| Message ID | 20221027172435.2687118-4-jamie@jamieiles.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | RISC-V: Dynamic ftrace support for RV32I | expand |
| Context | Check | Description |
|---|---|---|
| conchuod/cover_letter | success | Series has a cover letter |
| conchuod/tree_selection | success | Guessed tree name to be for-next |
| conchuod/patch_count | success | Link |
| conchuod/fixes_present | success | Fixes tag not required for -next series |
| conchuod/verify_signedoff | success | Signed-off-by tag matches author and committer |
| conchuod/header_inline | success | No static functions without inline keyword in header files |
| conchuod/verify_fixes | success | No Fixes tag |
| conchuod/checkpatch | success | total: 0 errors, 0 warnings, 0 checks, 25 lines checked |
| conchuod/kdoc | success | Errors and warnings before: 0 this patch: 0 |
| conchuod/source_inline | success | Was 0 now: 0 |
| conchuod/cc_maintainers | warning | 5 maintainers not CCed: ajones@ventanamicro.com palmer@dabbelt.com conor.dooley@microchip.com paul.walmsley@sifive.com aou@eecs.berkeley.edu |
| conchuod/build_warn_rv64 | fail | Errors and warnings before: 0 this patch: 0 |
| conchuod/module_param | success | Was 0 now: 0 |
On Thu, Oct 27, 2022 at 06:24:34PM +0100, Jamie Iles wrote: > The RISC-V ELF psABI states that both a0 and a1 are used for return > values so we should preserve them both in return_to_handler. This is > especially important for RV32 for functions returning a 64-bit quantity > otherwise the return value can be corrupted and undefined behaviour > results. > > Signed-off-by: Jamie Iles <jamie@jamieiles.com> > --- > arch/riscv/kernel/mcount.S | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/arch/riscv/kernel/mcount.S b/arch/riscv/kernel/mcount.S > index 4166909ce7b3..5aa29509f047 100644 > --- a/arch/riscv/kernel/mcount.S > +++ b/arch/riscv/kernel/mcount.S > @@ -29,6 +29,7 @@ > REG_S s0, 2*SZREG(sp) > REG_S ra, 3*SZREG(sp) > REG_S a0, 1*SZREG(sp) > + REG_S a1, 0*SZREG(sp) > addi s0, sp, 32 > .endm > > @@ -42,6 +43,7 @@ > REG_L ra, 3*SZREG(sp) > REG_L s0, 2*SZREG(sp) > REG_L a0, 1*SZREG(sp) > + REG_L a1, 0*SZREG(sp) > addi sp, sp, 4*SZREG > .endm > > @@ -71,9 +73,9 @@ ENTRY(return_to_handler) > mv a0, t6 > #endif > call ftrace_return_to_handler > - mv a1, a0 > + mv a2, a0 > RESTORE_RET_ABI_STATE > - jalr a1 > + jalr a2 > ENDPROC(return_to_handler) > #endif > > -- > 2.34.1 > Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
diff --git a/arch/riscv/kernel/mcount.S b/arch/riscv/kernel/mcount.S index 4166909ce7b3..5aa29509f047 100644 --- a/arch/riscv/kernel/mcount.S +++ b/arch/riscv/kernel/mcount.S @@ -29,6 +29,7 @@ REG_S s0, 2*SZREG(sp) REG_S ra, 3*SZREG(sp) REG_S a0, 1*SZREG(sp) + REG_S a1, 0*SZREG(sp) addi s0, sp, 32 .endm @@ -42,6 +43,7 @@ REG_L ra, 3*SZREG(sp) REG_L s0, 2*SZREG(sp) REG_L a0, 1*SZREG(sp) + REG_L a1, 0*SZREG(sp) addi sp, sp, 4*SZREG .endm @@ -71,9 +73,9 @@ ENTRY(return_to_handler) mv a0, t6 #endif call ftrace_return_to_handler - mv a1, a0 + mv a2, a0 RESTORE_RET_ABI_STATE - jalr a1 + jalr a2 ENDPROC(return_to_handler) #endif
The RISC-V ELF psABI states that both a0 and a1 are used for return values so we should preserve them both in return_to_handler. This is especially important for RV32 for functions returning a 64-bit quantity otherwise the return value can be corrupted and undefined behaviour results. Signed-off-by: Jamie Iles <jamie@jamieiles.com> --- arch/riscv/kernel/mcount.S | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)