in test in place of fr_assert

So failures give meaningful output.

3 months agoMake message_set_test compile again
Nick Porter [Thu, 12 Feb 2026 14:36:58 +0000 (14:36 +0000)] 
Make message_set_test compile again

3 months agoAllow message set to alloc messages greater than half initial ring buffer
Nick Porter [Fri, 30 Jan 2026 09:33:31 +0000 (09:33 +0000)] 
Allow message set to alloc messages greater than half initial ring buffer

This is optional so the message size protection can be retained for
exsiting network -> worker use

3 months agoTypo
Nick Porter [Fri, 30 Jan 2026 12:09:34 +0000 (12:09 +0000)] 
Typo

3 months agotests for ipv4prefix
Alan T. DeKok [Thu, 12 Feb 2026 02:30:05 +0000 (21:30 -0500)] 
tests for ipv4prefix

3 months agomask off ipv4 prefix bits, to be in compliance with RFC8044 Section 3.11
Alan T. DeKok [Thu, 12 Feb 2026 01:53:43 +0000 (20:53 -0500)] 
mask off ipv4 prefix bits, to be in compliance with RFC8044 Section 3.11

this shouldn't strictly be necessary, as the value-boxes should
already enforce this.  But layers of protection aren't bad.

3 months agoadd secret length to ascend_secret() function
Alan T. DeKok [Thu, 12 Feb 2026 01:44:13 +0000 (20:44 -0500)] 
add secret length to ascend_secret() function

all of the other password encode / decode functions already took
a secret length parameter

3 months agocorrect subtraction of pointers
Alan T. DeKok [Thu, 12 Feb 2026 01:39:45 +0000 (20:39 -0500)] 
correct subtraction of pointers