aboutsummaryrefslogtreecommitdiff
path: root/riscv-gcc
AgeCommit message (Collapse)AuthorFilesLines
2021-02-19Bump gcc for beKito Cheng1-0/+0
2020-11-24Fix last gcc bump.Jim Wilson1-0/+0
Commit ids changed on branch merge, so accidentally pointing at wrong commit id, on deleted branch.
2020-11-05Bump gccKito Cheng1-0/+0
Backport following patches from upstream: - RISC-V: Check multiletter extension has more than 1 letter - RISC-V: Add configure option: --with-multilib-generator to flexible config multi-lib settings. - RISC-V: Refine riscv_parse_arch_string - RISC-V: Extend syntax for the multilib-generator - RISC-V: Handle implied extension in multilib-generator - RISC-V: Add support for -mcpu option. - RISC-V: Define __riscv_cmodel_medany for PIC mode.
2020-10-15Bump gcc for zstd fix.Jim Wilson1-0/+0
2020-09-23Bump GCC to 10.2Kito Cheng1-0/+0
2020-06-17Bump gcc 10Kito Cheng1-0/+0
- Fix https://github.com/riscv/riscv-gcc/issues/190
2020-06-11Bump gccKito Cheng1-0/+0
Backport following patches: - RISC-V: Add shorten_memrefs pass. - RISC-V: Make unique SECCAT_SRODATA names start with .srodata (not .sdata2) - RISC-V: Update march parser - RISC-V: Handle implied extension for -march parser - Fix alignment for local variable [PR90811] - RISC-V: Optimize si to di zero-extend followed by left shift. - RISC-V: Make __divdi3 handle div by zero same as hardware. - RISC-V: Describe correct USEs for gpr_save pattern [PR95252] - RISC-V: Unify the output asm pattern between gpr_save and gpr_restore pattern.
2020-06-03Bump GCC to 10.1Kito Cheng1-0/+0
- With few bugfix backport: - riscv: Fix up riscv_atomic_assign_expand_fenv [PR94950] - testsuite: Require gnu-tm support for pr94856.C
2019-11-01Bump gcc to include fix for issue #530.Jim Wilson1-0/+0
2019-10-23Bump gcc to including fix for PR91860Kito Cheng1-0/+0
2019-10-01Bump gccKito Cheng1-0/+0
Backport following patches: - RISC-V: Fix bad insn splits with paradoxical subregs. - RISC-V: Fix more splitters accidentally calling gen_reg_rtx. - Fix for PR rtl-optimization/89795
2019-08-20Bump gcc to 9.2Kito Cheng1-0/+0
2019-08-07Bump gccKito Cheng1-0/+0
Backport following patches: - RISC-V: Handle g extension in multilib-generator - RISC-V: Handle extensions combination correctly in multilib-generator. - RISC-V: Promote type correctly for libcalls - RISC-V: Fix testcase on rv64 - RISC-V: Raise error on unexpected ISA string at end.
2019-07-25Bump riscv-gccKito Cheng1-0/+0
backport following patches: - RISC-V: Add -malign-data= option.
2019-07-17Bump riscv-gccKito Cheng1-0/+0
backport following patches: - RISC-V: Fix epilogue unwind info with fp and single sp adjust - RISC-V: Generalize -march support, add ELF attribute support. - RISC-V: Add libstdc++ check-abi support. - RISC-V: Fix %lo overflow with BLKmode references. - RISC-V: Add sifive-7 pipeline description. - RISC-V: Fix __riscv_compressed regression. - RISC-V: Promode modes of constant loads for store insns. - RISC-V: Short-forward-branch opt for SiFive 7 series cores. - RISC-V: Move STARTFILE_PREFIX_SPEC into target OS files. - RISC-V: Fix splitter for 32-bit AND on 64-bit target. - https://github.com/riscv/riscv-gcc/issues/161 - Fix typo in riscv_get_interrupt_type and riscv_merge_decl_attributes. - Correctly ignore empty C++ structs when flattening for ABI - Add --disable-tm-clone-registry libgcc configure option.
2019-02-27Upgrade to gcc-8.3.0 release.Jim Wilson1-0/+0
2018-10-09Bump gcc, included several backports from trunkKito Cheng1-0/+0
2018-08-09Bump gcc 8.2 with backports from trunkKito Cheng1-0/+0
2018-07-03Bump gcc-8 to include Kito's backports.Jim Wilson1-0/+0
2018-06-25Update GCC to the 8.1 releasePalmer Dabbelt1-0/+0
This has no additional patches, but it does result in a whole host of test suite failures.
2017-11-07Bump GCCv20171107Palmer Dabbelt1-0/+0
b731149757b9 RISC-V: Implement movmemsi 605bc7b5e06a RISC-V: Define MUSL_DYNAMIC_LINKER
2017-11-06I accidentally commited a staging branch of GCCPalmer Dabbelt1-0/+0
This has the same commits as the previous hash, it's just on the proper branch now (so it's round-tripped through upstream's trunk). I archived the old commit.
2017-11-04Bump GCCPalmer Dabbelt1-0/+0
b66926e93524 RISC-V: Emit "i" suffix for instructions with immediate operands a2fc54542b66 RISC-V: If -m[no-]strict-align is not passed, assume its value from -mtune f34a83e82258 RISC-V: Set SLOW_BYTE_ACCESS=1 7dde69e2c5f7 RISC-V: Handle non-legitimate address in riscv_legitimize_move 1751fbe7b9e8 RISC-V: Use "@minus{}2 GB" instead of "-2 GB" in invoke.texi 6d1f1f891869 RISC-V: Document the medlow and medany code models
2017-10-31Move to a GCC based on the 7.2.0 releasePalmer Dabbelt1-0/+0
Much like the recent binutils changes, I've now moved to a different GCC strategy. We have the following commits right now d2d1f783b2c1 RISC-V: Correct and improve the "-mabi" documentation d13dd0242604 RISC-V: Add Sign/Zero extend patterns for PIC loads 341375637a7d RISC-V: Add -mstrict-align option f47f9c2b3b90 RISC-V: Unify indention in riscv.md Note that this still fails a handful of the GCC regression tests, so it's not ready to go yet.
2017-10-11Bump GCC, to emit a NOP after movmemPalmer Dabbelt1-0/+0
2017-09-11Bump gcc for block-move support (#277)Andrew Waterman1-0/+0
2017-05-17Bump GCC, for an ICE fixPalmer Dabbelt1-0/+0
16210e6 RISC-V: Handle non-legitimate address in riscv_legitimize_move
2017-05-10Bump gccPalmer Dabbelt1-0/+0
ff03ebe RISC-V: Add Sign/Zero extend patterns for PIC loads 2fe94d5 RISC-V: Add -mstrict-align option f9771ad RISC-V: Unify indention in riscv.md
2017-05-02Use the GCC 7.1 releasePalmer Dabbelt1-0/+0
2017-04-26Move GCC to the 7 release branchPalmer Dabbelt1-0/+0
GCC has created a stable branch upstream for the 7 release series. This patch moves our GCC sources over to that branch, just like we did for binutils.
2017-03-30Bump binutils and GCC, passed the testsPalmer Dabbelt1-0/+0
2017-03-03bump everythingAndrew Waterman1-0/+0
2017-02-01bump; change linux default to RVCAndrew Waterman1-0/+0
2017-01-20Rework multilibs and bump everythingAndrew Waterman1-0/+0
2017-01-05bump gccAndrew Waterman1-0/+0
2016-12-22bump binutils, gcc, glibcAndrew Waterman1-0/+0
2016-12-12Multilibs go in /lib${XLEN}/${ABI}/, e.g. /lib64/lp64/Andrew Waterman1-0/+0
2016-12-12Incorporate new GCC flags and multilib supportAndrew Waterman1-0/+0
2016-12-06avoid non-standard predefined macrosAndrew Waterman1-0/+0
2016-11-28Use builtin linker script for newlib, not a custom oneAndrew Waterman1-0/+0
2016-10-25Fix interaction of -mno-float with -marchAndrew Waterman1-0/+0
2016-10-25Overhaul floating-point command-line argumentsAndrew Waterman1-0/+0
- -march controls which instructions may be emitted, and does not affect ABI. - -mno-float is equivalent to passing neither F for D to -march; -msingle-float is equivalent to passing F but not D to -march; and -mdouble-float is equivalent to passing F and D to -march. - -mfloat-abi controls which calling convention is used: "soft" means no args passed in registers; "single" means only single-precision values are passed in regisers; "double" means single- and double-precision values are passed in registers. - -mfloat-abi defaults to "soft" if the D extension is not present, or "double" if it is. - GCC will issue an error if -mfloat-abi requires an extension not provided by -march, but vice-versa is OK. Closes #187 (I hope). @kito-cheng can you test this and let me know what I need to fix?
2016-10-13Support an RV32F compilerAndrew Waterman1-0/+0
Soft-float calling convention is default, but an alternate hard-float calling convention (with doubles passed as in soft-float, but floats passed as in hard-float) is supported.
2016-09-12bump gcc and binutilsAndrew Waterman1-0/+0
2016-08-16bump binutils and gccAndrew Waterman1-0/+0
2016-07-16bump gccAndrew Waterman1-0/+0
2016-07-11Move to submodules for GCC, glibcPalmer Dabbelt1-0/+0