| History log of /src/usr.bin/xlint/lint1/tree.c |
| Revision | | Date | Author | Comments |
| 1.695 |
| 17-Sep-2025 |
rillig | lint: answer two questions from the code by writing tests for them
|
| 1.694 |
| 14-Sep-2025 |
rillig | lint: fix-up prefix commit by adding a rationale and an edge case check
|
| 1.693 |
| 14-Sep-2025 |
rillig | lint: warn about signed '<<' that ends up exactly in the msb
|
| 1.692 |
| 14-Sep-2025 |
rillig | lint: fix integer overflow in '<<' evaluation
Just in case that -ftrapv will someday cover '<<' as well, in addition
to the classic arithmetic operators.
|
| 1.691 |
| 14-Sep-2025 |
rillig | lint: do not preserve bit-field information in integer promotions
The bit-field information was wrongly added back in 2021. At that time,
the integer constraints had not been implemented yet, so it was a
straight-forward idea to implement value restrictions using the type
system and propagage these through the expressions.
Due to preserving the bit-field information, some lint warnings
contained wrong type information. These warnings are now fixed.
|
| 1.690 |
| 14-Sep-2025 |
rillig | lint: add more type details to several messages
In most expressions, the integer promotions are performed on the
operands, making the actual type different from the "visible" type of an
operand. To properly explain the situation, mention the type before the
integer promotions in case it differs from the resulting type.
|
| 1.689 |
| 07-Sep-2025 |
rillig | lint: clean up checks for '<<' and '>>'
|
| 1.688 |
| 06-Sep-2025 |
rillig | lint: reword message for bit-shift in trad-C90 migration mode
Mentioning the type names gives a hint about why the behavior differs;
the message only occurs when the type of the shift amount has higher
rank than the type of the shifted value.
|
| 1.687 |
| 31-Jul-2025 |
rillig | lint: fix return type of GCC's __atomic functions
|
| 1.686 |
| 11-Jul-2025 |
rillig | lint: fix warning about enum array index
Seen in openssh/fmt_scaled.c, scale_chars.
|
| 1.685 |
| 08-Jul-2025 |
rillig | lint: reword message about 'extra bits set to 0'
The word 'extra' in the previous wording was too unspecific to make
sense of. In the reworded message, focus on the problem instead of
mentioning only the effect.
|
| 1.684 |
| 16-May-2025 |
rillig | lint: clean up
|
| 1.683 |
| 04-May-2025 |
rillig | lint: add more details to message about discarding qualifiers
The name of the function (or, if not available, its type) is useful in
lines that contain more than one function call.
|
| 1.682 |
| 12-Apr-2025 |
rillig | lint: warn about constant assignment in operand to '&&' or '||'
Seen in bin/sh/redir.c:1084 for errno, which is fine.
Seen in usr.sbin/mopd/mopprobe/mopprobe.c:235, which looks suspicious
but hasn't been touched since 1997.
|
| 1.681 |
| 12-Apr-2025 |
rillig | lint: s/illegal/invalid/g
Lint does not provide legal advice.
|
| 1.680 |
| 10-Apr-2025 |
rillig | lint: remove now-unnecessary CONSTCOND comments
|
| 1.679 |
| 10-Apr-2025 |
rillig | lint: do not warn about constant operand of '!'
Like its related warning about a constant condition, the purpose of this
warning is unclear, and suppressing the warning requires special lint
comments, thus cluttering the code.
The message itself is still kept in err.c to keep the '-X 239' option
valid. These LINTFLAGS will be removed in a follow-up commit.
The lint comments /* CONSTCOND */, as well as its more verbose forms /*
CONSTANTCOND */ and /* CONSTANTCONDITION */ are no longer needed and can
be removed.
|
| 1.678 |
| 06-Apr-2025 |
rillig | lint: remove warning about 'constant in conditional context'
In a full NetBSD build, this warning occurs about 2400 times, making it
the top 9 warning overall. There is no evidence though that this
warning detects any real bugs, so the noise it makes is not warranted.
In this first step, the warning is still listed in the lint(7) manual
page, as removing it from there would immediately make all places with
"LINTFLAGS += -X 161" fail. These '-X 161' flags cannot be removed
first because then, the warning would step in and make the build fail.
The next step is to clean up the "-X 161" options from all LINTFLAGS,
and then the final step is to remove the warning from err.c.
|
| 1.677 |
| 21-Mar-2025 |
rillig | lint: don't warn when a cast increases the alignment from 1 to n
|
| 1.676 |
| 11-Mar-2025 |
rillig | lint: fix saturated multiplication in integer constraints check
After the change from ui_max_value to si_max_value in tree.c 1.675 from
2025-02-27, it was wrong to divide the maximum value by 2 once more.
|
| 1.675 |
| 27-Feb-2025 |
rillig | lint: fix integer overflow when multiplying two large signed numbers
Seen in ipsec-tools/crypto_openssl.c.
|
| 1.674 |
| 27-Feb-2025 |
rillig | lint: add details to message about conversion in 'case'
|
| 1.673 |
| 24-Feb-2025 |
rillig | lint: remove no-op variant of the unary '&' operator
Neither K&R 1978 nor any of the C standards mentions that &function or
&array would be a no-op. This no-op behavior produced wrong warnings in
sbin/gpt.
|
| 1.672 |
| 20-Feb-2025 |
rillig | lint: add detail to message about truncated constant
|
| 1.671 |
| 20-Feb-2025 |
rillig | lint: mention the assignment operator in 'illegal combination'
In source lines that contain both a 'return' statement as well as a
function-like macro, such as the C11 atomics, seeing the word 'init' in
the diagnostic helps to see that the type conflict is not in the
'return' statement but instead in some initializer.
Seen in userspace-rcu/wfcqueue.h:147.
|
| 1.670 |
| 05-Jan-2025 |
rillig | lint: clean up integer constraints
Since ic_con already handles conversions (and casts), there is no need to
unwrap the conversions (but not the casts) explicitly.
|
| 1.669 |
| 02-Jan-2025 |
rillig | lint: fix possible loss of accuracy in multiplication and division
|
| 1.668 |
| 02-Jan-2025 |
rillig | lint: fix assertion failure in pointer subtraction
|
| 1.667 |
| 02-Jan-2025 |
rillig | lint: fix a few wrong warnings about possible loss of accuracy
|
| 1.666 |
| 02-Jan-2025 |
rillig | lint: compute integer constraints for xor and minus
These operators allow lint to correctly track the possible values in
expressions that sign-extend an integer.
|
| 1.665 |
| 01-Jan-2025 |
rillig | lint: fix wrong warning about possible loss of accuracy
The expression "'0' + random() % 10" always fits in a signed or unsigned
char.
|
| 1.664 |
| 15-Dec-2024 |
rillig | lint: merge duplicate code for possibly confusing precedence
|
| 1.663 |
| 15-Dec-2024 |
rillig | lint: add details to message about possible precedence confusion
|
| 1.662 |
| 30-Nov-2024 |
rillig | lint: add back optimization for non-query mode
The main point of the optimization is to skip the expensive calls to
type_name when preparing the message details. This was not spelled out
explicitly in the comment.
While here, fix the stray 'previous declaration' message that occurred
whenever a function was first declared as 'static' and later defined
without 'static', see query 16.
|
| 1.661 |
| 29-Nov-2024 |
rillig | lint: remove premature optimization for non-query scenarios
|
| 1.660 |
| 23-Nov-2024 |
rillig | lint: use separate message when discarding a qualifier in call
|
| 1.659 |
| 23-Nov-2024 |
rillig | lint: add details to messages about discarding qualifiers
The related message 153 is left as-is, as it is used in two places, and
the second place is not only about discarded qualifiers.
|
| 1.658 |
| 13-Nov-2024 |
rillig | lint: add more details to 'statement not reached' message
In lib/libcompat/regexp/regexp.c, the FAIL macro expands to a compound
statement containing a function call statement and a return statement,
and the macro invocation is followed by a semicolon, forming an extra
empty statement. Which of these statements is unreachable now becomes
clear from the diagnostic, without having to inspect the preprocessed
source code.
|
| 1.657 |
| 31-Oct-2024 |
rillig | lint: allow enum constant named 'N_*' to exceed the array index
Seen in libxcb.
While here, add the name of the enum constant to the message, to quickly
decide whether the identifier is outside the range of expected enum
values, thus marking the number of enum constants.
|
| 1.656 |
| 12-Oct-2024 |
rillig | lint: reduce wrong warnings about loss of accuracy in signed '%'
|
| 1.655 |
| 11-Oct-2024 |
rillig | lint: sort integer constraint functions and branches
No functional change.
|
| 1.654 |
| 10-Oct-2024 |
rillig | lint: reduce debug logging for possible loss of accuracy
When checking whether a type can represent an expression, only do the
debug logging in the case where the expression cannot obviously be
represented by the type, as the successful cases are usually
uninteresting. While here, clean up the nesting in the debug log.
|
| 1.653 |
| 08-Oct-2024 |
rillig | lint: compute integer constraints for unsigned multiplication
This fixes the wrong warning about possible loss of accuracy in
libc/c16rtomb.
|
| 1.652 |
| 28-Sep-2024 |
rillig | lint: reword warning about cast between unrelated pointer types
The phrase "may be troublesome" was too unspecific to be useful.
|
| 1.651 |
| 19-Aug-2024 |
rillig | lint: remove unused integer constraint
The assignment in ic_cond was wrong, it should have been '&' instead of
'|', but as long as '~' is not involved in the integer constraints,
there is no way to demonstrate this bug.
|
| 1.650 |
| 18-Aug-2024 |
rillig | lint: add value propagation for '/' combined with bit operations
Fixes __SHIFTIN/__SHIFTOUT expressions, as in PR toolchain/58617.
|
| 1.649 |
| 10-Jul-2024 |
rillig | lint: add warning for '!(var = constant)'
This catches expressions like 'ATF_CHECK(errno = ENOENT)', which was
intended to use '==' instead of '='.
|
| 1.648 |
| 17-Jun-2024 |
rillig | branches: 1.648.2;
lint: accept C23 [[fallthrough]] statements
Needed by bind9/getaddresses.c:137.
|
| 1.647 |
| 09-Jun-2024 |
rillig | lint: fix usage marker for nested function calls
|
| 1.646 |
| 09-Jun-2024 |
rillig | lint: warn about lossy floating point constant to integer conversions
|
| 1.645 |
| 08-Jun-2024 |
rillig | lint: fix type of GCC-specific __FUNCTION__ to be an array
|
| 1.644 |
| 08-Jun-2024 |
rillig | lint: add details to warnings about negative constant to unsigned
|
| 1.643 |
| 12-May-2024 |
rillig | lint: don't call memcpy with null pointer
Even copying 0 bytes from a null pointer invokes undefined behavior.
|
| 1.642 |
| 11-May-2024 |
rillig | lint: warn when comparing an integer to a floating point constant
|
| 1.641 |
| 10-May-2024 |
rillig | lint: clean up conversion from floating point constant
|
| 1.640 |
| 03-May-2024 |
rillig | lint: measure the alignment in bytes, not bits
While measuring the alignment in bits makes sense when building a struct
type with bit-fields, in all other places it is more confusing than
helpful.
The only visible change is that in debug mode, the format of type sizes
and alignment changed. Since the size of all complete types is a
multiple of a byte (as defined in the C standard), sizes and alignments
are reported in bytes as well. Only while a struct is being built, the
type size may include an additional '+x' for the bits of a bit-field.
|
| 1.639 |
| 01-May-2024 |
rillig | lint: make 'offsetof(t, array-member)' a constant expression
The macro 'offsetof(t, m)' already expanded to a constant expression for
scalar members but not for arrays. This was because the macro expanded
to '(size_t)(((t *)0)->m)', which lint internally represents as
'addr(indir(ptr(0) + offset(m)))', and build_address simplifies
'addr(indir(x))' to 'x' if the types match. The types only match for
scalar types though, but not for arrays.
When build_address happens, the type information is incomplete,
therefore 'offsetof(t, array)' has to be simplified at a later point.
|
| 1.638 |
| 01-May-2024 |
rillig | lint: fix warning about out-of-bounds bit-field value
|
| 1.637 |
| 27-Apr-2024 |
rillig | lint: converting a null pointer to another pointer type is not narrowing
|
| 1.636 |
| 27-Apr-2024 |
rillig | lint: add query for conversion from void pointer to other pointer
|
| 1.635 |
| 12-Apr-2024 |
rillig | lint: clean up and speed up the check for snprintb
|
| 1.634 |
| 31-Mar-2024 |
rillig | lint: merge function call operators 'CALL' and 'ICALL'
|
| 1.633 |
| 30-Mar-2024 |
rillig | lint: in the query about implicit conversions, ignore casts
|
| 1.632 |
| 30-Mar-2024 |
rillig | lint: add query for implicit integer-to-floating conversion
|
| 1.631 |
| 30-Mar-2024 |
rillig | lint: do not convert array subscripts from size_t to ptrdiff_t
The C standards do not specify a fixed type for an array subscript, it
just has to be an integer type. Previously, query 4 fired for the
ubiquitous expression 'ptr[sz]' when sz had type 'size_t'.
The test platform_ilp32_long is unaffected by this change, as the
integer constant 0x80000000 has type 'unsigned int', while size_t is
'unsigned long' on those platforms, and even though the types 'unsigned
int' and 'unsigned long' have the same value space, there's still a
conversion, at least for now.
|
| 1.630 |
| 30-Mar-2024 |
rillig | lint: reword messages about array subscripts to sound more natural
|
| 1.629 |
| 29-Mar-2024 |
rillig | lint: clean up
|
| 1.628 |
| 27-Mar-2024 |
rillig | lint: clean up; extend overflow test
|
| 1.627 |
| 27-Mar-2024 |
rillig | lint: don't use 'long' in diagnostics
The size of 'long' differs between 64-bit and 32-bit platforms.
Eliminate this possible platform-dependency.
|
| 1.626 |
| 25-Mar-2024 |
rillig | lint: fix warnings about loss of accuracy on bit-field operations
|
| 1.625 |
| 19-Mar-2024 |
rillig | lint: keep invalid arguments in function calls
Previously, arguments of incomplete type or 'void' cleared all arguments
of the function call expression, requiring extra checks in later checks.
Invalid function calls are now exported to the .ln files, but that's
irrelevant in practice as these invalid function calls make lint1 fail,
after which xlint removes the .ln file.
|
| 1.624 |
| 12-Mar-2024 |
rillig | lint: fix warning about loss of conversion for unsigned bit-fields
Since decl.c 1.180 from 2021-05-02.
|
| 1.623 |
| 10-Mar-2024 |
rillig | lint: add details to the message about integer overflow
Having only the operator was too unspecific to be actionable, so add the
actual numbers and the data type.
|
| 1.622 |
| 10-Mar-2024 |
rillig | lint: saturate signed integer overflow
In array address calculations, this prevents a 'array subscript cannot
be negative' for large array subscripts.
|
| 1.621 |
| 10-Mar-2024 |
rillig | lint: fix integer overflow detection
Previously, an unsigned operation that had a negative result went
undetected in a few cases. Now, all results that are not representable
by their type are considered overflows.
The implementation of signed shift-right had been wrong for a few
commits.
|
| 1.620 |
| 10-Mar-2024 |
rillig | lint: clean up check for overflow in integer constants
|
| 1.619 |
| 10-Mar-2024 |
rillig | lint: detect more cases of integer overflow in constant expressions
For unsigned integers, detect when 'a + b' wraps around.
|
| 1.618 |
| 10-Mar-2024 |
rillig | lint: split integer overflow check into separate functions
The checks for unsigned and signed integers differ for each operator, so
there's no point having both parts in the same function.
|
| 1.617 |
| 10-Mar-2024 |
rillig | lint: remove wrong warning about overflow in unary '-' for unsigned
|
| 1.616 |
| 10-Mar-2024 |
rillig | lint: fix integer overflow in integer overflow check
|
| 1.615 |
| 10-Mar-2024 |
rillig | lint: in check for integer overflow, sort operators
|
| 1.614 |
| 09-Mar-2024 |
rillig | lint: fix excessive overflow warning after division by zero
|
| 1.613 |
| 09-Mar-2024 |
rillig | lint: merge duplicate code for checking array index
|
| 1.612 |
| 09-Mar-2024 |
rillig | lint: inline accessor macros for tnode_t
|
| 1.611 |
| 09-Mar-2024 |
rillig | lint: inline accessor macros for type_t
|
| 1.610 |
| 09-Mar-2024 |
rillig | lint: remove unneeded checks for left and right operands
|
| 1.609 |
| 03-Mar-2024 |
rillig | lint: clean up string parsing and snprintb check
|
| 1.608 |
| 03-Mar-2024 |
rillig | lint: check for unreachable bits and fields in snprintb formats
While here, clean up a few existing checks.
|
| 1.607 |
| 01-Mar-2024 |
rillig | lint: fix misleading initializer for string iterator
The field 'start' marks the start of the previous matching character,
not the current iterator position.
No binary change.
|
| 1.606 |
| 01-Mar-2024 |
rillig | lint: test format strings from snprintb calls
The functions snprintb and snprintb_m are specific to NetBSD, and their
format strings are tricky to get correct. Provide some assistance in
catching the most common mistakes.
|
| 1.605 |
| 08-Feb-2024 |
rillig | lint: clean up variable names, parameter order, comments
No functional change.
|
| 1.604 |
| 08-Feb-2024 |
rillig | lint: clean up redundant braces
No functional change.
|
| 1.603 |
| 05-Feb-2024 |
rillig | lint: make function call arguments directly accessible
Previously, the arguments of a function call expression were arranged in
a linear tree structure, from right to left. To allow easier access to
the arguments, store them in an array instead.
|
| 1.602 |
| 03-Feb-2024 |
rillig | lint: keep strings in their source representation
This allows further analysis depending on whether individual characters are
escaped as octal, hexadecimal or not at all.
|
| 1.601 |
| 01-Feb-2024 |
rillig | lint: reuse buffer editing code across lint1
Fix cat_strings to update the capacity of the buffer, to prevent
quadratic runtime when concatenating string literals.
|
| 1.600 |
| 01-Feb-2024 |
rillig | lint: use standard buffer for storing string values
No functional change.
|
| 1.599 |
| 29-Jan-2024 |
rillig | lint: do not remember content of wide string literals
The plain char literals are needed for checking printf/scanf format
strings; lint has no similar check for wide strings. These format
strings are checked by modern compilers, making this check less
relevant.
|
| 1.598 |
| 23-Jan-2024 |
rillig | lint: unconst parameters
These were leftovers from earlier refactorings and are no longer needed.
No binary change.
|
| 1.597 |
| 23-Jan-2024 |
rillig | lint: rename symt_t to symbol_kind
It was confusing to have two kinds of "symbol type" (s_type and s_symt),
so rename all related identifiers to be more distinctive.
No functional change.
|
| 1.596 |
| 21-Jan-2024 |
rillig | lint: clean up redundant const from parameters and local variables
These were leftovers from splitting large functions into smaller
functions, to ensure that variables were not unintentionally reassigned.
Those refactorings are finished, and the extra help from the compiler is
no longer necessary.
No binary change.
|
| 1.595 |
| 11-Jan-2024 |
rillig | lint: clean up enum constants for designators
In intializers and offsetof, both struct and union members are handled
in the same way, thus there is no need to distinguish them.
|
| 1.594 |
| 11-Jan-2024 |
rillig | lint: correctly set system-header flag on cast-expression
When a cast-expression comes partly from a system header, determine at
the ')' whether the whole cast-expression comes from the system header.
Previously, it was based on the operand, which contradicted the
documentation of tn_sys.
Mainly affects strict bool mode (where expressions from system headers
are handled more leniently), as well as query 9 for parenthesized return
expressions.
Discovered upon manual inspection, as calling expr_alloc_tnode should
never be necessary when creating an expression node with operands;
there's build_op for that purpose.
|
| 1.593 |
| 11-Jan-2024 |
rillig | lint: warn about integer overflow when folding constant INT_MIN / -1
|
| 1.592 |
| 09-Jan-2024 |
rillig | lint: allow complex offsetof(type, member-designator)
Both GCC 11 and Clang 8 accept member-designators that are not
identifiers but designator sequences, such as in 'offsetof(struct stat,
st_atim.tv_sec)', so make lint accept them as well.
|
| 1.591 |
| 07-Jan-2024 |
rillig | lint: fix crash for invalid __func__ (since 2023-01-29)
|
| 1.590 |
| 07-Jan-2024 |
rillig | lint: fix memory allocation names, eliminate double negation
|
| 1.589 |
| 06-Jan-2024 |
rillig | lint: remove redundant parentheses, braces and comments
Rename the functions for folding constant expressions, to make the
comments redundant.
|
| 1.588 |
| 06-Jan-2024 |
rillig | lint: remove redundant comments
|
| 1.587 |
| 03-Dec-2023 |
rillig | lint: in declarations, replace tab with space
Previously, in some cases, the keywords 'const' or 'struct' were
followed by a tab, which doesn't make sense.
No functional change.
|
| 1.586 |
| 03-Dec-2023 |
rillig | lint: re-wrap comments
No functional change.
|
| 1.585 |
| 03-Dec-2023 |
rillig | lint: move function body brace to the left
No functional change.
|
| 1.584 |
| 03-Dec-2023 |
rillig | lint: indent statement continuations consistently
No binary change.
|
| 1.583 |
| 02-Dec-2023 |
rillig | lint: extract nonportable char comparison to separate function
No functional change.
|
| 1.582 |
| 02-Dec-2023 |
rillig | lint: rename NOSCL to NO_SCL
For symmetry with NO_TSPEC. No functional change.
|
| 1.581 |
| 13-Sep-2023 |
rillig | lint: prevent invalid memory access when checking an expression
In check_expr_misc, the left and right operands of an expression were
accessed even in the case of CON (constant), STRING (string literal) and
NAME (identifier), which led to invalid values in pointer variables.
These invalid values were not used though, but technically they invoked
undefined behavior.
Precede each access to the operands with a check that the expression
indeed has operands, except in those cases where the operand is known to
have operands by only looking at the code of the current function.
|
| 1.580 |
| 12-Sep-2023 |
rillig | lint: mark unreachable function call arguments as used as well
Previously, in a '?:' expression with a constant condition, the branch
that is not taken was skipped but any identifiers in there were intended
to be marked as used. In function call expressions, this only worked
for the last argument, as the PUSH operator is not a binary operator
(see ops.def). Cover this case as well.
|
| 1.579 |
| 12-Sep-2023 |
rillig | lint: mark symbols that are only used in untaken '?:' branch as used
|
| 1.578 |
| 26-Aug-2023 |
rillig | lint: make diagnostics about ANSI C more international
|
| 1.577 |
| 08-Aug-2023 |
rillig | lint: in -a mode, do not warn about integer conversions from 'int'
Since tree.c 1.552 from 2023-07-08, lint warned about integer
conversions from 'int' or 'unsigned int' to smaller integer types. This
only affected 32-bit platforms where size_t is 'unsigned int' rather
than 'unsigned long', as on these platforms, the integer ranks of 'int'
and 'long' are the same, see INT_RANK in inittyp.c.
Discovered by lib/libkvm, which fails on i386 when lint generates any
warnings.
|
| 1.576 |
| 05-Aug-2023 |
rillig | lint: implement __builtin_offsetof for the simplest cases
Cases not covered:
1. C99 allows designators in the offsetof macro.
2. For packed types, the offset is likely to be incorrect.
|
| 1.575 |
| 02-Aug-2023 |
rillig | lint: fix plural form in message 150
|
| 1.574 |
| 02-Aug-2023 |
rillig | lint: distinguish between arguments and parameters
|
| 1.573 |
| 15-Jul-2023 |
rillig | lint: clean up duplicate code for finding struct/union members
|
| 1.572 |
| 15-Jul-2023 |
rillig | lint: properly initialize unnamed struct/union members
Previously, their 'containing type' was not initialized. The code for
initializing them is adapted from declarator_name.
|
| 1.571 |
| 15-Jul-2023 |
rillig | lint: fix member lookup after GCC statement expression
|
| 1.570 |
| 15-Jul-2023 |
rillig | lint: split struct_or_union_member into separate parts
|
| 1.569 |
| 15-Jul-2023 |
rillig | lint: fix use-after-free bug in GCC statement expressions
|
| 1.568 |
| 15-Jul-2023 |
rillig | lint: add detailed logging for finding memory allocation bugs
|
| 1.567 |
| 15-Jul-2023 |
rillig | tests/lint: demonstrate use-after-free in GCC statement expression
|
| 1.566 |
| 14-Jul-2023 |
rillig | lint: clean up redundant and verbose code
|
| 1.565 |
| 14-Jul-2023 |
rillig | lint: rename function for creating a new expression node
|
| 1.564 |
| 14-Jul-2023 |
rillig | lint: clean up comments, add a test for the '?:' operator
|
| 1.563 |
| 13-Jul-2023 |
rillig | lint: indent copyright lines consistently
|
| 1.562 |
| 13-Jul-2023 |
rillig | lint: clean up code related to lint comments
Remove type information from variable names, as the word 'flag' did not
indicate that the variables were negated.
Remove contradicting comments. Suppressing a warning for 'this line',
'this and the following line' and 'this statement or declaration' cannot
all be accurate at the same time.
|
| 1.561 |
| 12-Jul-2023 |
rillig | lint: clean up handling of __real__ and __imag__
These two operatos are not binary, therefore they don't need a right
operand. The questionable operands were a copy-and-paste mistake, as
the code was taken from the ++ and -- operands (tree.c 1.46 from
2008-04-25). The ++ and -- operands aren't binary either, but since
lint represents address calculations in their premultiplied form, the
expression ptr++ contains a hidden right operand specifying the number
of bytes by which to increment the pointer.
Creating an integer-constant-expression node with type 'long double'
didn't make sense either. Luckily, these expressions are only built but
not analyzed any further.
|
| 1.560 |
| 10-Jul-2023 |
rillig | lint: merge duplicate code for getting the name of an operator
|
| 1.559 |
| 10-Jul-2023 |
rillig | lint: clean up wchar_t and hash tables
|
| 1.558 |
| 09-Jul-2023 |
rillig | lint: in the warning about ordered enums, mention the affected type
|
| 1.557 |
| 09-Jul-2023 |
rillig | lint: clean up wording in diagnostics
Use the term 'parameter' as defined in C99 3.15.
|
| 1.556 |
| 09-Jul-2023 |
rillig | lint: clean up the wording of a few diagnostics
|
| 1.555 |
| 09-Jul-2023 |
rillig | lint: remove redundant '#' after 'argument' in diagnostics
|
| 1.554 |
| 08-Jul-2023 |
rillig | lint: warn about pointer casts between different kinds of types
Pointer casts from an integer type to a floating-point type and vice
versa get a 'maybe troublesome' warning now. The previous assumption
that all types of the same bit-size are convertible may have been valid
from a technical point of view, but still such code should get more
attention.
The rules for struct and union types could be made more fine-grained
later, if the need arises. To suppress this warning, it's always
possible to cast to an intermediate 'void *'.
|
| 1.553 |
| 08-Jul-2023 |
rillig | lint: do not use portable type sizes in integer constraints
This reverts the change from tree.c 1.547 from 2023-07-03. Back then, I
didn't know that the actual value from a type's 'portable size in bits'
was not supposed to be used.
|
| 1.552 |
| 08-Jul-2023 |
rillig | lint: warn about conversion from 128-bit to smaller integer types
|
| 1.551 |
| 08-Jul-2023 |
rillig | lint: clean up
PUSH nodes are not marked as binary, yet they have a left and a right
operand.
If none of the queries is enabled, omit the query from the debug log.
|
| 1.550 |
| 08-Jul-2023 |
rillig | lint: clarify the meaning of 'portable size in bits' of a type
No functional change.
|
| 1.549 |
| 07-Jul-2023 |
rillig | lint: only warn about traditional/C90 differences in migration mode
In C99 mode, there is no point warning about traditional C.
No change in the tests, as a complete test suite would require several
new test files, and migration mode is not used intensively.
|
| 1.548 |
| 07-Jul-2023 |
rillig | lint: only skip 'unused' warnings after errors, not other warnings
Previously, in -w mode, any warning suppressed further 'unused'
warnings, even though there was no need to do that. This can be seen in
the test gcc_attribute_var.c, where only the last unused variable from a
function was marked as unused, the others slipped through.
Fixed by counting the errors and the warnings separately and only
combining them if actually desired.
|
| 1.547 |
| 03-Jul-2023 |
rillig | lint: consistently use portable type size in integer constraints
Since tree.c 1.546 from 2023-07-03, lint no longer warned about possible
loss of accuracy when converting from 'long' to 'int' on an ILP32
platform that uses 'unsigned long' for size_t, when run in portable mode
(-p), which is enabled by default in the NetBSD build.
The integer constraints avoid false-positive warnings by looking at the
actual values an expression can take. The function can_represent is
guarded by a condition that uses the portable_size_in_bits, but then
internally used the opposite size_in_bits, which led to inconsistent
results.
The warning looks confusing though, as on an ILP32 platform, 'int' and
'long' have the same size and representation, therefore there cannot be
an actual loss of accuracy. The warning may need to be reworded to
explicitly mention the portability mode, in which sizeof(int) is assumed
to be 3 instead of 4, to catch possible loss of accuracy on other
platforms.
|
| 1.546 |
| 03-Jul-2023 |
rillig | lint: clean up redundant casts
|
| 1.545 |
| 03-Jul-2023 |
rillig | lint: rename uppercase QUAD to LLONG
No binary change.
|
| 1.544 |
| 02-Jul-2023 |
rillig | lint: rename 'quad' to 'signed int' or 'unsigned int'
No functional change.
|
| 1.543 |
| 02-Jul-2023 |
rillig | lint: clean up variable names
|
| 1.542 |
| 01-Jul-2023 |
rillig | lint: constify, reduce indentation
No functional change.
|
| 1.541 |
| 01-Jul-2023 |
rillig | lint: clean up duplicate and dead code for integer constants
No functional change.
|
| 1.540 |
| 01-Jul-2023 |
rillig | lint: clean up typos
|
| 1.539 |
| 01-Jul-2023 |
rillig | lint: fix initialization of unnamed union member
|
| 1.538 |
| 30-Jun-2023 |
rillig | lint: clean up names related to declaration levels
The previous prefix 'DK_' (declaration level kind) had a conflict with
the 'DK_' (designator kind) in init.c, so change the prefix to 'DLK_'.
The new name for dinfo_t is decl_level, which is more expressive.
No functional change.
|
| 1.537 |
| 30-Jun-2023 |
rillig | lint: fix handling of unnamed struct/union members
The support for unnamed struct/union members that was added in decl.c
1.60 from 2015-10-13 was simple but wrong. It didn't cover initializers
of these structures and computed wrong sizes for structures containing
anonymous unions. At that time, the handling of initializers was broken
as well, it was fixed 6 years later in init.c 1.229 from 2021-12-22.
Real-life examples for code that lint couldn't handle are:
* external/bsd/jemalloc/dist/src/jemalloc.c
* external/mit/xorg/lib/dri.old/Makefile
|
| 1.536 |
| 30-Jun-2023 |
rillig | lint: clean up tree.c
No functional change.
|
| 1.535 |
| 30-Jun-2023 |
rillig | lint: make alignof(incomplete enum) an error
|
| 1.534 |
| 30-Jun-2023 |
rillig | lint: use NULL for null pointer constants
No binary change.
|
| 1.533 |
| 30-Jun-2023 |
rillig | lint: add query for assigning an integer 0 to a pointer
|
| 1.532 |
| 29-Jun-2023 |
rillig | lint: use standard terminology for bit-field width
No functional change.
|
| 1.531 |
| 24-Jun-2023 |
rillig | lint: reduce memory allocations
The type val_t has the same size as the tn_s member in the same union.
No functional change.
|
| 1.530 |
| 24-Jun-2023 |
rillig | lint: don't warn about comparison between char and character constant
|
| 1.529 |
| 24-Jun-2023 |
rillig | lint: add query for comparing 'char' with plain integers
|
| 1.528 |
| 22-Jun-2023 |
rillig | lint: add query for comma operator
|
| 1.527 |
| 09-Jun-2023 |
rillig | lint: miscellaneous clean-ups
No binary change, except for line numbers in assertions.
|
| 1.526 |
| 09-Jun-2023 |
rillig | lint: indent local variables consistently
No binary change.
|
| 1.525 |
| 03-Jun-2023 |
rillig | lint: provide more detailed types when operands do not match
|
| 1.524 |
| 22-May-2023 |
rillig | lint: rename constant NOTSPEC to NO_TSPEC
It was too easy to misread the old name as NOT_SPEC instead of the
intended NO_TSPEC.
|
| 1.523 |
| 13-May-2023 |
rillig | lint: add query for chained assignments
|
| 1.522 |
| 10-May-2023 |
rillig | lint: make code simpler
No functional change.
|
| 1.521 |
| 09-May-2023 |
rillig | lint: track integer constraints through conditional expressions
|
| 1.520 |
| 09-May-2023 |
rillig | lint: preserve integer constraints on cast
|
| 1.519 |
| 22-Apr-2023 |
rillig | lint: allow cast from a type to a union containing that type
|
| 1.518 |
| 22-Apr-2023 |
rillig | lint: fix missing initialization for cast to union
The left operand of a unary AST node must not be NULL.
The previous code crashed lint when run with some query enabled, as
is_cast_redundant assumes that a non-null AST node has valid operands.
$ cat <<EOF > crash.c
double
demo(void) {
union u {
double *num;
} u;
u = (union u)&((double) { 0.0 });
return *u.num;
}
EOF
$ /usr/libexec/lint1 -w -S -g -q8 crash.c /dev/null
|
| 1.517 |
| 22-Apr-2023 |
rillig | lint: rename type_t.t_str to t_sou
The name 't_str' could have meant 'string' or 'struct'. Since both
struct and union types have members, rename it to 't_sou'.
No functional change.
|
| 1.516 |
| 22-Apr-2023 |
rillig | lint: don't warn about cast between union and one of its member types
Seen in src/sbin/newfs_udf/udf_core.c for context.anchors.
|
| 1.515 |
| 22-Apr-2023 |
rillig | lint: do not warn about parenthesized assignment in controlling expr
Same as for GCC and Clang.
|
| 1.514 |
| 22-Apr-2023 |
rillig | lint: fix typos in comments
|
| 1.513 |
| 14-Apr-2023 |
rillig | lint: document suppressed LDBL_MAX warning more accurately
The crucial point is not that lint is cross-compiled but that tools/lint
is run on the usr.bin/xlint code, and in that situation, the platform
mismatch happens.
|
| 1.512 |
| 12-Apr-2023 |
rillig | lint: suppress warnings about LDBL_MAX in cross-build mode
This fixes the cross build for sparc64 on x86_64, in which lint
complained:
warning: floating-point constant out of range [248]
warning: floating point overflow on operator '-' [142]
warning: floating-point constant out of range [248]
|
| 1.511 |
| 11-Apr-2023 |
rillig | lint: fix converting a complex floating-point constant
The complex '+' in msg_142 line 27 led to an invalid floating point
value when converting from 'double' to '_Complex double'.
|
| 1.510 |
| 11-Apr-2023 |
rillig | lint: fix wrong warning about overflow in floating constant expressions
|
| 1.509 |
| 11-Apr-2023 |
rillig | lint: condense code
No functional change.
|
| 1.508 |
| 11-Apr-2023 |
rillig | lint: don't wrongly warn about overflow in complex constants
Seen in lib/libm.
|
| 1.507 |
| 28-Mar-2023 |
rillig | lint: warn about extern declarations outside headers
https://mail-index.netbsd.org/tech-userlevel/2023/03/15/msg013727.html
|
| 1.506 |
| 24-Feb-2023 |
rillig | lint: order functions to to reduce forward declarations
No functional change.
|
| 1.505 |
| 22-Feb-2023 |
rillig | lint: fix confusing message about nonportable character constants
|
| 1.504 |
| 29-Jan-2023 |
rillig | lint: fix definition of __func__
|
| 1.503 |
| 29-Jan-2023 |
rillig | lint: demonstrate wrong declaration of __func__
|
| 1.502 |
| 29-Jan-2023 |
rillig | lint: don't warn about negative '>>' when the actual value is positive
|
| 1.501 |
| 29-Jan-2023 |
rillig | lint: inline an expression, clean up comments
No functional change.
|
| 1.500 |
| 29-Jan-2023 |
rillig | lint: untangle conditions in promote_c90
No functional change.
|
| 1.499 |
| 28-Jan-2023 |
rillig | lint: extract casting to a union into separate function
No functional change.
|
| 1.498 |
| 28-Jan-2023 |
rillig | lint: use stronger wording in comments of unimplemented offsetof
|
| 1.497 |
| 28-Jan-2023 |
rillig | lint: split usual arithmetic conversions into separate functions
No functional change.
|
| 1.496 |
| 28-Jan-2023 |
rillig | lint: split integer promotions into separate functions
No functional change.
|
| 1.495 |
| 28-Jan-2023 |
rillig | lint: remove commented code from new_tnode
The approach in that code was wrong anyway. If an expression is shifted
to the right, that doesn't change the declared type of the expression,
it only changes the possible range of values. A better way to implement
these restricted value spaces is by using integer_constraints.
No functional change.
|
| 1.494 |
| 22-Jan-2023 |
rillig | lint: prevent undefined behavior for signed '<<'
Found by manual code inspection, verified by MKSANITIZER=yes
USE_SANITIZER=undefined.
|
| 1.493 |
| 21-Jan-2023 |
rillig | lint: make code for strict bool mode more consistent
Guard all entry points with a check for Tflag and remove the now
redundant internal check for Tflag.
No functional change.
|
| 1.492 |
| 14-Jan-2023 |
rillig | lint: use fewer abbreviations
No binary change except for line numbers in assertions in decl.c.
|
| 1.491 |
| 14-Jan-2023 |
rillig | lint: split struct_or_union_member into two halves
No functional change.
|
| 1.490 |
| 13-Jan-2023 |
rillig | lint: remove custom memory allocator
Besides adding complexity, the custom memory allocator didn't invalidate
freed memory, which made it harder to find possible use-after-free bugs.
|
| 1.489 |
| 08-Jan-2023 |
rillig | lint: do not report usual arithmetic conversions for constants
|
| 1.488 |
| 08-Jan-2023 |
rillig | lint: extract usual arithmetic conversions to separate function
No functional change.
|
| 1.487 |
| 08-Jan-2023 |
rillig | lint: recognize enum constant named 'max' as a count of values
Seen in external/bsd/mdocml/dist/mdoc.h(50).
|
| 1.486 |
| 04-Jan-2023 |
rillig | lint: fix the result type of '?:' for void pointers
The change from 2015-07-29 was wrong since that rule only applies to
null pointer constants but not to other void pointers.
|
| 1.485 |
| 03-Jan-2023 |
rillig | lint: condense building of nodes for '?:' expressions
No binary change except for line numbers in assertions.
|
| 1.484 |
| 30-Nov-2022 |
rillig | lint: add measurement unit to local variable names
No binary change, except for line numbers in assertions.
|
| 1.483 |
| 01-Oct-2022 |
rillig | lint: inline local variable in build_function_argument
No functional change.
|
| 1.482 |
| 01-Oct-2022 |
rillig | lint: miscellaneous cleanup
No functional change.
|
| 1.481 |
| 01-Oct-2022 |
rillig | lint: add hyphen to adjective 'old-style'
|
| 1.480 |
| 28-Aug-2022 |
rillig | lint: inline previous_declaration for nonconstant message IDs
This adds back the compile-time printf validation in debug mode that was
missing before.
|
| 1.479 |
| 28-Aug-2022 |
rillig | lint: clean up visual clutter
No functional change.
|
| 1.478 |
| 28-Aug-2022 |
rillig | lint: rename functions to be clearer
No need anymore to keep external identifiers at the "6 significant
initial characters" mandated by C90.
|
| 1.477 |
| 25-Aug-2022 |
rillig | lint: remove explicit list of known GCC attributes
Most GCC attributes consist of a single identifier. Up to now, it was
necessary to list each of these identifiers in the grammar, even those
that only apply to a single target architecture.
Instead, parse the general form of attributes, matching the few
attributes that lint handles by name instead. While here, rename the
grammar rules to use the GCC terms.
To avoid conflicts between the global function 'printf' and the GCC
attribute of the same name, do not add GCC attributes to the symbol
table, and don't make these symbols 'extern' either.
ok christos@.
|
| 1.476 |
| 19-Aug-2022 |
rillig | lint: add more details to message about large bit-shifts
|
| 1.475 |
| 16-Jul-2022 |
rillig | lint: add more details to message about pointer alignment
The previous message 'may cause alignment problem' was not detailed
enough to be actionable, it didn't give the necessary insight to why
lint was complaining at all.
The new message 'increases alignment from 1 to 4' or 'from 2 to 8'
describes the potentially problematic conversion, and together with the
involved type names, it allows an informed decision about whether lint's
warning is warranted or not.
In a typical NetBSD build, this warning is in the top 10. The number of
these warnings depends on the architecture, it is typically between 800
and 1600.
|
| 1.474 |
| 08-Jul-2022 |
rillig | lint: fix edge cases in the query for redundant cast before assignment
Casting from and to _Bool is only allowed outside strict bool mode.
Outside strict bool mode, _Bool is an integer type, therefore return
early if any of the operands has type _Bool. In strict bool mode, even
casting from _Bool to _Bool is not allowed, as it is not needed in
practice.
Handle _Complex types before real floating-point types. Return early
for _Complex types, as these are floating-point types as well.
For pointer casts, not only flag casts to or from 'pointer to void', but
also casts between the same types.
In debug mode, when constructing the type '_Complex float', the type
name of '_Complex' occurs in the debug log. Outside of debug mode,
printing this type name is an error since this type keyword only occurs
internally, when constructing a type. At that point, it is not supposed
to occur in any user-visible message.
|
| 1.473 |
| 08-Jul-2022 |
rillig | lint: fix query for redundant cast before assignment
Previously, 'i = (int)dbl' was marked as redundant, even though it
performs a value conversion.
|
| 1.472 |
| 06-Jul-2022 |
rillig | lint: do not warn about 'may lose accuracy' in safe cases of '%'
The possible values of the expression 'a % b' for unsigned integers lie
between 0 and (b - 1). For signed integers, it's more complicated, so
ignore them for now.
|
| 1.471 |
| 05-Jul-2022 |
rillig | lint: add additional queries that are not enabled by default
In the last 18 months, several lint warnings have been made adjusted to
allow common usage patterns. For example, lint no longer warns about a
constant condition in the statement 'do { ... } while (false)' (message
161), as this pattern is well-known in statement-like macros, making it
unlikely that the 'false' is a mistake. Another example is casts
between unequal pointer types (message 247) for a few well-known
patterns that are unlikely to be bugs.
Occasionally, it is useful to query the code for patterns or events that
would not justify a warning. These patterns are modeled as predefined
queries that can be selected individually, in addition to and
independently of the existing warnings and errors.
New queries can be added as needed, in the same way as new warnings.
Queries that are deemed no longer used can be deactivated in the same
way as warnings that are no longer used.
As long as none of the queries is enabled, they produce a minimal
overhead of querying a single global variable. Computations that are
more expensive than a few machine instructions should be guarded by
any_query_enabled.
https://mail-index.netbsd.org/source-changes-d/2022/06/28/msg013716.html
ok christos@
|
| 1.470 |
| 03-Jul-2022 |
rillig | lint: add debug logging for struct and enum details
|
| 1.469 |
| 02-Jul-2022 |
rillig | lint: prevent undefined behavior when folding '<<' or '>>'
|
| 1.468 |
| 02-Jul-2022 |
rillig | lint: document usage of 'convert'
No functional change.
|
| 1.467 |
| 02-Jul-2022 |
rillig | lint: only convert the right operand of '<<' or '>>' in traditional C
Traditional C says: "Then the right operand is converted to int".
C90 dropped that sentence, probably because it didn't have any effect on
the result or the conditions for undefined behavior.
To stick to the wording of the specification, also convert UINT to INT.
While here, fix the call to 'convert'. The first argument being 'CVT'
means that the conversion comes from a cast-expression instead of an
implicit conversion. This prevents warnings for 'uint64_t << uint64_t'.
Keeping this unnecessary conversion in C90 or later would have generated
warnings for the functions at the bottom of msg_132.c.
|
| 1.466 |
| 01-Jul-2022 |
rillig | lint: add quotes around operator in message 138, remove message 240
Message 138 doesn't occur in practice, it was the last one that was
missing the quotes around the placeholder.
Message 240 was also missing the quotes, but it was not reachable. There
was also no apparent benefit in warning about 'assignment of different
structures' when the message about 'assignment type mismatch' serves the
same purpose.
|
| 1.465 |
| 01-Jul-2022 |
rillig | lint: restructure 'convert'
The conditions are now grouped by target type, highlighting that
conversions between integer and floating point types are not checked by
this function, in fact, they are not checked at all.
No functional change.
|
| 1.464 |
| 01-Jul-2022 |
rillig | lint: rename functions for value conversion
No functional change.
|
| 1.463 |
| 24-Jun-2022 |
rillig | lint: do not warn about pointer casts to array types
If the (recursive) element type of the array is compatible, that's good
enough. Even after the previous commits, this warning is the one that
occurs most in a standard NetBSD build, and it is generally ignored.
For now, focus on reducing the number of false positives to an
acceptable level.
|
| 1.462 |
| 24-Jun-2022 |
rillig | lint: allow pointer cast between compatible structs
Lint already allowed pointer casts from 'struct counter' to 'struct
counter_impl'. Now it also allows the other direction.
|
| 1.461 |
| 24-Jun-2022 |
rillig | lint: do not warn about pointer cast between sockaddr variants
|
| 1.460 |
| 24-Jun-2022 |
rillig | lint: allow pointer cast from char to struct/union
|
| 1.459 |
| 22-Jun-2022 |
rillig | lint: add quotes around placeholders for the remaining messages
Reword some of the messages slightly, exchanging brevity for clarity.
Message 138 is kept as-is, as it is not yet covered by any tests.
Message 240 is kep as-is, as it is unreachable.
|
| 1.458 |
| 21-Jun-2022 |
rillig | lint: merge duplicate code in typeok_minus
No functional change.
|
| 1.457 |
| 21-Jun-2022 |
rillig | lint: clean up comments, don't include unused header
No functional change.
|
| 1.456 |
| 21-Jun-2022 |
rillig | lint: use is_struct_or_union instead of comparing twice
No functional change.
|
| 1.455 |
| 21-Jun-2022 |
rillig | lint: add quotes and details to some more messages
|
| 1.454 |
| 19-Jun-2022 |
rillig | lint: add quotes around placeholders in 4 messages
|
| 1.453 |
| 15-Jun-2022 |
rillig | lint: rename mod_t.m_requires_bool to m_compares_with_zero
The operators NOT, LOGAND, LOGOR and QUEST only require _Bool in strict
bool mode, in default mode they accept any scalar expression and compare
it with zero. The new names are more accurate.
No functional change.
|
| 1.452 |
| 30-May-2022 |
rillig | lint: revert 'do not pre-multiply pointer expressions' from 2022-05-26
In tree.c 1.448, removing the pre-multiplication generated wrong
warnings about out-of-bounds array access.
|
| 1.451 |
| 30-May-2022 |
rillig | lint: fix assertion failure in '(unsigned long)(ptr) >> 12'
Since tree.c 1.449 from 2022-05-26.
|
| 1.450 |
| 29-May-2022 |
rillig | lint: fix wrong errors about sizeof of a bit-field (since 2022-05-26)
|
| 1.449 |
| 26-May-2022 |
rillig | lint: do not warn about 'uint32_t = uint64_t >> 32'
If all possible values fit into the destination type, there is no
possibility of losing accuracy.
Enhances PR 36668.
|
| 1.448 |
| 26-May-2022 |
rillig | lint: do not pre-multiply pointer expressions 'ptr + int'
In the AST, it is confusing to see that 'stderr == __sF + 304', it
should rather be 'stderr == __sF + 2'.
No functional change.
|
| 1.447 |
| 26-May-2022 |
rillig | lint: improve local variable and function names
No functional change.
|
| 1.446 |
| 26-May-2022 |
rillig | lint: reduce scope of local variables in build_plus_minus
No functional change.
|
| 1.445 |
| 26-May-2022 |
rillig | lint: re-order conditions for lossy conversions
Now that can_represent does more work, put it at the end of the
conditions.
No functional change.
|
| 1.444 |
| 26-May-2022 |
rillig | lint: do not warn about loss in accuracy if the actual value fits
The expression 'any & 0xff' can always be assigned to 'uint8_t' without
loss of any value bits. In the same way, '(any & 0xff) << 8' can always
be assigned to 'uint16_t'.
Previously, lint warned about these cases. Fix these wrong warnings by
tracking the possible values of integer expressions across a single
expression.
Fixes PR 36668, so that <sys/endian.h> does not need to be cluttered
with useless casts anymore.
|
| 1.443 |
| 26-May-2022 |
rillig | lint: remove long list of node types in switch statement
No functional change.
|
| 1.442 |
| 20-May-2022 |
rillig | lint: use __RCSID in lint mode as well
Since 1995-10-02, lint supports __asm statements and __asm modifiers.
No binary change.
|
| 1.441 |
| 18-May-2022 |
rillig | lint: do not warn about '==' on the right-hand side of ','
|
| 1.440 |
| 18-May-2022 |
rillig | lint: reduce negations in typeok_point, document check_expr_misc
No functional change.
|
| 1.439 |
| 30-Apr-2022 |
rillig | lint: inline macro 'sflag'
Mark all places where lint's C90 mode is stricter than its C99 mode.
Most of the situations in which lint produces only warnings instead of
errors covered by the "Constraints" sections in the relevant standards.
This doesn't prevent a specific compiler from accepting it though.
No functional change.
|
| 1.438 |
| 30-Apr-2022 |
rillig | lint: inline macro 'tflag'
The definition of the macro tested both allow_trad and allow_c90, but
there is only a single mode in which allow_c90 is false, therefore it
suffices to test only that.
While double-checking each occurrence of tflag individually, I learned
why lint performs lookups of struct members only by name, independently
of the struct in which they are declared. See typeok_arrow for details.
No functional change.
|
| 1.437 |
| 30-Apr-2022 |
rillig | lint: inline macro Sflag in cases of expected behavior
In the grammar rules 'enums_with_opt_comma' and 'block_item_list',
checking for allow_c99 was redundant since c99ism does not warn in C99
mode.
In the grammar rule 'designator', align the two actions structurally.
No functional change.
|
| 1.436 |
| 19-Apr-2022 |
rillig | lint: fix integer overflow on seeing -(uint64_t)INT64_MIN
Seen in external/cddl/osnet/dist/lib/libuutil/common/uu_strtoint.c,
which is protected by a NOLINT in ../Makefile.inc.
|
| 1.435 |
| 19-Apr-2022 |
rillig | lint: only warn about unsigned constants in C90 migration mode
For C99 code, the difference between traditional C and C90 is not
relevant.
Seen in sqlite3.c line 30448.
|
| 1.434 |
| 19-Apr-2022 |
rillig | lint: do not warn about converting (1234567L & 0xFF) to unsigned char
Due to the '& 0xFF', there is no possible loss of accuracy.
|
| 1.433 |
| 16-Apr-2022 |
rillig | lint: merge mod_t.m_test_context into m_requires_bool
These two flags mean exactly the same.
No functional change.
|
| 1.432 |
| 16-Apr-2022 |
rillig | lint: rename members of tnode_t to more closely match reality
The flags do not describe the left operand of the node but both, as for
most operators, either none or both operands are in test context or in
value context.
The one exception is the operator '?' from the '?:' conditional, for
which the left operand is in test context and the right operand is in
value context.
No binary change.
|
| 1.431 |
| 16-Apr-2022 |
rillig | lint: investigate why lint only warns about some constant conditions
Noticed in lex.c, in the macro kwdef, where lint complains about the
condition containing '||' but not about the conditions containing only
'>' or '=='.
|
| 1.430 |
| 16-Apr-2022 |
rillig | lint: prepare keyword table for C11
The C11 keywords had been listed as being C99 keywords. Instead of
initializing the individual fields separately, merge them by specifying
the year of the standard in which they appeared.
No binary change, except for line numbers in assertions.
|
| 1.429 |
| 16-Apr-2022 |
rillig | lint: migrate gflag to allow_gcc
No functional change.
|
| 1.428 |
| 16-Apr-2022 |
rillig | lint: clean up conditions for migration check between trad and C90
No functional change, as each of the conditions was effectively '!tflag
&& !sflag && !Sflag' (even if distributed over several statements).
|
| 1.427 |
| 15-Apr-2022 |
rillig | lint: in C99 mode, do not warn about non-prototype conversions
Message 259 is "argument #%d is converted from '%s' to '%s' due to
prototype", and it is intended to warn about compatibility between
traditional C where functions had no prototypes and standard C where
functions have prototypes.
Running lint in C99 mode is further away from traditional C than running
lint in C90 mode, so that warning doesn't make sense for C99. There are
still some inconsistencies in the 5 language version modes that lint
offers:
-t for traditional C
(no option) for migrating traditional C to C90
-s for C90 code
-S for C99 code
-Ac11 for C11 code
By disabling warning 259 in C99 mode, a typical NetBSD build produces
14.500 fewer warnings than before, of about 100.000 total.
Message 259 overlaps with message 298 "conversion from '%s' to '%s' may
lose accuracy, arg #%d", and in some cases of potentially lossy
conversions, lint now produces none of these messages. In some other
cases, these warnings were reported redundantly. The cases where
message 298 makes sense will be added back later, as needed.
|
| 1.426 |
| 15-Apr-2022 |
rillig | lint: merge conditions in check_integer_conversion
No functional change.
|
| 1.425 |
| 15-Apr-2022 |
rillig | lint: reduce indentation in check_integer_conversion
No functional change.
|
| 1.424 |
| 09-Apr-2022 |
rillig | lint: distinguish between storage class and declaration kind
These types overlap but are not the same.
No functional change.
|
| 1.423 |
| 09-Apr-2022 |
rillig | lint: extract is_member into separate function
No functional change.
|
| 1.422 |
| 09-Apr-2022 |
rillig | lint: disambiguate sym_t.s_value
Having a single variable for 4 different purposes with different types
makes the code unnecessarily hard to follow.
No functional change.
|
| 1.421 |
| 09-Apr-2022 |
rillig | lint: split CTCONST into BOOL_CONST and ENUM_CONST
Having a unified compile-time constant "storage class" made the code
more difficult to understand.
No functional change.
|
| 1.420 |
| 09-Apr-2022 |
rillig | lint: inline member access macros for sym_t
Having the 'u.' explicitly in the code serves as a reminder that these
members are only defined under certain conditions.
No functional change.
|
| 1.419 |
| 09-Apr-2022 |
rillig | lint: rename length to length_in_bits
No functional change.
|
| 1.418 |
| 03-Apr-2022 |
rillig | lint: fix crash after syntax error in GCC statement expression
Since cgram.y 1.226 from 2021-05-03.
|
| 1.417 |
| 02-Apr-2022 |
rillig | lint: clean up style
Remove outdated ARGSUSED (the one in tree.c had been wrong since 1995),
remove unused argument.
No functional change.
|
| 1.416 |
| 02-Apr-2022 |
rillig | lint: rename dinfo_t.d_next to d_enclosing
The name d_next gave the wrong impression that the declaration infos
would be independent of each other, but they are nested.
No functional change.
|
| 1.415 |
| 02-Apr-2022 |
rillig | lint: add debug logging for the declaration stack
To track down the wrong edge cases in decl_direct_abstract.c.
|
| 1.414 |
| 01-Apr-2022 |
rillig | lint: add type details to message about 'sizeof(function)'
The code in add_function is severely broken, it mixes up the return type
of the function with the argument types. For now, at least show the
guessed type in the diagnostic, to allow human readers quickly spot the
bug.
Extend the test cases in decl_direct_abstract.c to show that the
behavior differs unreasonably if the first parameter of the function is
equal to its return type.
|
| 1.413 |
| 01-Apr-2022 |
rillig | lint: improve determination of abstract typename
Still not perfect, but at least a step in the right direction. See
decl_direct_abstract.c for the missing edge cases.
See PR#22119.
|
| 1.412 |
| 01-Apr-2022 |
rillig | lint: replace crash with assertion failure
Triggered by 'sizeof(int())'.
GCC evaluates this expression to 1, without any warning.
Clang warns about an "invalid application of 'sizeof' to a function
type".
|
| 1.411 |
| 13-Mar-2022 |
rillig | lint: add braces around multi-line loop body
No functional change.
|
| 1.410 |
| 27-Feb-2022 |
rillig | lint: merge duplicate code for handling plain and wide strings
No functional change. As before, the string literals "1" "2" "3" are
not concatenated from left to right, instead concatenation starts with
"23" and then proceeds to "123".
|
| 1.409 |
| 27-Feb-2022 |
rillig | lint: C99 has been released, so refer to it by its proper name
|
| 1.408 |
| 27-Feb-2022 |
rillig | lint: unabbreviate two more members of sym_t
No functional change.
|
| 1.407 |
| 27-Feb-2022 |
rillig | lint: unabbreviate some fields in sym_t
No functional change.
|
| 1.406 |
| 27-Feb-2022 |
rillig | lint: encode lifetime of allocated memory in function names
No functional change.
|
| 1.405 |
| 27-Feb-2022 |
rillig | lint: encode lifetime of allocated memory in the function names
No functional change.
|
| 1.404 |
| 26-Feb-2022 |
rillig | lint: fix memory corruption in statement expressions (since 2021-12-17)
The commit that introduced the assertion failure looks innocent, it only
adds a few predefined functions for GCC mode. Nevertheless, before that
commit, lint consistently complained about 'error: void type illegal in
expression [109]', which doesn't make sense either.
This fix also removes the creative use of the initialization stack to
store the type of the statement expression. Having a separate stack for
these statement expressions makes the code easier to understand.
|
| 1.403 |
| 26-Feb-2022 |
rillig | lint: extract code for handling statement expressions from the grammar
This prepares the fix of the memory corruption bug that is demonstrated
in t_integration.sh, test case assertion_failures.
No functional change.
|
| 1.402 |
| 21-Dec-2021 |
rillig | lint: sync comment with recent change of parameter name
|
| 1.401 |
| 21-Dec-2021 |
rillig | lint: invert condition in build_name
No functional change.
|
| 1.400 |
| 17-Dec-2021 |
rillig | lint: in GCC mode, declare alloca and variants
The prototype declarations define the correct parameter types of these
functions so that they are no longer subject to the default argument
promotions (C11 6.5.2.2p6).
The GCC builtins are only recognized in GCC mode (-g).
|
| 1.399 |
| 16-Dec-2021 |
rillig | lint: narrow down parameter of build_name
Passing an arbitrary tokenizer symbol left too much freedom and
uncertainty to the caller, and 0 was a magic number in this context.
No functional change.
|
| 1.398 |
| 15-Dec-2021 |
rillig | lint: unexport struct_or_union_member
Strangely, lint2 didn't notice this unnecessary export.
|
| 1.397 |
| 06-Dec-2021 |
rillig | lint: fix return type of GCC's __builtin_alloca
Needed for libgmp.
|
| 1.396 |
| 04-Dec-2021 |
rillig | lint: add quotes around placeholders in message 123
|
| 1.395 |
| 16-Nov-2021 |
rillig | lint: fix check for function calls in strict bool mode
Previously, if a function call occurred in the controlling expression,
its return type could be any scalar, not just bool. This was against
the goal of strict bool mode, which makes bool a separate and
incompabile type to all other types. For example, it would allow
controlling expressions like 'strcmp(a, b)' without the usual '!= 0',
but only if at least one of 'a' and 'b' came from a macro definition
from a system header.
The fix is that the decision of whether the type of the controlling
expression may be scalar is no longer based on the operand types but on
the main operator of the controlling expression.
|
| 1.394 |
| 16-Nov-2021 |
rillig | tests/lint: explain the missing error message in strict bool mode
There are 2 separate bugs that lead to the missing error message.
The first bug is that when parsing a name, the information about whether
that name comes from a system header is not properly recorded for the
last token before switching from or to a system header.
The second bug is that for determining whether the type of a function
call expression is relaxed, the types of the function call arguments are
taken into account, even though the type only depends on the function
itself. The same idea applies to the operators '<<' and '>>', as their
result type only depends on the left operand, at least since C90.
|
| 1.393 |
| 01-Nov-2021 |
rillig | lint: enter full C90 compatibility mode
The C99 comment in tree.c:3468 has been there since 2017-03-06, without
anyone complaining that their compiler would not handle it.
Strangely, running GCC 10.3.0 in '-std=c90' mode does not complain about
declarations after statements, '-Wdeclaration-after-statement' is needed
separately.
No functional change.
|
| 1.392 |
| 01-Nov-2021 |
rillig | lint: move all declarations above statements
All code that is used by src/tools is supposed to be compatible with C90.
No functional change.
|
| 1.391 |
| 01-Nov-2021 |
rillig | lint: do not warn about array size mismatch in array[(int)enum]
The cast to 'int' explicitly converts the type away from being an enum.
|
| 1.390 |
| 01-Nov-2021 |
christos | simplify.
|
| 1.389 |
| 01-Nov-2021 |
rillig | lint: in the check for array[enum], allow enum constant NUM
When an enum type defines a constant NUM_VALUES, this constant is
usually not part of the enum values available to the application but
rather a handy place for defining the number of other enum values. Don't
warn about this case.
Seen in openpam_impl.h and several other places.
|
| 1.388 |
| 31-Oct-2021 |
rillig | lint: fix invalid memory access in array[enum] check
Lint checks whether in an expression 'array[enum]', the array size
matches the value of the maximum enum constant.
The previous tests for this check were missing the case where an enum
name was explicitly cast to an integer type and then used as an array
index. In this situation, the resulting type of the array index is a
plain 'int' without any information about its previous 'enum' history.
An entirely different case is when the 'enum' is implicitly converted to
an integer type, as in the test color_name_too_many. There, for the
final type of the array index, rn->tn_type->t_is_enum is true, which
means that rn->tn_type->t_enum is properly filled.
The bug was a simple typo, I had forgotten a tn_left indirection, which
is necessary to get the type before the implicit conversion.
Found and reported by Christos, triggered by src/lib/libperfuse/ops.c
1.89 line 1226 expression 'VTTOIF(vap->va_type)'.
|
| 1.387 |
| 31-Oct-2021 |
christos | Avoid null pointer exception hpcmips/mipsel src/lib/libperfuse/ops.c
|
| 1.386 |
| 30-Oct-2021 |
rillig | lint: warn if an enum name is used for mismatched array access
This helps to keep the enum definition and the straight-forward
implementation of a to_string or name function in sync.
The test for message 241 had to be adjusted because of exactly this bug.
That test defined a bit mask enum but accessed it like a value enum
type.
|
| 1.385 |
| 09-Oct-2021 |
rillig | lint: add more details to warning about nonportable character comparison
Especially when macros are involved, this makes it easier to see that
there is a real problem.
Seen in bin/sh/expand.c:273, where a plain char is compared to numbers
larger than 127.
|
| 1.384 |
| 09-Oct-2021 |
rillig | lint: warn that bsearch effectively discards 'const'
Just like strchr, bsearch takes a const pointer and returns a non-const
pointer into the same object.
|
| 1.383 |
| 26-Sep-2021 |
rillig | tests/lint: explain difference between i386 and sparc for 259
Seen in usr.bin/make/cond.c 1.278 from 2021-09-21, line 800, the call to
is_token, where unsigned char gets converted to unsigned int or unsigned
long, depending on the platform.
|
| 1.382 |
| 18-Sep-2021 |
jmcneill | use isfinite(3) instead of finite(3) for portability
Fixes tools build on macOS 11.6 arm64 hosts.
|
| 1.381 |
| 12-Sep-2021 |
rillig | lint: un-abbreviate rvused and rvdisc
No functional change.
|
| 1.380 |
| 05-Sep-2021 |
rillig | lint: reduce number of ifdef names
No functional change.
|
| 1.379 |
| 05-Sep-2021 |
rillig | lint: do not warn about comparison 'unsigned <= 0'
Seen in scanners generated by Flex, and about 50 occurrences in the
NetBSD src and xsrc tree, all of which are not suspicious of being bugs.
|
| 1.378 |
| 05-Sep-2021 |
rillig | lint: fix lint warnings
|
| 1.377 |
| 04-Sep-2021 |
rillig | lint: spell out abbreviations in comments
No functional change.
|
| 1.376 |
| 04-Sep-2021 |
rillig | lint: sort operators in typeok_op according to ops.def
Well, except for INIT, FARG, RETURN. These are listed at the very
bottom of the operators table but still behave much like ASSIGN.
No functional change.
|
| 1.375 |
| 04-Sep-2021 |
rillig | lint: extract typeok_point and typeok_arrow from typeok_op
No functional change.
|
| 1.374 |
| 04-Sep-2021 |
rillig | lint: clean up switch statement in typeok_op
In the old times where typeok_op was inlined into typeok, it was
necessary to write the complicated if-not-return-break. Not anymore.
No functional change.
|
| 1.373 |
| 04-Sep-2021 |
rillig | lint: make typeok_quest simpler
No functional change.
|
| 1.372 |
| 03-Sep-2021 |
rillig | lint: set the return type of __builtin_*_overflow to bool, not int
Needed for inetd.c in strict bool mode.
|
| 1.371 |
| 03-Sep-2021 |
rillig | lint: extract build_name_call from build_name
This reduces the indentation, providing enough space to write out the
full diagnostic in the code. It also prepares for supporting GCC
builtins like __builtin_add_overflow, which return _Bool instead of int.
No functional change.
|
| 1.370 |
| 02-Sep-2021 |
rillig | lint: split check_expr_misc into separate functions
No functional change.
|
| 1.369 |
| 02-Sep-2021 |
rillig | lint: split check_assign_types_compatible into separate functions
No functional change.
|
| 1.368 |
| 02-Sep-2021 |
rillig | lint: fix unintended warning for converting long long to signed char
|
| 1.367 |
| 02-Sep-2021 |
rillig | lint: move a condition into should_warn_about_prototype_conversion
No functional change.
|
| 1.366 |
| 02-Sep-2021 |
rillig | lint: clean up should_warn_about_prototype_conversion
|
| 1.365 |
| 01-Sep-2021 |
rillig | lint: remove array creation from expr_derive_type
It is only used for creating string types. The creation of other array
types happens in add_array.
This is a preparation to make all stored types const.
No functional change.
|
| 1.364 |
| 31-Aug-2021 |
rillig | lint: extract condition for message 259 into separate function
No functional change.
|
| 1.363 |
| 29-Aug-2021 |
rillig | lint: in C99 mode, make implicit function declarations an error
In tree.c 1.294 from 2021-06-28, I had already tried this, but at that
time, there were too many implicit function definitions in the NetBSD
tree. Most of them were GCC builtins, which lint did not recognize.
Therefore I had to revert to a warning in tree.c 1.302 from 2021-06-30.
In the meantime, lint has learnt to recognize compiler builtins, see
is_compiler_builtin, so try again now. The build logs from x86_64,
i386, sparc and sparc64 show no more implicit function declarations.
|
| 1.362 |
| 29-Aug-2021 |
rillig | lint: treat _mm_ as builtin function name prefix for all compilers
This prefix is not specific to GCC, it is also known by the Intel C
Compiler and the Microsoft C compiler.
|
| 1.361 |
| 29-Aug-2021 |
rillig | lint: add __sync_ and _mm_ as prefixes for builtin functions
These two additions cover all cases that occur in the current NetBSD
build on x86_64. This allows build_name to use the usual pattern 'if
Sflag then error else if sflag then warning'. That function currently
issues a warning in C99 as well, even though C99 prohibits implicit
function declarations.
|
| 1.360 |
| 28-Aug-2021 |
rillig | lint: remove double inversion from is_out_of_char_range
No functional change.
|
| 1.359 |
| 28-Aug-2021 |
rillig | lint: clean up check_integer_comparison
No functional change.
|
| 1.358 |
| 28-Aug-2021 |
rillig | lint: do not emit GCC builtin functions
Lint1 no longer emits declarations of GCC builtin functions and calls to
them.
Previously, lint generated 3421 useless warnings in a default NetBSD
build, like this:
__atomic_load_n, arg 1 used inconsistently
acl.c(216)[pointer to unsigned int]
rbtdb.c(921)[pointer to unsigned short]
This was because lint just doesn't understand that these functions are
type-generic, which is indeed unusual in C.
These useless warnings made the lint output more frightening than it
should actually be. Together with the strange formatting of the
diagnostics (space-space-tab after the main message, two spaces and two
colons between the occurrences, symbols are listed in hashcode order),
this creates the impression that lint is not intended to be a
user-friendly tool.
For now, fix the excess warnings, leaving the other items for later.
|
| 1.357 |
| 28-Aug-2021 |
rillig | lint: merge duplicate code in convert_constant_floating
|
| 1.356 |
| 28-Aug-2021 |
rillig | lint: use 'unsigned int' for bit-size of types in convert_integer
There was no need to have two separate magic values (0 and -1) to mean
the same.
No functional change.
|
| 1.355 |
| 28-Aug-2021 |
rillig | lint: use 'unsigned int' for bit-size of types
Lint does not need to support any types larger than 256 MB since they
don't occur in practice. Practically, such large types have never been
supported at all since the function type_size_in_bits used int for the
internal calculations, resulting in overflows.
|
| 1.354 |
| 28-Aug-2021 |
rillig | lint: remove redundant type casts from convert_constant_floating
Since tree.c 1.70 from 2012-03-27, lint has been using properly typed
constants for the min and max values, independent from the host
platform.
No functional change.
|
| 1.353 |
| 28-Aug-2021 |
rillig | lint: un-abbreviate unsigned integer types
No functional change.
|
| 1.352 |
| 25-Aug-2021 |
rillig | lint: rename local variable in plength
The assertion in build_plus_minus is safe since that function is only
called if the types are compatible. Pointer and floating are not
compatible.
No functional change.
|
| 1.351 |
| 23-Aug-2021 |
rillig | lint: expand macro in string concatenation
No functional change.
|
| 1.350 |
| 23-Aug-2021 |
rillig | lint: add quotes around placeholder in message 141
|
| 1.349 |
| 23-Aug-2021 |
rillig | lint: make overflow check for binary '+' simpler
No functional change.
|
| 1.348 |
| 23-Aug-2021 |
rillig | lint: remove unnecessary width parameter for msb
No functional change.
|
| 1.347 |
| 23-Aug-2021 |
rillig | lint: fix wrong warning about signed integer overflow in '-'
The newly added test cases in line 30 and 33 may or may not be worth a
warning since the expressions deal with unsigned integers, where the
behavior on overflow is well defined.
|
| 1.346 |
| 22-Aug-2021 |
rillig | lint: merge duplicate code for binary operator
No functional change.
|
| 1.345 |
| 22-Aug-2021 |
rillig | lint: fix folding of comparisons in constant expressions
|
| 1.344 |
| 21-Aug-2021 |
rillig | lint: extract check_prototype_conversion_integer
No functional change.
|
| 1.343 |
| 21-Aug-2021 |
rillig | lint: rename local variable '1' in fold_float
It looked too similar to the number 1.
|
| 1.342 |
| 21-Aug-2021 |
rillig | lint: fix spacing in tree.c
|
| 1.341 |
| 21-Aug-2021 |
rillig | lint: do not warn about '(void)arg' and similar expressions
In the current NetBSD build, 5260 of the 46264 total lint warnings are
about expressions that have a null effect. Most of these occurrences
follow well-established patterns, which makes the warnings bogus.
Remove these warnings.
|
| 1.340 |
| 19-Aug-2021 |
rillig | lint: change return type of 'msb' from int to bool
No functional change.
|
| 1.339 |
| 19-Aug-2021 |
rillig | lint: clean up formatting of 'fold'
|
| 1.338 |
| 19-Aug-2021 |
rillig | lint: fix wrong integer overflow warning for unsigned types
|
| 1.337 |
| 16-Aug-2021 |
rillig | lint: remove dead code from check_bad_enum_operation
There is a single caller of that function, and it checks the exact same
condition beforehand.
|
| 1.336 |
| 15-Aug-2021 |
rillig | lint: extend check for unconst functions
The functions memchr, strpbrk, strrchr and strstr effectively remove the
const qualifier of their first argument, just like strchr.
|
| 1.335 |
| 15-Aug-2021 |
rillig | lint: fix wrong warning about 'unsigned char >> constant'
|
| 1.334 |
| 14-Aug-2021 |
rillig | lint: use standard quoting style for messages 124 and 184
|
| 1.333 |
| 14-Aug-2021 |
rillig | lint: allow initialization of struct with constant member
The operator INIT, just like RETURN and FARG, initializes an object with
an expression. The target object of such an initialization may be a
struct with constant members.
The operator ASSIGN, on the other hand, is entirely different. It
overwrites the existing value of the object, and this is not allowed for
structs that have a constant member. Therefore it was wrong to use the
operator ASSIGN for initialization.
|
| 1.332 |
| 10-Aug-2021 |
rillig | lint: fix 3 of the 4 wrong messages about lvalue in initial assignment
|
| 1.331 |
| 09-Aug-2021 |
rillig | lint: warn about 'char * = strchr(const char *, int)'
Found in findcc.c, there are about 25 other instances of this
incongruency in the whole source tree.
For more examples of functions from the C Standard Library that
implicitly remove the 'const' qualifier from an argument, see the C++
include file 'cstring'.
|
| 1.330 |
| 03-Aug-2021 |
rillig | lint: merge almost duplicate code from 'sametype' into 'eqtype'
In 'sametype', the branch for comparing array types was unreachable
since it requires both tspecs to be the same, but t2 underwent the
array-to-pointer conversion.
Previously, lint warned about enum type mismatches, even without -e for
strict enum mode. Instead, it got the case for 'char *' wrong, which is
now fixed. Now lint behaves like GCC 10.3.0 in this regard. The
warning about enum mismatch is useful though, so it may be re-added in a
future commit.
|
| 1.329 |
| 03-Aug-2021 |
rillig | lint: union casts are only available as a GCC extension, not in C99
|
| 1.328 |
| 03-Aug-2021 |
rillig | lint: casting to a struct is not allowed in C99, only with GCC
|
| 1.327 |
| 03-Aug-2021 |
rillig | lint: clean up and move 'sametype'
This function is only used by lint1. That's good since the lint2 code
was completely broken, as it would regard any two struct types as being
the same.
Remove the large switch statement since it is unlikely that there will
be new type derivations in C anytime soon.
No functional change.
|
| 1.326 |
| 01-Aug-2021 |
rillig | lint: merge duplicate debugging code
The functions 'debug_node' and 'display_expression' were similar enough
to be merged.
Migrate debug_node to use the existing debug logging functions.
Remove the now unused option 'd' from the options string.
|
| 1.325 |
| 01-Aug-2021 |
rillig | lint: remove option -d, clean up debug logging
The command line option -d was not used by /usr/bin/lint, and it only
triggered a handful of debug messages. Move this debug logging over to
the compile-time -DDEBUG setting.
Move display_expression further up to avoid the forward declaration.
|
| 1.324 |
| 01-Aug-2021 |
rillig | lint: remove unreachable code from usual arithmetic conversions
|
| 1.323 |
| 01-Aug-2021 |
rillig | lint: fix usual arithmetic conversions for 128-bit integer types
|
| 1.322 |
| 31-Jul-2021 |
rillig | lint: extract debug logging to separate file
Lint currently has several different kinds of debug log:
* The -DDEBUG log is controlled at compile time.
* The -d command line options enables some other debug logging.
* The -DYYDEBUG log for parsing is controlled at compile time.
* The -y command line option only has an effect in -DYYDEBUG mode.
Extracting the logging into a separate file is a first step towards
unifying these logs and making the code for debug logging stand out less
than the current #ifdef DEBUG.
No functional change.
|
| 1.321 |
| 31-Jul-2021 |
rillig | lint: in has_constant_member, don't reuse variables
No functional change.
|
| 1.320 |
| 31-Jul-2021 |
rillig | lint: merge duplicate code for generating unqualified type
This is a preparation for fixing the wrong warnings in msg_115.c.
No functional change.
|
| 1.319 |
| 25-Jul-2021 |
rillig | lint: add type information to message about invalid cast
This probably doesn't matter in practice since the compiler is supposed
to catch this kind of error. Adding the type information mainly ensures
that parsing abstract type names works as specified.
|
| 1.318 |
| 20-Jul-2021 |
rillig | lint: use consistent naming scheme for functions that build nodes
No functional change.
|
| 1.317 |
| 20-Jul-2021 |
rillig | lint: split 'build' into build_binary and build_unary
No functional change.
|
| 1.316 |
| 15-Jul-2021 |
rillig | lint: allow pointer casts from 'char *' and 'unsigned char *'
For the sake of traditional code that did not yet migrate to using 'void
*' for arbitrary pointers.
In the standard NetBSD build, this reduces the number of lint warnings
by around 7000, of 57000 total.
|
| 1.315 |
| 15-Jul-2021 |
rillig | lint: replace call to ttos with type_name
Since tyname.c 1.20 from 2021-01-02, type_name is as simple to use as
ttos and more expressive. It can also be called multiple times without
invalidating the returned strings.
Used only in debug mode.
|
| 1.314 |
| 14-Jul-2021 |
rillig | lint: extract build_member_access from the grammar
No functional change.
|
| 1.313 |
| 06-Jul-2021 |
rillig | lint: rename type generic_association_types to generic_association
The word 'types' was misleading and unnecessary.
No functional change.
|
| 1.312 |
| 04-Jul-2021 |
rillig | lint: replace wrong assumption with citation from C99
No functional change.
|
| 1.311 |
| 04-Jul-2021 |
rillig | lint: document where to fix missing const in lvalue-to-rvalue
No functional change.
|
| 1.310 |
| 04-Jul-2021 |
rillig | lint: align messages for invalid left operand of '.' or '->'
|
| 1.309 |
| 04-Jul-2021 |
rillig | lint: clean up new_tnode
No functional change.
|
| 1.308 |
| 04-Jul-2021 |
rillig | lint: remove remaining support for lvalue casts
These had been GCC extensions until GCC 3.4, they were removed in GCC
4.0.
|
| 1.307 |
| 04-Jul-2021 |
rillig | lint: remove support for obsolete GCC-style lvalue casts
Since GCC 4.0, a cast results in an rvalue.
|
| 1.306 |
| 04-Jul-2021 |
rillig | lint: in strict bool mode, allow mixed types in generated C code
This allows flex lexers to be run through lint in strict bool mode.
|
| 1.305 |
| 04-Jul-2021 |
rillig | lint: remove outdated assertion
Since err.c 1.12 from 2000-07-06, lint allows to suppress individual
error messages. Suppressed error messages do not increment nerr.
Keeping nerr at 0 had triggered the assertion.
|
| 1.304 |
| 03-Jul-2021 |
rillig | lint: fix assertion failure in display_expression for bool constants
|
| 1.303 |
| 30-Jun-2021 |
rillig | lint: do not warn about use of implicitly declared GCC builtins
|
| 1.302 |
| 30-Jun-2021 |
rillig | lint: in C99 mode, only warn about implicit function declarations
Since tree.c 1.294 from 2021-06-28 (two days ago), lint errored out on
an implicit function declaration. In principle it is correct to do so
since C99 requires it, but in practice there are a several functions
that are not declared in the translation unit itself since they are
provided by the compiler. Typical examples for GCC and Clang are the
various functions named '__builtin_*' and '__atomic_*'.
For now, only warn about these but don't error out.
|
| 1.301 |
| 30-Jun-2021 |
rillig | lint: mention the name of an implicitly declared function
In the regular NetBSD builds, this happened in swab.c:65. That line
contains __predict_false, which may or may not be a macro. In other
cases, there may be more than one function call in a single line.
|
| 1.300 |
| 30-Jun-2021 |
rillig | lint: fix type name in message 101 for wrong member name
|
| 1.299 |
| 30-Jun-2021 |
rillig | lint: improve comment about INT keeping t_is_enum
|
| 1.298 |
| 29-Jun-2021 |
rillig | lint: rename xsign to convert_integer
The term sign-extend was too specific, the function actually does a
broader conversion.
No functional change.
|
| 1.297 |
| 29-Jun-2021 |
rillig | lint: do not sign-extend pointers
Sign extension only makes sense for integer types.
No idea why the code had tried to sign-extend pointers for the past 26
years.
|
| 1.296 |
| 29-Jun-2021 |
rillig | lint: document where to fix the wrong warning for '\xff'
|
| 1.295 |
| 29-Jun-2021 |
rillig | lint: fix typos in comment
|
| 1.294 |
| 28-Jun-2021 |
rillig | lint: in C99 mode, complain about implicitly declared functions
C99, foreword, p5, item 22 lists among the major changes from C90:
"remove implicit function declaration".
|
| 1.293 |
| 28-Jun-2021 |
rillig | lint: add type information to message 155 (type mismatch)
|
| 1.292 |
| 27-Jun-2021 |
rillig | lint: fix type comparison in _Generic selection expressions
In the newly added test comma_expression, there were two distinct type
objects for the tspec DOUBLE.
|
| 1.291 |
| 27-Jun-2021 |
rillig | lint: fix result type of _Generic expressions
|
| 1.290 |
| 20-Jun-2021 |
rillig | lint: remove unnecessary call to print_tnode
For the possible operators that occur in message 324, print_tnode is
equivalent to op_name, and the latter is simpler.
When the function print_node was added to the code base, it had another
use in init.c, for understanding how initialization works in lint. That
code has since been rewritten completely, therefore print_tnode is no
longer needed. For debugging, display_expression is the better choice
since it has multi-line output and does not suffer from a fixed-length
buffer.
No functional change.
|
| 1.289 |
| 20-Jun-2021 |
rillig | lint: rename val_t.v_unsigned to avoid confusion
The name v_unsigned suggested that the value would be interpreted as
unsigned, which was wrong. Whether a value is signed or unsigned is
decided by v_tspec instead.
Revert the previous commit for boolen constants since their value is
already interpreted as unsigned, and there is no need for any warning
about differences between traditional C and ANSI C since the _Bool type
has only been added ten years later in C99.
The code for printing a tree node was also confused by this struct
member, even with its old name v_ansiu. That code will be fixed in a
follow-up commit.
No functional change.
|
| 1.288 |
| 20-Jun-2021 |
rillig | lint: rename val_t.v_ansiu to v_unsigned
When lint was written in 1995, traditional C was still nearby since C90
had been around for only 5 years. 26 years later, almost all code
adheres to C90 or even C99 or C11, therefore "C90 or later" can safely
be assumed as the default.
No functional change.
|
| 1.287 |
| 15-Jun-2021 |
rillig | lint: replace array access with function calls
First and foremost, the test d_c99_complex_split accessed the array
qlmasks out-of-bounds, with an index of 128 for the type 'double
_Complex'. This invoked undefined behavior since the maximum allowed
index was 64.
Replacing the raw array accesses with function calls allows for bounds
checks to catch these errors early.
Determining the value bits for a 'double _Complex' does not make sense
at all since it is not an integer type. This means that lint didn't
handle these types correctly for several years. Support for int128_t
has been added in inittyp.c 1.12 from 2018-09-07, support for _Complex
has been added in inittyp.c 1.9 from 2008-04-26.
Determining the value bits for an int128_t would make sense, but the
unit tests don't contain examples for this type since at the moment all
unit tests must produce the same results on 32-bit and 64-bit platforms,
and the 32-bit platforms don't support int128_t.
|
| 1.286 |
| 15-Jun-2021 |
rillig | lint: extract convert_constant_to_floating
|
| 1.285 |
| 15-Jun-2021 |
rillig | lint: split convert_constant_check_range
No functional change.
|
| 1.284 |
| 15-Jun-2021 |
rillig | lint: merge duplicate code
|
| 1.283 |
| 15-Jun-2021 |
rillig | lint: extract convert_constant_check_range
No functional change.
|
| 1.282 |
| 15-Jun-2021 |
rillig | lint: extract convert_constant_floating
No functional change.
|
| 1.281 |
| 04-May-2021 |
rillig | lint: fix assertion failure when promoting a bit-field larger than int
|
| 1.280 |
| 18-Apr-2021 |
rillig | lint: rename parameter to expr
That parameter used to be used for a single purpose, later it got used
for checking the reachability as well, which made the name misleading.
|
| 1.279 |
| 18-Apr-2021 |
rillig | lint: remove redundant CONSTCOND
In do-while-0 loops, these are no longer needed since tree.c 1.202 from
2021-01-31.
|
| 1.278 |
| 18-Apr-2021 |
rillig | lint: extend documentation about handling initializations
|
| 1.277 |
| 17-Apr-2021 |
rillig | lint: do not warn about alignment when casting from incomplete struct
This removes a bunch of technically correct but practically useless
warnings from the regular NetBSD build.
|
| 1.276 |
| 10-Apr-2021 |
rillig | lint: prepare renaming of 'struct type'
It's confusing to have the same struct tag in both lint1 and lint2, with
mostly the same members, but also some differences. Before actually
changing this, I reviewed all occurrences of the word 'type' in the
code.
No functional change.
|
| 1.275 |
| 09-Apr-2021 |
rillig | lint: fix wrong warning about uninitialized _Complex variable
Seen in divxc3.c.
|
| 1.274 |
| 09-Apr-2021 |
rillig | lint: do not warn about pointer conversion to or from incomplete type
This cuts down the warnings in a NetBSD release build by about 38,000.
|
| 1.273 |
| 09-Apr-2021 |
rillig | lint: clean up the check for pointer conversions
No functional change.
|
| 1.272 |
| 08-Apr-2021 |
rillig | lint: don't warn about cast between pointers to compatible structs
|
| 1.271 |
| 06-Apr-2021 |
rillig | lint: for shift in C99 mode, do not warn about difference to pre-C90
C99 is too far away from traditional C to make this warning useful.
There are 3 different situations in which this warning is generated:
For '1 << (unsigned char)1', the result type is 'unsigned int' in
traditional C. The result type is unsigned because at least 1 of the
operators is unsigned, and it is 'unsigned int' because the usual
arithmetic promotions are applied.
For '1 >> (long)1', as well as for '1 << (long)1', the result type is
'long' in traditional C since the usual arithmetic promotions are
applied.
Omitting this warning in C99 mode reduces the amount of lint warnings in
a typical NetBSD release build by approximately 6800 of 107000 total.
|
| 1.270 |
| 06-Apr-2021 |
rillig | lint: reduce indentation of typeok_shift
No functional change.
|
| 1.269 |
| 06-Apr-2021 |
rillig | lint: add details to warning about too large shift amount
The previous message 'shift greater than size of object' was too short
to give reasonable hints, especially when the expressions involve
typedefs or macros.
|
| 1.268 |
| 06-Apr-2021 |
rillig | lint: fix wrong warning about losing accuracy when converting to _Bool
|
| 1.267 |
| 06-Apr-2021 |
rillig | lint: move check for strict bool mode into separate file
No functional change.
|
| 1.266 |
| 05-Apr-2021 |
rillig | lint: warn about for wrong type cast in argument to ctype.h functions
The argument to most of the functions from <ctype.h> "shall either be
representable as an 'unsigned char' or shall equal the value of the
macro EOF".
When confronted with the infamous warning 'array subscript has type
char', there are enough programmers who don't know the background of
that warning and thus fix it in a wrong way. Neither GCC nor Clang
explain its warning to target these programmers.
Both GCC and Clang warn about 'array subscript has type char', but they
ignore the other requirements of the <ctype.h> functions, even though
these are in the C standard library.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94182
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95177
https://stackoverflow.com/a/60696378
|
| 1.265 |
| 02-Apr-2021 |
rillig | lint: reword message about constant argument to '!'
In the other messages, operators are typically written in their literal
form as well. Using single quotes disambiguates them from normal
punctuation.
|
| 1.264 |
| 02-Apr-2021 |
rillig | lint: fix typo from previous commit in merge_qualifiers
|
| 1.263 |
| 02-Apr-2021 |
rillig | lint: clean up merge_qualifiers
No functional change.
|
| 1.262 |
| 02-Apr-2021 |
rillig | lint: rename local variable in build_colon
The name 'rtp' was confusing since it did not correspond to 'rn'. The
'r' meant 'result', not 'right-hand side'.
No functional change.
|
| 1.261 |
| 02-Apr-2021 |
rillig | lint: replace a new modtab with op_name
This reduces the types of variables that are passed around.
No functional change.
|
| 1.260 |
| 02-Apr-2021 |
rillig | lint: rename getopname to op_name
There are several functions called 'get' that allocate memory. Remove
this possible confusion.
No functional change.
|
| 1.259 |
| 02-Apr-2021 |
rillig | lint: add parentheses after sizeof, as required by share/misc/style
No functional change.
|
| 1.258 |
| 02-Apr-2021 |
rillig | lint: rename functions for duplicating types
No functional change.
|
| 1.257 |
| 02-Apr-2021 |
rillig | lint: name memory allocation functions consistently
No functional change.
|
| 1.256 |
| 02-Apr-2021 |
rillig | lint: rename getnode to expr_zalloc_tnode
The new name highlights that the returned memory is only valid in the
scope of the current expression. This was misleading before since the
other related functions all have a 't' (probably for 'temporary') in
their names.
Also encode in the function name that the returned memory is zeroed out
as that could not be inferred from the old name.
No functional change.
|
| 1.255 |
| 01-Apr-2021 |
rillig | lint: remove wrong assumption from comment
The size in bits of a struct or union is not measured at all at this
point since portable_size_in_bits only takes the broad type
classification (tspec_t), not the precise type information (type_t).
No functional change.
|
| 1.254 |
| 30-Mar-2021 |
rillig | lint: add type information for message about unknown member name
|
| 1.253 |
| 28-Mar-2021 |
rillig | lint: sprinkle const on function declarations
No functional change.
|
| 1.252 |
| 27-Mar-2021 |
rillig | lint: remove space between 'sizeof ('
No functional change.
|
| 1.251 |
| 26-Mar-2021 |
rillig | lint: add quotes around placeholder in message about undefined variable
Before: error: expected undefined [99]
After: error: 'expected' undefined [99]
Seen in external/mpl/bind, which for Clang defines in stdatomic.h:
> #define atomic_exchange_explicit(obj, desired, order) \
> __c11_atomic_exchange_explicit(obj, expected, order)
Note the mismatch between 'desired' and 'expected'.
|
| 1.250 |
| 26-Mar-2021 |
rillig | lint: in malloc calls, use 'sizeof *ptr' instead of 'sizeof(type)'
No functional change.
|
| 1.249 |
| 26-Mar-2021 |
rillig | lint: allow pointer cast to 'char *' and 'unsigned char *'
This reduces the number of warnings in the regular NetBSD build by 5560.
|
| 1.248 |
| 26-Mar-2021 |
rillig | lint: rename variables in check_pointer_conversion
No functional change.
|
| 1.247 |
| 26-Mar-2021 |
rillig | lint: extract decision for warning about pointer cast
This makes the code several lines longer but way more readable. In the
previous dense expression it was hard to see what was going on at all
and that there are two completely separate situations in which this
warning applies.
No functional change.
|
| 1.246 |
| 22-Mar-2021 |
rillig | lint: use precise type information in message about type mismatch
While here, reword the message, avoiding operators and parentheses.
Since 2021-01-02, providing the precise type name is as easy as the
broad type classification (just replace tspec_name with type_name), and
it's definitely more useful to the human readers.
|
| 1.245 |
| 21-Mar-2021 |
rillig | lint: invert 'rchflag', call it warn_about_unreachable instead
No functional change.
|
| 1.244 |
| 21-Mar-2021 |
rillig | lint: reduce number of places where 'reached' is set
When determining the reachability of a statement, the idea was that
whenever 'reached' was set to false, 'rchflg' (the abbreviation for "do
not warn about unreachable statements") would be reset as well.
In some (trivial) cases, this was done, but many more interesting cases
simply forgot to set this second variable. To prevent this in the
future, encapsulate this in a simple helper function.
Now even if a statement is reachable, 'rchflg' gets reset. This does
not hurt since as long as the current statement is reachable, the value
of 'rchflg' does not matter.
No functional change. There would be quite a big functional change
though if check_statement_reachable were to reset 'rchflg' instead of
'reached', as the comment already suggests. In that case, with the
current code, many legitimate warnings about unreachable statements
would be skipped, especially those involving 'if' statements, since
these didn't reset 'rchflg' properly before.
|
| 1.243 |
| 21-Mar-2021 |
rillig | lint: fix wrong 'statement not reached' in do-while loop
|
| 1.242 |
| 20-Mar-2021 |
rillig | lint: clean up new_name_node
No functional change.
|
| 1.241 |
| 20-Mar-2021 |
rillig | lint: move getopname over to tree.c
Except for the one use in print_tnode, the name of the operator is only
used in tree.c.
No functional change.
|
| 1.240 |
| 20-Mar-2021 |
rillig | lint: remove redundant operator properties table
It's enough to have modtab, which describes the properties of the
various operators. There is no need to have a second table imods that
holds the same content. Rather make modtab constant as well.
The only possible functional change is that the names of the internal
operators 'no-op', '++', '--', 'real', 'imag' and 'case' may appear in
diagnostics, where previously lint invoked undefined behavior by passing
a null pointer for a '%s' conversion specifier.
|
| 1.239 |
| 20-Mar-2021 |
rillig | lint: inline access to type properties in check_bad_enum_operation
No functional change.
|
| 1.238 |
| 20-Mar-2021 |
rillig | lint: in strict bool mode, check initialization as well
C99 6.7.8p11 says for initialization that "the same type constraints and
conversions as for simple assignments apply", so actually apply them.
(I had just forgotten this "operator" when I first implemented strict
bool mode.)
|
| 1.237 |
| 19-Mar-2021 |
rillig | lint: rename in_bit to in_bits
No functional change.
|
| 1.236 |
| 19-Mar-2021 |
rillig | lint: rename tsize to type_size_in_bits
The shorter name size_in_bits was already taken by the function-like
macro with argument type tspec_t.
No functional change.
|
| 1.235 |
| 18-Mar-2021 |
rillig | lint: make the debug log for nodes more readable
The operator NAME has the name 'name', therefore no special case is
needed.
Having the words 'with type' in the message makes the message easier to
find from the debug log. Given that the operator name is used unquoted,
the log message 'name: int value=111' was nearly impossible to find in
the code.
Replace the '()' with an actual word, to avoid any confusion about
whether the type name might be a function type without prototype.
Reduce the amount of '=' signs, instead use commas to separate the
properties of the node.
No functional change outside debug mode.
|
| 1.234 |
| 18-Mar-2021 |
rillig | lint: reword message about type mismatch in initialization
Using parentheses for quotes is unusual, furthermore the previous
message didn't follow proper grammar rules, sacrificing clarity for
brevity.
|
| 1.233 |
| 17-Mar-2021 |
rillig | lint: rename 'blklev' to 'block_level'
No functional change.
|
| 1.232 |
| 28-Feb-2021 |
rillig | lint: fix null pointer dereference on parse error
Fixes PR bin/22119.
|
| 1.231 |
| 28-Feb-2021 |
rillig | lint: document where in C99 the behavior of cconv is specified
|
| 1.230 |
| 28-Feb-2021 |
rillig | lint: replace wrong comment with assertion
The broad type of a value is indeed stored in the value itself, in the
member v_tspec. For nodes that refer to this value, it is redundantly
stored, it always equals tn->tn_type->t_tspec.
After initialization, neither tn->tn_type nor val->v_tspec are modified.
This is not ensured by the compiler but has to be analyzed manually.
No functional change.
|
| 1.229 |
| 28-Feb-2021 |
rillig | lint: rename is_nonzero to constant_is_nonzero
The new function name emphasizes that the given node must have the
operator CON.
No functional change.
|
| 1.228 |
| 28-Feb-2021 |
rillig | lint: rename members and access macros of the basic types
Having the measurement unit in the variable name prevents accidental
confusion between bits and bytes, especially since usually the word
'size' means the size in bytes, while 'width' means the size in bits, at
least for integer types.
No functional change.
|
| 1.227 |
| 28-Feb-2021 |
rillig | lint: do not warn about constant expressions involving sizeof
These expressions are indeed constant for a specific platform, but on
another platform their value may change. This makes them unsuspicious
and legitimate for portable code.
Seen in rump_syscalls.c, as 'sizeof(int) > sizeof(register_t)'.
|
| 1.226 |
| 28-Feb-2021 |
rillig | lint: rename parameter in function 'expr'
For symmetry with the function is_constcond_false.
No functional change.
|
| 1.225 |
| 28-Feb-2021 |
rillig | lint: add type information for 'incompatible struct pointers'
|
| 1.224 |
| 28-Feb-2021 |
rillig | lint: add type information to enum type mismatch
|
| 1.223 |
| 28-Feb-2021 |
rillig | lint: make messages for incompatible pointers more specific
Message 153 didn't state obviously which of the pointer types was the
one before conversion (or cast) and which was the resulting type.
Message 229 didn't have any type information at all.
|
| 1.222 |
| 28-Feb-2021 |
rillig | lint: add type information to 'possible pointer alignment problem [135]'
This warning occurs more than 7400 times in a regular NetBSD build, and
without giving any type information, leaves the reader clueless about
what the underlying issue might be. Add type information since that is
a no-brainer to implement.
|
| 1.221 |
| 28-Feb-2021 |
rillig | lint: skip alignment computation if possible
Testing a global variable is simpler than calling a 20-line function.
No functional change.
|
| 1.220 |
| 28-Feb-2021 |
rillig | lint: rename getbound to alignment_in_bits
No functional change.
|
| 1.219 |
| 27-Feb-2021 |
rillig | lint: rename confusing local variable
The variable name rtp is reserved for the type of the right-hand
operand.
No functional change.
|
| 1.218 |
| 22-Feb-2021 |
rillig | lint: change spelling of initialisation to initialization
That's the wording from the ISO C99 standard.
|
| 1.217 |
| 22-Feb-2021 |
rillig | lint: improve debug message and comment
|
| 1.216 |
| 21-Feb-2021 |
rillig | lint: extract check_non_constant_initializer from init_using_expr
No functional change.
|
| 1.215 |
| 21-Feb-2021 |
rillig | lint: always initialize return values of constant_addr
Before, the caller was responsible for initializing the return values
from the function. This was an unexpected burden.
Ensure that in each branch that returns true, both return values are
properly set.
Strangely, the only caller of that function, init_using_expr, uses
neither of the return values. It just tests whether the expression is
constant or not.
No functional change.
|
| 1.214 |
| 21-Feb-2021 |
rillig | lint: add type information to message about enum mismatch
|
| 1.213 |
| 21-Feb-2021 |
rillig | lint: indent node details in debug mode
|
| 1.212 |
| 20-Feb-2021 |
rillig | lint: fix lint warnings
No functional change.
|
| 1.211 |
| 20-Feb-2021 |
rillig | lint: do not warn about 'do { ... } while (false)' in strict bool mode
|
| 1.210 |
| 20-Feb-2021 |
rillig | lint: extend debugging for initializing objects
No functional change outside debug mode.
|
| 1.209 |
| 19-Feb-2021 |
rillig | lint: rename t_isenum and t_aincompl to be more expressive
No functional change.
|
| 1.208 |
| 19-Feb-2021 |
rillig | lint: rename str_t and its members to be more expressive
No functional change.
|
| 1.207 |
| 15-Feb-2021 |
rillig | lint: remove redundant comment
That comment was useful when there was no function is_null_pointer.
Back then, the code for testing a null pointer was written in-line,
which made it really hard to see what's going on. This is no longer the
case.
|
| 1.206 |
| 15-Feb-2021 |
rillig | lint: extract typeok_colon_pointer from typeok_colon
The subtype information is now only accessed if both operands are
actually pointers.
No functional change.
|
| 1.205 |
| 04-Feb-2021 |
rillig | lint: add back "due to prototype" to message 259
That message is only supposed to warn about compatibility to traditional
C, in case the function should ever be compiled without its prototype
being in effect. All other type checks are supposed to be in another
function, as documented, but that type check misses to report a few
error-prone type combinations (long to char, long to int).
30 years after the introduction of prototypes with C90, almost all
existing code uses prototypes. The warning has thus lost most of its
usefulness and can rather be confusing since a conversion from 'char' to
'long' is not problematic with prototypes in action, and the probability
of the code being backported to a pre-C90 compiler is diminishingly
small.
The words "due to prototype" now serve as a hint again. The proper fix
could be to suppress this warning in C99 mode since that's far enough
from traditional C.
|
| 1.204 |
| 31-Jan-2021 |
rillig | lint: reword message 259 about function argument conversion
The words "due to prototype" are an anachronism from the 1990s.
Nowadays every function is defined using a prototype, which makes these
words redundant.
|
| 1.203 |
| 31-Jan-2021 |
rillig | lint: make warning about function argument conversion more detailed
For every conversion it is useful to know both the source and the target
type since these are not always obvious from the code.
The only surprise is the warning in d_gcc_extension. The conversion
there is from 'double' to 'long double', which is a lossless conversion.
This may be a bug in lint.
|
| 1.202 |
| 31-Jan-2021 |
rillig | lint: don't warn about constant condition in 'do { } while (0)'
|
| 1.201 |
| 31-Jan-2021 |
rillig | lint: fix strange message about nested '==' operators
If one of the nested subexpressions is parenthesized, the author
probably knew how these expressions are evaluated. Therefore don't warn
in such a situation.
Maybe the original author once made a typo and tried to initialize
variables but instead compared them, like this:
int a, b, c;
a == b == c;
This would explain the text of the message, which still sounds strange.
At least it doesn't show up as often anymore.
|
| 1.200 |
| 31-Jan-2021 |
rillig | lint: mention the operator in messages about bit shifts
|
| 1.199 |
| 30-Jan-2021 |
rillig | lint: flatten has_side_effect
Since GCC performs tail call optimization, the generated code is
practically the same.
Replace redundant comments with open questions.
No functional change.
|
| 1.198 |
| 30-Jan-2021 |
rillig | lint: fix wrong 'expression has null effect'
|
| 1.197 |
| 30-Jan-2021 |
rillig | lint: extract has_side_effect from check_null_effect
No functional change.
|
| 1.196 |
| 30-Jan-2021 |
rillig | lint: add type information to message 275
Before: cast discards 'const' from pointer target type
After: cast discards 'const' from type 'pointer to const char'
Seen in sqlite3.c.
|
| 1.195 |
| 30-Jan-2021 |
rillig | lint: add type information to warning about troublesome casts
The previous warning text did not mention the actual types that are
involved in the type conversion. These types can be hard to see from
the source code as soon as macros are involved, and even in plain code,
one would have to follow the declarations, which is an unnecessary
burden. Lint already has all information about the involved types, so
there is no reason for omitting this crucial information.
Seen in external/mit/lua/dist/src/lvm.c and several other files.
Including the type information in the message immediately makes the
message scarier.
Before: pointer casts may be troublesome
After: pointer cast from 'pointer to struct TString' to 'pointer to
union GCUnion' may be troublesome
|
| 1.194 |
| 30-Jan-2021 |
rillig | lint: rename incompl to is_incomplete
No functional change.
|
| 1.193 |
| 30-Jan-2021 |
rillig | lint: fix type conversion for very large data types
Data types that are 4 GB or larger are an edge case. Nevertheless,
compute their size correctly.
|
| 1.192 |
| 30-Jan-2021 |
rillig | lint: remove redundant parentheses
No functional change.
|
| 1.191 |
| 24-Jan-2021 |
rillig | lint: reduce preprocessor magic for platform target types
|
| 1.190 |
| 24-Jan-2021 |
rillig | lint: extract duplicate code into is_null_pointer
No functional change.
|
| 1.189 |
| 24-Jan-2021 |
rillig | lint: fix wrong warning about null pointer comparison
|
| 1.188 |
| 24-Jan-2021 |
rillig | lint: fix local variable names in check_pointer_comparison
|
| 1.187 |
| 24-Jan-2021 |
rillig | lint: clean up typeok_eq and check_pointer_comparison
Since typeok_eq does not issue any diagnostics, prefix it with "is_".
By convention, the variable rt aliases rn->tn_type->t_tspec. Make it
obvious that in check_pointer_comparison, rt corresponds to the subtype
of the pointer.
|
| 1.186 |
| 24-Jan-2021 |
rillig | lint: document bug in typeok_eq
Since rt is an alias for rn->tn_type->t_tspec, it cannot be PTR and VOID
at the same time. This makes the condition unsatisfiable. Removing
that part of the code didn't show any change in behavior, as expected.
It may even be that fixing this obvious bug doesn't show any change in
behavior since that function is only used in a single place and
check_pointer_comparison performs its own checks before issuing any
warning.
At least the test cases added to msg_124.c all run as expected.
|
| 1.185 |
| 24-Jan-2021 |
rillig | lint: update data types in comments from int to bool
|
| 1.184 |
| 23-Jan-2021 |
rillig | lint: fix error message for relaxed bool operations in system headers
In strict mode, allowing 1 as bool constant expression is probably not
needed in practice since most comparisons are != 0 instead of == 0.
Furthermore, in the expression (flags & 0x0002) == true, comparing with
true is misleading since the '==' operator can never evaluate to true in
this case.
|
| 1.183 |
| 23-Jan-2021 |
rillig | lint: fix two wrong error messages in strict bool mode
The strict bool mode gets complicated because for system headers the
rules need to be relaxed since they cannot be changed easily, often not at all.
Still, if lint validates a program in strict bool mode, that program
must run with equal behavior regarding boolean expressions even on a
pre-C99 platform.
|
| 1.182 |
| 18-Jan-2021 |
rillig | lint: clean up code (mostly comments)
|
| 1.181 |
| 17-Jan-2021 |
rillig | lint: allow system headers to use int as bool, even in strict bool mode
|
| 1.180 |
| 17-Jan-2021 |
rillig | lint: rename typeok_amper to typeok_address
This is about the address-of operator, not about the '&' sign that is
also used for bitwise and.
|
| 1.179 |
| 17-Jan-2021 |
rillig | lint: add type information to message 126
|
| 1.178 |
| 17-Jan-2021 |
rillig | lint: flatten typeok_ordered_comparison
|
| 1.177 |
| 17-Jan-2021 |
rillig | lint: extend a few message comments
|
| 1.176 |
| 17-Jan-2021 |
rillig | lint: add test for incrementing const variable
|
| 1.175 |
| 17-Jan-2021 |
rillig | lint: clean up typeok_ordered_comparison
|
| 1.174 |
| 17-Jan-2021 |
rillig | lint: fix return type of conaddr
|
| 1.173 |
| 17-Jan-2021 |
rillig | lint: reduce scope of local variables in check_expr_misc
|
| 1.172 |
| 17-Jan-2021 |
rillig | lint: replace integer constants with bool constants
|
| 1.171 |
| 17-Jan-2021 |
rillig | lint: rename bitwise operators
When there are several variants of the AND operator, both of them should
get a distinguishing prefix, otherwise it's not clear which of the two
possible operators is meant by the plain AND.
|
| 1.170 |
| 17-Jan-2021 |
rillig | lint: rename operator STAR to INDIR
C99 calls this operator the "indirection operator". The word "star"
does not occur in the index of that standard.
|
| 1.169 |
| 17-Jan-2021 |
rillig | lint: rename operator AMPER to ADDR
There is no operator called 'ampersand'. The ampersand sign is merely
the textual representation of both the operator ADDR and the operator
BITAND.
|
| 1.168 |
| 17-Jan-2021 |
rillig | lint: rename build_ampersand to build_address
At that stage of analysis, the ampersand is no longer ambiguous, it has
already been resolved as the address-of operator, instead of the
bitwise-and operator.
|
| 1.167 |
| 17-Jan-2021 |
rillig | lint: extract fallback_symbol from new_name_node
The code was too close to the right margin, causing unnatural line
breaks. Furthermore it was getting too long and detailed.
|
| 1.166 |
| 17-Jan-2021 |
rillig | lint: flatten check_pointer_integer_conversion
|
| 1.165 |
| 17-Jan-2021 |
rillig | lint: in strict bool mode, don't treat bool as arithmetic type
|
| 1.164 |
| 17-Jan-2021 |
rillig | lint: rename mod_t members
|
| 1.163 |
| 16-Jan-2021 |
rillig | lint: refactor strict bool mode and improve comments
The previous comment above typeok_strict_bool_compatible was too hard to
understand.
|
| 1.162 |
| 16-Jan-2021 |
rillig | lint: in strict bool mode, perform the normal checks as well
|
| 1.161 |
| 16-Jan-2021 |
rillig | lint: simplify typeok_strict_bool_compatible
|
| 1.160 |
| 16-Jan-2021 |
rillig | lint: add missing warning for bool() == int
|
| 1.159 |
| 16-Jan-2021 |
rillig | lint: simplify typeok_strict_bool_assign
When that function is called, the nodes are already before_conversion.
|
| 1.158 |
| 16-Jan-2021 |
rillig | lint: remove redundant calls to before_conversion
|
| 1.157 |
| 16-Jan-2021 |
rillig | lint: replace integer constant expressions with true and false
LINTFLAGS=-gST make lint, with manual review.
The error messages from lint are all correct, they are not complete
though. The return value of a function returning bool may still be
compared to the integer 0.
|
| 1.156 |
| 16-Jan-2021 |
rillig | lint: in strict bool mode, integer constants do not have type bool
Previously, lint1 allowed integer constants such as 0 and 1 to be used
as bool constants. This was only half-baked since after fixing all
error messages from that strict mode, there may still be integer
literals in the code that should be replaced with true or false. This
would stop a migration from int to bool in the middle, leaving
inconsistent code around.
To find the remaining type inconsistencies, treat integers and bool as
completely incompatible, even for compile time constants.
|
| 1.155 |
| 16-Jan-2021 |
rillig | lint: prepare to make strict bool mode even stricter
Currently, strict bool mode still allows integer constant expressions to
be converted implicitly to bool. This is something that other languages
such as Go, Java, C#, Pascal don't allow.
By providing a custom implementation of <stdbool.h> that defines false
and true to custom bool constant identifiers, lint will cover these
cases as well.
To prepare for this, reword the rules and restructure the tests in
d_c99_bool_strict.c.
|
| 1.154 |
| 16-Jan-2021 |
rillig | lint: replace 0 and 1 with false and true, where appropriate
Change in behavior: Passing the option -h exactly 4294967296 times or
any multiple thereof is no longer equivalent to passing it never at all,
it is now equivalent to passing it once. See main2.c, hflag++ for the
actual change.
Other than that, no functional change intended.
A very large portion of the code already conformed to the requirements
of the strict bool mode. The only missing thing was using the constant
literals false and true instead of 0 and 1. For sure there are some
integer literals left that can be converted. For now, all literals that
appeared in the form " = 0" or " = 1" have been replaced.
|
| 1.153 |
| 15-Jan-2021 |
rillig | lint: merge duplicate code for non-zero detection
|
| 1.152 |
| 14-Jan-2021 |
rillig | lint: add type details to message for enum type mismatch
Before December 2020, it was cumbersome to add type information to a
message since the caller had to explicitly allocate buffers for the type
names. That's probably the reason why this crucial detail had been left
out of the warning.
|
| 1.151 |
| 12-Jan-2021 |
rillig | lint: add new check for strict bool mode
In strict bool mode, bool is considered incompatible with all other
scalar types, just as in Java, C#, Pascal.
The controlling expressions in if statements, while loops, for loops and
the '?:' operator must be of type bool. The logical operators work on
bool instead of int, the bitwise operators accept both integer and bool.
The arithmetic operators don't accept bool.
Since <stdbool.h> implements bool using C preprocessor macros instead of
predefining the identifiers "true" and "false", the integer constants 0
and 1 may be used in all contexts that require a bool expression.
Except from these, no implicit conversion between bool and scalar types
is allowed.
See usr.bin/tests/xlint/lint1/d_c99_bool_strict.c for more details.
The command line option -T has been chosen because all obvious choices
(-b or -B for bool, -s or -S for strict) are already in use. The -T may
stand for "types are checked strictly".
The default behavior of lint doesn't change. The strict bool check is
purely optional.
An example program for strict bool mode is usr.bin/make, which has been
using explicit comparisons such as p != NULL, ch != '\0' or n > 0 in
most places for a long time now, even before the refactoring in 2020.
|
| 1.150 |
| 11-Jan-2021 |
rillig | lint: split typeok into several functions
|
| 1.149 |
| 11-Jan-2021 |
rillig | lint: sprinkle a few const modifiers throughout the code
|
| 1.148 |
| 10-Jan-2021 |
rillig | lint: fix conversion of non-constant scalar to _Bool
|
| 1.147 |
| 10-Jan-2021 |
rillig | lint: fix conversion of constant expressions to _Bool
|
| 1.146 |
| 10-Jan-2021 |
rillig | lint: rename cvtcon to convert_constant
No functional change.
|
| 1.145 |
| 10-Jan-2021 |
rillig | lint: remove redundant braces in is_confusing_precedence
This nicely aligns the different branches, in which "l" and "r" are
swapped.
No functional change.
|
| 1.144 |
| 10-Jan-2021 |
rillig | lint: rename type classification macros
The previous names tspec_is_int and tspec_is_uint were confusing because
there are actually tspec_t constants called INT and UINT, these
classification macros return true for other integer types as well,
though.
While here, remove the prefix "tspec_" from these macros. It wasn't as
helpful as intended, in many cases it was obviously redundant, when it
was called as tspec_is_integer(tn->tn_type->t_tspec).
No functional change.
|
| 1.143 |
| 09-Jan-2021 |
rillig | lint: push down complexity from typeok to typeok_shr
Contrary to the comment in typeok, the types of the expressions before
promotions and conversions are not needed for SHL, SHLASS and SHRASS.
Move that code over to typeok_shr, the only place where it is actually
used. This removes another 3 variables from typeok.
|
| 1.142 |
| 09-Jan-2021 |
rillig | lint: move pointer subtypes from typeok to the sub-functions
The code in typeok is already complicated enough. Only few of the
checks actually examine the subtype of the pointer, the others don't
need to look at it. Therefore don't initialize the variables lstp
(left-hand side subtype) and rstp unless they are actually needed.
This reduces the number of variables in typeok and the number of
parameters to the sub-functions.
|
| 1.141 |
| 09-Jan-2021 |
rillig | lint: change return type of typeok to bool
No functional change.
|
| 1.140 |
| 09-Jan-2021 |
rillig | lint: split typeok into several smaller functions
This reduces the number of local variables from 16 to around 5, in most
of the smaller functions.
No functional change.
|
| 1.139 |
| 09-Jan-2021 |
rillig | lint: extract typeok_amper from typeok
No functional change.
|
| 1.138 |
| 09-Jan-2021 |
rillig | lint: extract typeok_incdec from typeok
No functional change.
|
| 1.137 |
| 09-Jan-2021 |
rillig | lint: fix crash for error 108 (invalid type of unary operator)
|
| 1.136 |
| 09-Jan-2021 |
rillig | lint: make target platform independent of host platform
If lint is run on a platform that has CHAR_BIT == 10, this doesn't
magically make an ILP32 platform have 40 bits per uint32_t.
At the moment, all of the supported platforms are either ILP32 or
I32LP64 anyway, and all of them have CHAR_BIT == 8 == CHAR_SIZE,
so nothing changes practically.
|
| 1.135 |
| 05-Jan-2021 |
rillig | lint: make check_precedence_confusion simpler
In C, only binary operators have possibly confusing precedence. All
binary operators have lower precedence than an explicit cast. When an
expression is parsed, the parentheses are associated with the innermost
possible node. This means that as soon as a cast operator is
parenthesized, its contained expression can no longer have confusing
precedence.
This allows the code to be written more succinct since the local
variables are no longer necessary.
|
| 1.134 |
| 05-Jan-2021 |
rillig | lint: extract code for determining possible precedence confusion
The function check_precedence_confusion was pretty long, and right in
the middle of that function was the complicated part of determining
which of the operand combinations are confusing and which aren't.
Extract this part into a separate function to document on which
information this decision is based. This makes it easier to understand
the code since there are fewer local variables around.
As a left-over from a previous commit, rop and rparn don't need to be
initialized twice, now that the assertion for a binary operator is in
place.
Remove the large and useless switch statement over all operator types.
This list was completely unsorted, for no apparent reason. To see the
list of operators, better look them up in ops.def, there was no need to
have this list duplicated here.
|
| 1.133 |
| 04-Jan-2021 |
rillig | lint: precedence confusion is only possible with binary operators
No functional change.
The operator table in ops.def states that every operator that has
possibly confusing precedence is also a binary operator, so assert that
instead of having two different code paths.
|
| 1.132 |
| 04-Jan-2021 |
rillig | lint: fix bug in "precedence confusion possible [169]"
|
| 1.131 |
| 04-Jan-2021 |
rillig | lint: document and demonstrate the bug in check_precedence_confusion
It took quite a while to get to the correct interpretation of this small
piece of code and to draw the right conclusions from it. Now the bug is
finally ready to be fixed, as already announced in the test.
|
| 1.130 |
| 04-Jan-2021 |
rillig | lint: add more rationale for removing effect-less code
|
| 1.129 |
| 04-Jan-2021 |
rillig | lint: finish the comments in check_precedence_confusion
|
| 1.128 |
| 04-Jan-2021 |
rillig | lint: replace LERROR with lint_assert in check_expr_misc
This reduces the visual clutter. There is no reason for anyone to
modify the code around the CALL operator, therefore the assertion is not
expected to fail anytime soon.
|
| 1.127 |
| 04-Jan-2021 |
rillig | lint: fix typos and other minor stylistic issues
|
| 1.126 |
| 04-Jan-2021 |
rillig | lint: fix bugs in dprint_node in debug mode (since today)
The node was dereferenced before the null check. GCC 5.5 didn't warn
about this obvious bug, not even with -Wall -Wextra -O2. Such a case
didn't occur though in the few tests that this function was used in.
The indentation for the nested nodes only needs to be set for a few
lines of code, make this region as small as possible.
There are nodes that use both tn_left and tn_right, even though they are
not defined as binary operators. An example is CALL, for which tn_left
is the address of the function name and tn_right, which are the
arguments, linked via PUSH nodes. CALL is not a binary operator since
it doesn't do any calculations with its arguments.
|
| 1.125 |
| 04-Jan-2021 |
rillig | lint: move dprint_node to the top of the file
It now resides right below dumpnode, which implements the same idea but
uses a fixed-size output buffer and prints everything in a single line,
which quickly gets hard to read. Maybe that's the reason why it had
been commented out since it got added in 2014.
|
| 1.124 |
| 04-Jan-2021 |
rillig | lint: in debug mode, print node tree for precedence
From the code alone, it is too difficult to see how the various internal
operators are combined and what properties they have. A simple tree
visualization helps to see all the details.
This is used to track down the typo in check_precedence_confusion, to
see whether it could have possibly had any influence at all.
|
| 1.123 |
| 04-Jan-2021 |
rillig | lint: add test for "precedence confusion possible [169]"
|
| 1.122 |
| 03-Jan-2021 |
rillig | lint: rename funcarg and funccall to longer names
From the previous short names, it was no obvious that these functions
create a new tree node.
The function named funccall in lint2 has been left as-is, since it has a
completely different prototype.
|
| 1.121 |
| 03-Jan-2021 |
rillig | lint: rename functions that create nodes
|
| 1.120 |
| 03-Jan-2021 |
rillig | lint: rename type.t_isfield to t_bitfield
|
| 1.119 |
| 03-Jan-2021 |
rillig | lint: remove redundant include of externs1.h
It is already included by lint1.h.
|
| 1.118 |
| 03-Jan-2021 |
rillig | lint: let gnuism and c99ism return void instead of int
The return value was only used in a single case. Duplicating the
condition for printing a message is ok in that case, since it makes all
other places in the code simpler.
The occasional "(void)" or "msg = " before the function call had hidden
the calls from check-msgs.lua, which didn't check the message texts in
such cases.
|
| 1.117 |
| 03-Jan-2021 |
rillig | lint: add type information to message 124 "illegal pointer combination"
|
| 1.116 |
| 02-Jan-2021 |
rillig | lint: fix lint warning 161 "constant in conditional context"
|
| 1.115 |
| 02-Jan-2021 |
rillig | lint: fix buffer truncation for type names
Previously, most type names had been cut off after 63 significant
characters. In some cases, 127 characters survived, or 255. And for
the debugging messages, sometimes even 1023. This inconsistency was
useless.
It was wrong in the first place to make the caller of the function
tyname responsible for handling the buffer. That's not something a
caller of such a simple function should do. These callers have better
things to do.
The API of the new function type_name is as simple as possible.
In the implementation, the name of the type is generated anew each time.
I just didn't know whether the type details could change, once the type
is initialized, and I didn't want to find out. To be on the safe side,
the resulting type name is cached, independently of the type it was
generated for. Using a trivial, unbalanced binary tree should be good
enough for now.
All this work is necessary to support adding new debug logging, without
being distracted by irrelevant implementation details such as these
buffer sizes. Adding new debug messages should be fun and easy; up to
now, it was overly bureaucratic.
|
| 1.114 |
| 01-Jan-2021 |
rillig | lint: rename styp and utyp
|
| 1.113 |
| 01-Jan-2021 |
rillig | lint: add missing redundant messages in source code
|
| 1.112 |
| 01-Jan-2021 |
rillig | lint: rename tokens for left and right parentheses
|
| 1.111 |
| 01-Jan-2021 |
rillig | lint: rename basic_type_name to tspec_name
The term "basic type" is precisely defined in C99 6.2.5p14, and tspec_t
is close to that, but nowhere identical.
|
| 1.110 |
| 01-Jan-2021 |
rillig | lint: clean up warn_incompatible_types
Splitting the code arbitrarily in separate phases made the code harder
to understand, both for humans as well as automated tools.
One of these tools, check-msgs.lua, couldn't check whether the comments
match the actual messages, and of course, the comments were wrong.
There was no good reason to deviate from the pattern followed by all the
rest of the code.
|
| 1.109 |
| 01-Jan-2021 |
rillig | lint: fix segmentation fault when checking returned enum types (211)
|
| 1.108 |
| 01-Jan-2021 |
rillig | lint: align comments in code with actual messages
Redundancy is bad. Especially in this case, separating the format
strings from the actual arguments prevents the compiler from
cross-checking them.
|
| 1.107 |
| 30-Dec-2020 |
rillig | lint: replace LERROR with lint_assert
This removes the redundancy of mentioning the function name in the error
message. This redundancy had been correct in all but 2 cases:
build_real_imag and tsize.
|
| 1.106 |
| 30-Dec-2020 |
rillig | lint: spell check comments
|
| 1.105 |
| 30-Dec-2020 |
rillig | lint: un-abbreviate s_field, s_keyw and s_xsym
|
| 1.104 |
| 30-Dec-2020 |
rillig | lint: un-abbreviate parenthesized and _strg
|
| 1.103 |
| 30-Dec-2020 |
rillig | lint: rename remaining _nxt members to _next
|
| 1.102 |
| 30-Dec-2020 |
rillig | lint: rename symt_t constants
There's no need to abbreviate them, furthermore FMOS was imprecise.
|
| 1.101 |
| 30-Dec-2020 |
rillig | lint: rename s_nxt to s_next
|
| 1.100 |
| 29-Dec-2020 |
rillig | lint: rename functions that had very short names
|
| 1.99 |
| 29-Dec-2020 |
rillig | lint: fix typo in comments
|
| 1.98 |
| 29-Dec-2020 |
rillig | lint: rename functions with very short names
|
| 1.97 |
| 29-Dec-2020 |
rillig | lint: fix indentation and alignment that used space-tab
|
| 1.96 |
| 29-Dec-2020 |
rillig | lint: spell check
|
| 1.95 |
| 29-Dec-2020 |
rillig | lint: remove redundant parentheses around return value
|
| 1.94 |
| 29-Dec-2020 |
rillig | lint: rename functions that had very short names
C99 guarantees that the first 31 characters of an identifier with
external linkage are significant. This removes the need to use
abbreviations for common words.
|
| 1.93 |
| 28-Dec-2020 |
rillig | lint: rename fields in mod_t
|
| 1.92 |
| 28-Dec-2020 |
rillig | lint: sort includes
|
| 1.91 |
| 28-Dec-2020 |
rillig | lint: realign code
|
| 1.90 |
| 28-Dec-2020 |
rillig | lint: rename tspec macros
|
| 1.89 |
| 28-Dec-2020 |
rillig | lint1: remove trailing whitespace
|
| 1.88 |
| 02-Jun-2020 |
christos | Fix return of alignof()
|
| 1.87 |
| 12-Jul-2019 |
christos | allow c9x struct casts (in new libXt)
|
| 1.86 |
| 07-Sep-2018 |
christos | recognize int128
|
| 1.85 |
| 29-Jun-2018 |
christos | add missing args for 123
|
| 1.84 |
| 06-Mar-2017 |
christos | branches: 1.84.10; 1.84.12;
fix typeof, add __builtin_offsetof
|
| 1.83 |
| 19-Aug-2016 |
christos | branches: 1.83.2;
Add union casts.
|
| 1.82 |
| 14-Oct-2015 |
christos | branches: 1.82.2;
more descriptive errors
|
| 1.81 |
| 28-Aug-2015 |
joerg | ~0 and -1 are the same for two-complement machines. ISO C says left
shifts of negative values are UB, so do the shift for the unsigned
equivalent and cast to int afterwards.
|
| 1.80 |
| 29-Jul-2015 |
christos | Make the return type of "? type1 : void *" or "? void * : type1" to be type1
instead of void *.
|
| 1.79 |
| 29-Jul-2015 |
christos | Be more specific about the type errors
|
| 1.78 |
| 09-Feb-2015 |
christos | Treat complex numbers like other floating numbers. This caused a core-dump
when linting libm complex code and assumed the size of the type larger than
the array size of value bitmaps.
|
| 1.77 |
| 20-Nov-2014 |
christos | allow c99 flexible arrays for packed structures.
|
| 1.76 |
| 17-Apr-2014 |
christos | - add a function to dump a node
- better diagnostics on abort
- allow converting a constant
- initialize right node now that we trash memory this makes a difference.
before it was NULL.
|
| 1.75 |
| 18-Feb-2014 |
christos | branches: 1.75.2;
add __extension__ and typeof
|
| 1.74 |
| 18-Feb-2014 |
christos | __real__ and __imag__ produce lvalues
|
| 1.73 |
| 19-Apr-2013 |
christos | make NOSTRICT behave the same a LINTED; use the new format.
|
| 1.72 |
| 02-Mar-2013 |
christos | obey constcond to !constcond
|
| 1.71 |
| 20-Apr-2012 |
christos | branches: 1.71.2;
add __PRETTY_FUNCTION__
|
| 1.70 |
| 27-Mar-2012 |
christos | more cross lint friendlyness
XXX: needs more constants converted double/float
|
| 1.69 |
| 21-Mar-2012 |
christos | -0 is not unary overflow.
|
| 1.68 |
| 05-Feb-2011 |
christos | branches: 1.68.4;
as promised make the last ops table auto-generated.
|
| 1.67 |
| 04-Feb-2011 |
christos | fix typo
|
| 1.66 |
| 18-Dec-2010 |
christos | branches: 1.66.2;
PR/44235: Martin Husemann: Fix core dump due to memory corruption.
Found by Henning Petersen
|
| 1.65 |
| 24-Nov-2010 |
christos | PR/44148: Henning Petersen: catstrg() does not set the resulting string length.
|
| 1.64 |
| 21-Mar-2010 |
christos | revert for now.
|
| 1.63 |
| 21-Mar-2010 |
christos | make lint recognize constant shift values and not warn if the number of bits
shifted results in a narrower type.
|
| 1.62 |
| 02-Oct-2009 |
christos | compute sizes by the new tsize() function.
|
| 1.61 |
| 02-Oct-2009 |
christos | remove debugging
|
| 1.60 |
| 02-Oct-2009 |
christos | recognize struct __packed x { }; in addition to struct x { } __packed;
|
| 1.59 |
| 02-May-2009 |
christos | Add __alignof__.
|
| 1.58 |
| 27-Apr-2009 |
ginsbach | Print left and right type and not left type twice. (Hi Christos!)
|
| 1.57 |
| 15-Apr-2009 |
christos | Lukemify (WARNS=4)
|
| 1.56 |
| 13-Apr-2009 |
christos | since we know the types, print what they are in error messages.
|
| 1.55 |
| 02-Mar-2009 |
christos | Don't produce type mismatch warnings if one side of ?: is "[qual] void *"
and the other side is "[qual] pointer".
|
| 1.54 |
| 16-Nov-2008 |
dholland | branches: 1.54.2;
WARNS=4
|
| 1.53 |
| 27-Sep-2008 |
matt | Add the complex types to the floating point promotion rules.
|
| 1.52 |
| 26-Sep-2008 |
matt | Teach lint about long double _Complex (C99)
|
| 1.51 |
| 31-Jul-2008 |
christos | Add Picky flag; this produces more warnings:
1. long a; int i; a = i * i;
suggests casting i to long, so that we gain precision in the multiplication.
2. warns about magnitude comparisons in enums.
3. warns about possible sign extension issues when integer types become widened.
|
| 1.50 |
| 03-May-2008 |
christos | branches: 1.50.2;
don't check for NaN if fpe.
|
| 1.49 |
| 02-May-2008 |
christos | Since we cannot guarantee that all machines do ieee math, or that they have
the proper math setup deal with SIGFPE directly.
|
| 1.48 |
| 27-Apr-2008 |
christos | print the types involved in an error.
|
| 1.47 |
| 27-Apr-2008 |
christos | reorder a couple of error messages.
|
| 1.46 |
| 25-Apr-2008 |
christos | preliminary _Complex support.
NB: Does not really understand type conversions between complex and doubles.
|
| 1.45 |
| 04-Mar-2008 |
christos | branches: 1.45.2;
PR/38142: YAMAMOTO Takashi: lint -S complains on variable length arrays
|
| 1.44 |
| 18-Oct-2006 |
he | branches: 1.44.8; 1.44.14;
Simplify the previous change. There's actually no need to treat constants
specially in ptconv(), the constants will be checked in convert() anyway.
|
| 1.43 |
| 18-Oct-2006 |
he | Partly redo when to emit warnings for argument type conversions.
Remove the apparently always true "styp(nt) != SHORT" part of the
innermost test. Allow atomatic conversion of literals as long as
they fit into the target type.
This should fix some of the lint issues in proplib on some of our
platforms.
Approved by christos.
|
| 1.42 |
| 22-Mar-2006 |
christos | Coverity CID 199: make sure that rtp is not NULL before dereferencing.
|
| 1.41 |
| 07-Apr-2005 |
christos | Factor out tyname() so that it can be used both by lint1 and lint2.
Since type_t is different between lint1.h and lint2.h include the
appropriate file depending on the pass. Make the argument mismatch
error print the type names of the types involved. Now that we have
a tyname() function we can fix the rest of the pass2 warnings to be
more explanatory, but not now.
|
| 1.40 |
| 02-Jan-2005 |
christos | Shift assignment operators were handled incorrectly. On 64 bit machines,
unsigned long x, y;
x <<= y;
always produces a warning because y is casted to int. Handle them instead
the same way as regular shifts.
|
| 1.39 |
| 02-Jan-2005 |
christos | say from what type to what time we are converting.
|
| 1.38 |
| 12-Sep-2004 |
yamt | recognize _Bool.
|
| 1.37 |
| 20-Jun-2004 |
jmc | Completely rework how tools/compat is done. Purge all uses/references to
_NETBSD_SOURCE as this makes cross building from older/newer versions of
NetBSD harder, not easier (and also makes the resulting tools 'different')
Wrap all required code with the inclusion of nbtool_config.h, attempt to
only use POSIX code in all places (or when reasonable test w. configure and
provide definitions: ala u_int, etc).
Reviewed by lukem. Tested on FreeBSD 4.9, Redhat Linux ES3, NetBSD 1.6.2 x86
NetBSD current (x86 and amd64) and Solaris 9.
Fixes PR's: PR#17762 PR#25944
|
| 1.36 |
| 30-May-2003 |
christos | branches: 1.36.2;
PR/21706: Krister Walfridsson: CVT nodes not handled in ?: expressions.
|
| 1.35 |
| 14-May-2003 |
wiz | constant usually has two n.
|
| 1.34 |
| 22-Oct-2002 |
christos | add support for ({}) gcc shit.
|
| 1.33 |
| 22-Oct-2002 |
christos | handle free-ing of temp symbols properly. Don't segv on bad node types.
|
| 1.32 |
| 22-Oct-2002 |
christos | add C9X/GCC compound literal expressions.
|
| 1.31 |
| 22-Oct-2002 |
christos | add variable array dimension.
|
| 1.30 |
| 22-Oct-2002 |
christos | handle gcc __FUNCTION__ and C9X __func__
|
| 1.29 |
| 22-Oct-2002 |
christos | a cast to a pointer is an l-value.
|
| 1.28 |
| 22-Oct-2002 |
christos | handle cast in the rhs of a - op.
|
| 1.27 |
| 21-Oct-2002 |
christos | Ignore cast size test in initializers; look at example in the comment.
Maybe there is a better way...
|
| 1.26 |
| 13-Sep-2002 |
christos | Minimize diffs with my C99 capable version [this commit does not include
C99 support.
- turn lerror() into a macro so that the filename and the line number of the
error are printed before we abort.
- recurse in type printing to provide the proper type name.
|
| 1.25 |
| 05-Feb-2002 |
thorpej | Replace u_quad_t with uint64_t and quad_t with int64_t, and use
<inttypes.h> to get those type definitions. These types are more
portable, and a little more sane to do autoconf tests for.
|
| 1.24 |
| 31-Jan-2002 |
tv | Use !finite() instead of isinf() in two places for better compatibility.
|
| 1.23 |
| 31-Jan-2002 |
tv | Protect __RCSID and __COPYRIGHT from being invoked if not defined.
|
| 1.22 |
| 17-Nov-2001 |
perry | Fix a bug in detecting overflow in unsigned multiplication.
XXX Note that the overflow code for many cases seems to be buggy. I've
only fixed one bug that was bothering me. A set of regression tests
and extensive testing are needed.
|
| 1.21 |
| 16-Sep-2001 |
wiz | Spell 'occurred' with two 'r's.
|
| 1.20 |
| 28-May-2001 |
lukem | cleanup (prior to more adding more features):
- convert to ANSI KNF
- remove trailing whitespace
- translate some comments from german into english
code compiles and runs clean, and tested by running "make lint" against
xlint source using previous and this lint produces same results.
|
| 1.19 |
| 28-Jul-1998 |
mycroft | Propagate the value and test contexts to the right-hand side of a
comma operator.
|
| 1.18 |
| 27-Jul-1998 |
mycroft | Search both sides of a comma operator for side effects.
|
| 1.17 |
| 27-Jul-1998 |
mycroft | Fix a recurring typo: comparision -> comparison.
|
| 1.16 |
| 27-Jul-1998 |
mycroft | Fix uses of uninitialized memory, and incorrect types for initializers.
From ITOH Yasufumi, PR 5861.
|
| 1.15 |
| 09-Apr-1998 |
tv | .y.c <sys.mk> rule fixes. Don't create a y.tab.h file unless asked for,
and use smarter creation of the header file.
|
| 1.14 |
| 22-Feb-1998 |
christos | WARNSify
|
| 1.13 |
| 12-Mar-1997 |
mycroft | Remove hack for (very old) GCC bug on i386.
|
| 1.12 |
| 02-Oct-1995 |
jpo | insert a conversion operator from argument type to parameter type also
if the subtypes of both argument and parameter do not match (use
eqtype() instead of comparing t_tspec fields only).
|
| 1.11 |
| 02-Oct-1995 |
jpo | removed some dubious warnings about conversion of constant operands of
bitwise operators
|
| 1.10 |
| 02-Oct-1995 |
jpo | allow register variables as left operand of POINT; fixes PR 1442
|
| 1.9 |
| 02-Oct-1995 |
jpo | don't print a warning about use of unitialized variables if one of the
enclosing compound statements had already an asm statement
|
| 1.8 |
| 02-Oct-1995 |
jpo | LINTED and CONSTCOND are now valid up to the next end of a
global or local declaration/definition/statement. Originally they were
valid on the current and next line, which made it hard to suppress
warnings in constructs with more then one line.
LONGLONG can now be used to suppress errors or warnings in the next
declaration, definition or statement.
|
| 1.7 |
| 02-Oct-1995 |
jpo | some minor bug fixes
|
| 1.6 |
| 02-Oct-1995 |
jpo | prototypes override old style function definitions
this is a gnu extension to ansi c
|
| 1.5 |
| 02-Oct-1995 |
jpo | added inline keywords
"inline" is enabled by -g, "__inline" and "__inline__" are always available
|
| 1.4 |
| 02-Oct-1995 |
jpo | prefixed members of dinfo_t with 'd_'
|
| 1.3 |
| 04-Jul-1995 |
cgd | hack around a bug in our gcc.
|
| 1.2 |
| 03-Jul-1995 |
cgd | RCS id cleanup
|
| 1.1 |
| 03-Jul-1995 |
cgd | branches: 1.1.1;
Initial revision
|
| 1.1.1.1 |
| 03-Jul-1995 |
cgd | lint(1) implementation, by Jochen Pohl. named 'xlint' for a similar
reason to why 'install' is named 'xinstall'.
|
| 1.36.2.1 |
| 22-Jun-2004 |
tron | Pull up revision 1.37 (requested by jmc in ticket #527):
Completely rework how tools/compat is done. Purge all uses/references to
_NETBSD_SOURCE as this makes cross building from older/newer versions of
NetBSD harder, not easier (and also makes the resulting tools 'different')
Wrap all required code with the inclusion of nbtool_config.h, attempt to
only use POSIX code in all places (or when reasonable test w. configure and
provide definitions: ala u_int, etc).
Reviewed by lukem. Tested on FreeBSD 4.9, Redhat Linux ES3, NetBSD 1.6.2 x86
NetBSD current (x86 and amd64) and Solaris 9.
Fixes PR's: PR#17762 PR#25944
|
| 1.44.14.1 |
| 24-Mar-2008 |
keiichi | sync with head.
|
| 1.44.8.1 |
| 23-Mar-2008 |
matt | sync with HEAD
|
| 1.45.2.1 |
| 18-May-2008 |
yamt | sync with head.
|
| 1.50.2.1 |
| 18-Sep-2008 |
wrstuden | Sync with wrstuden-revivesa-base-2.
|
| 1.54.2.1 |
| 13-May-2009 |
jym | Sync with HEAD.
Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
|
| 1.66.2.1 |
| 08-Feb-2011 |
bouyer | Sync with HEAD
|
| 1.68.4.3 |
| 22-May-2014 |
yamt | sync with head.
for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.
this commit was splitted into small chunks to avoid
a limitation of cvs. ("Protocol error: too many arguments")
|
| 1.68.4.2 |
| 23-May-2012 |
yamt | sync with head.
|
| 1.68.4.1 |
| 17-Apr-2012 |
yamt | sync with head
|
| 1.71.2.2 |
| 20-Aug-2014 |
tls | Rebase to HEAD as of a few days ago.
|
| 1.71.2.1 |
| 23-Jun-2013 |
tls | resync from head
|
| 1.75.2.1 |
| 10-Aug-2014 |
tls | Rebase.
|
| 1.82.2.1 |
| 20-Mar-2017 |
pgoyette | Sync with HEAD
|
| 1.83.2.1 |
| 21-Apr-2017 |
bouyer | Sync with HEAD
|
| 1.84.12.2 |
| 13-Apr-2020 |
martin | Mostly merge changes from HEAD upto 20200411
|
| 1.84.12.1 |
| 10-Jun-2019 |
christos | Sync with HEAD
|
| 1.84.10.2 |
| 30-Sep-2018 |
pgoyette | Ssync with HEAD
|
| 1.84.10.1 |
| 28-Jul-2018 |
pgoyette | Sync with HEAD
|
| 1.648.2.1 |
| 02-Aug-2025 |
perseant | Sync with HEAD
|
