public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: David Laight <david.laight.linux@gmail.com>
To: "Thomas Weißschuh" <linux@weissschuh.net>
Cc: Willy Tarreau <w@1wt.eu>, Daniel Palmer <daniel@thingy.jp>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 5/7] tools/nolibc: handle 64-bit system call arguments on MIPS N32
Date: Sat, 18 Apr 2026 17:32:44 +0100	[thread overview]
Message-ID: <20260418173244.0f9c970a@pumpkin> (raw)
In-Reply-To: <e0c9710e-67b5-49c8-a496-d6b7f30563e7@weissschuh.net>

On Sat, 18 Apr 2026 13:54:55 +0200
Thomas Weißschuh <linux@weissschuh.net> wrote:

> Hey David,
> 
> Apr 18, 2026 13:14:46 David Laight <david.laight.linux@gmail.com>:
> 
> > On Sat, 18 Apr 2026 12:20:00 +0200
> > Thomas Weißschuh <linux@weissschuh.net> wrote:
> >  
> >> The N32 system call ABI expects 64-bit values directly in registers.
> >> This does not work on nolibc currently, as a 'long' is only 32 bits
> >> wide. Switch the system call wrappers to use 'long long' instead which
> >> can handle 64-bit values on N32. As on N64 'long' and 'long long' are
> >> the same, this does not change the behavior there.
> >>
> >> Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
> >> ---
> >> tools/include/nolibc/arch-mips.h | 94 +++++++++++++++++++++-------------------
> >> 1 file changed, 49 insertions(+), 45 deletions(-)
> >>
> >> diff --git a/tools/include/nolibc/arch-mips.h b/tools/include/nolibc/arch-mips.h
> >> index bb9d580ea1b1..557ef34d9df8 100644
> >> --- a/tools/include/nolibc/arch-mips.h
> >> +++ b/tools/include/nolibc/arch-mips.h
> >> @@ -55,6 +55,8 @@
> >> #define _NOLIBC_SYSCALL_STACK_RESERVE "addiu $sp, $sp, -32\n"
> >> #define _NOLIBC_SYSCALL_STACK_UNRESERVE "addiu $sp, $sp, 32\n"
> >>
> >> +#define _NOLIBC_SYSCALL_REG register long
> >> +
> >> #else /* _ABIN32 || _ABI64 */
> >>
> >> /* binutils, GCC and clang disagree about register aliases, use numbers instead. */
> >> @@ -66,12 +68,14 @@
> >> #define _NOLIBC_SYSCALL_STACK_RESERVE
> >> #define _NOLIBC_SYSCALL_STACK_UNRESERVE
> >>
> >> +#define _NOLIBC_SYSCALL_REG register long long
> >> +
> >> #endif /* _ABIO32 */  
> >
> > Since you need to use a #define, did you think about:
> > #define _NOLIBC_SYSCALL_REG(var, reg) register long long var __asm__ (reg)
> > to shorten the lines and the repetitive pattern.  
> 
> I didn't think of this.
> Personally I am fine with both variants.
> The parameterized macro is a bit weird
> is it breaks the normal syntax.
> Let's wait what Willy thinks.

Actually you could take it one stage further:
#define _NOLIBC_SYSCALL_REG(reg) register long long _##reg __asm__ (#reg)
So you'd have:
	_NOLIBC_SYSCALL_REG(a0) = (cast)(arg1);
I'd then use "+r" (_v0), "+r" (_a3).

For the variables that get pushed use the [name] syntax to avoid counting,
they also don't need (well shouldn't need) a local variable - the cast
can be included on the asm line. Giving:
	"sw %[arg5], 16($sp)\n"
	...
	[arg5]"r"((long)(arg5), ...
(Does look like those are all 32bit only, so the (long) cast is correct.)

	David

> 
> >> #define __nolibc_syscall0(num)                                                \
> >> ({                                                                            \
> >> -   register long _num __asm__ ("v0") = (num);                            \
> >> -   register long _arg4 __asm__ ("a3");                                   \
> >> +   _NOLIBC_SYSCALL_REG _num __asm__ ("v0")  = (num);                     \
> >> +   _NOLIBC_SYSCALL_REG _arg4 __asm__ ("a3");                             \  
> >
> >     __NOLIBC_SYSCALL_REG(_num, "v0") = (num);
> >     __NOLIBC_SYSCALL_REG(_arg4, "a3");
> >
> > ...
> >
> >     David  
> 
> 


  reply	other threads:[~2026-04-18 16:32 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-18 10:19 [PATCH 0/7] tools/nolibc: large file support Thomas Weißschuh
2026-04-18 10:19 ` [PATCH 1/7] tools/nolibc: also handle _llseek system call Thomas Weißschuh
2026-04-18 11:23   ` David Laight
2026-04-18 11:56     ` Thomas Weißschuh
2026-04-18 16:03       ` David Laight
2026-04-18 10:19 ` [PATCH 2/7] tools/nolibc: add __nolibc_arg_to_reg() Thomas Weißschuh
2026-04-18 10:19 ` [PATCH 3/7] tools/nolibc: cast pointers returned from system calls through integers Thomas Weißschuh
2026-04-18 10:19 ` [PATCH 4/7] tools/nolibc: handle 64-bit system call arguments on x32 Thomas Weißschuh
2026-04-18 10:20 ` [PATCH 5/7] tools/nolibc: handle 64-bit system call arguments on MIPS N32 Thomas Weißschuh
2026-04-18 11:14   ` David Laight
2026-04-18 11:54     ` Thomas Weißschuh
2026-04-18 16:32       ` David Laight [this message]
2026-04-18 10:20 ` [PATCH 6/7] tools/nolibc: open files with O_LARGEFILE Thomas Weißschuh
2026-04-18 10:20 ` [PATCH 7/7] selftests/nolibc: test large file support Thomas Weißschuh
2026-04-18 14:01 ` [PATCH 0/7] tools/nolibc: " Daniel Palmer

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=20260418173244.0f9c970a@pumpkin \
    --to=david.laight.linux@gmail.com \
    --cc=daniel@thingy.jp \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@weissschuh.net \
    --cc=w@1wt.eu \
    /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