erasure-code/consistency: Allow consistency checker to be able to deal with non-4k aligned buffers

A parity returned to the client will be the size of shard 0, which may not be 4k aligned in optimised erasure coding. A parity returned through the encode function will always be 4k aligned. This change makes it so we truncate the encoded data down to the size of the client write before comparing them together in the consistency check so we know we are comparing them like-for-like

Signed-off-by: Jon Bailey <[email protected]>

Author: JonBailey1993

src/erasure-code/consistency/ConsistencyChecker.cc

@@ -5,6 +5,7 @@
 #include "ECReader.h"
 #include "ECEncoder.h"
 #include "ECEncoderSwitch.h"
+#include "osd/ECUtil.h"
 
 using ConsistencyChecker = ceph::consistency::ConsistencyChecker;
 
@@ -115,6 +116,30 @@ bool ConsistencyChecker::check_object_consistency(const std::string& oid,
     return false;
   }
 
+  // Encode always returns 4k aligned data
+  // A client read will always return parity with the same size as shard 0
+  // We confirm the encoded data is the same or larger than the client read
+  ceph_assert(outbl->length() >= data_and_parity.second.length());
+  // We check the difference is not larger than the page size multipled by the
+  //    number of parities
+  ceph_assert(outbl->length() - data_and_parity.second.length()
+                <= (encoder.get_m() * encoder.get_chunk_size()) - encoder.get_m());
+  // We truncate the encoded parity to the size of the client read for comparison
+  if (outbl->length() != data_and_parity.second.length())
+  {
+    const int aligned_shard_length = outbl->length() / encoder.get_m();
+    const int shard_data_length = data_and_parity.second.length()
+                                  / encoder.get_m();
+    ceph::bufferlist newdata;
+    for (int i = 0; i < encoder.get_m(); i++)
+    {
+      ceph::bufferlist bl;
+      bl.substr_of(*outbl, i * aligned_shard_length, shard_data_length);
+      newdata.append(bl);
+    }
+    outbl = std::move(newdata);
+  }
+
   return buffers_match(outbl.value(), data_and_parity.second);
 }
 

src/erasure-code/consistency/ECEncoder.cc

@@ -198,6 +198,17 @@ int ECEncoder<SInfo>::get_m()
 {
   return stripe_info->get_m();
 }
+
+/**
+ * Return chunksize for the stripe
+ *
+ * @returns int Chunksize for the stripe
+ */
+template <typename SInfo>
+int ECEncoder<SInfo>::get_chunk_size()
+{
+  return stripe_info->get_chunk_size();
+}
 }
 }
 

src/erasure-code/consistency/ECEncoder.h

@@ -29,6 +29,7 @@ class ECEncoder {
     std::optional<ceph::bufferlist> do_encode(ceph::bufferlist inbl);
     int get_k(void);
     int get_m(void);
+    int get_chunk_size(void);
 };
 }
 }

src/erasure-code/consistency/ECEncoderSwitch.cc

@@ -57,3 +57,17 @@ int ECEncoderSwitch::get_m()
     return encoder_legacy.get_m();
   }
 }
+
+/**
+ * Return chunksize for the stripe from the correct version of the encoder
+ *
+ * @returns int Chunksize for stripe
+ */
+int ECEncoderSwitch::get_chunk_size()
+{
+  if (optimizations_enabled) {
+    return encoder_optimized.get_chunk_size();
+  } else {
+    return encoder_legacy.get_chunk_size();
+  }
+}

src/erasure-code/consistency/ECEncoderSwitch.h

@@ -26,6 +26,7 @@ class ECEncoderSwitch {
     std::optional<ceph::bufferlist> do_encode(ceph::bufferlist inbl);
     int get_k(void);
     int get_m(void);
+    int get_chunk_size(void);
 };
 }
 }