summaryrefslogtreecommitdiff
path: root/utils
AgeCommit message (Collapse)Author
2024-05-15Added a check to enable Thumb-2 floating point code to be treated differentlyDavid Boddie
to Thumb code. This requires Thumb-2 builds to pass the -2 option to tl.
2024-04-22Merge pull request #11 from dboddie/thumb2-compiler-changesCharles Forsyth
Thumb compiler changes to support Thumb-2
2024-04-22Merge pull request #9 from dboddie/add-data2textsCharles Forsyth
Add data2texts utility
2023-11-16Fixed support for the status flag in some AND instructions.David Boddie
2023-11-16Added a definition for the temporary floating point register.David Boddie
2023-11-16Reassigned the temporary floating point register.David Boddie
2023-11-16Adapted code from 5l to simplify the FP comparison case.David Boddie
Annotated the integer/float conversion encodings.
2023-11-16Used modified VFP instruction sequences from 5l.David Boddie
2023-11-16Unsuccessfully tried to implement reasonable sets of instructions.David Boddie
2023-11-16Made single and double FP register usage consistent.David Boddie
Worked around double to long handling in the _d2l function.
2023-11-16Adjusted transfers between doubles and register pairs.David Boddie
2023-11-16Fixed register encoding, or at least made our use of registers consistent.David Boddie
2023-11-16Added a note about the number of floating point registers.David Boddie
Hopefully fix float-int transfer and comparison instructions.
2023-11-16Enabled assembly of constant strings into the text segment.David Boddie
2023-11-16Imposed limitations on supported instruction encodings.David Boddie
2023-11-16Build the Thumb suite by default.David Boddie
Create a library directory for Thumb-native Inferno.
2023-11-16Moved division and modulo operator support into the compiler.David Boddie
Fixed constant encoding in case 61.
2023-11-16Added support for AND instruction variants.David Boddie
2023-11-16Tidied up the thumbopfp function and added a case for ACMPD.David Boddie
2023-11-16Brought the order of operations in line with the VFP instructions in 5l for ↵David Boddie
easier comparison. Fixed encoding of MOVD and related operations so that their execution causes a trap to occur.
2023-11-16Added support for more floating point operations.David Boddie
Fixed bit flipping for a MOVD instruction.
2023-11-16Added entries to the optab for combinations that occur when compiling bits ↵David Boddie
of a kernel. Split ACMPF handling like the VFP support does in 5l.
2023-11-16Started trying to add missing instructions.David Boddie
2023-11-16Changed an entry in the opcode table to reflect the direction of conversion.David Boddie
Added references to instruction information.
2023-11-16Handled loads and stores using different base registers with slightly ↵David Boddie
different instructions.
2023-11-16Speculatively enable more FP instructions.David Boddie
2023-11-16Worked on converting ARM instructions to Thumb instructions for floating ↵David Boddie
point handling.
2023-11-16Removed unused array entry.David Boddie
2023-11-16Renumbered FP opcodes and imported cases from the arm.c assembler.David Boddie
Added a layer of indirection to float constants to encode them correctly in Thumb instructions.
2023-11-16Prevented change to regular ARM mode.David Boddie
2023-11-16Added FP instructions from the optab.c file.David Boddie
2023-11-16Fixed typo.David Boddie
2023-11-16Update outlstring function in tc to match othersDavid Boddie
2023-11-16Added a variant of data2s that puts data in the text section.David Boddie
2021-09-30Update outlstring function in tc to match othersDavid Boddie
2021-09-30install tl as tl not 5lCharles Forsyth
2021-03-23update NOTICEs and other legal bits to reflect changes to Plan 9's licence ↵Charles Forsyth
and confirm MIT-template for Inferno base and revisions
2021-03-11remove obsolete All rights reserved. fix email addressCharles Forsyth
2021-03-07Merged in utils-riscv (pull request #10)Richard Miller
RISC-V compiler and linker corrections Approved-by: Charles Forsyth
2021-02-03Merged in Linux/spim (pull request #11)David Boddie
Fixes for spim architecture running on a Linux host Approved-by: Charles Forsyth
2021-01-01Renamed "mips" structure member to avoid reserved symbol conflictsPaul Boddie
2020-12-08il: align data segment between small and large varsRichard Miller
2020-12-08jc: fix opcode for vlong shiftsRichard Miller
2020-12-08jc: use MOVWU for unsigned int register loadRichard Miller
2020-11-09Merged in dboddie/inferno-os-git/fix-armv3 (pull request #6)David Boddie
Fix pre-ARMv4 code generation Approved-by: Charles Forsyth <charles.forsyth@gmail.com>
2020-11-09Add toolchain for riscv (ia ic il) and riscv64 (ja jc jl)Richard Miller
Because the rv64 ISA is very nearly a proper superset of rv32, the compilers ic and jc are actually the same program, which compiles to .i or .j depending on how it is invoked; similarly for ia/ja and il/jl. It is also possible to invoke ia/ic/il with a '-j' option to specify 64-bit behaviour.
2020-11-09Add riscv and riscv64 support to libmachRichard Miller
2020-07-18Fixed case 60 which used an ARMv4 instruction regardless of version.David Boddie
Fixed case 22 to avoid using the pre/post-indexing flag with non-load instructions.
2018-12-21Merged in yk/inferno-os/AIX-power (pull request #17)Charles Forsyth
AIX-power port Approved-by: Charles Forsyth <charles.forsyth@gmail.com>
2017-07-15Reverted casts in asm.c, changing types in l.h to compensate.David Boddie
Increased the size of the dbuf array to match the total size of obuf and ibuf.
generated by cgit v1.2.3 (git 2.46.0) at 2025-12-10 03:37:57 +0000