maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/test/Profile/Inputs/c-general.proftext

158 lines
607 B
Plaintext
Raw Normal View History

CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
simple_loops
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
16515
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
4
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
100
100
75
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
conditionals
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
74917022372782735
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
11
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
100
50
50
33
33
16
99
100
99
100
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
early_exits
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
44128811889290
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
9
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
0
51
1
25
1
25
1
0
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
jumps
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
2016037664281362839
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
22
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
1
0
1
0
0
1
0
1
2
3
2
0
3
0
1
1
1
10
0
10
9
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
switches
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
2745195701975551402
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
19
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
1
1
15
7
1
0
2
2
3
3
4
4
0
4
4
5
1
0
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
big_switch
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
10218718452081869619
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
17
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
32
32
1
0
1
1
11
11
1
1
15
15
1
1
2
2
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
boolean_operators
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
291222909838
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
8
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
100
34
66
17
34
33
50
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
boolop_loops
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
9760565944591
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
9
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
50
51
50
26
50
51
50
26
CodeGen: Handle binary conditional operators in PGO instrumentation This treats binary conditional operators in the same way as ternary conditional operators for instrumentation based profiling. llvm-svn: 206021
2014-04-11 14:10:10 +08:00
conditional_operator
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
848
CodeGen: Handle binary conditional operators in PGO instrumentation This treats binary conditional operators in the same way as ternary conditional operators for instrumentation based profiling. llvm-svn: 206021
2014-04-11 14:10:10 +08:00
3
1
0
1
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
do_fallthrough
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
16586
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
4
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
10
2
8
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
main
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
0
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
1
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
[PGO] Revert r255366: solution incomplete, not handling lambda yet llvm-svn: 255368
2015-12-12 04:23:12 +08:00
c-general.c:static_func
InstrProf: Calculate a better function hash The function hash should change when control flow changes. This patch hashes the type of each AST node that affects counters, rather than just counting how many there are. These types are combined into a small enumerator that currently has 16 values. The new hash algorithm packs the enums for consecutively visited types into a `uint64_t`. In order to save space for new types, the types are assumed to be 6-bit values (instead of 4-bit). In order to minimize overhead for functions with little control flow, the `uint64_t` is used directly as a hash if it never fills up; if it does, it's passed through an MD5 context. <rdar://problem/16435801> llvm-svn: 206397
2014-04-17 00:03:27 +08:00
4
CodeGen: Include a function hash in instrumentation based profiling The hash itself is still the number of counters, which isn't all that useful, but this separates the API changes from the actual implementation of the hash and will make it easier to transition to the ProfileData library once it's implemented. llvm-svn: 204186
2014-03-19 05:58:06 +08:00
2
Revert "CodeGen: Use a binary format for instrumentation based profiling" I've clearly done something wrong with how to get this to link correctly. Reverting for now. This reverts commit r203711. llvm-svn: 203712
2014-03-13 05:06:31 +08:00
1
10