Re: [PATCH v2] LoongArch: Add jump-label implementation

[Peter Zijlstra]

On Wed, May 10, 2023 at 05:34:33PM +0800, WANG Xuerui wrote:
> Hi Peter,
> 
> My 2 cents:
> 
> On 2023/5/10 17:27, Peter Zijlstra wrote:
> > On Wed, May 10, 2023 at 05:16:46PM +0800, Youling Tang wrote:
> > > Add jump-label implementation based on the ARM64 version.
> > > 
> > > Signed-off-by: Youling Tang <tangyouling@xxxxxxxxxxx>
> > 
> > > <snip>
> > > 
> > > +	if (type == JUMP_LABEL_JMP)
> > > +		insn = larch_insn_gen_b(jump_entry_code(entry), jump_entry_target(entry));
> > > +	else
> > > +		insn = larch_insn_gen_nop();
> > > +
> > > +	larch_insn_patch_text(addr, insn);
> > > +}
> > 
> > This all implies Loongarch is fine with the nop<->b transition (much
> > like arm64 is), but I found no actual mention of what transitions are
> > valid for the architecture in your inst.c file -- perhaps you could put
> > a small comment there to elucidate the occasional reader that doesn't
> > have your arch manual memorized?
> 
> Do you mean by "valid transition" something like "what kind of
> self-modification is architecturally sound, taking ICache /
> micro-architecture behavior etc. into consideration"?

Yes that. There are definitely architectures that have limitations on
what instructions can be hot-patched in the face of concurrent execution
without jumping through too many hoops.