1. LogTM: Log-Based Transactional Memory Kevin E. Moore, Jayaram Bobba, Michelle J. Moravan, Mark D. Hill, & David A. Wood Presented by Colleen Lewis

2. Credits Animations from the original LogTM HPCA presentation Original graphs modified for readability

3. Big Picture Hardware transaction motivation Per thread log Optimize commits (Hardware)

4. Design Decisions Version Management Eager – write in place Lazy – write on commit Conflict Detection Eager – detect at read/write time Lazy – detect at commit time

5. Transaction Logs Pointer to the beginning of the log Pointer to the end of the log Read and Write bits for each cache line

6. 2/15/06HPCA-12 12-------------- --------------23 34-------------- 0 Transaction Log Example 00 40 C0 1000 1040 1080 Initial State LogBase = LogPointer TM count > 0 Data BlockVA Log Base Log Ptr TM count 1000 1 00 R W 00 00

7. 2/15/06HPCA-12 10001048 -- 34------------ 12-------------- --------------23 34-------------- 0 Transaction Log Example 56-------------- 00 40 C0 1000 1040 1080 Store r2, (c0) /* r2 = 56 */ –Set W bit for block (c0) –Store address (c0) and old data on the log –Increment Log Ptr to 1048 –Update memory Data BlockVA Log Base Log Ptr TM count 1000 1 00 R W 00 01 c0

8. 2/15/06HPCA-12 12-------------- --------------23 56-------------- Transaction Log Example 00 40 C0 1000 1040 1080 Commit transaction –Clear R & W for all blocks –Reset Log Ptr to Log Base (1000) –Clear TM count Data BlockVA Log Base Log Ptr TM count 1000 0 00 R W 00 00 34------------c0 -- 0 00 1 1 1048

9. 2/15/06HPCA-12 1 1090 Transaction Log Example 12-------------- --------------23 34-------------- 00 40 C0 1000 1040 1080 Abort transaction –Replay log entries to “undo” the transaction –Reset Log Ptr to Log Base (1000) –Clear R & W bits for all blocks –Clear TM count Data BlockVA Log Base Log Ptr TM count 1000 1048 0 00 R W 00 00 c034------------ -- 0 00 1 56-------------- 1000

10. Conflict Detection Checked at every read/write Directory forwards read requests Directory can have “sticky” data Individual nodes responsible for detecting conflicts Needs Transaction mode bit Overflow bit

11. 2/15/06HPCA-12 I [old] M@P0 [old] I (--) [none]M (--) [old]M (-W) [new] Conflict Detection (example) Directory TM mode Overflow 0 0 P1 I (--) [none] TM mode Overflow 0 0 1 P0 store –P0 sends get exclusive (GETX) request –Directory responds with data (old) –P0 executes store P0 GETX DATA

12. 2/15/06HPCA-12 M (-W) [new] Conflict Detection (example) Directory TM mode Overflow 0 0 P1 I (--) [none] TM mode Overflow 0 0 M@P0 [old] 1 In-cache transaction conflict –P1 sends get shared (GETS) request –Directory forwards to P0 –P0 detects conflict and sends NACK P0 GETS Fwd_GETS Conflict! NACK

13. 2/15/06HPCA-12 M (-W) [new] I (--) [none] M@P0 [old] M sticky @P0 [new] Conflict Detection (example) Directory TM mode Overflow 0 0 P1 I (--) [none] TM mode Overflow 0 0 1 Cache overflow –P0 sends put exclusive (PUTX) request –Directory acknowledges –P0 sets overflow bit –P0 writes data back to memory P0 PUTX ACK DATA 1

14. 2/15/06HPCA-12 Conflict Detection (example) Directory I (--) [none] TM mode Overflow 0 0 P1 I (--) [none] TM mode Overflow 0 0 M@P0 [old] 1 Out-of-cache conflict –P1 sends GETS request –Directory forwards to P0 –P0 detects a (possible) conflict –P0 sends NACK P0 M (--) [old]M (-W) [new] M sticky @P0 [new] I (--) [none] 1 GETS Fwd_GETS Conflict! NACK 1

15. 2/15/06HPCA-12 Conflict Detection (example) Directory I (--) [none] TM mode Overflow 0 0 P1 I (--) [none] TM mode Overflow 0 0 M@P0 [old] 1 Commit –P0 clears TM mode and Overflow bits P0 M (--) [old]M (-W) [new] M sticky @P0 [new] I (--) [none] 1 0 0

16. 2/15/06HPCA-12 M sticky @P0 [new] S(P1) [new] 0 0 0 Conflict Detection (example) Directory I (--) [none] TM mode Overflow 0 P1 I (--) [none] TM mode Overflow 0 0 Lazy cleanup –P1 sends GETS request –Directory forwards request to P0 –P0 detects no conflict, sends CLEAN –Directory sends Data to P1 P0 M (--) [old]M (-W) [new] I (--) [none] GETS Fwd_GETS CLEAN DATA S (--) [new]

17. False Positives? What if P0 has started a new transaction without cleaning the sticky data?

18. M (-W) [new] I (--) [none] M@P0 [old] M sticky @P0 [new] False Positive Example Directory TM mode Overflow 0 0 P1 I (--) [none] TM mode Overflow 0 0 1 Cache overflow P0 sends put exclusive (PUTX) request Directory acknowledges P0 sets overflow bit P0 writes data back to memory P0 PUTX ACK DATA 1

19. False Positive Example Directory I (--) [none] TM mode Overflow 0 0 P1 I (--) [none] TM mode Overflow 0 0 M@P0 [old] 1 Commit P0 clears TM mode and Overflow bits Start New Transaction P0 set TM mode Eventually overflow Set overflow bits P0 M (--) [old]M (-W) [new] M sticky @P0 [new] I (--) [none] 1 0 0 1 1

20. Conflict Detection (example) Directory I (--) [none] TM mode Overflow 0 0 P1 I (--) [none] TM mode Overflow 0 0 M@P0 [old] 1 Out-of-cache conflict P1 sends GETS request Directory forwards to P0 P0 detects a (possible) conflict P0 sends NACK P0 M (--) [old]M (-W) [new] M sticky @P0 [new] I (--) [none] 1 GETS Fwd_GETS Conflict! NACK 1

21. Conflict Resolution and Deadlock Avoidance Options Wait – risk deadlock? Abort – risk livelock? Current Behavior Wait Abort if waiting on a logically younger process Future Behavior? Software contention manager

22. Evaluation 32 SPARC processors Solaris 9 OS SIMICS – full system simulator Magic no-ops Tests Micro-benchmarks SPLASH suite

23. Microbenchmarks High Contention / Short Transactions Comparing: EXP - TTS locks with exponential backoff MCS – SW Queue based locks BEGIN_TRANSACTION(); new_total = total.count + 1; private_data[id].count++; total.count = new_total; COMMIT_TRANSACTION();

25. SPLASH2 Benchmark Results

26. Data presented as: PARMACS locks execution time LogTM execution time Modified version: LogTM execution time PARMACS locks execution time

27. SPLASH2 Benchmark Results

28. Conclusions Optimize commits Aborts handled by software Stall to avoid wasting work Allow sticky data because overflow is rare Good performance on microbenchmark False sharing has a big impacts on LogTM

29. Questions?