aboutsummaryrefslogtreecommitdiff
path: root/core
diff options
context:
space:
mode:
authorDavies Liu <davies@databricks.com>2015-08-14 22:30:35 -0700
committerDavies Liu <davies.liu@gmail.com>2015-08-14 22:30:35 -0700
commit7c1e56825b716a7d703dff38254b4739755ac0c4 (patch)
tree0de34f9c8b81c8787f47ca8973b1bc94e62962ef /core
parent71a3af8a94f900a26ac7094f22ec1216cab62e15 (diff)
downloadspark-7c1e56825b716a7d703dff38254b4739755ac0c4.tar.gz
spark-7c1e56825b716a7d703dff38254b4739755ac0c4.tar.bz2
spark-7c1e56825b716a7d703dff38254b4739755ac0c4.zip
[SPARK-9725] [SQL] fix serialization of UTF8String across different JVM
The BYTE_ARRAY_OFFSET could be different in JVM with different configurations (for example, different heap size, 24 if heap > 32G, otherwise 16), so offset of UTF8String is not portable, we should handler that during serialization. Author: Davies Liu <davies@databricks.com> Closes #8210 from davies/serialize_utf8string.
Diffstat (limited to 'core')
0 files changed, 0 insertions, 0 deletions