Revert "Clarify comment about compaction_readahead_size's sanitizatio… (#11773)

Summary:
…n change (https://github.com/facebook/rocksdb/issues/11755)"

This reverts commit 45131659.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/11773

Reviewed By: ajkr

Differential Revision: D48832320

Pulled By: hx235

fbshipit-source-id: 96cef26a885134360766a83505f6717598eac6a9

HISTORY.md

@@ -22,7 +22,6 @@
 ### Behavior Changes
 * Statistics `rocksdb.sst.read.micros` now includes time spent on multi read and async read into the file
 * For Universal Compaction users, periodic compaction (option `periodic_compaction_seconds`) will be set to 30 days by default if block based table is used.
-* `Options::compaction_readahead_size` will be sanitized to 2MB when set to 0 under non-direct IO since we have moved prefetching responsibility to page cache for compaction read with readhead size equal to `Options::compaction_readahead_size` under non-direct IO (#11631)
 
 ### Bug Fixes
 * Fix a bug in FileTTLBooster that can cause users with a large number of levels (more than 65) to see errors like "runtime error: shift exponent .. is too large.." (#11673).

include/rocksdb/options.h

@@ -951,13 +951,10 @@ struct DBOptions {
   enum AccessHint { NONE, NORMAL, SEQUENTIAL, WILLNEED };
   AccessHint access_hint_on_compaction_start = NORMAL;
 
-  // The size RocksDB uses to perform readahead during compaction read.
-  // If set zero, RocksDB will sanitize it to be 2MB during db open.
-  // If you're
+  // If non-zero, we perform bigger reads when doing compaction. If you're
   // running RocksDB on spinning disks, you should set this to at least 2MB.
   // That way RocksDB's compaction is doing sequential instead of random reads.
   //
-  //
   // Default: 0
   //
   // Dynamically changeable through SetDBOptions() API.