| Age | Commit message (Collapse) | Author |
|---|
|
|
|
Sometimes the beam_bool pass wants to know whether an
y register will be killed by the code that follows and
will do (effectively):
beam_utils:is_killed({y,Y}, Code, L)
When asked to calculate the liveness for an y register,
beam_utils:is_killed/3 will loop forever if the code
includes a receive loop.
Since this rarely occurs, fix the problem in the simplest
and most conservative way.
Reported-by: Christopher Williams
|
|
When gc_bif instructions occurred outside of a block,
beam_utils:check_liveness/3 did not take into account
that the instruction could do a garbage collection, and
could falsely report that an x register would be killed.
That could cause the beam_dead pass to make the code
unsafe by removing the assignment to an x register that
would subsequently be referenced by the garbage collector.
Reported-by: Christopher Williams
|
|
Moving of allocation instructions upwards in the instruction
stream (in order to enable further optimizations) in beam_block,
is implemented with the assumption that if a register {x,X}
contains a valid term, then all other x register with lower
numbers than X also contain valid terms. That assumption is
true after code generation.
The beam_utils:live_opt/1 optimization, however, may invalidate
that assumption. For instance, if a receive statement exports a
variable that is used, but the return value of the receive statement
is not used, then {x,1} but not {x,0} contains a valid term at the
end of the receive statement. If the receive statement is
followed by
{bif,self,{f,0},[],{x,0}}.
{test_heap,NumberOfWords,2}.
moving the allocation upwards will produce
{test_heap,NumberOfWords,2}.
{bif,self,{f,0},[],{x,0}}.
which will cause the beam_validator pass to scream loudly that
{x,0} is not live at the test_heap instruction.
Fix the problem by doing the optimizations in reverse order.
Reported-by: Jim Engquist
|
|
* bjorn/compiler-bin-generators/OTP-8864:
core_lint: Enforce that tail segments only occur at the end
Don't generate multiple tail segments in binary matching
Factor out some of the code for binary generators
Add tests for tail segments in binary generators
|
|
Code such as
foo(A) -> <<A:0>>.
would cause a compiler crash.
|
|
|
|
The code for generating the string table (which is now
only used for bit syntax matching) in a BEAM file is quite
complicated and potentially expensive when compiling modules
with many thousands of clauses doing bit syntax matching.
Simplify and optimize the code using bit syntax and
binary:match/2 instead of the list operations in the
original code.
|
|
warn_unused_import works correctly (does not give warnings when overridden).
Local call in guard gives its own error pointing out the local/imported
function.
Use of the phrase "overridden auto-imported bif" instead of "redefined
auto-imported bif" in textual error messages.
|
|
|
|
Wrote and changed some tests in stdlib:erl_lint_SUITE
nowarn_bif_clash is obsoleted but will remove warnings and errors about bif
clashes. The recommended way is to use no_auto_import directives instead.
Hopefully erlang.erl is the only user in the world of nowarn_bif_clash.
|
|
Added only a few testcases in compiler:error_SUITE and guard_SUITE
The new behaviour of warnings and errors when overriding autoimported BIF's:
Bifs that were autoimported before R14 are dangerous because old code
using them and overriding them in exports can start behaving
differently. For newly added autoimports this can't happen to the new
code that wants to (or dont want to) use them, why only warnings are
added for the BIFs autoimported after the compilator change. Errors
are issued only for code that could have worked in one way in R13 and
now will behave in a different way.
If overriding autoimport with local function:
- if explicit -compile directive supresses autoimport
-> no message
else
- if called from inside module
- if pre R14 autoimported bif
-> error
else
-> warning
else
-> no message
If overriding autoimport with import directive
- if explicit -compile directive supresses autoimport
-> no message
else (regardless of actual usage)
- if pre R14 autoimported bif
-> error
else
-> warning
Calls of local functions or imports overriding autoimported functions
(either post R14 or by using explicit -compile supressions of
autoimport) always goes to the local function or the imported.
The compileation errors are added to not let code like this silently
and disastrously change its semantic (probably to an infinite loop)
between R13 and R14:
----------
-module(m).
-export([length/1]).
length(X) ->
...
Y = length(Z),
....
----------
The user has to select if he/she wants to call length in 'erlang' explicitly
or if the overriding semantics is desired, in which case the -compile
directive has to be used.
-compile({no_auto_import,[F/A]}). Is added to allow to override the
autoimports so that code gets unanbiguous. The directive will remove
an autoimport even if there is no local function or import overriding,
because any other behaviour would be inconsistent and confusing.
record_info and module_info can never be overridden.
|
|
* bg/compiler-cover-and-clean:
v3_life: Remove clause that cannot match in match_fail/3
v3_life tests: Cover exception handling code in v3_life:function/1
beam_type: Remove redundant clause
v3_core tests: Cover make_bool_switch_guard/5
v3_core tests: Cover handling of pattern aliases
v3_core: Remove a clause in is_simple/1 that cannot match
v3_core: Remove unused support for generating compilation errors
Remove stray support for the put_literal/2 instruction
Remove stray support for the bs_bits_to_bytes2/2 instruction
Remove the bs_bits_to_bytes/3 instruction
Cover handling of 'math' BIFs
beam_bool: Remove a clause in live_regs/1 that cannot match
beam_bool: Cover handling of bs_context_to_binary in initialized_regs/2
beam_bool: Remove a clause in initialized_regs/2 that cannot match
beam_block: Remove a clause that will never be executed
erts: Stop supporting non-literal empty tuples
compile: Remove code that is only executed on Solaris
Do not cover-analyze core_scan
core_SUITE_data: Don't ignore *.core files in this directory
OTP-8636 bg/compiler-cover-and-clean
|
|
|
|
|
|
|
|
Extend float_SUITE to cover the handling of all BIFs in
the 'math' module. This change will increase test coverage
of erl_bifs and erl_type.
|
|
Add a test case (derived from dets_v9:scan_skip/8), to cover the
clause in initialized_regs/2 that handles bs_context_to_binary.
|
|
Compiling from *.core files is not a supported feature, and writing
test cases to improve the coverage of core_scan will not help
us to find more bugs in the compiler.
|
|
|
|
Add the gc_bif's to the VM.
Add infrastructure for gc_bif's (guard bifs that can gc) with two and.
three arguments in VM (loader and VM).
Add compiler support for gc_bif with three arguments.
Add compiler (and interpreter) support for new guard BIFs.
Add testcases for new guard BIFs in compiler and emulator.
|
|
-opaque declarations should not be retained in the attributes
(because they will be loaded along with the code and are not
useful).
While at it, filter away those Dialyzer attributes as early
as possible - in v3_kernel.
|
|
* bg/opt-receive:
Test that gen_server:call/2,3 are fast even with a huge message queue
erts: Add tests for the receive optimization
Update primary bootstrap
erts: Implement recv_mark/1 and recv_set/1 for real
compiler tests: Cover the error handling code in beam_receive
compiler test: Test optimization of receive statements
Optimize selective receives in the presence of a large message queue
Introduce the new recv_mark/1 and recv_mark/1 instructions
Compile tests that communicate with R12 nodes with the r12 option
Move p_run/2 to test_lib
gen: Inline wait_resp_mon/2 to help the compiler optimize
OTP-8623 bg/opt-receive
reveive statements that can only read out a newly created reference are now
specially optimized so that it will execute in constant time regardless of
the number of messages in the receive queue for the process. That
optimization will benefit calls to gen_server:call(). (See gen:do_call/4
for an example of a receive statement that will be optimized.)
|
|
|
|
We don't attempt to run the generated code, but use beam_disasm
and check for the presence or absence (as appropriate) of the
recv_mark/1 and recv_set/1 instructions.
|
|
It can be useful for other test suites.
|
|
* bg/compiler-suppress-result-ignored:
compiler tests: Eliminate "result of expression is ignored" warnings
Silence warnings for expressions that are assigned to "_"
OTP-8602 bg/compiler-suppress-result-ignored
It is now possible to suppress the warning in code such as
"list_to_integer(S), ok" by assigning the ignored value "_" like this: "_ =
list_to_integer(S), ok".
|
|
Original patch from YAMASHINA Hio posted to erlang-patches@
on Tue Jun 12 11:27:53 CEST 2007:
http://www.erlang.org/pipermail/erlang-patches/2007-June/000182.html
http://fleur.hio.jp/pub/erlang/record2.patch
Only had to do minor changes to port the patch to the
current R14A development tree.
Also added compiler/record_SUITE:nested_access/2 to test
nested record access with or without parentheses.
With this change the following will work.
-record(nrec0, {name = <<"nested0">>}).
-record(nrec1, {name = <<"nested1">>, nrec0=#nrec0{}}).
-record(nrec2, {name = <<"nested2">>, nrec1=#nrec1{}}).
nested_access() ->
N0 = #nrec0{},
N1 = #nrec1{},
N2 = #nrec2{},
<<"nested0">> = N0#nrec0.name,
<<"nested1">> = N1#nrec1.name,
<<"nested2">> = N2#nrec2.name,
<<"nested0">> = N1#nrec1.nrec0#nrec0.name,
<<"nested0">> = N2#nrec2.nrec1#nrec1.nrec0#nrec0.name,
<<"nested1">> = N2#nrec2.nrec1#nrec1.name,
<<"nested0">> = ((N2#nrec2.nrec1)#nrec1.nrec0)#nrec0.name,
N1a = N2#nrec2.nrec1#nrec1{name = <<"nested1a">>},
<<"nested1a">> = N1a#nrec1.name,
N2a = N2#nrec2.nrec1#nrec1.nrec0#nrec0{name = <<"nested0a">>},
N2b = ((N2#nrec2.nrec1)#nrec1.nrec0)#nrec0{name = <<"nested0a">>},
<<"nested0a">> = N2a#nrec0.name,
N2a = N2b,
ok.
Signed-off-by: Tuncer Ayaz <[email protected]>
|
|
* bg/deprecations:
test suites: Remove incidental use of deprecated concat_binary/1
Postpone removal of concat_binary/1
Remove deprecated lists:flat_length/1
OTP-8584 bg/deprecations
|
|
|
|
There is currently no zero-cost way to silence the warning
"the result of the expression is ignored", which is issued
for code such as:
list_to_integer(S),
ok
Such code can be useful for assertions or input validation.
Teach the compiler to silence the warning for expressions
that are explicitly assigned to to the "_" variable,
such as:
_ = list_to_integer(S),
ok
Implement it by having the v3_core pass annotate calls in
Core Erlang like this:
let <_> = ( call 'erlang':'list_to_integer'(S) -| ['result_not_wanted'] )
in 'ok'
and modifiy sys_core_fold to suppress the warning for any call
having the annotation.
We deliberately do not make it possible to silence the warnings
for expressions like:
{build,an,unnecessary,term}, ok
or
is_list(L), ok
because we don't know of any real-world scenarios in which that would
be useful.
|
|
Don't use the deprecated concat_binary/1 BIF in test suites
that don't specifically test concat_binary/1 itself.
|
|
* bg/compiler-inliner:
pmod_SUITE: Again test inlining parameterized modules
compiler tests: Cope with missing args in function_clause for native code
compiler tests: Compile a few more modules with 'inline'
Consistently rewrite an inlined function_clause exception to case_clause
compiler tests: Test the 'inline' option better
compiler: Suppress bs_context_to_binary/1 for a literal operand
compiler: Fix binary matching bug in the inliner
sys_core_inline: Don't generated multiple compiler_generated annos
OTP-8552 bg/compiler-inliner
Several problems in the inliner have been fixed.
|
|
Commit 91de9d0670c6fe1cff08cefa6e1c396effba47b8 stopped testing
inlining of parameterized modules, because of a bug in the inliner.
|
|
Native-compiled code generates a different stack trace for
function_clause exceptions - instead of the arguments for the
function, only the arity is reported. Accept missing arguments
if the test suite is native-compiled.
|
|
Since a function_clause exception in an inlined function will
be changed to a case_clause exception, we must test for both.
|
|
Clone some test suites and compile them with the 'inline' option
to test inlining more thorughly.
|
|
* bg/compiler-remove-r11-support:
compiler: Don't support the no_binaries option
erts: Don't support the put_string/3 instruction
compiler: Don't support the no_constant_pool option
compiler: Don't support the r11 option
test_server: Don't support communication with R11 nodes
binary_SUITE: Don't test bit-level binary roundtrips with R11 nodes
erts: Test compatibility of funs with R12 instead of R11
OTP-8531 bg/compiler-remove-r11-support
|
|
The no_binaries option terminates the compiler with an error
if any bit syntax is used in the module being compiled.
(It used to be implied by the removed r11 option.)
|
|
The no_constant_pool option was implied by the r11 option. It turns
off the usage of the constant (literal) pool, so that BEAM
instructions that use constants can be loaded in an R11 system.
Since the r11 option has been removed, there is no need to
retain the no_constant_pool option.
|
|
The r11 option was used to generate BEAM modules that could
loaded both on both R11 and R12/R13 to facilitate testing
that R11 and R13 nodes could communicate with each other.
Since R14 is only required to be compatible with R12 and
R13 nodes, the r11 option is no longer needed.
|
|
modules. (Thanks to Jebu Ittiachen.)
|
|
* cf/compile_warning_as_error:
Add option -Werror in erlc(1)
compile: add flag warnings_as_errors to treat warnings as errors
compile.erl: remove trailing whitespace
OTP-8382 The -Werror option for erlc and the compiler option
warnings_as_errors will cause warnings to be treated as errors.
(Thanks to Christopher Faulet.)
|
|
With this flag, warnings are treated as errors, like gcc flag '-Werror'.
|
|
* bg/compiler-beam_validator:
beam_validator: fix incorrect assumptions about GC guard BIFs
OTP-8378 In rare circumstances when using garbaging collecting guard BIFs,
the validation pass (beam_validator) would signal that the code
was unsafe, when it in fact was correct. (Thanks to Kiran
Khaladkar.)
|
|
The beam_validator pass incorrectly assumes that a GC guard
BIF (such as length/1) may first do a garbage collection
and then fail. That assumption is not correct (guards BIF
only do garbage collection when it is known that the BIF
call will succeed), and will cause the compiler to reject
valid programs.
Modify the beam_validator to assume that if the branch is
taken for a gc_bif instruction, all registers are unchanged
and no garbage collection has occurred. Also add a comment
in the emulator about that assumption.
|
|
* bg/on_load:
Test on_load functions that don't return 'ok'
Change the expected return value for on_load functions
OTP-8339 The expected return value for an on_load function has been
changed. (See the section about code loading in the Reference
manual.)
|
|
An on_load function is supposed to return 'true' to indicate
that the module should be loaded, and 'false' if it should be
unloaded. But returning any other term, as well as causing an
exception, will also unload the module.
Since we don't like boolean values mixed with other values,
change the expected return value as follows:
* If 'ok' is returned, the module will remain loaded and become
callable.
* If any other value is returned (or an exception is generated),
the module will be unloaded. Also, if the returned value is
not an atom, send a warning message to the error_logger
(using error_logger:warning_msg/2).
The new interpretation of the return value means that an on_load
function can now directly return the return value from
erlang:load_nif/2.
|
|
The following code (by Simon Cornish)
bad(XDo1, XDo2, Do3) ->
Do1 = (XDo1 =/= []),
Do2 = (XDo2 =/= []),
CH1 = if Do1 == true;
Do1 == false,Do2==false,Do3 == blah ->
ch1;
true ->
no
end,
CH2 = if Do1 == true;
Do1 == false,Do2==false,Do3 == xx ->
ch2;
true ->
no
end,
{CH1,CH2}.
is optimized by beam_bool even though the optimization is not
safe. The trouble is that an assignment to {y,0} no longer
occurs on all paths leading to its use.
The bug is in dst_regs/2 which is supposed to return a set
of all registers assigned in a code block, but it ignores
registers assigned in 'move' instructions.
Fix the bug by taking 'move' instructions into account. This change
is safe since it can only cause more registers to be added
to the MustBeKilled and MustBeUnused sets in ensure_opt_safe/6,
which means that it can only cause the optimization to be turned
off for code that used to be optimized.
|
|
The following code crashes beam_bool:
bad(XDo1, XDo2, Do3) ->
Do1 = (XDo1 =/= []),
Do2 = (XDo2 =/= []),
if Do1 =:= true;
Do1 =:= false, Do2 =:= false, Do3 =:= delete ->
no
end.
(Reported by Simon Cornish; minimized by Kostis Sagonas.)
For the moment fix the bug in the simplest and safest way possible
(basically, instead of crashing just don't do the optimization).
In a future major release (e.g. R14), the following improvements
could be considered:
* In beam_bool, it should be possible to move the Do1 and Do2
expressions to the pre-block and still optimize the expression
in the 'if' statement.
* In sys_core_fold, it should be possible to eliminate the
try/catch around the guard expression in the 'if', because
none of the guard tests can actually fail.
|
