On Tue, Aug 29, 2023 at 07:26:54AM +0100, Conor Dooley wrote:
> On Tue, Aug 29, 2023 at 01:56:34PM +0800, Guo Ren wrote:
> > On Mon, Aug 28, 2023 at 4:56 AM Nam Cao <[email protected]> wrote:
> > >
> > > uprobes expects is_trap_insn() to return true for any trap instructions,
> > > not just the one used for installing uprobe. The current default
> > > implementation only returns true for 16-bit c.ebreak if C extension is
> > > enabled. This can confuse uprobes if a 32-bit ebreak generates a trap
> > > exception from userspace: uprobes asks is_trap_insn() who says there is no
> > > trap, so uprobes assume a probe was there before but has been removed, and
> > > return to the trap instruction. This cause an infinite loop of entering
> > > and exiting trap handler.
> > >
> > > Instead of using the default implementation, implement this function
> > > speficially for riscv which checks for both ebreak and c.ebreak.
> > >
> > > Fixes: 74784081aac8 ("riscv: Add uprobes supported")
> > > Signed-off-by: Nam Cao <[email protected]>
> > > ---
> > > arch/riscv/kernel/probes/uprobes.c | 10 ++++++++++
> > > 1 file changed, 10 insertions(+)
> > >
> > > diff --git a/arch/riscv/kernel/probes/uprobes.c b/arch/riscv/kernel/probes/uprobes.c
> > > index 194f166b2cc4..91f4ce101cd1 100644
> > > --- a/arch/riscv/kernel/probes/uprobes.c
> > > +++ b/arch/riscv/kernel/probes/uprobes.c
> > > @@ -3,6 +3,7 @@
> > > #include <linux/highmem.h>
> > > #include <linux/ptrace.h>
> > > #include <linux/uprobes.h>
> > > +#include <asm/insn.h>
> > >
> > > #include "decode-insn.h"
> > >
> > > @@ -17,6 +18,15 @@ bool is_swbp_insn(uprobe_opcode_t *insn)
> > > #endif
> > > }
> > >
> > > +bool is_trap_insn(uprobe_opcode_t *insn)
> > > +{
> > > +#ifdef CONFIG_RISCV_ISA_C
>
> > Can we remove the CONFIG_RISCV_ISA_C? As you said, "uprobes expects
> > is_trap_insn() to return true for any trap instructions". So userspace
> > wouldn't be limited by CONFIG_RISCV_ISA_C.
>
> Isn't the RISCV_ISA_C required because there's a different encoding for
> EBREAK vs C_EBREAK? That said, this should be using IS_ENABLED() not
> #ifdef, since the definition for riscv_insn_is_c_ebreak() is provided
> unconditionally afaict.
Sorry, was too quick that I missed the last sentence.
Now I'm not sure what you mean. But I agree with Guo Ren here, users can use
compressed instructions but kernel does not have it enabled. So we should
always check c.ebreak regardless of RISCV_ISA_C.
Best regards,
Nam
On Tue, Aug 29, 2023 at 10:18:30AM +0200, Nam Cao wrote:
> Now I'm not sure what you mean. But I agree with Guo Ren here, users can use
> compressed instructions but kernel does not have it enabled. So we should
> always check c.ebreak regardless of RISCV_ISA_C.
I think I was just being dumb, apologies for the noise.