[PATCH v4 12/12] RISC-V: fix auipc-jalr addresses in patched alternatives

[Heiko Stuebner]

Am Mittwoch, 7. Dezember 2022, 21:48:08 CET schrieb Conor Dooley:
> On Wed, Dec 07, 2022 at 07:08:21PM +0100, Heiko Stuebner wrote:
> > From: Heiko Stuebner <heiko.stuebner at vrull.eu>
> > 
> > Alternatives live in a different section, so addresses used by call
> > functions will point to wrong locations after the patch got applied.
> > 
> > Similar to arm64, adjust the location to consider that offset.
> > 
> > Signed-off-by: Heiko Stuebner <heiko.stuebner at vrull.eu>
> > ---
> >  arch/riscv/include/asm/alternative.h |  3 ++
> >  arch/riscv/kernel/alternative.c      | 56 ++++++++++++++++++++++++++++
> >  arch/riscv/kernel/cpufeature.c       |  5 ++-
> >  3 files changed, 63 insertions(+), 1 deletion(-)
> > 
> > diff --git a/arch/riscv/include/asm/alternative.h b/arch/riscv/include/asm/alternative.h
> > index 6511dd73e812..1bd4027d34ca 100644
> > --- a/arch/riscv/include/asm/alternative.h
> > +++ b/arch/riscv/include/asm/alternative.h
> > @@ -27,6 +27,9 @@ void __init apply_boot_alternatives(void);
> >  void __init apply_early_boot_alternatives(void);
> >  void apply_module_alternatives(void *start, size_t length);
> >  
> > +void riscv_alternative_fix_offsets(void *alt_ptr, unsigned int len,
> > +				   int patch_offset);
> > +
> >  struct alt_entry {
> >  	void *old_ptr;		 /* address of original instruciton or data  */
> >  	void *alt_ptr;		 /* address of replacement instruction or data */
> > diff --git a/arch/riscv/kernel/alternative.c b/arch/riscv/kernel/alternative.c
> > index a7d26a00beea..c29e198ed9df 100644
> > --- a/arch/riscv/kernel/alternative.c
> > +++ b/arch/riscv/kernel/alternative.c
> > @@ -15,6 +15,8 @@
> >  #include <asm/vendorid_list.h>
> >  #include <asm/sbi.h>
> >  #include <asm/csr.h>
> > +#include <asm/insn.h>
> > +#include <asm/patch.h>
> >  
> >  struct cpu_manufacturer_info_t {
> >  	unsigned long vendor_id;
> > @@ -53,6 +55,60 @@ static void __init_or_module riscv_fill_cpu_mfr_info(struct cpu_manufacturer_inf
> >  	}
> >  }
> >  
> > +static u32 riscv_instruction_at(void *p)
> > +{
> > +	u16 *parcel = p;
> > +
> > +	return (unsigned int)parcel[0] | (unsigned int)parcel[1] << 16;
> 
> I feel bad for not mentioning this before - can we replace this magic 16
> with something self documenting?
> 
> > +}
> > +
> > +static void riscv_alternative_fix_auipc_jalr(void *ptr, u32 insn1, u32 insn2, int patch_offset)
> > +{
> > +	/* pick the original auipc + jalr */
> > +	u32 call[2] = { insn1, insn2 };
> > +	s32 imm;
> > +
> > +	/* get and adjust new target address */
> > +	imm = riscv_insn_extract_utype_itype_imm(insn1, insn2);
> > +	imm -= patch_offset;
> > +
> > +	/* update instructions */
> > +	riscv_insn_insert_utype_itype_imm(call, imm);
> > +
> > +	/* patch the call place again */
> > +	patch_text_nosync(ptr, call, sizeof(u32) * 2);
> > +}
> 
> Obv. I have not left R-b tags on stuff without trying to understand
> what's going on, but from a lay-observer point of view, I think these
> function names & flow does a good job of explaining some of the black
> magic in this neck of the woods.
> 
> > +
> > +void riscv_alternative_fix_offsets(void *alt_ptr, unsigned int len,
> > +				      int patch_offset)
> > +{
> > +	int num_instr = len / sizeof(u32);
> 
> instr...
> 
> > +	int i;
> > +
> > +	/*
> > +	 * stop one instruction before the end, as we're checking
> > +	 * for auipc + jalr
> > +	 */
> > +	for (i = 0; i < num_instr; i++) {
> > +		u32 inst = riscv_instruction_at(alt_ptr + i * sizeof(u32));
> 
> ...inst...
> 
> > +
> > +		/* may be the start of an auipc + jalr pair */
> > +		if (riscv_insn_is_auipc(inst) && i < num_instr - 1) {
> 
> ...insn.
> 
> Is there a reason for that?

I guess more of a generational issue - with the code spanning
too much time :-)

So poll question, what would be preferred?
I think I remember seeing all of them somewhere, so I'm unsure what
to standardize on.

> 
> Also, I've gotten myself slightly confused about the loop. You "stop one
> instruction before the end" but the main loop goes from 0 -> num_instr.
> The inner loop then checks for i < num_instr - 1. What am I missing that
> prevents the outer loop from stopping at num_instr - 1 instead?

The idea with this is to allow a
	if(riscv_insn_is_jal(inst))
		riscv_alternative_fix_jal(...)

etc, and everything else is a single instruction so needs one more
loop iteration, only for auipc+jalr do we want to stop one earlier.

So to get this alternatives_fix_offsets() main entry point I
made the loop do the one iteration more again.


> > +			u32 inst2 = riscv_instruction_at(alt_ptr + (i + 1) * sizeof(u32));
> > +
> > +			if (!riscv_insn_is_jalr(inst2))
> > +				continue;
> > +
> > +			/* call will use ra register */
> 
> Super minor, but "call will use ra register" is a little unclear. As
> written, it makes perfect sense when you've been staring at this code,
> but not so much if you're passing through.. How about:
> /* if this instruction pair is a call, it will use the ra register */

sure :-)

> 
> > +			if (RV_EXTRACT_RD_REG(inst) != 1)
> > +				continue;
> >
> 
> All minor stuff though, so you can re-add my R-b either way:
> Reviewed-by: Conor Dooley <conor.dooley at microchip.com>

thanks :-)

Heiko