| Age | Commit message (Collapse) | Author |
|---|
|
|
|
* rc/r14-gc-fix:
fix 64-bit issues in the garbage collection
OTP-9488
|
|
Use a union for pointer type conversion to avoid compiler warnings
about strict-aliasing violations with gcc-4.1. gcc >= 4.2 does not
emit the warning.
TODO: Reconsider use of union once gcc-4.1 is obsolete?
|
|
We discovered that if a single Erlang process tried to grow above 32
GB (i.e., more 64-bit words than can be counted by a 32-bit number),
the VM failed to find the next larger heap size, even though there
were plenty more heap sizes left to pick from and even though we had a
lot more memory available on the machine. (Obviously, this is only
applicable on 64-bit Erlang.)
It turned out to be due to some 'int' variables in the heap resizing
parts of erl_gc.c not being properly updated to 'Uint' or 'Sint'. Once
that was fixed, I got segfaults instead as soon as the heap got larger
than 2^32 words, due to even more 'int' declarations in the same file,
but now in the GC code.
After fixing this as well, I successfully ran an Erlang node in which
a single Erlang process had a heap so large that I'm not at liberty to
divulge the exact size, but I think the scientific term is
"humongous", and I'm confident that there are no further immediate
problems with very very large individual process heaps.
|
|
|
|
* ms/inet-socket-domain-error:
inet: error if fd does not match socket domain
OTP-9455
|
|
|
|
* pg/fix-hibernate-scheduling-with-hipe:
Fix bug related to hibernate and HiPE (clear F_HIBERNATE_SCHED flag)
OTP-9452
|
|
* bjorn/parallel-make/OTP-9451: (28 commits)
erl_interface: Support parallel make
dialyzer: Remove special-case build in the top Makefile
pcre: Rename Makefile.in to pcre.mk and include it
cos*/src/Makefile: Support parallel make
ic: Support parallel make
orber: Support parallel make
.gitignore: Ignore IDL-GENERATED
public_key: Support parallel make
ssh: Support parallel make
os_mon: Support parallel make
diameter: Support parallel make
snmp: Support parallel make
megaco: Support parallel make
megaco/src/flex/Makefile.in: Support parallel make
*/c_src/Makefile*: Support parallel make
eunit: Support parallel make
gs: Support parallel make
common_test Makefile: Support parallel make
erts/emulator/Makefile.in: Support parallel make
erts: Fix dependency generation
...
|
|
To avoid issues with parallel make and to slightly speed up the
build process, avoid a recursive make by replacing pcre/Makefile.in
with pcre/pcre.mk and including it from the main emulator Makefile.
|
|
|
|
|
|
* dev:
Fix binary construction with huge literal sizes
beam_load.c: Add overflow check of tag values
beam_makeops: Add some sanity checks
Fix construction of <<0:((1 bsl 32)-1)>>
|
|
Constructing binaries using the bit syntax with literals sizes
that would not fit in an Uint will either cause an emulator crash
or the loading to be aborted.
Use the new TAG_o tag introduced in the previous commit to make sure
that the attempt to create huge binary literals will generate a
system_limit exception at run-time.
|
|
The handling of large values for other tags than TAG_i (integer) is
buggy. Any tag value equal to or greater than 2^40 (5 bytes) will
abort loading. Tag values fitting in 5 bytes will be truncated to 4
bytes values.
Those bugs cause real problems because the bs_init2/6 and
bs_init_bits/6 instructions unfortunately use TAG_u to encode literal
sizes (using TAG_i would have been a better choice, but it is too late
to change that now). Any binary size that cannot fit in an Uint
should cause a system_limit exception at run-time, but instead the
buggy handling will either cause an emulator crash (for values in the
range 2^32 to 2^40-1) or abort loading.
In this commit, implement overflow checking of tag values as a
preparation for fixing the binary construction instructions. If any
tag value cannot fit in an Uint (except for TAG_i), change the
tag to the special TAG_o overflow tag.
|
|
We want to make sure that a tag/type name is not defined more than
once and that we don't define too many primitive tags. Primitive
tags must be named with lowercase letters (or they will be confused
with variable names in transformations in the ops.tab file).
|
|
Attempting to construct <<0:((1 bsl 32)-1)>>, the largest bitstring
allowed in a 32 bit emulator, would cause an emulator crash because
of integer overflow.
Fix the problem by using an Uint64 to avoid integer overflow.
Do not attempt to handle construction of <<0:((1 bsl 64)-1>> in
a 64-bit emulator, because that will certainly cause the emulator
to terminate anyway because of insufficient memory.
|
|
|
|
* sverker/allocator-aoff/OTP-9424:
New allocator: Address order first fit (aoff)
|
|
|
|
* sverker/fun_SUITE-refc_dist-gcfix:
Fix test case fun_SUITE:refc_dist
|
|
|
|
* sverker/testcase/OTP-9422:
Test case for OTP-9422
|
|
|
|
It failed sometimes depending on GC invocation.
|
|
|
|
alloc_no of sbmbc_low_alloc was set to ERTS_ALC_A_STANDARD_LOW
|
|
|
|
|
|
Conflicts:
erts/emulator/test/nif_SUITE.erl
erts/emulator/test/nif_SUITE_data/nif_SUITE.c
sverker/enif_make_int64-halfword/OTP-9394
|
|
* sverker/enif_make_int64-halfword/OTP-9394:
Fix halfword bug in enif_make_int64
|
|
|
|
* rickard/sbmbc/OTP-9339:
Use separate memory carriers for small blocks
|
|
* sverker/ets_delete-deadlock-race/OTP-9423:
Fix bug in ets:delete for write_concurrency that could lead to deadlock
|
|
|
|
* rickard/atomics-api/OTP-9014:
Use new atomic API in runtime system
Improve ethread atomics
|
|
* lukas/erts/enif_make_reverse_list/OTP-9392:
Add version comment
Rename enif_get_reverse_list to enif_make_reverse_list
Remove extra allocated heap fragment
Added enif_get_reverse_list to nif API
|
|
A trace matchspec with 'enable_trace' or 'disable_trace' in body could
cause an emulator crash if a concurrent process altered the trace
setting of the traced function by calling erlang:trace_pattern.
The effect was a deallocation of the binary holding the matchspec
program while it was running. Fixed by increasing reference count of
ms-binary in the cases when 'enable_trace' or 'disable_trace' may
cause a system block that may alter the ongoing trace.
The paradox here is that db_prog_match() is using erts_smp_block_system()
to do 'enable_trace' and 'disable_trace' in a safe (atomic) way. But that
also have the (non-atomic) effect that racing thread might block the
system and change the trace settings with erlang:trace_pattern.
|
|
Relocking in ets_delete_1() and remove_named_tab() was done by
unlocking the table without clearing the is_thread_safe flag. A racing
thread could then read-lock the table and then incorrectly
write-unlock the table as db_unlock() looked at is_thread_safe to
determine which kind of lock to unlock.
Several fixes:
1. Make db_unlock() use argument 'kind' instead of is_thread_safe to
determine lock type.
2. Make relock logic use db_lock() and db_unlock() instead of directly
accessing lock primitives.
3. Do ownership transfer earlier in ets_delete_1 to avoid racing owner
process to also start deleting the same table.
|
|
concat_binary/1 was deprecated in R13B04, but already in
the R10B-2 release, the documentation recommends using
list_to_binary/1 instead.
|
|
|
|
|
|
The bug was creating an invalid bignum instead of a small integer,
causing strange comparing behavior (=:= failed but == succeeded).
|
|
|
|
|
|
All uses of the old deprecated atomic API in the runtime system
have been replaced with the use of the new atomic API. In a lot of
places this change imply a relaxation of memory barriers used.
|
|
The ethread atomics API now also provide double word size atomics.
Double word size atomics are implemented using native atomic
instructions on x86 (when the cmpxchg8b instruction is available)
and on x86_64 (when the cmpxchg16b instruction is available). On
other hardware where 32-bit atomics or word size atomics are
available, an optimized fallback is used; otherwise, a spinlock,
or a mutex based fallback is used.
The ethread library now performs runtime tests for presence of
hardware features, such as for example SSE2 instructions, instead
of requiring this to be determined at compile time.
There are now functions implementing each atomic operation with the
following implied memory barrier semantics: none, read, write,
acquire, release, and full. Some of the operation-barrier
combinations aren't especially useful. But instead of filtering
useful ones out, and potentially miss a useful one, we implement
them all.
A much smaller set of functionality for native atomics are required
to be implemented than before. More or less only cmpxchg and a
membar macro are required to be implemented for each atomic size.
Other functions will automatically be constructed from these. It is,
of course, often wise to implement more that this if possible from a
performance perspective.
|
|
|
|
The efile driver will now use chunked data on list_dir. This will
lessen the number of sends to prim_file and hence improve
performance. This method is utilized in both direct and async cases.
|
|
If an IPv4 fd is opened as an IPv6 socket, unexpected behaviour can
occur. For example, if an IPv4 UDP socket is opened and passed into
Erlang as an IPv6 socket, the first 3 bytes (corresponding to 1 byte
representing the protocol family, 2 bytes set to the port) are stripped
from the payload. The cause of the UDP payload truncation happens in
inet_drv.c:packet_inet_input when a call to inet_get_address fails
silently because the family is set to PF_INET6 but the buffer len is
the size of an IPv4 struct sockaddr_in.
Prevent this behaviour by checking that the protocol family of the file
descriptor matches the family of the requested Erlang socket.
{ok, S1} = gen_udp:open(0, [binary, inet]),
{ok, FD} = inet:getfd(S1),
{ok, Port} = inet:port(S1),
{ok, S} = gen_udp:open(Port, [binary, {fd, FD}, inet6]),
{ok, C} = gen_udp:open(0, [binary]),
Msg = <<1,2,3,4,5>>,
gen_udp:send(C, "127.0.0.1", Port, Msg),
receive
{udp, S, _, _, Msg} -> ok;
{udp, S, _, _, NewMsg} -> {error, Msg, NewMsg}
end.
This test results in: {error,<<1,2,3,4,5>>,<<4,5>>}
Thanks to Andrew Tunnell-Jones for finding the bug and the test case!
|
