Merge #15402: Granular invalidateblock and RewindBlockIndex

519b0bc5dc5155b6f7e2362c2105552bb7618ad0 Make last disconnected block BLOCK_FAILED_VALID, even when aborted (Pieter Wuille)
8d220417cd7bc34464e28a4861a885193ec091c2 Optimization: don't add txn back to mempool after 10 invalidates (Pieter Wuille)
9ce9c37004440d6a329874dbf66b51666d497dcb Prevent callback overruns in InvalidateBlock and RewindBlockIndex (Pieter Wuille)
9bb32eb571a846b66ed3bac493f55cee11a3a1b9 Release cs_main during InvalidateBlock iterations (Pieter Wuille)
9b1ff5c742dec0a6e0d6aab29b0bb771ad6d8135 Call InvalidateBlock without cs_main held (Pieter Wuille)
241b2c74ac8c4c3000e778554da1271e3f293e5d Make RewindBlockIndex interruptible (Pieter Wuille)
880ce7d46b51835c00d77a366ec28f54a05239df Call RewindBlockIndex without cs_main held (Pieter Wuille)
436f7d735f1c37e77d42ff59d4cbb1bd76d5fcfb Release cs_main during RewindBlockIndex operation (Pieter Wuille)
1d342875c21b5d0a17cf4d176063bb14b35b657e Merge the disconnection and erasing loops in RewindBlockIndex (Pieter Wuille)
32b2696ab4b079db736074b57bbc24deaee0b3d9 Move erasure of non-active blocks to a separate loop in RewindBlockIndex (Pieter Wuille)
9d6dcc52c6cb0cdcda220fddccaabb0ffd40068d Abstract EraseBlockData out of RewindBlockIndex (Pieter Wuille)

Pull request description:

  This PR makes a number of improvements to the InvalidateBlock (`invalidateblock` RPC) and RewindBlockIndex functions, primarily around breaking up their long-term cs_main holding. In addition:
  * They're made safely interruptible (`bitcoind` can be shutdown, and no progress in either will be lost, though if incomplete, `invalidateblock` won't continue after restart and will need to be called again)
  * The validation queue is prevented from overflowing (meaning `invalidateblock` on a very old block will not drive bitcoind OOM) (see #14289).
  * `invalidateblock` won't bother to move transactions back into the mempool after 10 blocks (optimization).

  This is not an optimal solution, as we're relying on the scheduler call sites to make sure the scheduler doesn't overflow. Ideally, the scheduler would guarantee this directly, but that needs a few further changes (moving the signal emissions out of cs_main) to prevent deadlocks.

  I have manually tested the `invalidateblock` changes (including interrupting, and running with -checkblockindex and -checkmempool), but haven't tried the rewinding (which is probably becoming increasingly unnecessary, as very few pre-0.13.1 nodes remain that would care to upgrade).

Tree-SHA512: 692e42758bd3d3efc2eb701984a8cb5db25fbeee32e7575df0183a00d0c2c30fdf72ce64c7625c32ad8c8bdc56313da72a7471658faeb0d39eefe39c4b8b8474

1 files changed, 3 insertions, 3 deletions

diff --git a/src/rpc/blockchain.cpp b/src/rpc/blockchain.cpp
index 72fec2eaa5..1827aec637 100644
--- a/src/rpc/blockchain.cpp
+++ b/src/rpc/blockchain.cpp
@@ -1579,15 +1579,15 @@ static UniValue invalidateblock(const JSONRPCRequest& request)
     uint256 hash(ParseHashV(request.params[0], "blockhash"));
     CValidationState state;
 
+    CBlockIndex* pblockindex;
     {
         LOCK(cs_main);
-        CBlockIndex* pblockindex = LookupBlockIndex(hash);
+        pblockindex = LookupBlockIndex(hash);
         if (!pblockindex) {
             throw JSONRPCError(RPC_INVALID_ADDRESS_OR_KEY, "Block not found");
         }
-
-        InvalidateBlock(state, Params(), pblockindex);
     }
+    InvalidateBlock(state, Params(), pblockindex);
 
     if (state.IsValid()) {
         ActivateBestChain(state, Params());