| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
A large 64-bit immediate number will be stored as SMALL_BIG_EXT by ETS
compressed format. When uncompressing, the SMALL_BIG_EXT was first
decoded as as bignum (by bytes_to_big) and then turned into a small
(by big_norm). This works for normal "binary_to_term" as
decoded_size() over-estimates the needed heap size. But for ETS no
over-estimation is done as the real term size is known and stored in
DbTerm.
Fixed by preventing bytes_to_big() from writing bignum digit when the
number is seen to fit in an immediate.
|
|
Also added more information when waiting for test proc to die
|
|
* rickard/alloc-opt/OTP-7775:
Optimize memory allocation
Conflicts:
erts/aclocal.m4
erts/emulator/hipe/hipe_bif_list.m4
erts/preloaded/ebin/erl_prim_loader.beam
erts/preloaded/ebin/erlang.beam
erts/preloaded/ebin/init.beam
erts/preloaded/ebin/otp_ring0.beam
erts/preloaded/ebin/prim_file.beam
erts/preloaded/ebin/prim_inet.beam
erts/preloaded/ebin/prim_zip.beam
erts/preloaded/ebin/zlib.beam
|
|
A number of memory allocation optimizations have been implemented. Most
optimizations reduce contention caused by synchronization between
threads during allocation and deallocation of memory. Most notably:
* Synchronization of memory management in scheduler specific allocator
instances has been rewritten to use lock-free synchronization.
* Synchronization of memory management in scheduler specific
pre-allocators has been rewritten to use lock-free synchronization.
* The 'mseg_alloc' memory segment allocator now use scheduler specific
instances instead of one instance. Apart from reducing contention
this also ensures that memory allocators always create memory
segments on the local NUMA node on a NUMA system.
|
|
|
|
* dev:
Cleanup ETS bif's in hipe:erl_bif_types.erl (for dialyzer)
Conflicts:
erts/emulator/beam/erl_db.c
|
|
* pan/erl-bif-types/OTP-9496:
Cleanup ETS bif's in hipe:erl_bif_types.erl (for dialyzer)
|
|
* dev:
Add a high value to test data to catch type cast truncation errors
|
|
|
|
|
|
This commit is a preparation for introducing location information
(filename/line number) in stacktraces in exceptions. Currently
a stack trace looks like:
[{Mod1,Function1,Arity1},
.
.
.
{ModN,FunctionN,ArityN}]
Add a forth element to each tuple that can be used indication
the filename and line number of the source file:
[{Mod1,Function1,Arity1,Location1},
.
.
.
{ModN,FunctionN,ArityN,LocationN}]
In this commit, the fourth element will just be an empty list,
and we will change all code that look at or manipulate stacktraces.
|
|
|
|
|
|
|
|
* sverker/ets_halfword_highmem/OTP-8941:
HALFWORD ETS Fix copyright year in some source files
Fix vm crash in kernel test case seq_trace_SUITE:call
remove NIF compile warning: no previous prototype for ‘nif_init’
Refuse to load NIF library on wrong VM variant (halfword/fullword)
HALFWORD ETS match spec heap fragment optimization
HALFWORD ETS removed eheap and improved test case t_match_spec_run
HALFWORD ETS Further match spec optimization to minimize copying and garbage
HALFWORD ETS db_prog_match optimization
HALFWORD ETS Fix segv for match spec with several function and guards
HALFWORD Make system_info mseg_alloc report both low/high mem
HALFWORD Fix segv caused by erlang:halt
HALFWORD Make more allocators use high mem (binary, fixed and driver)
HALFWORD ETS 32-bit arch fixes and other cleanups
HALFWORD ETS nicer update_element
HALFWORD ETS Real matching on relative terms
HALFWORD first stab at high mem alloc
HALFWORD ETS relative terms
Conflicts:
erts/emulator/test/driver_SUITE.erl
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Did not properly take care of case when TryMeElse restarted
with next match clause.
|
|
|
|
In halfword emulator, make ETS use a variant of the internal term
format that uses relative offsets instead of absolute pointers. This
will allow storage in high memory (>4G). Preprocessor macros (like
list_val_rel(TERM,BASE)) are used to make normal (fullword) emulator
almost completely unchanged while still reusing most of the code.
|
|
Ensure that all threads potentially accessing an ETS-table have dropped
all references to the table before deallocating it.
|
|
Waiting for table sys_dist to stablize after slave node has been stopped.
|
|
|
|
The compressed format is using a slighty modified variant of the extern format
(term_to_binary). To not worsen key lookup's too much, the top tuple itself
and the key element are not compressed. Table objects with only immediate
non-key elements will therefor not gain anything (but actually consume one
extra word for "alloc_size").
|
|
* pan/ets_binary_overhead/OTP-8762:
Remove binary overhead counter from ets objects
|
|
As the overhead counter got larger and never really was needed in ets objects,
I removed them.
A few stray comments of XXX:PaN type from halfword dev removed in the process.
|
|
|
|
* rickard/ethread-rewrite/OTP-8544:
Rewrite ethread library
|
|
Large parts of the ethread library have been rewritten. The
ethread library is an Erlang runtime system internal, portable
thread library used by the runtime system itself.
Most notable improvement is a reader optimized rwlock
implementation which dramatically improve the performance of
read-lock/read-unlock operations on multi processor systems by
avoiding ping-ponging of the rwlock cache lines. The reader
optimized rwlock implementation is used by miscellaneous
rwlocks in the runtime system that are known to be read-locked
frequently, and can be enabled on ETS tables by passing the
`{read_concurrency, true}' option upon table creation. See the
documentation of `ets:new/2' for more information.
The ethread library can now also use the libatomic_ops library
for atomic memory accesses. This makes it possible for the
Erlang runtime system to utilize optimized atomic operations
on more platforms than before. Use the
`--with-libatomic_ops=PATH' configure command line argument
when specifying where the libatomic_ops installation is
located. The libatomic_ops library can be downloaded from:
http://www.hpl.hp.com/research/linux/atomic_ops/
The changed API of the ethread library has also caused
modifications in the Erlang runtime system. Preparations for
the to come "delayed deallocation" feature has also been done
since it depends on the ethread library.
Note: When building for x86, the ethread library will now use
instructions that first appeared on the pentium 4 processor. If
you want the runtime system to be compatible with older
processors (back to 486) you need to pass the
`--enable-ethread-pre-pentium4-compatibility' configure command
line argument when configuring the system.
|
|
* sverker/one_offheap_list/OTP-8737:
One off-heap list, to eliminate two words per ETS object.
|
|
Merging the three off-heap lists (binaries, funs and externals) into
one list. This reduces memory consumption by two words (pointers) per
ETS object.
|
|
erl_db_tree.c incorrectly used NIL (empty list) as "lastkey" to mark
start of the iteration. A real NIL key could then cause a
select or match iteration to be restarted over and over again if the
last key before a trap happended to be NIL.
Changed NIL to THE_NON_VALUE. Should be ok as the initial key value can
never be put into any continuation tuple.
|
|
fragments was created. This will mainly benefit NIFs that return
large compound terms.
|
|
|
