Adar Dembo has posted comments on this change. Change subject: rpc: reduce the number of recv syscalls ......................................................................
Patch Set 4: (6 comments) http://gerrit.cloudera.org:8080/#/c/2943/4/src/kudu/rpc/connection.cc File src/kudu/rpc/connection.cc: Line 449 Heh, this old loop never actually looped. http://gerrit.cloudera.org:8080/#/c/2943/4/src/kudu/rpc/transfer.cc File src/kudu/rpc/transfer.cc: Line 78: cur_offset_(4) { Should this also be kMsgLengthPrefixLength? Line 122: // Receive message body plus next length prefix. Maybe add some more context here on the optimization being done? Line 135: } else { If we're in here, 0 < extra < kMsgLengthPrefixLength, right? Line 145: if (cur_offset_ >= total_length_) { How can cur_offset_ be > total_length_? Don't L129 and L131 guarantee that cur_offset_ is always <= total_length_? http://gerrit.cloudera.org:8080/#/c/2943/4/src/kudu/rpc/transfer.h File src/kudu/rpc/transfer.h: Line 70: Status ReceiveBuffer(Socket &socket, gscoped_ptr<InboundTransfer>& next); The pass-by-ref isn't google style guide friendly, but I imagine you're trying to remain consistent with the existing style of the file here. -- To view, visit http://gerrit.cloudera.org:8080/2943 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: I2adbe51b51267deb43b837c8ffae09cf8f54d251 Gerrit-PatchSet: 4 Gerrit-Project: kudu Gerrit-Branch: master Gerrit-Owner: Todd Lipcon <[email protected]> Gerrit-Reviewer: Adar Dembo <[email protected]> Gerrit-Reviewer: Binglin Chang <[email protected]> Gerrit-Reviewer: Henry Robinson <[email protected]> Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Mike Percy <[email protected]> Gerrit-HasComments: Yes
