Re: [PATCH V11 14/22] LoongArch: Add signal handling support

[WANG Xuerui]

On 5/19/22 00:54, Huacai Chen wrote:
> Hi, Eric,
>
> On Thu, May 19, 2022 at 12:40 AM Eric W. Biederman
> <ebiederm@xxxxxxxxxxxx> wrote:
> > Huacai Chen <chenhuacai@xxxxxxxxxxx> writes:
> >
> > > Add ucontext/sigcontext definition and signal handling support for
> > > LoongArch.
> > >
> > > Cc: Eric Biederman <ebiederm@xxxxxxxxxxxx>
> > > Cc: Al Viro <viro@xxxxxxxxxxxxxxxxxx>
> > > Signed-off-by: Huacai Chen <chenhuacai@xxxxxxxxxxx>
> > > ---
> > >   arch/loongarch/include/uapi/asm/sigcontext.h |  44 ++
> > >   arch/loongarch/include/uapi/asm/signal.h     |  13 +
> > >   arch/loongarch/include/uapi/asm/ucontext.h   |  35 ++
> > >   arch/loongarch/kernel/signal.c               | 566 +++++++++++++++++++
> > >   4 files changed, 658 insertions(+)
> > >   create mode 100644 arch/loongarch/include/uapi/asm/sigcontext.h
> > >   create mode 100644 arch/loongarch/include/uapi/asm/signal.h
> > >   create mode 100644 arch/loongarch/include/uapi/asm/ucontext.h
> > >   create mode 100644 arch/loongarch/kernel/signal.c
> > >
> > > diff --git a/arch/loongarch/include/uapi/asm/sigcontext.h b/arch/loongarch/include/uapi/asm/sigcontext.h
> > > new file mode 100644
> > > index 000000000000..be3d3c6ac83e
> > > --- /dev/null
> > > +++ b/arch/loongarch/include/uapi/asm/sigcontext.h
> > > @@ -0,0 +1,44 @@
> > > +/* SPDX-License-Identifier: GPL-2.0+ WITH Linux-syscall-note */
> > > +/*
> > > + * Author: Hanlu Li <lihanlu@xxxxxxxxxxx>
> > > + *         Huacai Chen <chenhuacai@xxxxxxxxxxx>
> > > + *
> > > + * Copyright (C) 2020-2022 Loongson Technology Corporation Limited
> > > + */
> > > +#ifndef _UAPI_ASM_SIGCONTEXT_H
> > > +#define _UAPI_ASM_SIGCONTEXT_H
> > > +
> > > +#include <linux/types.h>
> > > +#include <linux/posix_types.h>
> > > +
> > > +/* FP context was used */
> > > +#define SC_USED_FP           (1 << 0)
> > > +/* Address error was due to memory load */
> > > +#define SC_ADDRERR_RD                (1 << 30)
> > > +/* Address error was due to memory store */
> > > +#define SC_ADDRERR_WR                (1 << 31)
> > > +
> > > +struct sigcontext {
> > > +     __u64   sc_pc;
> > > +     __u64   sc_regs[32];
> > > +     __u32   sc_flags;
> > > +     __u64   sc_extcontext[0] __attribute__((__aligned__(16)));
> > > +};
> > > +
> > > +#define CONTEXT_INFO_ALIGN   16
> > > +struct _ctxinfo {
> > > +     __u32   magic;
> > > +     __u32   size;
> > > +     __u64   padding;        /* padding to 16 bytes */
> > > +};
> > This is probably something I a missing but what is struct _ctxinfo and
> > why is it in a uapi header?
> >
> > I don't see anything else in the uapi implementation using it.
> This is used by get_ctx_through_ctxinfo() in signal.c and I think
> similar function is also needed by userspace.
>
> Its name is once before called context_info but conflict with another
> software, then I want to use ctx_info but conflict with another kernel
> struct. :(
> > Symbols that start with an underscore "_" are reserved and should not
> > be used in general, and especially not in uapi header files.
> Then, maybe we can use sctx_info here?

Actually it seems to me that this struct is the header of each 
sc_extcontext; i.e. one struct sigcontext with 0~N trailing 
"sc_extcontext", and each "sc_extcontext" consists of one "struct 
_ctxinfo" header plus the real content. For now there's only the FPU 
context but the LSX/LASX/LBT contexts will come later. The "magic" and 
"size" fields also matches the definitions following.

So I'd suggest something like "struct sc_extcontext_head" while naming 
the individual contexts like "struct sc_extcontext_foo", e.g. "struct 
sc_extcontext_fpu". The "sc_extcontext" part could use some further 
abbreviation but the end result should be something intuitive.

> Huacai
> > Eric