JosiahWI

This tests that the correct document data is written to cache by aggWrite. It also covers some fields of the Doc struct used for reading the data: the magic value,...

Fix building tests and benchmarks together

3

Fixes #14692 The cause of the failure was defining CATCH_CONFIG_RUNNER in two places. This fixes the issue by defining it only in unittests if both tests and benchmarks are to...

``` b5faf95ecc11:/trafficserver # cmake --build build --target inkcache [0/2] Re-checking globbed directories... [302/306] Building CXX object src/iocore/cache/CMakeFiles/inkcache.dir/Stripe.cc.o FAILED: src/iocore/cache/CMakeFiles/inkcache.dir/Stripe.cc.o /usr/bin/c++ -DDEBUG -DOPENSSL_API_COMPAT=10002 -DOPENSSL_IS_OPENSSL3 -DPACKAGE_NAME="\"Apache Traffic Server\"" -DPACKAGE_VERSION=\"10.1.0\" -D_DEBUG -Dlinux -I/trafficserver/include...

The one I noticed is in agg_copy; in debug mode, every time bytes are copied to the aggregation buffer to prepare for writing, it's counted as a backlog failure instead...

This refactors `Stripe::agg_copy` with the goal of making it a bit more understandable. I've tried to improve some variable names and extract a few functions where it makes sense. There...

This does not add or remove any tests or change the implementation of the test. ## To do Ready for Review. ## How to test `minetest --run-unittests --test-module "[nodedef]"`

Decouple disk from stripe

1

This moves the following variables out of `Stripe`: - [x] `fd` - [x] `path` - [x] `disk` - [ ] `hash_text`

TSan: ink atomic queue not so atomic

1

``` WARNING: ThreadSanitizer: data race (pid=9915) Atomic write of size 8 at 0xffff933d5bb0 by main thread (mutexes: write M0, write M1, write M2, write M3, write M4): #0 ink_atomiclist_push(InkAtomicList*, void*)...

``` WARNING: ThreadSanitizer: data race (pid=9915) Write of size 8 at 0xffff937ffd70 by thread T2 (mutexes: write M0): #0 ts::Histogram::operator()(unsigned long) /trafficserver/include/tsutil/Histogram.h:163 (traffic_server+0x9f8674) (BuildId: 7206bc40bbaaebae6ec8626a5203b22a73f3cecf) #1 EThread::Metrics::record_loop_time(long) (traffic_server+0xed3824) (BuildId: 7206bc40bbaaebae6ec8626a5203b22a73f3cecf)...

The issue was originally reported by TSan as a shutdown race. That report is easily reproducible, but I don't have a TSan build right now to do so. There were...