Andrey Kuzmin (1): Add support for waiting for other jobs by name Fabrice Bacchella (1): rewrote libhdfs engine Jens Axboe (43): server: rewrite message handling server: make the io log transmit use the new infrastructure server: cleanup and proper error returns Fixup and improve per-thread data server: comment sk_entry struct and remove hole server: ensure each connection sets up its own sk_out server: comments, and clear sk_out->sk when we close the 'sk' server: cleanup exported functions server: don't pass sk_out into accept loop server: create sk_out_key earlier client: make SEND_ETA timeout non-fatal server: bump listen() backlog Merge branch 'server' Rework 'dump_cmdline' First stab at adding job options to json output stat: work around 'numjobs' oddity in dumping job output client: cleanup json output client/server: pass back job options to client client: fix segfault for !json output client/server: ensure we don't overrun memory for long option values parse: fix segfault if posval option doesn't have a value verify: dump failed header, if verification fails client: fix double free of command on timeout gclient: don't free pdu on iolog return Ensure that command line options also end up in json output client/server: various bug fixes Fix compile warning for !zlib client/server: transparent handling of storing compressed logs configure: fix zlib typo server: remove leftover debug statement server: code cleanups client: iolog cleanups server: don't store command tag on the stack server: command tag cleanup backend: count iterative bytes for progress backend: only terminate multiple loops if we did no IO Make options mask a 64-bit type parse: ensure that option group/category hold 64-bit gfio: update for 64-bit option category/group backend: continue on failed wait_for() check, not break arch: add ffs64 and ffz64 gfio: use ffz64 for 64-bit types Fio 2.3 shoichi.sawada@toshiba.co.jp (1): Fix runtime of terse output