summaryrefslogtreecommitdiffhomepage
AgeCommit message (Collapse)AuthorFilesLines
2024-05-08tools/unitctl: enable passing IP addresses to the 'instances new' commandAva Hahn7-81/+235
* use path seperator constant from rust std package * pass a ControlSocket into deploy_new_container instead of a string * parse and validate a ControlSocket from argument to instances new * conditionally mount control socket only if its a unix socket * use create_image in a way that actually pulls nonpresent images * possibly override container command if TCP socket passed in * handle more weird error cases * add a ton of validation cases in the CLI command handler * add a nice little progress bar :) Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-05-08tools/unitctl: clean up control socket implsAva Hahn1-87/+87
Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-05-08tools/unitctl: Update host_path() to account for OSX special behaviourAva Hahn2-9/+32
Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-05-08tools/unitctl: Improve quality of life on osxAva Hahn2-9/+23
* unit-client-rs Mac build fix * elaborate in Readme on build requirements with examples for Mac users. Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-05-08tools/unitctl: Readme fixesAva Hahn1-12/+12
* fix Unit spelling in Readme * remove trailiing whitespace Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-05-08tools/unitctl: temporarily ignore issues with autogenerated readmeAva Hahn1-0/+1
Suggested-by: Andrew Clayton <a.clayton@nginx.com> Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-05-08tools/unitctl: Add new functionality to README.md and fmt codeAva Hahn6-108/+140
Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-05-08tools/unitctl: Add Docker deployment functionalityAva Hahn3-10/+108
* add UnitdDockerError type * write complete procedure to deploy unit via docker * additional tweaks verifying it fails peacefully * print important information in client Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-05-08tools/unitctl: API Plumbing for docker deploymentsAva Hahn4-27/+91
* refactored "instance" command out of enum * plumbed through function stub from client library * error handling Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-05-08tools/unitctl: Initial Docker ProceduresAva Hahn17-185/+990
* move UnitdProcess serialization logic into UnitdProcess * filter out docker processes from process output on Linux * initial implementation of a UnitdContainer type * initial implementation of a docker container search for unitd * pull out custom openapi future executor and use same tokio runtime as docker client * refactor openapi client to not manage its own tokio runtime * process mount points per docker container * correctly output docker container info in relevant unitd instances * create UnitdProcess from UnitdContainer * UnitdProcess now owns UnitdContainer * get and parse container details from docker API * introduce procedure to rewrite file paths based on docker container mounts * test path rewrite facilities * apply path rewrite to unix socket Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-05-07auto, perl: Fix building the Perl language module with clangAndrew Clayton1-0/+6
When we added -fno-strict-overflow to the CFLAGS back in c1e3f02f9 ("Compile with -fno-strict-overflow") we inadvertently broke building the Perl language module with clang, e.g $ make CC build/src/perl/nxt_perl_psgi-perl.o clang: error: argument unused during compilation: '-fno-strict-overflow' [-Werror,-Wunused-command-line-argument] This is due to for example on Apline $ perl -MExtUtils::Embed -e ccflags -D_REENTRANT -D_GNU_SOURCE -D_GNU_SOURCE -fwrapv -fno-strict-aliasing -pipe -fstack-protector-strong -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 Where on clang the -fwrapv causes the -fno-strict-overflow to be discarded resulting in the above error. We can get around that by simply appending -Qunused-arguments to the Perl CFLAGS. This fixes things for _some_ systems, as there is actually another issue with building this with clang on Fedora (and probably Red Hat) in that there the Perl ccflags & ldopts have been heavily modified and uses flags simply not only not in clang (which we can work around as above) but also incompatible flags, e.g $ make perl CC build/src/perl/nxt_perl_psgi-perl.o clang: error: optimization flag '-ffat-lto-objects' is not supported [-Werror,-Wignored-optimization-argument] There doesn't seem to be an easy workaround like -Qunused-arguments for this. While we could work around it in some way, I'm not sure it's worth the effort right now. On Red Hat & Fedora GCC _is_ the system compiler. This could be revisited if we find people trying to build this on Red Hat/Fedora with clang... For comparison this is the Alpine Perl ccflags & ldops $ perl -MExtUtils::Embed -e ccflags -D_REENTRANT -D_GNU_SOURCE -D_GNU_SOURCE -fwrapv -fno-strict-aliasing -pipe -fstack-protector-strong -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 alpine:~$ $ perl -MExtUtils::Embed -e ldopts -rdynamic -Wl,-rpath,/usr/lib/perl5/core_perl/CORE -fstack-protector-strong -L/usr/local/lib -L/usr/lib/perl5/core_perl/CORE -lperl -lpthread -ldl -lm -lcrypt -lutil -lc Fedora $ perl -MExtUtils::Embed -e ccflags -D_REENTRANT -D_GNU_SOURCE -O2 -flto=auto -ffat-lto-objects -fexceptions -g -grecord-gcc-switches -pipe -Wall -Wno-complain-wrong-lang -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -fstack-protector-strong -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -fwrapv -fno-strict-aliasing -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 $ perl -MExtUtils::Embed -e ldopts -Wl,--enable-new-dtags -Wl,-z,relro -Wl,--as-needed -Wl,-z,now -specs=/usr/lib/rpm/redhat/redhat-hardened-ld -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -Wl,--build-id=sha1 -Wl,-z,relro -Wl,--as-needed -Wl,-z,now -specs=/usr/lib/rpm/redhat/redhat-hardened-ld -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -Wl,--build-id=sha1 -fstack-protector-strong -L/usr/local/lib -L/usr/lib64/perl5/CORE -lperl -lpthread -lresolv -ldl -lm -lcrypt -lutil -lc Fixes: c1e3f02f9 ("Compile with -fno-strict-overflow") Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-05-07Convert 0-sized arrays to true flexible array membersAndrew Clayton4-10/+10
Declaring a 0-sized array (e.g 'char arr[0];') as the last member of a structure is a GNU extension that was used to implement flexible array members (FAMs) before they were standardised in C99 as simply '[]'. The GNU extension itself was introduced to work around a hack of declaring 1-sized arrays to mean a variable-length object. The advantage of the 0-sized (and true FAMs) is that they don't count towards the size of the structure. Unit already declares some true FAMs, but it also declared some 0-sized arrays. Converting these 0-sized arrays to true FAMs is not only good for consistency but will also allow better compiler checks now (as in a C99 FAM *must* be the last member of a structure and the compiler will warn otherwise) and in the future as doing this fixes a bunch of warnings (treated as errors in Unit by default) when compiled with -O2 -Warray-bounds -Wstrict-flex-arrays -fstrict-flex-arrays=3 (Note -Warray-bounds is enabled by -Wall and -Wstrict-flex-arrays seems to also be enabled via -Wall -Wextra, the -02 is required to make -fstrict-flex-arrays more effective, =3 is the default on at least GCC 14) such as CC build/src/nxt_upstream.o src/nxt_upstream.c: In function ‘nxt_upstreams_create’: src/nxt_upstream.c:56:18: error: array subscript i is outside array bounds of ‘nxt_upstream_t[0]’ {aka ‘struct nxt_upstream_s[]’} [-Werror=array-bounds=] 56 | string = nxt_str_dup(mp, &upstreams->upstream[i].name, &name); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In file included from src/nxt_upstream.c:9: src/nxt_upstream.h:55:48: note: while referencing ‘upstream’ 55 | nxt_upstream_t upstream[0]; | ^~~~~~~~ Making our flexible array members proper C99 FAMs and ensuring any >0 sized trailing arrays in structures are really normal arrays will allow to enable various compiler options (such as the above and more) that will help keep our array usage safe. Changing 0-sized arrays to FAMs should have no effect on structure layouts/sizes (they both have a size of 0, although doing a sizeof() on a FAM will result in a compiler error). Looking at pahole(1) output for the nxt_http_route_ruleset_t structure for the [0] and [] cases... $ pahole -C nxt_http_route_ruleset_t /tmp/build/src/nxt_http_route.o typedef struct { uint32_t items; /* 0 4 */ /* XXX 4 bytes hole, try to pack */ nxt_http_route_rule_t * rule[]; /* 8 0 */ /* size: 8, cachelines: 1, members: 2 */ /* sum members: 4, holes: 1, sum holes: 4 */ /* last cacheline: 8 bytes */ } nxt_http_route_ruleset_t; $ pahole -C nxt_http_route_ruleset_t build/src/nxt_http_route.o typedef struct { uint32_t items; /* 0 4 */ /* XXX 4 bytes hole, try to pack */ nxt_http_route_rule_t * rule[]; /* 8 0 */ /* size: 8, cachelines: 1, members: 2 */ /* sum members: 4, holes: 1, sum holes: 4 */ /* last cacheline: 8 bytes */ } nxt_http_route_ruleset_t; Also checking with the size(1) command on the effected object files shows no changes to their sizes $ for file in build/src/nxt_upstream.o \ build/src/nxt_upstream_round_robin.o \ build/src/nxt_h1proto.o \ build/src/nxt_http_route.o \ build/src/nxt_http_proxy.o \ build/src/python/*.o; do \ size -G /tmp/${file} $file; echo; done text data bss total filename 640 418 0 1058 /tmp/build/src/nxt_upstream.o 640 418 0 1058 build/src/nxt_upstream.o text data bss total filename 929 351 0 1280 /tmp/build/src/nxt_upstream_round_robin.o 929 351 0 1280 build/src/nxt_upstream_round_robin.o text data bss total filename 11707 8281 16 20004 /tmp/build/src/nxt_h1proto.o 11707 8281 16 20004 build/src/nxt_h1proto.o text data bss total filename 8319 3101 0 11420 /tmp/build/src/nxt_http_route.o 8319 3101 0 11420 build/src/nxt_http_route.o text data bss total filename 1495 1056 0 2551 /tmp/build/src/nxt_http_proxy.o 1495 1056 0 2551 build/src/nxt_http_proxy.o text data bss total filename 4321 2895 0 7216 /tmp/build/src/python/nxt_python_asgi_http-python.o 4321 2895 0 7216 build/src/python/nxt_python_asgi_http-python.o text data bss total filename 4231 2266 0 6497 /tmp/build/src/python/nxt_python_asgi_lifespan-python.o 4231 2266 0 6497 build/src/python/nxt_python_asgi_lifespan-python.o text data bss total filename 12051 6090 8 18149 /tmp/build/src/python/nxt_python_asgi-python.o 12051 6090 8 18149 build/src/python/nxt_python_asgi-python.o text data bss total filename 28 1963 432 2423 /tmp/build/src/python/nxt_python_asgi_str-python.o 28 1963 432 2423 build/src/python/nxt_python_asgi_str-python.o text data bss total filename 5818 3518 0 9336 /tmp/build/src/python/nxt_python_asgi_websocket-python.o 5818 3518 0 9336 build/src/python/nxt_python_asgi_websocket-python.o text data bss total filename 4391 2089 168 6648 /tmp/build/src/python/nxt_python-python.o 4391 2089 168 6648 build/src/python/nxt_python-python.o text data bss total filename 9095 5909 152 15156 /tmp/build/src/python/nxt_python_wsgi-python.o 9095 5909 152 15156 build/src/python/nxt_python_wsgi-python.o Link: <https://lwn.net/Articles/908817/> Link: <https://people.kernel.org/kees/bounded-flexible-arrays-in-c> Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-05-07auto/wasm: No need to explicitly set -fno-strict-aliasing nowAndrew Clayton1-2/+1
Since commit 0b5223e1c ("Disable strict-aliasing in clang by default") we explicitly always build with -fno-strict-aliasing so there's no need to set it independently in auto/modules/wasm Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-05-02Add a GitHub workflow to check for whitespace issuesAndrew Clayton1-0/+48
If it fails you can check the 'git log --check' output of the workflow to see what the issue is. E.g --- 93ec0133 Oops... README.md:1: trailing whitespace. +# NGINX Unit Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-30tools: Add unitctl section to the READMETaryn Musgrave1-0/+12
Co-developed-by: Ava Hahn <a.hahn@f5.com> Signed-off-by: Ava Hahn <a.hahn@f5.com> [ Tweak subject and cli => unitctl in README - Andrew ] Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-30tools: Add unitctl CLIAva Hahn54-0/+7212
* Pull in entire unit-rust-sdk project * not included: CLA, COC, License * not included: duplicate openapi spec * not included: CI workflows * not included: changelog tooling * not included: commitsar tooling * not included: OpenAPI Web UI feature * update links in unitctl manpage * remove IDE configuration from .gitignore * rename Containerfile.debian to Dockerfile * simplify call to uname * keep Readmes and Makefiles to 80 character lines * outline specifically how to build unitctl for any desired target, and where to then find the binary for use * remove a section on the vision of the CLI which was superfluous given the state of completeness of the code and its use in unit * remove out of date feature proposals from readme * makefile: do not run when Rustup is not present * bump mio version to latest * generate openapi client library on demand * generate-openapi only runs when not present * generate-openapi now a dependency of binary build targets * deleted autogenerated code * reverted readme and Cargo document to autogenerated state * add additional build requirement to Readme Co-developed-by: Elijah Zupancic <e.zupancic@f5.com> Signed-off-by: Elijah Zupancic <e.zupancic@f5.com> Signed-off-by: Ava Hahn <a.hahn@f5.com> Reviewed-by: Andrew Clayton <a.clayton@nginx.com> # non rust stuff [ tools/cli => tools/unitctl and subject tweak - Andrew ] Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-25Tighten up some string arraysAndrew Clayton3-10/+13
This is the normal way of declaring such things. Reviewed-by: Zhidao HONG <z.hong@f5.com> Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-25configuration: Constify more pointersAndrew Clayton2-24/+28
This continues the patch series constifying various pointers in the configuration sub-system. This is done as a separate commit as it involved a _slightly_ more invasive change in nxt_conf_get_string(). While it takes a value parameter that is never modified, simply making it const results in CC build/src/nxt_conf.o src/nxt_conf.c: In function ‘nxt_conf_get_string’: src/nxt_conf.c:170:20: error: assignment discards ‘const’ qualifier from pointer target type [-Werror=discarded-qualifiers] 170 | str->start = value->u.str.start; | ^ due to the assignment operator. Making value const will allow for numerous other constification and seeing as we are not modifying it, seems worthwhile. We can get around the warning by casting ->u.{str,string}.start Reviewed-by: Zhidao HONG <z.hong@f5.com> Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-25php: Constify some local static variablesAndrew Clayton1-6/+6
A common pattern was to declare variables in functions like static nxt_str_t ... Not sure why static, as they were being treated more like string literals, let's actually make them constants (qualifier wise). Reviewed-by: Zhidao HONG <z.hong@f5.com> Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-25Constify a bunch of static local variablesAndrew Clayton10-74/+79
A common pattern was to declare variables in functions like static nxt_str_t ... Not sure why static, as they were being treated more like string literals (and of course they are _not_ thread safe), let's actually make them constants (qualifier wise). This handles core code conversion. Reviewed-by: Zhidao HONG <z.hong@f5.com> Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-25configuration: Constify numerous pointersAndrew Clayton3-68/+74
Mark numerous function argument pointers as 'const' in the configuration sub-system. This also does the same with a few functions in src/nxt_conf_validation.c that are required to accomplish the below, attacking the rest is an exercise for another day... While this is a worthwhile hardening exercise in its own right, the main impetus for this is to 'constify' some local function variables which are currently defined with 'static' storage class and turn them into 'static const', which will be done in a subsequent patch. Reviewed-by: Zhidao HONG <z.hong@f5.com> Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-23Fix some trailing whitespace and long lines in the READMEAndrew Clayton1-3/+4
Fixes: a48fbc035 ("Add additional information to the README") Reviewed-by: Ava Hahn <a.hahn@f5.com> Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-19wasm-wc: Bump the rustls crate from 0.21.10 to 0.21.11dependabot[bot]1-2/+2
Bumps <https://github.com/rustls/rustls> from 0.21.10 to 0.21.11. "This release corrects a denial-of-service condition in rustls::ConnectionCommon::complete_io(), reachable via network input. If a close_notify alert is received during a handshake, complete_io() did not terminate. Callers which do not call complete_io() are not affected." The wasm-wasi-component language module is not effected by this as it doesn't handle client connections, Unit does. Link: Release notes <https://github.com/rustls/rustls/releases> Link: Commits <https://github.com/rustls/rustls/compare/v/0.21.10...v/0.21.11> Signed-off-by: dependabot[bot] <support@github.com> Reviewed-by: Andrew Clayton <a.clayton@nginx.com> [ Tweaked commit message/subject - Andrew ] Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-18Elaborate on docker image differencesAva Hahn1-1/+3
* This commit adds a warning to readers to clarify that they should be aware of our different image tags before pulling their image.
2024-04-18Add additional information to the READMEAva Hahn1-2/+23
* expand on docker instructions * identify API documentation * identify WASM documentation Acked-by: Andrew Clayton <a.clayton@nginx.com> Signed-off-by: Ava Hahn <a.hahn@f5.com>
2024-04-17Fixes: 64934e59f ("HTTP: Introduce quoted target marker in HTTP parsing")Zhidao HONG1-2/+2
Reviewed-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-11HTTP: Rewrote url target section in nxt_h1p_peer_header_send()Zhidao HONG3-5/+74
Previously, proxy request was constructed based on the `r->target` field. However, r->target will remain unchanged in the future, even in cases of URL rewriting because of the requirement change for $request_uri that will be changed to constant. To accommodate this, the r->target should be designed to be constant, but Unit needs to pass a changeable URL to the upstream server. Based on the above, the proxy module can't depend on r->target.
2024-04-11HTTP: Introduce quoted target marker in HTTP parsingZhidao HONG2-12/+5
The quoted_target field is to indentify URLs containing percent-encoded characters. It can be used in places where you might need to generate new URL, such as in the proxy module. It will be used in the subsequent commit.
2024-04-10Tests: error report corrected for unknown variables in "response_headers"Andrei Zeliankou1-7/+6
For more information please see https://github.com/nginx/unit/pull/1191
2024-04-10HTTP: Added variable validation to the response_headers optionZhidao HONG1-1/+12
This is to improve error messages for response headers configuration. Take the configuration as an example: { "response_headers": { "a": "$b" } } Previously, when applying it the user would see this error message: failed to apply previous configuration After this change, the user will see this improved error message: the previous configuration is invalid: Unknown variable "b" in the "a" value
2024-04-10Tests: compatibility with OpenSSL 3.2.0Andrei Zeliankou1-1/+5
OpenSSL 3.2.0 generates X.509v3 certificates by default. These certificates, even self-signed, cannot sign other certificates unless "CA:TRUE" is explicitly set in the basicConstraints extension. As a result, tests attempting this are currently failing. Fix is to provide "CA:TRUE" in the basicConstraints for self-signed root certificates used in "openssl ca" commands. Closes: https://github.com/nginx/unit/issues/1202 Tested-by: Andrew Clayton <a.clayton@nginx.com> Reviewed-by: Andrew Clayton <a.clayton@nginx.com>
2024-04-10Tests: added $request_uri tests with proxyAndrei Zeliankou1-1/+100
This patch consist of 3 tests: 1. Ensure that $request_uri won't change while proxying the request. 2. Same as 1, but modifying the request using the "rewrite" directive. 3. Same as 2, but with rewrite containing a percent-encoded string.
2024-04-10Docs: njs (lowercase) is more preferred way to mentionAndrei Zeliankou2-9/+9
2024-04-10njs (lowercase) is more preferred way to mentionAndrei Zeliankou3-5/+5
2024-04-09Wasm-wc: Bump the h2 crate from 0.4.2 to 0.4.4dependabot[bot]1-2/+2
Bumps h2 <https://github.com/hyperium/h2> from 0.4.2 to 0.4.4. Limit number of CONTINUATION frames for misbehaving connections. Link: Changelog <https://github.com/hyperium/h2/blob/master/CHANGELOG.md> Link: Commits <https://github.com/hyperium/h2/compare/v0.4.2...v0.4.4> Signed-off-by: dependabot[bot] <support@github.com> Reviewed-by: Andrew Clayton <a.clayton@nginx.com> [ Tweaked commit message/subject - Andrew ] Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-14Wasm-wc: Fix application restartsAndrew Clayton1-3/+12
Liam reported a problem when trying to restart wasm-wasi-component based applications using the /control/applications/APPLICATION_NAME/restart endpoint. The application would become unresponsive. What was happening was the old application process(es) weren't exit(3)ing and so while we were starting new application processes, the old ones were still hanging around in a non-functioning state. When we are terminating an application it must call exit(3). So that's what we do. We use the return value of nxt_unit_run() as the exit status. Due to exit(3)ing we also need to now explicitly handle the return on error case. Reported-by: Liam Crilly <liam@nginx.com> Fixes: 20ada4b5c ("Wasm-wc: Core of initial Wasm component model language module support") Closes: https://github.com/nginx/unit/issues/1179 Tested-by: Liam Crilly <liam@nginx.com> Tested-by: Danielle De Leo <d.deleo@f5.com> Co-developed-by: Dan Callahan <d.callahan@f5.com> Signed-off-by: Dan Callahan <d.callahan@f5.com> Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-14Rebuild wasm-wasi-component when any of its dependencies changeAndrew Clayton1-1/+8
Have cargo run if for example src/wasm-wasi-component/src/lib.rs is changed, or any of the other files that should perhaps trigger a rebuild. Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-13Build with -std=gnu11 (C11 with GNU extensions)Alejandro Colomar1-2/+3
Currently Unit doesn't specify any specific C standard for compiling and will thus be compiled under whatever the compiler happens to default to. Current releases of GCC and Clang (13.x & 17.x respectively at the time of writing) default to gnu17 (C17 + GNU extensions). Our oldest still-supported system is RHEL/CentOS 7, that comes with GCC 4.8.5 which defaults to gnu90. Up until now this hasn't really been an issue and we have been able to use some C99 features that are implemented as GNU extensions in older compilers, e.g - designated initializers - flexible array members - trailing comma in enum declaration (compiles with -std=c89, warns with -std=c89 -pedantic) - snprintf(3) - long long (well we test for it but don't actually use it) - bool / stdbool.h - variadic macros However there are a couple of C99 features that aren't GNU extensions that would be handy to be able to use, i.e - The ability to declare variables inside for () loops, e.g for (int i = 0; ...; ...) - C99 inline functions (not to be confused with what's available with -std=gnu89). However, if we are going to switch up to C99, then perhaps we should just leap frog to C11 instead (the Linux Kernel did in fact make the switch from gnu89 to gnu11 in March '22). C17 is perhaps still a little new and is really just C11 + errata. GCC 4.8 as in RHEL 7 has *some* support for C11, so while we can make full use of C99, we couldn't yet make full use of C11, However RHEL 7 is EOL on June 30th 2024, after which we will no longer have that restriction and in the meantime we can restrict ourselves to the supported set of features (or implement fallbacks where appropriate). It can only be a benefit that we would be compiling Unit consistently under the same language standard. This will also help give the impression that Unit is a modern C code base. It is also worth noting the following regarding GCC "A version with corrections integrated was prepared in 2017 and published in 2018 as ISO/IEC 9899:2018; it is known as C17 and is supported with -std=c17 or -std=iso9899:2017; the corrections are also applied with - std=c11, and the only difference between the options is the value of STDC_VERSION." Suggested-by: Andrew Clayton <a.clayton@nginx.com> Acked-by: Andrew Clayton <a.clayton@nginx.com> [ Andrew wrote the commit message ] Signed-off-by: Alejandro Colomar <alx@kernel.org> Link: <https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e8c07082a810fbb9db303a2b66b66b8d7e588b53> Link: <https://www.ibm.com/blog/announcement/ibm-is-announcing-red-hat-enterprise-linux-7-is-going-end-of-support-on-30-june-2024/> Link: <https://gcc.gnu.org/onlinedocs/gcc-8.1.0/gcc/Standards.html#C-Language> Cc: Dan Callahan <d.callahan@f5.com> Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-13Add a GitHub workflow status badge for our CI to the READMEAndrew Clayton1-0/+1
This shows the current state of our CI builds and points to the Unit workflows page. Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-13Add a repostatus badge to the READMEAndrew Clayton1-0/+2
This is set to 'active' and is described here <https://www.repostatus.org/> This is now a requirement of F5/NGINX OSS projects. Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-12Tests: NJS cacheable variables with access logAndrei Zeliankou1-0/+18
Reproduces issue https://github.com/nginx/unit/issues/1169.
2024-03-12NJS: loader should be registered using njs_vm_set_module_loader()Andrei Zeliankou6-25/+19
This change makes NJS module incompatible with NJS older than 0.8.3. Therefore, the configuration version check has been adjusted accordingly. This change was introduced in NJS 0.8.3 here: <https://hg.nginx.com/njs/rev/ad1a7ad3c715>
2024-03-11Fix the security-alert email link in the READMEAndrew Clayton1-2/+2
This fixes the security-alert email link to actually open up in an email client, rather than it trying to make the email link a part of the Unit repository URL. Fixes: fa42d858a ("Adding GitHub-styled README and CONTRIBUTING files in Markdown.") Signed-off-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-11Avoiding arithmetic ops with NULL pointer in nxt_unit_mmap_getAndrei Zeliankou1-0/+6
Found by UndefinedBehaviorSanitizer. Reviewed-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-11Initialize port_impl only when it is neededAndrei Zeliankou1-1/+1
Found by UndefinedBehaviorSanitizer. Reviewed-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-11Avoiding arithmetic ops with NULL pointer in nxt_port_mmap_getAndrei Zeliankou1-0/+6
Can be reproduced by test/test_settings.py::test_settings_send_timeout with enabled UndefinedBehaviorSanitizer. Reviewed-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-11Avoiding arithmetic ops with NULL pointer in nxt_http_arguments_parseAndrei Zeliankou1-0/+6
Can be reproduced by test/test_variables.py::test_variables_dynamic_arguments with enabled UndefinedBehaviorSanitizer: src/nxt_http_request.c:961:17: runtime error: applying zero offset to null pointer #0 0x1050d95a4 in nxt_http_arguments_parse nxt_http_request.c:961 #1 0x105102bf8 in nxt_http_var_arg nxt_http_variables.c:621 #2 0x104f95d74 in nxt_var_interpreter nxt_var.c:507 #3 0x104f98c98 in nxt_tstr_query nxt_tstr.c:265 #4 0x1050abfd8 in nxt_router_access_log_writer nxt_router_access_log.c:194 #5 0x1050d81f4 in nxt_http_request_close_handler nxt_http_request.c:838 #6 0x104fcdc48 in nxt_event_engine_start nxt_event_engine.c:542 #7 0x104fba838 in nxt_thread_trampoline nxt_thread.c:126 #8 0x18133e030 in _pthread_start+0x84 (libsystem_pthread.dylib:arm64e+0x7030) #9 0x181338e38 in thread_start+0x4 (libsystem_pthread.dylib:arm64e+0x1e38) SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior src/nxt_http_request.c:961:17 Reviewed-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-11Fixed undefined behaviour in left shift of int valueAndrei Zeliankou1-4/+4
Found by UndefinedBehaviorSanitizer: src/nxt_random.c:151:31: runtime error: left shift of 140 by 24 places cannot be represented in type 'int' #0 0x104f78968 in nxt_random nxt_random.c:151 #1 0x104f58a98 in nxt_shm_open nxt_port_memory.c:377 #2 0x10503e24c in nxt_controller_conf_send nxt_controller.c:617 #3 0x105041154 in nxt_controller_process_request nxt_controller.c:1109 #4 0x104fcdc48 in nxt_event_engine_start nxt_event_engine.c:542 #5 0x104f27254 in main nxt_main.c:35 #6 0x180fbd0dc (<unknown module>) SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior src/nxt_random.c:151:31 Reviewed-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-11NJS: avoiding arithmetic ops with NULL pointer in r->argsAndrei Zeliankou1-2/+3
Can be reproduced by test/test_rewrite.py::test_rewrite_njs with enabled UndefinedBehaviorSanitizer: src/nxt_http_js.c:169:52: runtime error: applying zero offset to null pointer #0 0x10255b044 in nxt_http_js_ext_get_args nxt_http_js.c:169 #1 0x102598ad0 in njs_value_property njs_value.c:1175 #2 0x10259c2c8 in njs_vm_object_prop njs_vm.c:1398 #3 0x102559d74 in nxt_js_call nxt_js.c:445 #4 0x1023c0da0 in nxt_tstr_query nxt_tstr.c:276 #5 0x102516ec4 in nxt_http_rewrite nxt_http_rewrite.c:56 #6 0x1024fd86c in nxt_http_request_action nxt_http_request.c:565 #7 0x1024d71b0 in nxt_h1p_request_body_read nxt_h1proto.c:998 #8 0x1023f5c48 in nxt_event_engine_start nxt_event_engine.c:542 #9 0x1023e2838 in nxt_thread_trampoline nxt_thread.c:126 #10 0x18133e030 in _pthread_start+0x84 (libsystem_pthread.dylib:arm64e+0x7030) #11 0x181338e38 in thread_start+0x4 (libsystem_pthread.dylib:arm64e+0x1e38) SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior src/nxt_http_js.c:169:52 Same fix was introduced in NJS: <http://hg.nginx.org/njs/rev/4fba78789fe4> Reviewed-by: Andrew Clayton <a.clayton@nginx.com>
2024-03-11Router: match when pattern and tested string are both zero lengthAndrei Zeliankou1-0/+4
Otherwise, undefined behaviour will be triggered. Can be reproduced by test/test_routing.py::test_routes_match_host_empty with enabled UndefinedBehaviorSanitizer: src/nxt_http_route.c:2141:17: runtime error: applying zero offset to null pointer #0 0x100562588 in nxt_http_route_test_rule nxt_http_route.c:2091 #1 0x100564ed8 in nxt_http_route_handler nxt_http_route.c:1574 #2 0x10055188c in nxt_http_request_action nxt_http_request.c:570 #3 0x10052b1a0 in nxt_h1p_request_body_read nxt_h1proto.c:998 #4 0x100449c38 in nxt_event_engine_start nxt_event_engine.c:542 #5 0x100436828 in nxt_thread_trampoline nxt_thread.c:126 #6 0x18133e030 in _pthread_start+0x84 (libsystem_pthread.dylib:arm64e+0x7030) #7 0x181338e38 in thread_start+0x4 (libsystem_pthread.dylib:arm64e+0x1e38) SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior src/nxt_http_route.c:2141:17 Reviewed-by: Andrew Clayton <a.clayton@nginx.com>
generated by cgit (git 2.34.1) at 2024-11-17 01:54:58 +0000