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 always read from standard input.
28 Print a summary of command line options.
30 .. option:: --check-prefix prefix
32 FileCheck searches the contents of ``match-filename`` for patterns to match.
33 By default, these patterns are prefixed with "``CHECK:``". If you'd like to
34 use a different prefix (e.g. because the same input file is checking multiple
35 different tool or options), the :option:`--check-prefix` argument allows you
36 to specify a specific prefix to match.
38 .. option:: --input-file filename
40 File to check (defaults to stdin).
42 .. option:: --strict-whitespace
44 By default, FileCheck canonicalizes input horizontal whitespace (spaces and
45 tabs) which causes it to ignore these differences (a space will match a tab).
46 The :option:`--strict-whitespace` argument disables this behavior.
50 Show the version number of this program.
55 If :program:`FileCheck` verifies that the file matches the expected contents,
56 it exits with 0. Otherwise, if not, or if an error occurs, it will exit with a
62 FileCheck is typically used from LLVM regression tests, being invoked on the RUN
63 line of the test. A simple example of using FileCheck from a RUN line looks
68 ; RUN: llvm-as < %s | llc -march=x86-64 | FileCheck %s
70 This syntax says to pipe the current file ("``%s``") into ``llvm-as``, pipe
71 that into ``llc``, then pipe the output of ``llc`` into ``FileCheck``. This
72 means that FileCheck will be verifying its standard input (the llc output)
73 against the filename argument specified (the original ``.ll`` file specified by
74 "``%s``"). To see how this works, let's look at the rest of the ``.ll`` file
79 define void @sub1(i32* %p, i32 %v) {
83 %0 = tail call i32 @llvm.atomic.load.sub.i32.p0i32(i32* %p, i32 %v)
87 define void @inc4(i64* %p) {
91 %0 = tail call i64 @llvm.atomic.load.add.i64.p0i64(i64* %p, i64 1)
95 Here you can see some "``CHECK:``" lines specified in comments. Now you can
96 see how the file is piped into ``llvm-as``, then ``llc``, and the machine code
97 output is what we are verifying. FileCheck checks the machine code output to
98 verify that it matches what the "``CHECK:``" lines specify.
100 The syntax of the "``CHECK:``" lines is very simple: they are fixed strings that
101 must occur in order. FileCheck defaults to ignoring horizontal whitespace
102 differences (e.g. a space is allowed to match a tab) but otherwise, the contents
103 of the "``CHECK:``" line is required to match some thing in the test file exactly.
105 One nice thing about FileCheck (compared to grep) is that it allows merging
106 test cases together into logical groups. For example, because the test above
107 is checking for the "``sub1:``" and "``inc4:``" labels, it will not match
108 unless there is a "``subl``" in between those labels. If it existed somewhere
109 else in the file, that would not count: "``grep subl``" matches if "``subl``"
110 exists anywhere in the file.
112 The FileCheck -check-prefix option
113 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
115 The FileCheck :option:`-check-prefix` option allows multiple test
116 configurations to be driven from one `.ll` file. This is useful in many
117 circumstances, for example, testing different architectural variants with
118 :program:`llc`. Here's a simple example:
122 ; RUN: llvm-as < %s | llc -mtriple=i686-apple-darwin9 -mattr=sse41 \
123 ; RUN: | FileCheck %s -check-prefix=X32
124 ; RUN: llvm-as < %s | llc -mtriple=x86_64-apple-darwin9 -mattr=sse41 \
125 ; RUN: | FileCheck %s -check-prefix=X64
127 define <4 x i32> @pinsrd_1(i32 %s, <4 x i32> %tmp) nounwind {
128 %tmp1 = insertelement <4 x i32>; %tmp, i32 %s, i32 1
131 ; X32: pinsrd $1, 4(%esp), %xmm0
134 ; X64: pinsrd $1, %edi, %xmm0
137 In this case, we're testing that we get the expected code generation with
138 both 32-bit and 64-bit code generation.
140 The "CHECK-NEXT:" directive
141 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
143 Sometimes you want to match lines and would like to verify that matches
144 happen on exactly consecutive lines with no other lines in between them. In
145 this case, you can use "``CHECK:``" and "``CHECK-NEXT:``" directives to specify
146 this. If you specified a custom check prefix, just use "``<PREFIX>-NEXT:``".
147 For example, something like this works as you'd expect:
151 define void @t2(<2 x double>* %r, <2 x double>* %A, double %B) {
152 %tmp3 = load <2 x double>* %A, align 16
153 %tmp7 = insertelement <2 x double> undef, double %B, i32 0
154 %tmp9 = shufflevector <2 x double> %tmp3,
156 <2 x i32> < i32 0, i32 2 >
157 store <2 x double> %tmp9, <2 x double>* %r, align 16
161 ; CHECK: movl 8(%esp), %eax
162 ; CHECK-NEXT: movapd (%eax), %xmm0
163 ; CHECK-NEXT: movhpd 12(%esp), %xmm0
164 ; CHECK-NEXT: movl 4(%esp), %eax
165 ; CHECK-NEXT: movapd %xmm0, (%eax)
169 "``CHECK-NEXT:``" directives reject the input unless there is exactly one
170 newline between it and the previous directive. A "``CHECK-NEXT:``" cannot be
171 the first directive in a file.
173 The "CHECK-NOT:" directive
174 ~~~~~~~~~~~~~~~~~~~~~~~~~~
176 The "``CHECK-NOT:``" directive is used to verify that a string doesn't occur
177 between two matches (or before the first match, or after the last match). For
178 example, to verify that a load is removed by a transformation, a test like this
183 define i8 @coerce_offset0(i32 %V, i32* %P) {
184 store i32 %V, i32* %P
186 %P2 = bitcast i32* %P to i8*
187 %P3 = getelementptr i8* %P2, i32 2
191 ; CHECK: @coerce_offset0
196 FileCheck Pattern Matching Syntax
197 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
199 The "``CHECK:``" and "``CHECK-NOT:``" directives both take a pattern to match.
200 For most uses of FileCheck, fixed string matching is perfectly sufficient. For
201 some things, a more flexible form of matching is desired. To support this,
202 FileCheck allows you to specify regular expressions in matching strings,
203 surrounded by double braces: ``{{yourregex}}``. Because we want to use fixed
204 string matching for a majority of what we do, FileCheck has been designed to
205 support mixing and matching fixed string matching with regular expressions.
206 This allows you to write things like this:
210 ; CHECK: movhpd {{[0-9]+}}(%esp), {{%xmm[0-7]}}
212 In this case, any offset from the ESP register will be allowed, and any xmm
213 register will be allowed.
215 Because regular expressions are enclosed with double braces, they are
216 visually distinct, and you don't need to use escape characters within the double
217 braces like you would in C. In the rare case that you want to match double
218 braces explicitly from the input, you can use something ugly like
219 ``{{[{][{]}}`` as your pattern.
224 It is often useful to match a pattern and then verify that it occurs again
225 later in the file. For codegen tests, this can be useful to allow any register,
226 but verify that that register is used consistently later. To do this,
227 :program:`FileCheck` allows named variables to be defined and substituted into
228 patterns. Here is a simple example:
233 ; CHECK: notw [[REGISTER:%[a-z]+]]
234 ; CHECK: andw {{.*}}[[REGISTER]]
236 The first check line matches a regex ``%[a-z]+`` and captures it into the
237 variable ``REGISTER``. The second line verifies that whatever is in
238 ``REGISTER`` occurs later in the file after an "``andw``". :program:`FileCheck`
239 variable references are always contained in ``[[ ]]`` pairs, and their names can
240 be formed with the regex ``[a-zA-Z][a-zA-Z0-9]*``. If a colon follows the name,
241 then it is a definition of the variable; otherwise, it is a use.
243 :program:`FileCheck` variables can be defined multiple times, and uses always
244 get the latest value. Variables can also be used later on the same line they
245 were defined on. For example:
249 ; CHECK: op [[REG:r[0-9]+]], [[REG]]
251 Can be useful if you want the operands of ``op`` to be the same register,
252 and don't care exactly which register it is.
254 FileCheck Expressions
255 ~~~~~~~~~~~~~~~~~~~~~
257 Sometimes there's a need to verify output which refers line numbers of the
258 match file, e.g. when testing compiler diagnostics. This introduces a certain
259 fragility of the match file structure, as "``CHECK:``" lines contain absolute
260 line numbers in the same file, which have to be updated whenever line numbers
261 change due to text addition or deletion.
263 To support this case, FileCheck allows using ``[[@LINE]]``,
264 ``[[@LINE+<offset>]]``, ``[[@LINE-<offset>]]`` expressions in patterns. These
265 expressions expand to a number of the line where a pattern is located (with an
266 optional integer offset).
268 This way match patterns can be put near the relevant test lines and include
269 relative line number references, for example:
273 // CHECK: test.cpp:[[@LINE+4]]:6: error: expected ';' after top level declarator
274 // CHECK-NEXT: {{^int a}}
275 // CHECK-NEXT: {{^ \^}}
276 // CHECK-NEXT: {{^ ;}}