1 FileCheck - Flexible pattern matching file verifier
2 ===================================================
7 :program:`FileCheck` *match-filename* [*--check-prefix=XXX*] [*--strict-whitespace*]
12 :program:`FileCheck` reads two files (one from standard input, and one
13 specified on the command line) and uses one to verify the other. This
14 behavior is particularly useful for the testsuite, which wants to verify that
15 the output of some tool (e.g. :program:`llc`) contains the expected information
16 (for example, a movsd from esp or whatever is interesting). This is similar to
17 using :program:`grep`, but it is optimized for matching multiple different
18 inputs in one file in a specific order.
20 The ``match-filename`` file specifies the file that contains the patterns to
21 match. The file to verify is read from standard input unless the
22 :option:`--input-file` option is used.
29 Print a summary of command line options.
31 .. option:: --check-prefix prefix
33 FileCheck searches the contents of ``match-filename`` for patterns to
34 match. By default, these patterns are prefixed with "``CHECK:``".
35 If you'd like to use a different prefix (e.g. because the same input
36 file is checking multiple different tool or options), the
37 :option:`--check-prefix` argument allows you to specify one or more
38 prefixes to match. Multiple prefixes are useful for tests which might
39 change for different run options, but most lines remain the same.
41 .. option:: --input-file filename
43 File to check (defaults to stdin).
45 .. option:: --strict-whitespace
47 By default, FileCheck canonicalizes input horizontal whitespace (spaces and
48 tabs) which causes it to ignore these differences (a space will match a tab).
49 The :option:`--strict-whitespace` argument disables this behavior. End-of-line
50 sequences are canonicalized to UNIX-style ``\n`` in all modes.
52 .. option:: --implicit-check-not check-pattern
54 Adds implicit negative checks for the specified patterns between positive
55 checks. The option allows writing stricter tests without stuffing them with
58 For example, "``--implicit-check-not warning:``" can be useful when testing
59 diagnostic messages from tools that don't have an option similar to ``clang
60 -verify``. With this option FileCheck will verify that input does not contain
61 warnings not covered by any ``CHECK:`` patterns.
65 Show the version number of this program.
70 If :program:`FileCheck` verifies that the file matches the expected contents,
71 it exits with 0. Otherwise, if not, or if an error occurs, it will exit with a
77 FileCheck is typically used from LLVM regression tests, being invoked on the RUN
78 line of the test. A simple example of using FileCheck from a RUN line looks
83 ; RUN: llvm-as < %s | llc -march=x86-64 | FileCheck %s
85 This syntax says to pipe the current file ("``%s``") into ``llvm-as``, pipe
86 that into ``llc``, then pipe the output of ``llc`` into ``FileCheck``. This
87 means that FileCheck will be verifying its standard input (the llc output)
88 against the filename argument specified (the original ``.ll`` file specified by
89 "``%s``"). To see how this works, let's look at the rest of the ``.ll`` file
94 define void @sub1(i32* %p, i32 %v) {
98 %0 = tail call i32 @llvm.atomic.load.sub.i32.p0i32(i32* %p, i32 %v)
102 define void @inc4(i64* %p) {
106 %0 = tail call i64 @llvm.atomic.load.add.i64.p0i64(i64* %p, i64 1)
110 Here you can see some "``CHECK:``" lines specified in comments. Now you can
111 see how the file is piped into ``llvm-as``, then ``llc``, and the machine code
112 output is what we are verifying. FileCheck checks the machine code output to
113 verify that it matches what the "``CHECK:``" lines specify.
115 The syntax of the "``CHECK:``" lines is very simple: they are fixed strings that
116 must occur in order. FileCheck defaults to ignoring horizontal whitespace
117 differences (e.g. a space is allowed to match a tab) but otherwise, the contents
118 of the "``CHECK:``" line is required to match some thing in the test file exactly.
120 One nice thing about FileCheck (compared to grep) is that it allows merging
121 test cases together into logical groups. For example, because the test above
122 is checking for the "``sub1:``" and "``inc4:``" labels, it will not match
123 unless there is a "``subl``" in between those labels. If it existed somewhere
124 else in the file, that would not count: "``grep subl``" matches if "``subl``"
125 exists anywhere in the file.
127 The FileCheck -check-prefix option
128 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
130 The FileCheck :option:`-check-prefix` option allows multiple test
131 configurations to be driven from one `.ll` file. This is useful in many
132 circumstances, for example, testing different architectural variants with
133 :program:`llc`. Here's a simple example:
137 ; RUN: llvm-as < %s | llc -mtriple=i686-apple-darwin9 -mattr=sse41 \
138 ; RUN: | FileCheck %s -check-prefix=X32
139 ; RUN: llvm-as < %s | llc -mtriple=x86_64-apple-darwin9 -mattr=sse41 \
140 ; RUN: | FileCheck %s -check-prefix=X64
142 define <4 x i32> @pinsrd_1(i32 %s, <4 x i32> %tmp) nounwind {
143 %tmp1 = insertelement <4 x i32>; %tmp, i32 %s, i32 1
146 ; X32: pinsrd $1, 4(%esp), %xmm0
149 ; X64: pinsrd $1, %edi, %xmm0
152 In this case, we're testing that we get the expected code generation with
153 both 32-bit and 64-bit code generation.
155 The "CHECK-NEXT:" directive
156 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
158 Sometimes you want to match lines and would like to verify that matches
159 happen on exactly consecutive lines with no other lines in between them. In
160 this case, you can use "``CHECK:``" and "``CHECK-NEXT:``" directives to specify
161 this. If you specified a custom check prefix, just use "``<PREFIX>-NEXT:``".
162 For example, something like this works as you'd expect:
166 define void @t2(<2 x double>* %r, <2 x double>* %A, double %B) {
167 %tmp3 = load <2 x double>* %A, align 16
168 %tmp7 = insertelement <2 x double> undef, double %B, i32 0
169 %tmp9 = shufflevector <2 x double> %tmp3,
171 <2 x i32> < i32 0, i32 2 >
172 store <2 x double> %tmp9, <2 x double>* %r, align 16
176 ; CHECK: movl 8(%esp), %eax
177 ; CHECK-NEXT: movapd (%eax), %xmm0
178 ; CHECK-NEXT: movhpd 12(%esp), %xmm0
179 ; CHECK-NEXT: movl 4(%esp), %eax
180 ; CHECK-NEXT: movapd %xmm0, (%eax)
184 "``CHECK-NEXT:``" directives reject the input unless there is exactly one
185 newline between it and the previous directive. A "``CHECK-NEXT:``" cannot be
186 the first directive in a file.
188 The "CHECK-NOT:" directive
189 ~~~~~~~~~~~~~~~~~~~~~~~~~~
191 The "``CHECK-NOT:``" directive is used to verify that a string doesn't occur
192 between two matches (or before the first match, or after the last match). For
193 example, to verify that a load is removed by a transformation, a test like this
198 define i8 @coerce_offset0(i32 %V, i32* %P) {
199 store i32 %V, i32* %P
201 %P2 = bitcast i32* %P to i8*
202 %P3 = getelementptr i8* %P2, i32 2
206 ; CHECK: @coerce_offset0
211 The "CHECK-DAG:" directive
212 ~~~~~~~~~~~~~~~~~~~~~~~~~~
214 If it's necessary to match strings that don't occur in a strictly sequential
215 order, "``CHECK-DAG:``" could be used to verify them between two matches (or
216 before the first match, or after the last match). For example, clang emits
217 vtable globals in reverse order. Using ``CHECK-DAG:``, we can keep the checks
218 in the natural order:
222 // RUN: %clang_cc1 %s -emit-llvm -o - | FileCheck %s
224 struct Foo { virtual void method(); };
225 Foo f; // emit vtable
226 // CHECK-DAG: @_ZTV3Foo =
228 struct Bar { virtual void method(); };
230 // CHECK-DAG: @_ZTV3Bar =
232 ``CHECK-NOT:`` directives could be mixed with ``CHECK-DAG:`` directives to
233 exclude strings between the surrounding ``CHECK-DAG:`` directives. As a result,
234 the surrounding ``CHECK-DAG:`` directives cannot be reordered, i.e. all
235 occurrences matching ``CHECK-DAG:`` before ``CHECK-NOT:`` must not fall behind
236 occurrences matching ``CHECK-DAG:`` after ``CHECK-NOT:``. For example,
244 This case will reject input strings where ``BEFORE`` occurs after ``AFTER``.
246 With captured variables, ``CHECK-DAG:`` is able to match valid topological
247 orderings of a DAG with edges from the definition of a variable to its use.
248 It's useful, e.g., when your test cases need to match different output
249 sequences from the instruction scheduler. For example,
253 ; CHECK-DAG: add [[REG1:r[0-9]+]], r1, r2
254 ; CHECK-DAG: add [[REG2:r[0-9]+]], r3, r4
255 ; CHECK: mul r5, [[REG1]], [[REG2]]
257 In this case, any order of that two ``add`` instructions will be allowed.
259 If you are defining `and` using variables in the same ``CHECK-DAG:`` block,
260 be aware that the definition rule can match `after` its use.
262 So, for instance, the code below will pass:
266 ; CHECK-DAG: vmov.32 [[REG2:d[0-9]+]][0]
267 ; CHECK-DAG: vmov.32 [[REG2]][1]
271 While this other code, will not:
275 ; CHECK-DAG: vmov.32 [[REG2:d[0-9]+]][0]
276 ; CHECK-DAG: vmov.32 [[REG2]][1]
280 While this can be very useful, it's also dangerous, because in the case of
281 register sequence, you must have a strong order (read before write, copy before
282 use, etc). If the definition your test is looking for doesn't match (because
283 of a bug in the compiler), it may match further away from the use, and mask
286 In those cases, to enforce the order, use a non-DAG directive between DAG-blocks.
288 The "CHECK-LABEL:" directive
289 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
291 Sometimes in a file containing multiple tests divided into logical blocks, one
292 or more ``CHECK:`` directives may inadvertently succeed by matching lines in a
293 later block. While an error will usually eventually be generated, the check
294 flagged as causing the error may not actually bear any relationship to the
295 actual source of the problem.
297 In order to produce better error messages in these cases, the "``CHECK-LABEL:``"
298 directive can be used. It is treated identically to a normal ``CHECK``
299 directive except that FileCheck makes an additional assumption that a line
300 matched by the directive cannot also be matched by any other check present in
301 ``match-filename``; this is intended to be used for lines containing labels or
302 other unique identifiers. Conceptually, the presence of ``CHECK-LABEL`` divides
303 the input stream into separate blocks, each of which is processed independently,
304 preventing a ``CHECK:`` directive in one block matching a line in another block.
309 define %struct.C* @C_ctor_base(%struct.C* %this, i32 %x) {
311 ; CHECK-LABEL: C_ctor_base:
312 ; CHECK: mov [[SAVETHIS:r[0-9]+]], r0
313 ; CHECK: bl A_ctor_base
314 ; CHECK: mov r0, [[SAVETHIS]]
315 %0 = bitcast %struct.C* %this to %struct.A*
316 %call = tail call %struct.A* @A_ctor_base(%struct.A* %0)
317 %1 = bitcast %struct.C* %this to %struct.B*
318 %call2 = tail call %struct.B* @B_ctor_base(%struct.B* %1, i32 %x)
322 define %struct.D* @D_ctor_base(%struct.D* %this, i32 %x) {
324 ; CHECK-LABEL: D_ctor_base:
326 The use of ``CHECK-LABEL:`` directives in this case ensures that the three
327 ``CHECK:`` directives only accept lines corresponding to the body of the
328 ``@C_ctor_base`` function, even if the patterns match lines found later in
329 the file. Furthermore, if one of these three ``CHECK:`` directives fail,
330 FileCheck will recover by continuing to the next block, allowing multiple test
331 failures to be detected in a single invocation.
333 There is no requirement that ``CHECK-LABEL:`` directives contain strings that
334 correspond to actual syntactic labels in a source or output language: they must
335 simply uniquely match a single line in the file being verified.
337 ``CHECK-LABEL:`` directives cannot contain variable definitions or uses.
339 FileCheck Pattern Matching Syntax
340 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
342 The "``CHECK:``" and "``CHECK-NOT:``" directives both take a pattern to match.
343 For most uses of FileCheck, fixed string matching is perfectly sufficient. For
344 some things, a more flexible form of matching is desired. To support this,
345 FileCheck allows you to specify regular expressions in matching strings,
346 surrounded by double braces: ``{{yourregex}}``. Because we want to use fixed
347 string matching for a majority of what we do, FileCheck has been designed to
348 support mixing and matching fixed string matching with regular expressions.
349 This allows you to write things like this:
353 ; CHECK: movhpd {{[0-9]+}}(%esp), {{%xmm[0-7]}}
355 In this case, any offset from the ESP register will be allowed, and any xmm
356 register will be allowed.
358 Because regular expressions are enclosed with double braces, they are
359 visually distinct, and you don't need to use escape characters within the double
360 braces like you would in C. In the rare case that you want to match double
361 braces explicitly from the input, you can use something ugly like
362 ``{{[{][{]}}`` as your pattern.
367 It is often useful to match a pattern and then verify that it occurs again
368 later in the file. For codegen tests, this can be useful to allow any register,
369 but verify that that register is used consistently later. To do this,
370 :program:`FileCheck` allows named variables to be defined and substituted into
371 patterns. Here is a simple example:
376 ; CHECK: notw [[REGISTER:%[a-z]+]]
377 ; CHECK: andw {{.*}}[[REGISTER]]
379 The first check line matches a regex ``%[a-z]+`` and captures it into the
380 variable ``REGISTER``. The second line verifies that whatever is in
381 ``REGISTER`` occurs later in the file after an "``andw``". :program:`FileCheck`
382 variable references are always contained in ``[[ ]]`` pairs, and their names can
383 be formed with the regex ``[a-zA-Z][a-zA-Z0-9]*``. If a colon follows the name,
384 then it is a definition of the variable; otherwise, it is a use.
386 :program:`FileCheck` variables can be defined multiple times, and uses always
387 get the latest value. Variables can also be used later on the same line they
388 were defined on. For example:
392 ; CHECK: op [[REG:r[0-9]+]], [[REG]]
394 Can be useful if you want the operands of ``op`` to be the same register,
395 and don't care exactly which register it is.
397 FileCheck Expressions
398 ~~~~~~~~~~~~~~~~~~~~~
400 Sometimes there's a need to verify output which refers line numbers of the
401 match file, e.g. when testing compiler diagnostics. This introduces a certain
402 fragility of the match file structure, as "``CHECK:``" lines contain absolute
403 line numbers in the same file, which have to be updated whenever line numbers
404 change due to text addition or deletion.
406 To support this case, FileCheck allows using ``[[@LINE]]``,
407 ``[[@LINE+<offset>]]``, ``[[@LINE-<offset>]]`` expressions in patterns. These
408 expressions expand to a number of the line where a pattern is located (with an
409 optional integer offset).
411 This way match patterns can be put near the relevant test lines and include
412 relative line number references, for example:
416 // CHECK: test.cpp:[[@LINE+4]]:6: error: expected ';' after top level declarator
417 // CHECK-NEXT: {{^int a}}
418 // CHECK-NEXT: {{^ \^}}
419 // CHECK-NEXT: {{^ ;}}