[llvm] 941eab1 - [RISCV] Make parseNormalizedArchString only accept [a-z0-9_]. (#90815)

[via llvm-commits]

Author: Craig Topper
Date: 2024-05-02T09:39:52-07:00
New Revision: 941eab102f4ae524bb66bcd9bffafc605b20ef7c

URL: https://github.com/llvm/llvm-project/commit/941eab102f4ae524bb66bcd9bffafc605b20ef7c
DIFF: https://github.com/llvm/llvm-project/commit/941eab102f4ae524bb66bcd9bffafc605b20ef7c.diff

LOG: [RISCV] Make parseNormalizedArchString only accept [a-z0-9_]. (#90815)

Previously we only rejected upper case characters. We should instead
reject anything except lower case, numbers, and underscore. Other
characters will likely confuse the extension sorting.

Added: 
    

Modified: 
    llvm/lib/TargetParser/RISCVISAInfo.cpp
    llvm/unittests/TargetParser/RISCVISAInfoTest.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/TargetParser/RISCVISAInfo.cpp b/llvm/lib/TargetParser/RISCVISAInfo.cpp
index 42a66746d6455f..e4ff4bcfe25a5c 100644
--- a/llvm/lib/TargetParser/RISCVISAInfo.cpp
+++ b/llvm/lib/TargetParser/RISCVISAInfo.cpp
@@ -429,9 +429,11 @@ RISCVISAInfo::parseFeatures(unsigned XLen,
 
 llvm::Expected<std::unique_ptr<RISCVISAInfo>>
 RISCVISAInfo::parseNormalizedArchString(StringRef Arch) {
-  if (llvm::any_of(Arch, isupper))
+  // RISC-V ISA strings must be [a-z0-9_]
+  if (!llvm::all_of(
+          Arch, [](char C) { return isDigit(C) || isLower(C) || C == '_'; }))
     return createStringError(errc::invalid_argument,
-                             "string must be lowercase");
+                             "string may only contain [a-z0-9_]");
 
   // Must start with a valid base ISA name.
   unsigned XLen = 0;

diff  --git a/llvm/unittests/TargetParser/RISCVISAInfoTest.cpp b/llvm/unittests/TargetParser/RISCVISAInfoTest.cpp
index 82c8ee7f57d32d..facaa75ad912a0 100644
--- a/llvm/unittests/TargetParser/RISCVISAInfoTest.cpp
+++ b/llvm/unittests/TargetParser/RISCVISAInfoTest.cpp
@@ -20,11 +20,12 @@ bool operator==(const RISCVISAUtils::ExtensionVersion &A,
   return A.Major == B.Major && A.Minor == B.Minor;
 }
 
-TEST(ParseNormalizedArchString, RejectsUpperCase) {
-  for (StringRef Input : {"RV32", "rV64", "rv32i2P0", "rv64i2p0_A2p0"}) {
+TEST(ParseNormalizedArchString, RejectsInvalidChars) {
+  for (StringRef Input :
+       {"RV32", "rV64", "rv32i2P0", "rv64i2p0_A2p0", "rv32e2.0"}) {
     EXPECT_EQ(
         toString(RISCVISAInfo::parseNormalizedArchString(Input).takeError()),
-        "string must be lowercase");
+        "string may only contain [a-z0-9_]");
   }
 }
 
@@ -37,8 +38,8 @@ TEST(ParseNormalizedArchString, RejectsInvalidBaseISA) {
 }
 
 TEST(ParseNormalizedArchString, RejectsMalformedInputs) {
-  for (StringRef Input : {"rv64i2p0_", "rv32i2p0__a2p0", "rv32e2.0", "rv64e2p",
-                          "rv32i", "rv64ip1"}) {
+  for (StringRef Input :
+       {"rv64i2p0_", "rv32i2p0__a2p0", "rv64e2p", "rv32i", "rv64ip1"}) {
     EXPECT_EQ(
         toString(RISCVISAInfo::parseNormalizedArchString(Input).takeError()),
         "extension lacks version in expected format");