linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Mark Rutland <mark.rutland@arm.com>
To: Catalin Marinas <catalin.marinas@arm.com>
Cc: linux-arm-kernel@lists.infradead.org, will.deacon@arm.com,
	marc.zyngier@arm.com, linux@dominikbrodowski.net,
	hch@infradead.org, james.morse@arm.com, viro@zeniv.linux.org.uk,
	linux-fsdevel@vger.kernel.org, dave.martin@arm.com
Subject: Re: [PATCHv3 08/19] arm64: convert raw syscall invocation to C
Date: Tue, 19 Jun 2018 15:48:45 +0100	[thread overview]
Message-ID: <20180619144845.22mgd4req5cnp5tv@lakrids.cambridge.arm.com> (raw)
In-Reply-To: <20180619142125.esqcbknxd2pqeotb@localhost>

On Tue, Jun 19, 2018 at 03:21:25PM +0100, Catalin Marinas wrote:
> On Mon, Jun 18, 2018 at 01:02:59PM +0100, Mark Rutland wrote:
> > diff --git a/arch/arm64/kernel/syscall.c b/arch/arm64/kernel/syscall.c
> > new file mode 100644
> > index 000000000000..b463b962d597
> > --- /dev/null
> > +++ b/arch/arm64/kernel/syscall.c
> > @@ -0,0 +1,30 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +
> > +#include <linux/nospec.h>
> > +#include <linux/ptrace.h>
> > +
> > +long do_ni_syscall(struct pt_regs *regs);
> > +
> > +typedef long (*syscall_fn_t)(unsigned long, unsigned long,
> > +			     unsigned long, unsigned long,
> > +			     unsigned long, unsigned long);
> > +
> > +static void __invoke_syscall(struct pt_regs *regs, syscall_fn_t syscall_fn)
> > +{
> > +	regs->regs[0] = syscall_fn(regs->regs[0], regs->regs[1],
> > +				   regs->regs[2], regs->regs[3],
> > +				   regs->regs[4], regs->regs[5]);
> > +}
> > +
> > +asmlinkage void invoke_syscall(struct pt_regs *regs, unsigned int scno,
> > +			       unsigned int sc_nr,
> > +			       syscall_fn_t syscall_table[])
> > +{
> > +	if (scno < sc_nr) {
> > +		syscall_fn_t syscall_fn;
> > +		syscall_fn = syscall_table[array_index_nospec(scno, sc_nr)];
> > +		__invoke_syscall(regs, syscall_fn);
> > +	} else {
> > +		regs->regs[0] = do_ni_syscall(regs);
> > +	}
> > +}
> 
> While reviewing the subsequent patch, it crossed my mind that we no
> longer have any callers to do_ni_syscall() outside this file. Can we not
> more it here and have a similar implementation to __invoke_syscall() for
> consistency, i.e. set regs->regs[0].

I think so, though it complicates the logic in do_ni_syscall(), since
there are two places we may return a value.

Would you be happy if instead I made __invoke_syscall() return the
result, and have invoke_syscall() place the result in regs->regs[0] in
either case, e.g. as below?

Thanks,
Mark.

---->8----
// SPDX-License-Identifier: GPL-2.0

#include <linux/errno.h>
#include <linux/nospec.h>
#include <linux/ptrace.h>
#include <linux/syscalls.h>

#include <asm/compat.h>

long compat_arm_syscall(struct pt_regs *regs);

static long do_ni_syscall(struct pt_regs *regs)
{
#ifdef CONFIG_COMPAT
	long ret;
	if (is_compat_task()) {
		ret = compat_arm_syscall(regs);
		if (ret != -ENOSYS)
			return ret;
	}
#endif

	return sys_ni_syscall();
}

typedef long (*syscall_fn_t)(unsigned long, unsigned long,
			     unsigned long, unsigned long,
			     unsigned long, unsigned long);

static long __invoke_syscall(struct pt_regs *regs, syscall_fn_t syscall_fn)
{
	return syscall_fn(regs->regs[0], regs->regs[1], regs->regs[2],
			  regs->regs[3], regs->regs[4], regs->regs[5]);
}

asmlinkage void invoke_syscall(struct pt_regs *regs, unsigned int scno,
			       unsigned int sc_nr,
			       syscall_fn_t syscall_table[])
{
	long ret;

	if (scno < sc_nr) {
		syscall_fn_t syscall_fn;
		syscall_fn = syscall_table[array_index_nospec(scno, sc_nr)];
		ret = __invoke_syscall(regs, syscall_fn);
	} else {
		ret = do_ni_syscall(regs);
	}

	regs->regs[0] = ret;
}

  reply	other threads:[~2018-06-19 14:48 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-18 12:02 [PATCHv3 00/19] arm64: invoke syscalls with pt_regs Mark Rutland
2018-06-18 12:02 ` [PATCHv3 01/19] arm64: consistently use unsigned long for thread flags Mark Rutland
2018-06-19 10:49   ` Catalin Marinas
2018-06-19 11:19     ` Mark Rutland
2018-06-18 12:02 ` [PATCHv3 02/19] arm64: move SCTLR_EL{1,2} assertions to <asm/sysreg.h> Mark Rutland
2018-06-19 11:02   ` [PATCHv3 02/19] arm64: move SCTLR_EL{1, 2} " Catalin Marinas
2018-06-19 11:48   ` [PATCHv3 02/19] arm64: move SCTLR_EL{1,2} " Mark Rutland
2018-06-18 12:02 ` [PATCHv3 03/19] arm64: introduce sysreg_clear_set() Mark Rutland
2018-06-19 11:44   ` Catalin Marinas
2018-06-19 11:47   ` Marc Zyngier
2018-06-18 12:02 ` [PATCHv3 04/19] arm64: kill config_sctlr_el1() Mark Rutland
2018-06-19 11:44   ` Catalin Marinas
2018-06-18 12:02 ` [PATCHv3 05/19] arm64: kill change_cpacr() Mark Rutland
2018-06-19 11:45   ` Catalin Marinas
2018-06-18 12:02 ` [PATCHv3 06/19] arm64: move sve_user_{enable,disable} to <asm/fpsimd.h> Mark Rutland
2018-06-19 12:00   ` [PATCHv3 06/19] arm64: move sve_user_{enable, disable} " Catalin Marinas
2018-06-19 12:19   ` Dave Martin
2018-06-18 12:02 ` [PATCHv3 07/19] arm64: remove sigreturn wrappers Mark Rutland
2018-06-19 12:47   ` Catalin Marinas
2018-06-18 12:02 ` [PATCHv3 08/19] arm64: convert raw syscall invocation to C Mark Rutland
2018-06-19 13:33   ` Catalin Marinas
2018-06-19 14:21   ` Catalin Marinas
2018-06-19 14:48     ` Mark Rutland [this message]
2018-06-19 14:55       ` Catalin Marinas
2018-06-19 14:58         ` Mark Rutland
2018-06-18 12:03 ` [PATCHv3 09/19] arm64: convert syscall trace logic " Mark Rutland
2018-06-19 14:32   ` Catalin Marinas
2018-06-19 15:14     ` Mark Rutland
2018-06-18 12:03 ` [PATCHv3 10/19] arm64: convert native/compat syscall entry " Mark Rutland
2018-06-19 12:18   ` Dave Martin
2018-06-19 13:15     ` Mark Rutland
2018-06-20  9:21       ` Dave Martin
2018-06-19 15:20   ` Catalin Marinas
2018-06-18 12:03 ` [PATCHv3 11/19] arm64: don't reload GPRs after apply_ssbd Mark Rutland
2018-06-19 15:23   ` Catalin Marinas
2018-06-18 12:03 ` [PATCHv3 12/19] arm64: zero GPRs upon entry from EL0 Mark Rutland
2018-06-19 15:24   ` Catalin Marinas
2018-06-18 12:03 ` [PATCHv3 13/19] kernel: add ksys_personality() Mark Rutland
2018-06-19 15:45   ` Catalin Marinas
2018-06-19 15:53     ` Mark Rutland
2018-06-18 12:03 ` [PATCHv3 14/19] kernel: add kcompat_sys_{f,}statfs64() Mark Rutland
2018-06-18 12:03 ` [PATCHv3 15/19] arm64: remove in-kernel call to sys_personality() Mark Rutland
2018-06-19 15:46   ` Catalin Marinas
2018-06-18 12:03 ` [PATCHv3 16/19] arm64: use {COMPAT,}SYSCALL_DEFINE0 for sigreturn Mark Rutland
2018-06-19 15:48   ` Catalin Marinas
2018-06-18 12:03 ` [PATCHv3 17/19] arm64: use SYSCALL_DEFINE6() for mmap Mark Rutland
2018-06-19 15:50   ` Catalin Marinas
2018-06-18 12:03 ` [PATCHv3 18/19] arm64: convert compat wrappers to C Mark Rutland
2018-06-19 15:51   ` Catalin Marinas
2018-06-18 12:03 ` [PATCHv3 19/19] arm64: implement syscall wrappers Mark Rutland
2018-06-19 16:13   ` Catalin Marinas
2018-06-19 15:38 ` [PATCHv3 00/19] arm64: invoke syscalls with pt_regs Mark Rutland
2018-06-19 16:14   ` Catalin Marinas

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180619144845.22mgd4req5cnp5tv@lakrids.cambridge.arm.com \
    --to=mark.rutland@arm.com \
    --cc=catalin.marinas@arm.com \
    --cc=dave.martin@arm.com \
    --cc=hch@infradead.org \
    --cc=james.morse@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux@dominikbrodowski.net \
    --cc=marc.zyngier@arm.com \
    --cc=viro@zeniv.linux.org.uk \
    --cc=will.deacon@arm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).