messages from 2014-09-22 18:40:55 to 2015-07-05 00:18:29 UTC [more...]
[PATCH] test/ruby/test_process.rb: test for fd=3 usability in child
2015-07-05 0:18 UTC
[PATCH] enumerator.c: provide enumerator.rb, too
2015-07-04 21:20 UTC
[PATCH] load.c: prevent enumerator load search
2015-07-04 21:14 UTC
[PATCH] load.c: use fstring for loaded features and expanded load path
2015-07-04 5:29 UTC
[PATCH] string.c: ensure String#freeze resizes internal buffer
2015-07-04 5:25 UTC
[PATCH] vm.c: reduce branches for always-set VM fields
2015-07-04 0:34 UTC
[PATCH] socket: support accept `sock_nonblock: (true|false)'
2015-07-02 1:36 UTC
[PATCH 1/3] reflect changes in Rack::Utils::HTTP_STATUS_CODES
2015-06-30 22:19 UTC (3+ messages)
` [PATCH 2/3] reduce constants and optimize for Ruby 2.2
` [PATCH 3/3] http_response: reduce size of multi-line header path
[PATCH] move RB_GC_GUARD responsibility to rb_add_method_iseq
2015-06-30 20:55 UTC
[PATCH] struct.c: speedup for big structs
2015-06-30 20:11 UTC (2+ messages)
[PATCH] reflect changes in Rack::Utils::HTTP_STATUS_CODES
2015-06-30 0:35 UTC
[PATCH] io.c: reopen stdio streams correctly when given "w+"
2015-06-29 23:20 UTC
[PATCH] delay `exception: false' checks for minor speedup
2015-06-29 21:44 UTC
delay exception checks
2015-06-29 21:00 UTC
[PATCH] unordered
2015-06-29 19:42 UTC
spew
2015-06-29 17:45 UTC
[PATCH] ext/openssl/ossl_ssl.c: raise if kwargs given in blocking mode
2015-06-24 17:29 UTC
[PATCH] st.c: use ccan linked-list
2015-06-24 7:34 UTC
[PATCH] string.c (rb_str_justify): use RB_GC_GUARD
2015-06-24 3:43 UTC
[PATCH 1/2] enum.c: remove volatile, use RB_GC_GUARD
2015-06-24 3:37 UTC (2+ messages)
` [PATCH 2/2] enum.c (enum_minmax): simplify return value creation
[PATCH] use Array instead of custom struct for generic ivars
2015-06-23 23:20 UTC
[PATCH] enum.c: remove volatile, use RB_GC_GUARD
2015-06-23 3:02 UTC
[PATCH] socket: cache common socket families in fptr->mode
2015-06-23 2:11 UTC
[PATCH] socket: cache common socket families in fptr->mode
2015-06-23 2:04 UTC
[PATCH] dir.c (check_dirname): avoid volatile, use return value
2015-06-23 0:39 UTC
[PATCH] struct.c: speedup for big structs
2015-06-22 22:00 UTC
[PATCH] compile.c: convert to use ccan/list
2015-06-17 23:46 UTC
[PATCH] ext/socket/ancdata.c (bsock_recvmsg_internal): reduce stack use
2015-06-17 20:55 UTC
[PATCH] use %+PRIsVALUE instead of rb_inspect
2015-06-17 20:55 UTC
[PATCH] socket: allow explicit buffer for recv and recv_nonblock
2015-06-10 0:23 UTC
[PATCH] socket: allow exception-free nonblocking sendmsg/recvmsg
2015-06-05 23:16 UTC
[PATCH] socket: allow exception-free nonblocking sendmsg/recvmsg
2015-06-05 0:01 UTC
[PATCH] socket: allow exception-free nonblocking sendmsg/recvmsg
2015-06-04 22:08 UTC
[PATCH] socket: allow exception-free nonblocking sendmsg/recvmsg
2015-06-04 22:07 UTC
[PATCH] variable.c: cleanup removal of special constant support (part2)
2015-06-02 20:14 UTC
[PATCH] variable.c: remove generic ivar support for special constants
2015-06-02 3:05 UTC
[PATCH] variable.c: remove generic ivar support for special constants
2015-06-02 3:04 UTC
[PATCH] sendmsg_nonblock "exception: false"
2015-06-02 2:35 UTC
[PATCH] ext/socket/ancdata.c (bsock_sendmsg_internal): use rb_scan_args
2015-06-02 2:25 UTC
[PATCH] socket: nonblocking sendmsg/recvmsg can be exception-free
2015-06-02 1:51 UTC
[PATCH] avoid compatibility table with generic ivars
2015-05-29 0:39 UTC
[PATCH] lib/*: use monotonic clock for timeouts
2015-05-27 21:11 UTC
[PATCH] openssl: use RB_GC_GUARD instead of volatile
2015-05-27 0:43 UTC
[PATCH] socket: avoid redundant fcntl on Linux
2015-05-26 20:47 UTC
[PATCH 1/2] variable.c: extract common functions for generic ivar indices
2015-05-23 1:24 UTC (2+ messages)
` [PATCH 2/2] variable.c: use indices for generic ivars
[PATCH 1/2] variable.c: extract common functions for generic ivar indices
2015-05-21 2:51 UTC (2+ messages)
` [PATCH 2/2] variable.c: use indices for generic ivars
[PATCH] variable.c: generic_iv_tbl is unavoidable
2015-05-12 21:05 UTC
[PATCH] ext/socket/init.c (rsock_s_recvfrom_nonblock): avoid redundant fcntl
2015-05-12 0:23 UTC
[PATCH 1/2] ext/socket/init.c: use SOCK_NONBLOCK if available
2015-05-12 0:06 UTC (2+ messages)
` [PATCH 2/2] socket: accept_nonblock supports "nonblock: false" kwarg
[PATCH] ext/socket/init.c: use SOCK_NONBLOCK if available
2015-05-11 20:48 UTC
[PATCH 1/3] lib/webrick/utils.rb: simplify by avoiding fcntl
2015-05-11 20:17 UTC (3+ messages)
` [PATCH 2/3] lib/drb/*.rb: avoid redundant fcntl call
` [PATCH 3/3] lib/webrick/server.rb: "
[PATCH] thread.c (thread_start_func_2): warn on uncaught exception
2015-05-08 21:41 UTC
[PATCH 1/2] http_request: support rack.hijack by default
2015-05-07 21:43 UTC (2+ messages)
` [PATCH 2/2] avoid extra allocation for hijack proc creation
[PATCH] favor kgio_wait_readable for single FD over select
2015-05-07 20:24 UTC
[PATCH] lib/*: use monotonic clock for timeouts
2015-05-06 22:29 UTC
[PATCH] ext/socket/ancdata.c (bsock_recvmsg_internal): GC guard
2015-05-06 21:22 UTC
[PATCH] ext/socket/ancdata.c (bsock_recvmsg_internal): GC guard
2015-05-06 21:20 UTC
proxy_pass race fix
2015-04-21 4:13 UTC
[PATCH] stringio: avoid STR_NOFREE due to constant string
2015-04-17 1:40 UTC
[PATCH] connect_nonblock(..., exception: false) does not raise EISCONN
2015-04-16 2:25 UTC
[PATCH] experimental test to disable buffer sharing
2015-04-14 20:36 UTC
[PATCH] lib/net/*: use io/wait methods instead of IO.select
2015-04-13 2:22 UTC
[PATCH] wbuf: fix writev
2015-04-07 21:52 UTC
[PATCH] lib/net/protocol.rb (rbuf_fill): avoid exception with read_nonblock
2015-04-07 19:35 UTC
[PATCH] connect_nonblock supports "exception: false"
2015-04-01 20:42 UTC
[PATCH 0/5] new ccan/list functionality
2015-03-30 6:27 UTC (11+ messages)
` [RESEND PATCH 1/5] list: list_add_after and list_add_before functions
` [RESEND PATCH 2/5] list: list_swap to exchange elements
` [PATCH 3/5] list: new list_for_each{,_safe}_off_dir_ macros
` [PATCH 4/5] list: add list_for_each_rev_off macro
` [PATCH 5/5] list: add list_for_each_rev_safe{,_off} macros
` [ccan] [PATCH 5/5] list: add list_for_each_rev_safe{, _off} macros
[WIP] use rb_get_kwargs for exception:false
2015-03-12 3:00 UTC
[PATCH] st.c (st_numhash): mix float value
2015-01-22 0:49 UTC
[PATCH] fix flonum hashing regression from r45384
2015-01-21 22:15 UTC (2+ messages)
` flonum hash fix regression benchmark results
[PATCH] add flonum hash key benchmark
2015-01-21 20:54 UTC
[PATCH] thread: micro-optimize thread create/join
2015-01-17 2:49 UTC
wip-flonum-test-fix
2015-01-12 21:21 UTC
[PATCH] symbol.c: fix memory leak from global fstr hash
2015-01-01 1:48 UTC
[PATCH] symbol.c (rb_gc_free_dsymbol): delete from global fstr hash
2015-01-01 0:48 UTC
[PATCH] uri optimizations, too ugly
2014-12-22 21:53 UTC
[PATCH] uri optimizations, too ugly
2014-12-22 21:03 UTC
[PATCH] compile.c: skip opt_* insns for fstr args with block given
2014-12-16 21:13 UTC
[PATCH] compile.c: skip opt_* insns for fstr args with block given
2014-12-16 21:09 UTC
avoid duplicate block
2014-12-06 0:52 UTC
[RFC] struct: avoid all O(n) behavior on access
2014-12-06 0:23 UTC
[PATCH] struct: initial cut at defining optimized methods
2014-12-04 0:29 UTC
[PATCH] iseq.c: remove unused rb_iseq_new_with_bopt
2014-12-04 0:23 UTC
[PATCH] mostly fix rb_iseq_load
2014-12-03 22:05 UTC
[PATCH] iseq.c (iseq_data_to_ary): keep hidden variables
2014-12-01 22:17 UTC
[PATCH] rb_iseq_load-fix-v5@r48597
2014-11-27 2:58 UTC
[PATCH 1/2] compile.c (iseq_calc_param_size): hoist out of iseq_set_arguments
2014-11-26 8:02 UTC (2+ messages)
` [PATCH 2/2] wip - rb_iseq_load fix v4
[PATCH 1/2] compile.c (iseq_build_callinfo_from_hash): hoist out
2014-11-25 1:31 UTC (2+ messages)
` [PATCH 2/2] fix rb_iseq_load for C extensions (v3)
[PATCH] fix rb_iseq_load for C extensions
2014-11-23 5:55 UTC (2+ messages)
` [WIP] rb_iseq_load bugs
[PATCH] fix rb_iseq_load, a work-in-progress
2014-11-22 0:02 UTC
[PATCH 1/2] accept_nonblock supports "exception: false"
2014-11-20 20:17 UTC (2+ messages)
` [PATCH 2/2] openssl: support accept_nonblock(exception: false)
[PATCH 1/2] compile.c: hide "literal" optimizations behind peephole optimize
2014-11-16 9:21 UTC (3+ messages)
` [PATCH 2/2] opt_str_lit-v8
` [PATCH 2/2 v2] opt_str_lit-v9
[PATCH] hopefully fix signed-vs-unsigned warnings on all compilers
2014-11-12 21:50 UTC
[PATCH] opt_str_lit-v7
2014-11-06 0:20 UTC
[PATCH 4/4] opt_str_lit_yoda: account for method redefinition in arg
2014-11-05 23:18 UTC
[PATCH 3/4] template/opt_method.h.tmpl: explicitly cast to unsigned
2014-11-05 23:18 UTC
[PATCH 2/4] remove rb_undo_opt_str_lit
2014-11-05 23:17 UTC
[PATCH] gc.c (struct heap_page): trivial packing
2014-11-03 1:21 UTC
[PATCH 1/3] lib/uri/rfc2396_parser.rb (initialize): reduce bytecode size
2014-11-02 20:12 UTC (3+ messages)
` [PATCH 2/3] support ENV.freeze
` [PATCH 3/3] opt_str_lit*: avoid string literal allocations
[PATCH] thread.c: reinitialize keeping mutexes on fork
2014-10-28 21:47 UTC
[PATCH v4] opt_str_lit*: avoid string literal allocations
2014-10-26 0:51 UTC (2+ messages)
[PATCH] signal.c: avoid repeated disable/enable interrupt
2014-10-25 8:30 UTC
[PATCH] support ENV.freeze
2014-10-20 20:23 UTC
[RFC] opt_str_lit: optimize string literals in many cases
2014-10-18 2:51 UTC
[PATCH 01/18] compile.c: move "literal" optimizations to peephole optimize
2014-10-18 2:42 UTC (18+ messages)
` [PATCH 02/18] add generic and flexible opt_str_lit insn
` [PATCH 03/18] compile.c: optimize << and == using putstring_for
` [PATCH 04/18] opt_str_lit: further optimizations and cleanups
` [PATCH 05/18] opt_str_lit: optimize allocations for +, %, * and === calls
` [PATCH 06/18] vm: automatically define optimized method enums
` [PATCH 07/18] fix mismerge
` [PATCH 08/18] optimize string allocations for sub/gsub/tr/tr_s(!)
` [PATCH 09/18] compile.c (opt_str_lit_1): hoist out of iseq_peephole_optimize
` [PATCH 10/18] test/ruby/test_string.rb: cleanup allocation tests
` [PATCH 11/18] test_string: cleanup tests
` [PATCH 12/18] opt_method.inc.tmpl: flatten
` [PATCH 13/18] opt_str_lit: switch to type mask for raw class comparisons
` [PATCH 14/18] opt_str_lit: optimize delete
` [PATCH 15/18] opt_str_lit: optimize include?
` [PATCH 16/18] opt_str_lit: disable optimization on mismatch
` [PATCH 17/18] opt_str_lit: avoid allocations for Time#strftime
` [PATCH 18/18] opt_str_lit: optimize a lot more easy cases (untested)
[PATCH 01/13] compile.c: move "literal" optimizations to peephole optimize
2014-10-17 6:19 UTC (13+ messages)
` [PATCH 02/13] add generic and flexible opt_str_lit insn
` [PATCH 03/13] compile.c: optimize << and == using putstring_for
` [PATCH 04/13] opt_str_lit: further optimizations and cleanups
` [PATCH 05/13] opt_str_lit: optimize allocations for +, %, * and === calls
` [PATCH 06/13] vm: automatically define optimized method enums
` [PATCH 07/13] fix mismerge
` [PATCH 08/13] optimize string allocations for sub/gsub/tr/tr_s(!)
` [PATCH 09/13] compile.c (opt_str_lit_1): hoist out of iseq_peephole_optimize
` [PATCH 10/13] test/ruby/test_string.rb: cleanup allocation tests
` [PATCH 11/13] test_string: cleanup tests
` [PATCH 12/13] opt_method.inc.tmpl: flatten
` [PATCH 13/13] opt_str_lit: switch to type mask for raw class comparisons
[PATCH 1/4] compile.c (opt_str_lit_1): hoist out of iseq_peephole_optimize
2014-10-14 21:17 UTC (4+ messages)
` [PATCH 2/4] test/ruby/test_optimization.rb (test_hash_aset_with): assert assignment
` [PATCH 3/4] test/ruby/test_optimization.rb: redefinition tests for string
` [PATCH 4/4] test/ruby/test_string.rb: cleanup allocation tests
[PATCH] opt_str_lit: one instruction, many optimizations
2014-10-14 2:42 UTC
wip method mask optimization
2014-10-13 19:41 UTC
[PATCH] vm: automatically define optimized method enums
2014-10-11 20:31 UTC
benchmarks for opt_str_lit series
2014-10-10 23:01 UTC
[PATCH] opt_str_lit: one instruction, many optimizations
2014-10-10 0:46 UTC
[PATCH 1/2] compile.c: move "literal" optimizations to peephole optimize
2014-10-09 4:03 UTC (3+ messages)
` [PATCH 2/2] add generic and flexible putstring_for insn
` [PATCH] compile.c: optimize << and == using putstring_for
[PATCH] benchmark/bm_hash_aref_sym*.rb: force static symbols
2014-10-09 3:37 UTC
[PATCH] compile.c: move some "literal" opts to peephole
2014-10-08 1:56 UTC
[PATCH 1/2] optimize: "yoda literal" == string
2014-10-07 0:30 UTC (2+ messages)
` [PATCH 2/2] attempt to optimize obj << "#{interp} string"
[PATCH] optimize: foo == "literal string"
2014-10-06 2:27 UTC
[PATCH 1/3] list: list_add_after and list_add_before functions
2014-10-05 12:23 UTC (3+ messages)
` [PATCH 2/3] list: list_swap to exchange elements
` [PATCH 3/3] list: add list_for_each_rev_safe
[PATCH] make OPT_SUPPORT_JOKE a proper VM option
2014-10-05 10:19 UTC
[PATCH] st.c: use ccan linked-list
2014-10-04 6:21 UTC (7+ messages)
` benchmarks on Intel(R) Xeon(R) CPU E3-1230 v3 @ 3.30GHz
` bench results on AMD Phenom II X4 945
` [PATCH 2/1] st.c: fix up st_foreach* for ccan linked-list
` [PATCH 2/1] st.c: simplify st_foreach/st_foreach_check
` st-ll v2 bench results on AMD FX-8320
[PATCH] ext/zlib/zlib.c: TypedData conversion
2014-10-02 20:43 UTC (2+ messages)
` [PATCH v2] "
[PATCH] st.c (new_size): use next_pow2 function
2014-10-02 18:55 UTC
[PATCH] ruby.h (rb_data_type_t): dsize may be FIXNUM
2014-09-28 2:24 UTC
welcome to spew
2014-09-22 18:40 UTC
page: | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).