MSP430: Stop using setCmpLibcallCC #142708
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Contributor
Author
This stack of pull requests is managed by Graphite. Learn more about stacking. |
Member
|
@llvm/pr-subscribers-backend-msp430 Author: Matt Arsenault (arsenm) ChangesThis appears to only be useful for the eq/ne cases, and only for Full diff: https://github.com/llvm/llvm-project/pull/142708.diff 1 Files Affected:
diff --git a/llvm/lib/Target/MSP430/MSP430ISelLowering.cpp b/llvm/lib/Target/MSP430/MSP430ISelLowering.cpp
index 5589cea6e675d..8c55f77d062b7 100644
--- a/llvm/lib/Target/MSP430/MSP430ISelLowering.cpp
+++ b/llvm/lib/Target/MSP430/MSP430ISelLowering.cpp
@@ -151,104 +151,99 @@ MSP430TargetLowering::MSP430TargetLowering(const TargetMachine &TM,
// EABI Libcalls - EABI Section 6.2
const struct {
const RTLIB::Libcall Op;
- const char * const Name;
- const ISD::CondCode Cond;
+ const char *const Name;
} LibraryCalls[] = {
- // Floating point conversions - EABI Table 6
- { RTLIB::FPROUND_F64_F32, "__mspabi_cvtdf", ISD::SETCC_INVALID },
- { RTLIB::FPEXT_F32_F64, "__mspabi_cvtfd", ISD::SETCC_INVALID },
- // The following is NOT implemented in libgcc
- //{ RTLIB::FPTOSINT_F64_I16, "__mspabi_fixdi", ISD::SETCC_INVALID },
- { RTLIB::FPTOSINT_F64_I32, "__mspabi_fixdli", ISD::SETCC_INVALID },
- { RTLIB::FPTOSINT_F64_I64, "__mspabi_fixdlli", ISD::SETCC_INVALID },
- // The following is NOT implemented in libgcc
- //{ RTLIB::FPTOUINT_F64_I16, "__mspabi_fixdu", ISD::SETCC_INVALID },
- { RTLIB::FPTOUINT_F64_I32, "__mspabi_fixdul", ISD::SETCC_INVALID },
- { RTLIB::FPTOUINT_F64_I64, "__mspabi_fixdull", ISD::SETCC_INVALID },
- // The following is NOT implemented in libgcc
- //{ RTLIB::FPTOSINT_F32_I16, "__mspabi_fixfi", ISD::SETCC_INVALID },
- { RTLIB::FPTOSINT_F32_I32, "__mspabi_fixfli", ISD::SETCC_INVALID },
- { RTLIB::FPTOSINT_F32_I64, "__mspabi_fixflli", ISD::SETCC_INVALID },
- // The following is NOT implemented in libgcc
- //{ RTLIB::FPTOUINT_F32_I16, "__mspabi_fixfu", ISD::SETCC_INVALID },
- { RTLIB::FPTOUINT_F32_I32, "__mspabi_fixful", ISD::SETCC_INVALID },
- { RTLIB::FPTOUINT_F32_I64, "__mspabi_fixfull", ISD::SETCC_INVALID },
- // TODO The following IS implemented in libgcc
- //{ RTLIB::SINTTOFP_I16_F64, "__mspabi_fltid", ISD::SETCC_INVALID },
- { RTLIB::SINTTOFP_I32_F64, "__mspabi_fltlid", ISD::SETCC_INVALID },
- // TODO The following IS implemented in libgcc but is not in the EABI
- { RTLIB::SINTTOFP_I64_F64, "__mspabi_fltllid", ISD::SETCC_INVALID },
- // TODO The following IS implemented in libgcc
- //{ RTLIB::UINTTOFP_I16_F64, "__mspabi_fltud", ISD::SETCC_INVALID },
- { RTLIB::UINTTOFP_I32_F64, "__mspabi_fltuld", ISD::SETCC_INVALID },
- // The following IS implemented in libgcc but is not in the EABI
- { RTLIB::UINTTOFP_I64_F64, "__mspabi_fltulld", ISD::SETCC_INVALID },
- // TODO The following IS implemented in libgcc
- //{ RTLIB::SINTTOFP_I16_F32, "__mspabi_fltif", ISD::SETCC_INVALID },
- { RTLIB::SINTTOFP_I32_F32, "__mspabi_fltlif", ISD::SETCC_INVALID },
- // TODO The following IS implemented in libgcc but is not in the EABI
- { RTLIB::SINTTOFP_I64_F32, "__mspabi_fltllif", ISD::SETCC_INVALID },
- // TODO The following IS implemented in libgcc
- //{ RTLIB::UINTTOFP_I16_F32, "__mspabi_fltuf", ISD::SETCC_INVALID },
- { RTLIB::UINTTOFP_I32_F32, "__mspabi_fltulf", ISD::SETCC_INVALID },
- // The following IS implemented in libgcc but is not in the EABI
- { RTLIB::UINTTOFP_I64_F32, "__mspabi_fltullf", ISD::SETCC_INVALID },
-
- // Floating point comparisons - EABI Table 7
- { RTLIB::OEQ_F64, "__mspabi_cmpd", ISD::SETEQ },
- { RTLIB::UNE_F64, "__mspabi_cmpd", ISD::SETNE },
- { RTLIB::OGE_F64, "__mspabi_cmpd", ISD::SETGE },
- { RTLIB::OLT_F64, "__mspabi_cmpd", ISD::SETLT },
- { RTLIB::OLE_F64, "__mspabi_cmpd", ISD::SETLE },
- { RTLIB::OGT_F64, "__mspabi_cmpd", ISD::SETGT },
- { RTLIB::OEQ_F32, "__mspabi_cmpf", ISD::SETEQ },
- { RTLIB::UNE_F32, "__mspabi_cmpf", ISD::SETNE },
- { RTLIB::OGE_F32, "__mspabi_cmpf", ISD::SETGE },
- { RTLIB::OLT_F32, "__mspabi_cmpf", ISD::SETLT },
- { RTLIB::OLE_F32, "__mspabi_cmpf", ISD::SETLE },
- { RTLIB::OGT_F32, "__mspabi_cmpf", ISD::SETGT },
-
- // Floating point arithmetic - EABI Table 8
- { RTLIB::ADD_F64, "__mspabi_addd", ISD::SETCC_INVALID },
- { RTLIB::ADD_F32, "__mspabi_addf", ISD::SETCC_INVALID },
- { RTLIB::DIV_F64, "__mspabi_divd", ISD::SETCC_INVALID },
- { RTLIB::DIV_F32, "__mspabi_divf", ISD::SETCC_INVALID },
- { RTLIB::MUL_F64, "__mspabi_mpyd", ISD::SETCC_INVALID },
- { RTLIB::MUL_F32, "__mspabi_mpyf", ISD::SETCC_INVALID },
- { RTLIB::SUB_F64, "__mspabi_subd", ISD::SETCC_INVALID },
- { RTLIB::SUB_F32, "__mspabi_subf", ISD::SETCC_INVALID },
- // The following are NOT implemented in libgcc
- // { RTLIB::NEG_F64, "__mspabi_negd", ISD::SETCC_INVALID },
- // { RTLIB::NEG_F32, "__mspabi_negf", ISD::SETCC_INVALID },
-
- // Universal Integer Operations - EABI Table 9
- { RTLIB::SDIV_I16, "__mspabi_divi", ISD::SETCC_INVALID },
- { RTLIB::SDIV_I32, "__mspabi_divli", ISD::SETCC_INVALID },
- { RTLIB::SDIV_I64, "__mspabi_divlli", ISD::SETCC_INVALID },
- { RTLIB::UDIV_I16, "__mspabi_divu", ISD::SETCC_INVALID },
- { RTLIB::UDIV_I32, "__mspabi_divul", ISD::SETCC_INVALID },
- { RTLIB::UDIV_I64, "__mspabi_divull", ISD::SETCC_INVALID },
- { RTLIB::SREM_I16, "__mspabi_remi", ISD::SETCC_INVALID },
- { RTLIB::SREM_I32, "__mspabi_remli", ISD::SETCC_INVALID },
- { RTLIB::SREM_I64, "__mspabi_remlli", ISD::SETCC_INVALID },
- { RTLIB::UREM_I16, "__mspabi_remu", ISD::SETCC_INVALID },
- { RTLIB::UREM_I32, "__mspabi_remul", ISD::SETCC_INVALID },
- { RTLIB::UREM_I64, "__mspabi_remull", ISD::SETCC_INVALID },
-
- // Bitwise Operations - EABI Table 10
- // TODO: __mspabi_[srli/srai/slli] ARE implemented in libgcc
- { RTLIB::SRL_I32, "__mspabi_srll", ISD::SETCC_INVALID },
- { RTLIB::SRA_I32, "__mspabi_sral", ISD::SETCC_INVALID },
- { RTLIB::SHL_I32, "__mspabi_slll", ISD::SETCC_INVALID },
- // __mspabi_[srlll/srall/sllll/rlli/rlll] are NOT implemented in libgcc
-
+ // Floating point conversions - EABI Table 6
+ {RTLIB::FPROUND_F64_F32, "__mspabi_cvtdf"},
+ {RTLIB::FPEXT_F32_F64, "__mspabi_cvtfd"},
+ // The following is NOT implemented in libgcc
+ //{ RTLIB::FPTOSINT_F64_I16, "__mspabi_fixdi" },
+ {RTLIB::FPTOSINT_F64_I32, "__mspabi_fixdli"},
+ {RTLIB::FPTOSINT_F64_I64, "__mspabi_fixdlli"},
+ // The following is NOT implemented in libgcc
+ //{ RTLIB::FPTOUINT_F64_I16, "__mspabi_fixdu" },
+ {RTLIB::FPTOUINT_F64_I32, "__mspabi_fixdul"},
+ {RTLIB::FPTOUINT_F64_I64, "__mspabi_fixdull"},
+ // The following is NOT implemented in libgcc
+ //{ RTLIB::FPTOSINT_F32_I16, "__mspabi_fixfi" },
+ {RTLIB::FPTOSINT_F32_I32, "__mspabi_fixfli"},
+ {RTLIB::FPTOSINT_F32_I64, "__mspabi_fixflli"},
+ // The following is NOT implemented in libgcc
+ //{ RTLIB::FPTOUINT_F32_I16, "__mspabi_fixfu" },
+ {RTLIB::FPTOUINT_F32_I32, "__mspabi_fixful"},
+ {RTLIB::FPTOUINT_F32_I64, "__mspabi_fixfull"},
+ // TODO The following IS implemented in libgcc
+ //{ RTLIB::SINTTOFP_I16_F64, "__mspabi_fltid" },
+ {RTLIB::SINTTOFP_I32_F64, "__mspabi_fltlid"},
+ // TODO The following IS implemented in libgcc but is not in the EABI
+ {RTLIB::SINTTOFP_I64_F64, "__mspabi_fltllid"},
+ // TODO The following IS implemented in libgcc
+ //{ RTLIB::UINTTOFP_I16_F64, "__mspabi_fltud" },
+ {RTLIB::UINTTOFP_I32_F64, "__mspabi_fltuld"},
+ // The following IS implemented in libgcc but is not in the EABI
+ {RTLIB::UINTTOFP_I64_F64, "__mspabi_fltulld"},
+ // TODO The following IS implemented in libgcc
+ //{ RTLIB::SINTTOFP_I16_F32, "__mspabi_fltif" },
+ {RTLIB::SINTTOFP_I32_F32, "__mspabi_fltlif"},
+ // TODO The following IS implemented in libgcc but is not in the EABI
+ {RTLIB::SINTTOFP_I64_F32, "__mspabi_fltllif"},
+ // TODO The following IS implemented in libgcc
+ //{ RTLIB::UINTTOFP_I16_F32, "__mspabi_fltuf" },
+ {RTLIB::UINTTOFP_I32_F32, "__mspabi_fltulf"},
+ // The following IS implemented in libgcc but is not in the EABI
+ {RTLIB::UINTTOFP_I64_F32, "__mspabi_fltullf"},
+
+ // Floating point comparisons - EABI Table 7
+ {RTLIB::OEQ_F64, "__mspabi_cmpd"},
+ {RTLIB::UNE_F64, "__mspabi_cmpd"},
+ {RTLIB::OGE_F64, "__mspabi_cmpd"},
+ {RTLIB::OLT_F64, "__mspabi_cmpd"},
+ {RTLIB::OLE_F64, "__mspabi_cmpd"},
+ {RTLIB::OGT_F64, "__mspabi_cmpd"},
+ {RTLIB::OEQ_F32, "__mspabi_cmpf"},
+ {RTLIB::UNE_F32, "__mspabi_cmpf"},
+ {RTLIB::OGE_F32, "__mspabi_cmpf"},
+ {RTLIB::OLT_F32, "__mspabi_cmpf"},
+ {RTLIB::OLE_F32, "__mspabi_cmpf"},
+ {RTLIB::OGT_F32, "__mspabi_cmpf"},
+
+ // Floating point arithmetic - EABI Table 8
+ {RTLIB::ADD_F64, "__mspabi_addd"},
+ {RTLIB::ADD_F32, "__mspabi_addf"},
+ {RTLIB::DIV_F64, "__mspabi_divd"},
+ {RTLIB::DIV_F32, "__mspabi_divf"},
+ {RTLIB::MUL_F64, "__mspabi_mpyd"},
+ {RTLIB::MUL_F32, "__mspabi_mpyf"},
+ {RTLIB::SUB_F64, "__mspabi_subd"},
+ {RTLIB::SUB_F32, "__mspabi_subf"},
+ // The following are NOT implemented in libgcc
+ // { RTLIB::NEG_F64, "__mspabi_negd" },
+ // { RTLIB::NEG_F32, "__mspabi_negf" },
+
+ // Universal Integer Operations - EABI Table 9
+ {RTLIB::SDIV_I16, "__mspabi_divi"},
+ {RTLIB::SDIV_I32, "__mspabi_divli"},
+ {RTLIB::SDIV_I64, "__mspabi_divlli"},
+ {RTLIB::UDIV_I16, "__mspabi_divu"},
+ {RTLIB::UDIV_I32, "__mspabi_divul"},
+ {RTLIB::UDIV_I64, "__mspabi_divull"},
+ {RTLIB::SREM_I16, "__mspabi_remi"},
+ {RTLIB::SREM_I32, "__mspabi_remli"},
+ {RTLIB::SREM_I64, "__mspabi_remlli"},
+ {RTLIB::UREM_I16, "__mspabi_remu"},
+ {RTLIB::UREM_I32, "__mspabi_remul"},
+ {RTLIB::UREM_I64, "__mspabi_remull"},
+
+ // Bitwise Operations - EABI Table 10
+ // TODO: __mspabi_[srli/srai/slli] ARE implemented in libgcc
+ {RTLIB::SRL_I32, "__mspabi_srll"},
+ {RTLIB::SRA_I32, "__mspabi_sral"},
+ {RTLIB::SHL_I32, "__mspabi_slll"},
+ // __mspabi_[srlll/srall/sllll/rlli/rlll] are NOT implemented in libgcc
};
- for (const auto &LC : LibraryCalls) {
+ for (const auto &LC : LibraryCalls)
setLibcallName(LC.Op, LC.Name);
- if (LC.Cond != ISD::SETCC_INVALID)
- setCmpLibcallCC(LC.Op, LC.Cond);
- }
if (STI.hasHWMult16()) {
const struct {
|
Contributor
Author
|
ping |
arsenm
force-pushed
the
users/arsenm/msp430/stop-using-setCmpLibcallCC
branch
from
e4b22e5 to
c80d2a0
Compare
arsenm
force-pushed
the
users/arsenm/msp430/add-fcmp-test
branch
from
09ade25 to
3bdf2e2
Compare
Contributor
Author
Merge activity
|
arsenm
force-pushed
the
users/arsenm/msp430/add-fcmp-test
branch
from
3bdf2e2 to
676f67f
Compare
arsenm
force-pushed
the
users/arsenm/msp430/stop-using-setCmpLibcallCC
branch
2 times, most recently
from
fb14b7b to
4870d19
Compare
This appears to only be useful for the eq/ne cases, and only for ARM libcalls. There's no change in the new fcmp test output.
arsenm
force-pushed
the
users/arsenm/msp430/stop-using-setCmpLibcallCC
branch
from
4870d19 to
418f111
Compare
tomtor
pushed a commit
to tomtor/llvm-project
that referenced
this pull request
Jun 14, 2025
This appears to only be useful for the eq/ne cases, and only for ARM libcalls. This is setting it to the default values, and there's no change in the new fcmp test output.
akuhlens
pushed a commit
to akuhlens/llvm-project
that referenced
this pull request
Jun 24, 2025
This appears to only be useful for the eq/ne cases, and only for ARM libcalls. This is setting it to the default values, and there's no change in the new fcmp test output.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This appears to only be useful for the eq/ne cases, and only for
ARM libcalls. This is setting it to the default values, and there's
no change in the new fcmp test output.