Scott Lystig Fritchie
fd7dad0714
Coverage is about as good as it's going to get
2015-03-02 20:20:16 +09:00
Scott Lystig Fritchie
da2bad564f
Getting closer to understanding why test coverage appears so poor, part 2
2015-03-02 20:20:16 +09:00
Scott Lystig Fritchie
342a972543
Getting closer to understanding why test coverage appears so poor
2015-03-02 20:20:16 +09:00
Scott Lystig Fritchie
b4f2d314c7
More single chain manager simulation tests
2015-03-02 20:20:16 +09:00
Scott Lystig Fritchie
e717d797b3
Move almost all test code to test/* modules
2015-03-02 20:20:16 +09:00
Scott Lystig Fritchie
01e3325b81
Tiny refactoring of random number gen
2015-03-02 20:20:16 +09:00
Scott Lystig Fritchie
d2f93e919e
Single chain manager simulation test: no bad projection transitions!
2015-03-02 20:20:16 +09:00
Scott Lystig Fritchie
057f958bb1
WIP: chain manager simulation test
2015-03-02 20:20:16 +09:00
Scott Lystig Fritchie
410c8ff7ce
WIP: chain manager simulation test
2015-03-02 20:20:16 +09:00
Scott Lystig Fritchie
b8c87b23ad
WIP: chain manager simulation test
2015-03-02 20:20:16 +09:00
Scott Lystig Fritchie
4ebc80dc39
Add src/machi_util.erl
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
a81552ed82
Makefile un-derp'ing
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
a5dc72834f
Fix proj0_test for concuerror, yay!
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
4969e019b2
Fix proj0_test for concuerror, yay!
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
e50e669b79
TODO left off here
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
97c5789b44
WIP: eunit tests pass, but Concuerror loops forever then errs on max retries on proj0_test
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
f7447e8953
WIP: done (I hope) adding Lamport clocks
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
ee7bc2645b
WIP: in the middle of adding Lamport clocks
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
b443a15542
register op name sanity: write and _read_
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
921d90a69b
WIP: enforce wedging and new projection writes
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
bebce51ab9
WIP: minimal write-once projection store in FLU
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
34c8c6490a
WIP: add Name arg to start_link()
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
2d3a29471d
Minimal FLU0 single register, plus Concuerror tests
2015-03-02 20:20:15 +09:00
Scott Lystig Fritchie
f378204a91
Add fledgling log implementation based on CORFU papers (corfurl stuff)
2015-03-02 20:20:07 +09:00
Scott Lystig Fritchie
370f70303d
Merge branch 'merge/tango-prototype'
2015-03-02 20:07:25 +09:00
Scott Lystig Fritchie
94ebd4bb6f
Rename prototype/tango-prototype -> prototype/tango
2015-03-02 20:06:45 +09:00
Scott Lystig Fritchie
3f3f3e4f5d
Update README.tango.md with latest checkpoint implementation fix notes
2015-03-02 20:03:46 +09:00
Scott Lystig Fritchie
c5ed355dac
Rename tango readme
2015-03-02 20:03:46 +09:00
Scott Lystig Fritchie
8da46f78fe
BAH! Checkpoint is quite broken, see new README.tango.md
2015-03-02 20:03:46 +09:00
Scott Lystig Fritchie
7bf98fa648
All tests pass, but checkpointing does not truncate history
2015-03-02 20:03:46 +09:00
Scott Lystig Fritchie
fed2f43783
WIP: all but queue checkpointing now passes
2015-03-02 20:03:46 +09:00
Scott Lystig Fritchie
0b3bb3ee7c
WIP: tango_oid_test now passes
2015-03-02 20:03:46 +09:00
Scott Lystig Fritchie
a0bb7ee23d
WIP: tango_oid refactoring, all broken: infinite loop
2015-03-02 20:03:46 +09:00
Scott Lystig Fritchie
9a3ac02413
WIP: first round of tango_oid refactoring, all broken horribly
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
03f071316c
Gadz, more sequencer cleanup. corfurl_test now passes
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
30fc62ab22
Gadz, more sequencer cleanup. corfurl_sequencer_test now passes
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
b8c051c89f
Fix broken sequencer semantics.
...
It occurred to me today that I implemented the sequencer incorrectly and
hadn't yet noticed because I don't have any tests that are
complex/interleaved/perhaps-non-deterministic to find the problem.
The problem is that the sequencer's current implementation only keeps
track of the last LPN for any Tango stream.
The fix is to do what the paper actually says: the sequencer keeps a
*list* of the last $K$ LPNs for each stream. Derp. Yes, that's really
necessary to avoid a pretty simple race condition with 2 actors
simultaneously updating a single Tango stream.
1st commit: fix the implementation and the smoke test. The
broken-everything-else will be repaired in later commits.
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
940012cef1
Add checkpoint support for tango_dt_map
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
4cf8ac7ed8
Add checkpoint support for tango_dt_queue
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
970eb263db
Fix bug in backpointer handling, derp!
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
004a18d948
Add checkpoint support for tango_dt_register
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
7b9c94553c
Add skeleton support for single-page checkpointing
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
1c1e1368dd
Added src/tango_dt_queue.erl plus test
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
6caeaeb6b5
Ha! Damn quick and easy to add tango_dt_map.erl
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
df53ec0a4e
Refactor register DT into tango_dt.erl and tango_dt_register.erl
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
c068057c96
Add missing func corfurl_client:append_page/3, then fix tango_dt_register_test
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
cdeddbb582
Heh, demonstrate a concurrency bug that I knew was there, yay, fixit time!
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
18b38c249e
First draft of tango_dt_register
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
6067e26201
Change semantics of OID map, silly me, to match what's needed
2015-03-02 20:03:45 +09:00
Scott Lystig Fritchie
436c6ac14b
Minor type fixup
2015-03-02 20:03:45 +09:00
