Add references to the specific ABI documents

Expcept for L4RE and Xtensa these were obtained from rust-lang#131319

I could not find an open link to the Xtensa documentation, but the
signedness was confirmed by on of the Xtensa developers in
llvm/llvm-project#115967 (comment)

Co-authored-by: Taiki Endo <[email protected]>

library/core/src/ffi/mod.rs

@@ -92,13 +92,69 @@ pub type c_ssize_t = isize;
 mod c_char_definition {
     cfg_if! {
         // These are the targets on which c_char is unsigned. Usually the
-        // signedness is the same for all target_os values on a given
-        // architecture but there are some exceptions (see isSignedCharDefault()
-        // in clang/lib/Driver/ToolChains/Clang.cpp):
-        // - PowerPC uses unsigned char for all targets except Darwin
-        // - Arm/AArch64 uses unsigned char except for Darwin and Windows
-        // Note: the L4RE kernel builds with -funsigned-char on all targets, but
-        //  we only have a target for userspace apps so it follows the architectures.
+        // signedness is the same for all target_os values on a given architecture
+        // but there are some exceptions (see isSignedCharDefault() in clang).
+        //
+        // aarch64:
+        //   Section 10 "Arm C and C++ language mappings" in Procedure Call Standard for the Arm®
+        //   64-bit Architecture (AArch64) says C/C++ char is unsigned byte.
+        //   https://github.com/ARM-software/abi-aa/blob/2024Q3/aapcs64/aapcs64.rst#arm-c-and-c-language-mappings
+        // arm:
+        //   Section 8 "Arm C and C++ Language Mappings" in Procedure Call Standard for the Arm®
+        //   Architecture says C/C++ char is unsigned byte.
+        //   https://github.com/ARM-software/abi-aa/blob/2024Q3/aapcs32/aapcs32.rst#arm-c-and-c-language-mappings
+        // csky:
+        //   Section 2.1.2 "Primary Data Type" in C-SKY V2 CPU Applications Binary Interface
+        //   Standards Manual says ANSI C char is unsigned byte.
+        //   https://github.com/c-sky/csky-doc/blob/9f7121f7d40970ba5cc0f15716da033db2bb9d07/C-SKY_V2_CPU_Applications_Binary_Interface_Standards_Manual.pdf
+        //   Note: this doesn't seem to match Clang's default (https://github.com/rust-lang/rust/issues/129945).
+        // hexagon:
+        //   Section 3.1 "Basic data type" in Qualcomm Hexagon™ Application
+        //   Binary Interface User Guide says "By default, the `char` data type is unsigned."
+        //   https://docs.qualcomm.com/bundle/publicresource/80-N2040-23_REV_K_Qualcomm_Hexagon_Application_Binary_Interface_User_Guide.pdf
+        // msp430:
+        //   Section 2.1 "Basic Types" in MSP430 Embedded Application Binary
+        //   Interface says "The char type is unsigned by default".
+        //   https://www.ti.com/lit/an/slaa534a/slaa534a.pdf
+        //   Note: this doesn't seem to match Clang's default (https://github.com/rust-lang/rust/issues/129945).
+        // powerpc/powerpc64:
+        //   - PPC32 SysV: "Table 3-1 Scalar Types" in System V Application Binary Interface PowerPC
+        //     Processor Supplement says ANSI C char is unsigned byte
+        //     https://refspecs.linuxfoundation.org/elf/elfspec_ppc.pdf
+        //   - PPC64 ELFv1: Section 3.1.4 "Fundamental Types" in 64-bit PowerPC ELF Application
+        //     Binary Interface Supplement 1.9 says ANSI C is unsigned byte
+        //     https://refspecs.linuxfoundation.org/ELF/ppc64/PPC-elf64abi.html#FUND-TYPE
+        //   - PPC64 ELFv2: Section 2.1.2.2 "Fundamental Types" in 64-Bit ELF V2 ABI Specification
+        //     says char is unsigned byte
+        //     https://openpowerfoundation.org/specifications/64bitelfabi/
+        //   - AIX: XL C for AIX Language Reference says "By default, char behaves like an unsigned char."
+        //     https://www.ibm.com/docs/en/xl-c-aix/13.1.3?topic=specifiers-character-types
+        // riscv32/riscv64:
+        //   C/C++ type representations section in RISC-V Calling Conventions
+        //   page in RISC-V ELF psABI Document says "char is unsigned."
+        //   https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/draft-20240829-13bfa9f54634cb60d86b9b333e109f077805b4b3/riscv-cc.adoc#cc-type-representations
+        // s390x:
+        //   - ELF: "Table 1.1.: Scalar types" in ELF Application Binary Interface s390x Supplement
+        //     Version 1.6.1 categorize ISO C char in unsigned integer
+        //     https://github.com/IBM/s390x-abi/releases/tag/v1.6.1
+        //   - z/OS: XL C/C++ Language Reference says: "By default, char behaves like an unsigned char."
+        //     https://www.ibm.com/docs/en/zos/3.1.0?topic=specifiers-character-types
+        // Xtensa:
+        //   - "The char type is unsigned by default for Xtensa processors."
+        //
+        // On the following operating systems, c_char is signed by default, regardless of architecture.
+        // Darwin (macOS, iOS, etc.):
+        //   Apple targets' c_char is signed by default even on arm
+        //   https://developer.apple.com/documentation/xcode/writing-arm64-code-for-apple-platforms#Handle-data-types-and-data-alignment-properly
+        // Windows:
+        //   Windows MSVC C++ Language Reference says "Microsoft-specific: Variables of type char
+        //   are promoted to int as if from type signed char by default, unless the /J compilation
+        //   option is used."
+        //   https://learn.microsoft.com/en-us/cpp/cpp/fundamental-types-cpp?view=msvc-170#character-types)
+        // L4RE:
+        //   The kernel builds with -funsigned-char on all targets (but useserspace follows the
+        //   architecture defaults). As we only have a target for userspace apps so there are no
+        //   special cases for L4RE below.
         if #[cfg(all(
             not(windows),
             not(target_vendor = "apple"),