| Age | Commit message (Collapse) | Author |
|---|
|
We only tested that ValueTest.asn1 would compile, no that the
values were correct.
|
|
Don't build a record using the tuple syntax. It is bad in case
we would want to change the record definition later.
|
|
|
|
* beam_utils:joineven/1 -> beam_utils:join_even/1
* beam_utils:split_even/1 -> beam_utils:split_even/1
|
|
The get_map_elements instruction has been removed from all blocks by the
mandatory beam_split pass and thus only needs handling by the outer loop.
|
|
OTP-12394
* RoadRunnr/crypto-ecc-gen-key:
[crypto] enhance generate_key for ECC keys
|
|
|
|
* bjorn/compiler/beam_dead/OTP-12393:
Update the comments that explain what beam_dead does
Improve optimization of bs_start_match2
Extend count_bits_matched/3 to handle the UTF instructions
misc_SUITE: Cover the exception handling code in beam_dead
Generalize optimizations using shortcut_rel_op/4
beam_dead: Optimize branches from relational conditionals
|
|
|
|
HiPE on ARM is currently severely broken if the rest of the VM is
compiled to run in Thumb mode -- calling native code quickly ends
up executing code in the wrong mode and crashing the VM. This is
a problem on e.g. Ubuntu which configures its system GCC to generate
Thumb by default. It can also be triggered by overriding CC or
CFLAGS when compiling the VM.
There were three issues that caused the breakage:
1. Assembly-coded functions in hipe_arm_glue.S weren't explicitly
tagged as functions, preventing the linker from generating the
correct mode-switching call instructions for calls from C to
these functions.
Fixed by tagging those symbols as functions.
2. A few BIF wrappers were so simple that they performed tailcalls
to the C BIFs. This fails to switch mode when C is in Thumb.
Fixed by performing ordinary recursive calls when C is in Thumb.
3. The assembly-coded source files weren't explicitly tagged as ARM.
Tested with the HiPE testsuite on ARMv7, with the VM built as ARM
and as Thumb. Also manually inspected the object code for the beam
executable and checked that call sites from C to HiPE's ARM runtime
code and vice versa used the correct mode-switching instructions.
|
|
hipe_rtl:phi_remove_pred/2 can produce a #move{} instruction with
floating-point temporaries as operands, even though such moves MUST
be #fmove{} instructions.
Added type checks to the #move{} and #fmove{} constructor and setter
functions to ensure that similar mishaps cannot happen again.
|
|
RTL can produce an #fconv{} instruction with an immediate operand, but
the backends unconditionally access the operand as a temporary. This
results in broken representation in the backends and eventually they
crash.
|
|
No need to check for fail label zero for get_map_elements in beam_split.
get_map_elements is always used in pattern matching and never in a body.
|
|
The normalization in beam_flatten can never be reached for get_map_elements
since it is always handled in the mandatory beam_split pass.
|
|
This commit covers 'split_block_label_used' in the beam_bool pass for Maps.
|
|
This reverts changes in test_server from commit bb7fb3c31d1939ab39fb14322d11a59bc34d8f42.
|
|
|
|
While we are, clean up the comments and rearrange the code for
clarity. Also add a test to cover the last uncovered line in
beam_dead.erl.
|
|
Amend the test suite to call beam_dead as originally intended (and not
beam_block), and modify the input data so that the exception will
occur within the try ... catch block in function/2.
|
|
Better optimizations with less code.
|
|
The BEAM compiler translates code such as:
is_hex_digit(D) when $0 =< D, D =< $9 -> true;
is_hex_digit(D) when $a =< D, D =< $z -> true;
is_hex_digit(D) when $A =< D, D =< $Z -> true;
is_hex_digit(_) -> false.
to something like this:
L0: test is_ge L1 {x,0} 48
test is_ge L1 57 {x,0}
move true {x,0}
return.
L1: test is_ge L2 {x,0} 97
test is_ge L2 122 {x,0}
move true {x,0}
return
L2: test is_ge L3 {x,0} 65
test is_ge L3 90 {x,0}
move true {x,0}
return
L3: move false {x,0}
return
We can see that tests will be repeated even if they cannot possibly
succeed. For instance, if we pass in {x,0} equal to 32, the first
test that {x,0} is greater than or equal to 48 at L0 will fail.
The control will transfer to L1, where it will be tested whether
{x,0} is greater than 97. That test will fail and control
will pass to L2, where again the test will fail.
The compiler can do better by short-circuiting repeating tests:
L0: test is_ge L3 {x,0} 48
test is_ge L1 57 {x,0}
move true {x,0}
return.
L1: test is_ge L2 {x,0} 97
test is_ge L3 122 {x,0}
move true {x,0}
return
L2: test is_ge L3 {x,0} 65
test is_ge L3 90 {x,0}
move true {x,0}
return
L3: move false {x,0}
return
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* tombenner/doc_fixes:
Fix grammar and formatting issues
Fix typo ("to use to use")
Add a comma after "For example" when appropriate
Use colons before <code> when appropriate
Move periods inside parenthetical sentences
Add an apostrophe to contractions of "let us"
|
|
* derek121/mnesia-doc-fixes:
Fix grammar
|
|
This fixes a bug in httpc where redirection URIs could lead to bad requests if they
contained fragments.
|
|
* lemenkov/use_os_getenv_2:
fix missing include
Start using os:getenv/2 fun
Introduce os:getenv/2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
If a DCD dump is desired on-demand, use the function
mnesia_controller:snapshot_dcd(Tables). Tables must be a list of
tables that have a local disc_copy, otherwise an error will be
returned. Once the operation actually executes, any table that doesn't
have a local disc_copy is ignored.
Specifically, the dump_log worker record has been changed to allow an
arity-0 fun instead of the default log dump. This fun will be executed
as if it were a normal log dump, and must return 'dumped'. This could
also be used to e.g. insert a backup operation between log dumps.
|
|
Setting the new Mnesia parameter 'dump_disc_copies_at_startup' to
'false' will completely disable the DCD dumping while tables are being
loaded. If it is set to 'true' (the default), the same test will now
be performed as for normal dumps, i.e., using the 'dc_dump_limit'
parameter. Previously, the test performed at load time was different
from the one used at runtime, and caused a lot of unnecessary dumping
which slowed down the startup.
|
|
It seems like the commit 5a3c4668908254ee930c8db2e5cf9741945f9b2b also
broke the incl_dirs option and friends when given as relative paths.
When a cover spec contains a relative path it is looked-up in the
directory with the test results, not in the .cover file directory.
|
|
|
|
(The support in erl_parse got removed when 'packages' were removed, since
the dot notation was overlaid on the existing mnemosyne access syntax.)
|
|
|
|
* mikpe/fix-eacces-spelling:
fix eacces spelling
|
|
* arekinath/aes-evp/OTP-12380:
crypto: use EVP for AES-CBC
|
|
wxFileDialog:getPaths(FD) always return the empty list
on mac when using file dialog.
|
