Commit Graph
1243 Commits
Author SHA1 Message Date
Michal Nowak 531c96b8ed Update the source code formatting using clang-format-17 2023-10-17 17:56:31 +02:00
Michal Nowak 771fed4a14 Update sources to Clang 15 formatting 2022-11-29 10:30:34 +01:00
Michal Nowak a584a8f88f Update clang to version 14
(cherry picked from commit 1c45a9885a)
2022-06-16 18:11:03 +02:00
Mark Andrews 2a9ab8a732 Don't try to set IPV6_V6ONLY on OpenBSD
OpenBSD IPv6 sockets are always IPv6-only, so the socket option is read-only (not modifiable)
2022-05-02 14:09:31 +10:00
Ondřej Surý 79b7804ce8 Consistenly use UNREACHABLE() instead of ISC_UNREACHABLE()
In couple places, we have missed INSIST(0) or ISC_UNREACHABLE()
replacement on some branches with UNREACHABLE().  Replace all
ISC_UNREACHABLE() or INSIST(0) calls with UNREACHABLE().
2022-03-28 23:28:05 +02:00
Ondřej SurýandOndřej Surý b624be2544 Remove use of the inline keyword used as suggestion to compiler
Historically, the inline keyword was a strong suggestion to the compiler
that it should inline the function marked inline.  As compilers became
better at optimising, this functionality has receded, and using inline
as a suggestion to inline a function is obsolete.  The compiler will
happily ignore it and inline something else entirely if it finds that's
a better optimisation.

Therefore, remove all the occurences of the inline keyword with static
functions inside single compilation unit and leave the decision whether
to inline a function or not entirely on the compiler

NOTE: We keep the usage the inline keyword when the purpose is to change
the linkage behaviour.

(cherry picked from commit 20f0936cf2)
2022-03-25 09:37:18 +01:00
Ondřej SurýandOndřej Surý 75f9dd8e82 Simplify way we tag unreachable code with only ISC_UNREACHABLE()
Previously, the unreachable code paths would have to be tagged with:

    INSIST(0);
    ISC_UNREACHABLE();

There was also older parts of the code that used comment annotation:

    /* NOTREACHED */

Unify the handling of unreachable code paths to just use:

    UNREACHABLE();

The UNREACHABLE() macro now asserts when reached and also uses
__builtin_unreachable(); when such builtin is available in the compiler.

(cherry picked from commit 584f0d7a7e)
2022-03-25 09:33:51 +01:00
Ondřej SurýandOndřej Surý 673e53f81d Add FALLTHROUGH macro for __attribute__((fallthrough))
Gcc 7+ and Clang 10+ have implemented __attribute__((fallthrough)) which
is explicit version of the /* FALLTHROUGH */ comment we are currently
using.

Add and apply FALLTHROUGH macro that uses the attribute if available,
but does nothing on older compilers.

In one case (lib/dns/zone.c), using the macro revealed that we were
using the /* FALLTHROUGH */ comment in wrong place, remove that comment.

(cherry picked from commit fe7ce629f4)
2022-03-25 09:30:16 +01:00
Ondřej Surý 25732d818d Remove usage of deprecated ATOMIC_VAR_INIT() macro
The C17 standard deprecated ATOMIC_VAR_INIT() macro (see [1]).  Follow
the suite and remove the ATOMIC_VAR_INIT() usage in favor of simple
assignment of the value as this is what all supported stdatomic.h
implementations do anyway:

  * MacOSX.plaform: #define ATOMIC_VAR_INIT(__v) {__v}
  * Gcc stdatomic.h: #define ATOMIC_VAR_INIT(VALUE)	(VALUE)

1. http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2018/p1138r0.pdf

(cherry picked from commit f251d69eba)
2022-03-17 21:44:04 +01:00
Ondřej Surý be47b2e5e4 Add couple missing braces around single-line statements
The clang-format-15 has new option InsertBraces that could add missing
branches around single line statements.  Use that to our advantage
without switching to not-yet-released LLVM version to add missing braces
in couple of places.
2022-03-17 18:31:36 +01:00
Ondřej Surý 2bf7921c7e Update the copyright information in all files in the repository
This commit converts the license handling to adhere to the REUSE
specification.  It specifically:

1. Adds used licnses to LICENSES/ directory

2. Add "isc" template for adding the copyright boilerplate

3. Changes all source files to include copyright and SPDX license
   header, this includes all the C sources, documentation, zone files,
   configuration files.  There are notes in the doc/dev/copyrights file
   on how to add correct headers to the new files.

4. Handle the rest that can't be modified via .reuse/dep5 file.  The
   binary (or otherwise unmodifiable) files could have license places
   next to them in <foo>.license file, but this would lead to cluttered
   repository and most of the files handled in the .reuse/dep5 file are
   system test files.

(cherry picked from commit 58bd26b6cf)
2022-01-11 12:22:09 +01:00
Ondřej Surý 3d4ec2f244 Add isc_time_add and isc_time_subtract unit test
The isc_time_add() and isc_time_subtract() didn't have a unit test, add
the unit test with couple of edge case vectors to check whether overflow
and underflow is correctly handled.

(cherry picked from commit 04511736a0)
2021-10-21 09:51:53 +02:00
Ondřej Surý b58611a08b Fix isc_time_add() overflow
The isc_time_add() could overflow when t.seconds + i.seconds == UINT_MAX
and t.nanoseconds + i.nanoseconds >= NS_PER_S.

Fix the overflow in isc_time_add(), and simplify the ISC_R_RANGE checks
both in isc_time_add() and isc_time_subtract() functions.

(cherry picked from commit 8c05f12bc8)
2021-10-21 09:51:48 +02:00
Ondřej Surý 093cd31ae2 Update the source code formatting using clang-format-13
clang-format-13 fixed some of the formatting that clang-format-12 got
wrong.  Update the formatting.

(cherry picked from commit ed95f9fba3)
2021-10-12 11:31:55 +02:00
Aram Sargsyan 18514a43d5 Remove dead code
Remove dead code from the USE_DEVPOLL branch in libisc's unix/socket.c
2021-09-08 12:52:33 +00:00
Ondřej SurýandMichał Kępień 607f8d114e Disable the Path MTU Discover on UDP Sockets
Instead of disabling the fragmentation on the UDP sockets, we now
disable the Path MTU Discovery by setting IP(V6)_MTU_DISCOVER socket
option to IP_PMTUDISC_OMIT on Linux and disabling IP(V6)_DONTFRAG socket
option on FreeBSD.  This option sets DF=0 in the IP header and also
ignores the Path MTU Discovery.

As additional mitigation on Linux, we recommend setting
net.ipv4.ip_no_pmtu_disc to Mode 3:

    Mode 3 is a hardend pmtu discover mode. The kernel will only accept
    fragmentation-needed errors if the underlying protocol can verify
    them besides a plain socket lookup. Current protocols for which pmtu
    events will be honored are TCP, SCTP and DCCP as they verify
    e.g. the sequence number or the association. This mode should not be
    enabled globally but is only intended to secure e.g. name servers in
    namespaces where TCP path mtu must still work but path MTU
    information of other protocols should be discarded. If enabled
    globally this mode could break other protocols.
2021-08-19 07:20:15 +02:00
Ondřej SurýandOndřej Surý c546545d32 Disable setting the thread affinity
It was discovered that setting the thread affinity on both the netmgr
and netthread threads lead to inconsistent recursive performance because
sometimes the netmgr and netthread threads would compete over single
resource and sometimes not.

Removing setting the affinity causes a slight dip in the authoritative
performance around 5% (the measured range was from 3.8% to 7.8%), but
the recursive performance is now consistently good.

(cherry picked from commit a9e6a7ae57)
2021-07-13 15:47:13 +02:00
Ondřej Surý d115a9ae2a Disable the PMTUD also on the old socket UDP code
Instead of just disabling the PMTUD mechanism on the UDP sockets, we
now set IP_DONTFRAG (IPV6_DONTFRAG) flag.  That means that the UDP
packets won't get ever fragmented.  If the ICMP packets are lost the
UDP will just timeout and eventually be retried over TCP.
2021-06-23 21:06:05 +02:00
Ondřej Surý b0e7511001 Update the source code formatting using clang-format-12
clang-format now tries to keep the type-cast on the same line as the
variable.  Update the formatting.
2021-06-13 08:19:44 +02:00
Matthijs Mekking 72deed194d Use isdigit instead of checking character range
When looking for key files, we could use isdigit rather than checking
if the character is within the range [0-9].

Use (unsigned char) cast to ensure the value is representable in the
unsigned char type (as suggested by the isdigit manpage).

Change " & 0xff" occurrences to the recommended (unsigned char) type
cast.

(cherry picked from commit 1998ad6c776a9c17c27788b17765dee90d9e25df)
2021-05-05 18:23:53 +02:00
Ondřej SurýandOndřej Surý 97a5559ae3 Cleanup the isc_<*>mgr_createinc() constructors
Previously, the taskmgr, timermgr and socketmgr had a constructor
variant, that would create the mgr on top of existing appctx.  This was
no longer true and isc_<*>mgr was just calling isc_<*>mgr_create()
directly without any extra code.

This commit just cleans up the extra function.

(cherry picked from commit 3388ef36b3)
2021-04-19 15:57:40 +02:00
Ondřej SurýandOndřej Surý 08055b742c Cleanup the public vs private ISCAPI remnants
Since all the libraries are internal now, just cleanup the ISCAPI remnants
in isc_socket, isc_task and isc_timer APIs.  This means, there's one less
layer as following changes have been done:

 * struct isc_socket and struct isc_socketmgr have been removed
 * struct isc__socket and struct isc__socketmgr have been renamed
   to struct isc_socket and struct isc_socketmgr
 * struct isc_task and struct isc_taskmgr have been removed
 * struct isc__task and struct isc__taskmgr have been renamed
   to struct isc_task and struct isc_taskmgr
 * struct isc_timer and struct isc_timermgr have been removed
 * struct isc__timer and struct isc__timermgr have been renamed
   to struct isc_timer and struct isc_timermgr
 * All the associated code that dealt with typing isc_<foo>
   to isc__<foo> and back has been removed.

(cherry picked from commit 16fe0d1f41)
2021-04-19 15:24:10 +02:00
Patrick McLeanandEvan Hunt c5c9c9b83f Add isc_time_now_hires function to get current time with high resolution
The current isc_time_now uses CLOCK_REALTIME_COARSE which only updates
on a timer tick. This clock is generally fine for millisecond accuracy,
but on servers with 100hz clocks, this clock is nowhere near accurate
enough for microsecond accuracy.

This commit adds a new isc_time_now_hires function that uses
CLOCK_REALTIME, which gives the current time, though it is somewhat
expensive to call. When microsecond accuracy is required, it may be
required to use extra resources for higher accuracy.

(cherry picked from commit ebced74b19)
2021-03-20 11:59:21 -07:00
Ondřej Surý a35a666a7c Reformat sources using clang-format-11
(cherry picked from commit 7ba18870dc)
2020-12-08 19:34:05 +01:00
Ondřej Surý dbf2d0b15f Clean the last remnant of ISC_PLATFORM_HAVEIPV6 macro
In set_sndbuf() we were using ISC_PLATFORM_HAVEIPV6 macro that doesn't
exist anymore, because we assume that IPv6 support is always available.

(cherry picked from commit 96ac91a18a)
2020-10-08 09:03:08 +02:00
Mark Andrews ebf16118df Add ISO time stamps to the microsecond
(cherry picked from commit 519b070618)
2020-10-01 00:14:45 +10:00
Mark Andrews c4edcaf140 It appears that you can't change what you are polling for while connecting.
WARNING: ThreadSanitizer: data race
    Read of size 8 at 0x000000000001 by thread T1 (mutexes: write M1):
    #0 epoll_ctl <null>
    #1 watch_fd lib/isc/unix/socket.c:704:8
    #2 wakeup_socket lib/isc/unix/socket.c:897:11
    #3 process_ctlfd lib/isc/unix/socket.c:3362:3
    #4 process_fds lib/isc/unix/socket.c:3275:10
    #5 netthread lib/isc/unix/socket.c:3516:10

    Previous write of size 8 at 0x000000000001 by thread T2 (mutexes: write M2):
    #0 connect <null>
    #1 isc_socket_connect lib/isc/unix/socket.c:4737:7
    #2 resquery_send lib/dns/resolver.c:2892:13
    #3 fctx_query lib/dns/resolver.c:2202:12
    #4 fctx_try lib/dns/resolver.c:4300:11
    #5 resquery_connected lib/dns/resolver.c:3130:4
    #6 dispatch lib/isc/task.c:1152:7
    #7 run lib/isc/task.c:1344:2

    Location is file descriptor 513 created by thread T2 at:
    #0 connect <null>
    #1 isc_socket_connect lib/isc/unix/socket.c:4737:7
    #2 resquery_send lib/dns/resolver.c:2892:13
    #3 fctx_query lib/dns/resolver.c:2202:12
    #4 fctx_try lib/dns/resolver.c:4300:11
    #5 resquery_connected lib/dns/resolver.c:3130:4
    #6 dispatch lib/isc/task.c:1152:7
    #7 run lib/isc/task.c:1344:2

    Mutex M1 (0x000000000016) created at:
    #0 pthread_mutex_init <null>
    #1 isc__mutex_init lib/isc/pthreads/mutex.c:288:8
    #2 setup_thread lib/isc/unix/socket.c:3584:3
    #3 isc_socketmgr_create2 lib/isc/unix/socket.c:3825:3
    #4 create_managers bin/named/main.c:932:11
    #5 setup bin/named/main.c:1223:11
    #6 main bin/named/main.c:1523:2

    Mutex M2 is already destroyed.

    Thread T1 'isc-socket-1' (running) created by main thread at:
    #0 pthread_create <null>
    #1 isc_thread_create lib/isc/pthreads/thread.c:73:8
    #2 isc_socketmgr_create2 lib/isc/unix/socket.c:3826:3
    #3 create_managers bin/named/main.c:932:11
    #4 setup bin/named/main.c:1223:11
    #5 main bin/named/main.c:1523:2

    Thread T2 (running) created by main thread at:
    #0 pthread_create <null>
    #1 isc_thread_create lib/isc/pthreads/thread.c:73:8
    #2 isc_taskmgr_create lib/isc/task.c:1434:3
    #3 create_managers bin/named/main.c:915:11
    #4 setup bin/named/main.c:1223:11
    #5 main bin/named/main.c:1523:2

    SUMMARY: ThreadSanitizer: data race in epoll_ctl

(cherry picked from commit c37b251eb9)
2020-09-23 14:22:15 +10:00
Ondřej Surý 0ad77036d9 Handle the errors from sysconf() call in isc_meminfo_totalphys()
isc_meminfo_totalphys() would return invalid memory size when sysconf()
call would fail, because ((size_t)-1 * -1) is very large number.

(cherry picked from commit 79ca724d46)
2020-09-21 10:58:37 +02:00
Evan Hunt df698d73f4 update all copyright headers to eliminate the typo 2020-09-14 16:50:58 -07:00
Mark Andrews 2c1b6b2168 Silence REVERSE_INULL warning (CID 306652)
(cherry picked from commit 584dbffab1)
2020-09-10 07:54:19 +10:00
Mark Andrews e6332e4a67 watch_fd also requires thread->fdlock[lockid] to be held
(cherry picked from commit 22f499cdc4)
2020-09-03 07:14:45 +10:00
Mark Andrews eadfe4b673 remove dead code
(cherry picked from commit e923e62f6c)
2020-09-03 07:14:45 +10:00
Ondřej Surý 9b9fee13fa Handle EPROTO errno from recvmsg
It was discovered, that some systems might set EPROTO instead of EACCESS
on recvmsg() call causing spurious syslog messages from the socket
code.  This commit returns soft handling of EPROTO errno code to the
socket code. [GL #1928]

(cherry picked from commit e0380d437d)
2020-08-28 20:49:01 +02:00
Mark Andrews 14fe6e77a7 Always check the return from isc_refcount_decrement.
Created isc_refcount_decrement_expect macro to test conditionally
the return value to ensure it is in expected range.  Converted
unchecked isc_refcount_decrement to use isc_refcount_decrement_expect.
Converted INSIST(isc_refcount_decrement()...) to isc_refcount_decrement_expect.

(cherry picked from commit bde5c7632a)
2020-07-31 12:54:47 +10:00
Petr MenšíkandMark Andrews ac79d68765 Remove few lines in unix socket handling
Reuse the same checks two times, make difference minimal.

(cherry picked from commit 72d81c4768)
2020-07-24 13:47:26 +10:00
Mark Andrews f771d75c9b Address potential double unlock in process_fd
(cherry picked from commit cc0089c66b)
2020-07-14 07:35:17 +10:00
Mark Andrews dd32fb9284 Make the stdatomic shim and mutexatomic type complete
The stdatomic shims for non-C11 compilers (Windows, old gcc, ...) and
mutexatomic implemented only and minimal subset of the atomic types.
This commit adds 16-bit operations for Windows and all atomic types as
defined in standard.

(cherry picked from commit bccea5862d)
2020-07-08 10:29:59 +10:00
Matthijs Mekking 7915327aac Move dst key printtime in separate function
I'd like to use the same functionality (pretty print the datetime
of keytime metadata) in the 'rndc dnssec -status' command.  So it is
better that this logic is done in a separate function.

Since the stdtime.c code have differernt files for unix and win32,
I think the "#ifdef WIN32" define can be dropped.

(cherry picked from commit 9e03f8e8fe)
2020-07-01 09:57:44 +02:00
Witold Kręcicki bbf1422614 Fix possible deadlock in unix/socket.c
In process_fd we lock sock->lock and then internal_accept locks mgr->lock,
in isc_sockmgr_render* functions we lock mgr->lock and then lock sock->lock,
that can cause a deadlock when accessing stats. Unlock sock->lock early in
all the internal_{send,recv,connect,accept} functions instead of late
in process_fd.
2020-06-08 15:30:10 +02:00
Witold Kręcicki 444a16bff9 Don't set UDP recv/send buffer sizes - use system defaults (unless explicitly defined)
(cherry picked from commit fa02f6438b)
2020-05-01 17:47:19 +02:00
Ondřej Surý e017574b74 Correct the typecast of .tv_sec in isc_stdtime_get() 2020-03-25 22:10:10 +01:00
Ondřej Surý 2bb2a10ba4 Fix the tv_nsec check in isc_stdtime_get()
(cherry picked from commit 0d06a62dd1)
2020-03-25 21:19:55 +01:00
Ondřej Surý 0345dac44c Use clock_gettime() instead of gettimeofday() for isc_stdtime function
This also removes Solaris 2.8 broken gettimeofday() workaround

(cherry picked from commit e691b89a9a)
2020-03-19 10:17:26 +01:00
Ondřej Surý 67464af0bb Fixup the headers formatting 2020-03-11 10:23:35 +01:00
Ondřej Surý 60c6ff4ece Fix the deeper symlinks to .clang-format.headers 2020-03-11 10:21:54 +01:00
Ondřej Surý f3c2274479 Use the new sorting rules to regroup #include headers 2020-03-11 08:55:12 +00:00
Mark Andrews 5d64049301 Fix lists of installed header files 2020-03-06 13:00:04 +11:00
Witold Kręcicki e4d39f57ff Fix a race in isc_socket destruction.
There was a very slim chance of a race between isc_socket_detach and
process_fd: isc_socket_detach decrements references to 0, and before it
calls destroy gets preempted. Second thread calls process_fd, increments
socket references temporarily to 1, and then gets preempted, first thread
then hits assertion in destroy() as the reference counter is now 1 and
not 0.

(cherry picked from commit 81ba0fe0e6)
2020-03-03 09:26:54 +01:00
Evan Hunt 11a0d771f9 fix spelling errors reported by Fossies.
(cherry picked from commit ba0313e649)
2020-02-21 07:05:31 +00:00
Evan Hunt 25496b2299 adjust the clang-format penalties to reduce string breaking
this corrects some style glitches such as:
```
        long_function_call(arg, arg2, arg3, arg4, arg5, "str"
                                                        "ing");
```
...by adjusting the penalties for breaking strings and call
parameter lists.

(cherry picked from commit 0002377dca)
2020-02-17 14:43:46 -08:00