Linux SNPS ARC Archive on lore.kernel.org
 help / color / Atom feed
From: Vineet Gupta <Vineet.Gupta1@synopsys.com>
To: Adhemerval Zanella <adhemerval.zanella@linaro.org>,
	"libc-alpha@sourceware.org" <libc-alpha@sourceware.org>
Cc: "linux-snps-arc@lists.infradead.org"
	<linux-snps-arc@lists.infradead.org>
Subject: Re: [PATCH v6 07/13] ARC: Linux Syscall Interface
Date: Sat, 30 May 2020 02:02:50 +0000
Message-ID: <5024a581-4018-14df-b75c-17aeaf360e46@synopsys.com> (raw)
In-Reply-To: <248259c7-8263-805d-56cd-9db397eb9c04@linaro.org>

On 5/29/20 9:49 AM, Adhemerval Zanella via Libc-alpha wrote:
> 
> 
> On 22/04/2020 22:41, Vineet Gupta via Libc-alpha wrote:
>> Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
> 
> As prior patches we do not use DCO, but rather copyright assignment.
> 

>> diff --git a/sysdeps/unix/sysv/linux/arc/arch-syscall.h b/sysdeps/unix/sysv/linux/arc/arch-syscall.h
>> new file mode 100644
>> index 000000000000..2b017eb5bbaa
>> --- /dev/null
>> +++ b/sysdeps/unix/sysv/linux/arc/arch-syscall.h
>> @@ -0,0 +1,303 @@
>> +/* AUTOGENERATED by update-syscall-lists.py.  */
>> +#define __NR_accept 202
>> +#define __NR_accept4 242
>> +#define __NR_acct 89
>> +#define __NR_add_key 217

>> +#define __NR_vmsplice 75
>> +#define __NR_wait4 260
>> +#define __NR_waitid 95
>> +#define __NR_write 64
>> +#define __NR_writev 66
> 
> Ok, although I did not confered it was done with lastest Linux kernel 
> releases used by build-many-glibcs.py. You might double check it.

I did use 5.6 kernel at the time, will double check though. Thx.

>> diff --git a/sysdeps/unix/sysv/linux/arc/clone.S b/sysdeps/unix/sysv/linux/arc/clone.S

>> +
>> +/* int clone(int (*fn)(void *), void *child_stack,
>> +           int flags, void *arg, ...
>> +           < pid_t *ptid, struct user_desc *tls, pid_t *ctid > );
>> +
>> + NOTE: I'm assuming that the last 3 args are NOT var-args and in case all
>> +	3 are not relevant, caller will nevertheless pass those as NULL.
> 
> Maybe we should update our clone prototype, although this might incur 
> in a ABI break. The assumption here should be ok, since all other
> implementations do the same.
> 
>> +
>> + clone syscall in kernel (ABI: CONFIG_CLONE_BACKWARDS)
>> +
>> +  int sys_clone(unsigned long int clone_flags,
>> +	        unsigned long int newsp,
>> +		int __user *parent_tidptr,
>> +		void *tls,
>> +		int __user *child_tidptr).  */
>> +
>> +ENTRY (__clone)
>> +	cmp	r0, 0		; @fn can't be NULL
>> +	cmp.ne	r1, 0		; @child_stack can't be NULL
>> +	bz	.L__sys_err
>> +
>> +	; save some of the orig args
>> +	; r0 containg @fn will be clobbered AFTER syscall (with ret val)
>> +	; rest are clobbered BEFORE syscall due to different arg ordering
>> +	mov	r10, r0		; @fn
>> +	mov	r11, r3		; @args
>> +	mov	r12, r2		; @clone_flags
>> +	mov	r9,  r5		; @tls
>> +
>> +	; adjust libc args for syscall
>> +
>> +	mov 	r0, r2		; libc @flags is 1st syscall arg
>> +	mov	r2, r4		; libc @ptid
>> +	mov	r3, r5		; libc @tls
>> +	mov	r4, r6		; libc @ctid
>> +	mov	r8, __NR_clone
>> +	ARC_TRAP_INSN
>> +
>> +	cmp	r0, 0		; return code : 0 new process, !0 parent
>> +	blt	.L__sys_err2	; < 0 (signed) error
>> +	jnz	[blink]		; Parent returns
>> +
>> +	; ----- child starts here ---------
>> +
>> +	; Setup TP register (only recent kernels v4.19+ do that)
>> +	and.f	0, r12, CLONE_SETTLS
>> +	mov.nz	r25, r9
> 
> Do you still need to set it since the minimum supported kernel
> for ARC is 5.1 ?

Right we can drop this.


> It should be safe for internal glibc usage, since for both pthread
> and posix_spawn it blocks all signals including SIGCANCEL and SIGXID.
> However this is still small race window if this is called directly 
> with pthread cancellation or g*uid in multithread.
> 
>> +
>> +	; child jumps off to @fn with @arg as argument, and returns here
>> +	jl.d	[r10]
>> +	mov	r0, r11
>> +
>> +	; exit() with result from @fn (already in r0)
>> +	mov	r8, __NR_exit
>> +	ARC_TRAP_INSN
>> +	; In case it ever came back
>> +	flag	1
>> +
>> +.L__sys_err:
> 
> Can't you use L(...) macro as other architectures? 

OK will do.

> 
>> diff --git a/sysdeps/unix/sysv/linux/arc/mmap_internal.h b/sysdeps/unix/sysv/linux/arc/mmap_internal.h
>> new file mode 100644
>> index 000000000000..19aa078dd45e
>> --- /dev/null
>> +++ b/sysdeps/unix/sysv/linux/arc/mmap_internal.h
>> @@ -0,0 +1,27 @@
>> +/* mmap - map files or devices into memory.  Linux/ARC version.
>> +   Copyright (C) 2017-2020 Free Software Foundation, Inc.
>> +   This file is part of the GNU C Library.
>> +
>> +   The GNU C Library is free software; you can redistribute it and/or
>> +   modify it under the terms of the GNU Lesser General Public
>> +   License as published by the Free Software Foundation; either
>> +   version 2.1 of the License, or (at your option) any later version.
>> +
>> +   The GNU C Library is distributed in the hope that it will be useful,
>> +   but WITHOUT ANY WARRANTY; without even the implied warranty of
>> +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
>> +   Lesser General Public License for more details.
>> +
>> +   You should have received a copy of the GNU Lesser General Public
>> +   License along with the GNU C Library; if not, see
>> +   <https://www.gnu.org/licenses/>.  */
>> +
>> +#ifndef MMAP_ARC_INTERNAL_H
>> +#define MMAP_ARC_INTERNAL_H
>> +
>> +/* 8K is default but determine the shift dynamically with getpagesize.  */
>> +#define MMAP2_PAGE_UNIT -1
>> +
>> +#include_next <mmap_internal.h>
>> +
>> +#endif
> 
> Ok.  I would be interesting if kernel could consolidate de mmap
> implementation to return EINVAL for invalid offsets. A lot of
> architectures have its own implementation to handle.



>> diff --git a/sysdeps/unix/sysv/linux/arc/pt-vfork.S b/sysdeps/unix/sysv/linux/arc/pt-vfork.S
>> new file mode 100644
>> index 000000000000..1cc893170070
>> --- /dev/null
>> +++ b/sysdeps/unix/sysv/linux/arc/pt-vfork.S
>> @@ -0,0 +1 @@
>> +/* Not needed.  */
> 
> This file is not required since "nptl: Remove vfork IFUNC-based
> forwarder from libpthread [BZ #20188]" (41d6f74e6cb6a9).

Removed.


>> diff --git a/sysdeps/unix/sysv/linux/arc/sigrestorer.S b/sysdeps/unix/sysv/linux/arc/sigrestorer.S

>> +
>> +#include <sys/syscall.h>
>> +#include <sysdep.h>
>> +#include <tcb-offsets.h>
>> +
>> +/* Note the NOP has to be outside body.  */
>> +	nop
> 
> Wy exactly do you need it here?

It has to do with how ARC gdb recognized sigrestorer. This acts a special marker
used in conjunction with the syscall id. This goes back to days when cfi pseudo
ops were not suppported in toolchain, but the gdb code lingered.

> 
>> +ENTRY (__default_rt_sa_restorer)
>> +	mov r8, __NR_rt_sigreturn
>> +	ARC_TRAP_INSN
>> +	j_s     [blink]
> 
> Ok.



>> diff --git a/sysdeps/unix/sysv/linux/arc/sysctl.mk b/sysdeps/unix/sysv/linux/arc/sysctl.mk
>> new file mode 100644
>> index 000000000000..05a4e2f8c99b
>> --- /dev/null
>> +++ b/sysdeps/unix/sysv/linux/arc/sysctl.mk
>> @@ -0,0 +1 @@
>> +# ARC doesn't support sysctl.
> 
> This file is not required since "Linux: Remove <sys/sysctl.h>
> and the sysctl function" (076f09afbac1a).

Removed.

_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

  reply index

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-23  1:41 [PATCH v6 00/13] glibc port to ARC processors Vineet Gupta
2020-04-23  1:41 ` [PATCH v6 01/13] ARC: ABI Implementation Vineet Gupta
     [not found]   ` <88508d10-2d29-026a-bb54-ad607154ab87@linaro.org>
2020-05-27 22:15     ` Vineet Gupta
2020-05-29 13:56       ` Adhemerval Zanella
     [not found]     ` <a56a35d4-3e9e-9a88-4be5-8553d5f11ad3@synopsys.com>
     [not found]       ` <87mu5jxkv7.fsf@oldenburg2.str.redhat.com>
2020-06-04 19:01         ` Vineet Gupta
2020-06-04 23:56           ` Vineet Gupta
2020-04-23  1:41 ` [PATCH v6 02/13] ARC: startup and dynamic linking code Vineet Gupta
     [not found]   ` <17957ee6-2bc1-f43b-f184-f0703ba2765f@linaro.org>
2020-05-28  1:14     ` Vineet Gupta
2020-04-23  1:41 ` [PATCH v6 03/13] ARC: Thread Local Storage support Vineet Gupta
     [not found]   ` <4f7a67fb-6f96-57e6-b827-d1ab5dd6733f@linaro.org>
2020-05-28  1:36     ` Vineet Gupta
2020-06-01 18:53       ` Adhemerval Zanella
2020-04-23  1:41 ` [PATCH v6 04/13] ARC: Atomics and Locking primitives Vineet Gupta
     [not found]   ` <03f4a9b3-b1ca-90fa-0b6a-609a3135267d@linaro.org>
2020-04-24  7:23     ` Vineet Gupta
     [not found]     ` <20200427215938.14136-1-vgupta@synopsys.com>
2020-04-27 22:13       ` [PATCH] semaphore: consolidate arch headers into a generic one Vineet Gupta
     [not found]       ` <ac93c301-36d3-b20a-d31c-50c1f3264c68@linaro.org>
2020-05-05 22:59         ` Vineet Gupta
2020-05-08 13:32           ` Adhemerval Zanella
2020-04-23  1:41 ` [PATCH v6 05/13] ARC: math soft float support Vineet Gupta
2020-04-23  1:41 ` [PATCH v6 06/13] ARC: hardware floating point support Vineet Gupta
2020-05-29 14:12   ` Adhemerval Zanella
2020-05-29 22:28     ` Vineet Gupta
2020-05-29 23:50       ` Vineet Gupta
2020-06-02 17:51         ` Joseph Myers
     [not found]         ` <07887c48-7e07-9f89-035d-3f336a16f2da@synopsys.com>
2020-06-02 18:13           ` static inline math functions (was Re: [PATCH v6 06/13] ARC: hardware floating point support) Joseph Myers
2020-06-02 18:35             ` Adhemerval Zanella
2020-06-05  4:44         ` [PATCH v6 06/13] ARC: hardware floating point support Vineet Gupta
2020-06-05 17:22           ` Adhemerval Zanella
2020-06-02 17:48       ` Joseph Myers
2020-04-23  1:41 ` [PATCH v6 07/13] ARC: Linux Syscall Interface Vineet Gupta
2020-05-29 16:49   ` Adhemerval Zanella
2020-05-30  2:02     ` Vineet Gupta [this message]
2020-06-03 19:46     ` Vineet Gupta
2020-06-03 20:04       ` Adhemerval Zanella
2020-06-03 20:17         ` Vineet Gupta
2020-06-04 11:06           ` Adhemerval Zanella
2020-04-23  1:41 ` [PATCH v6 08/13] ARC: Linux ABI Vineet Gupta
2020-04-23  1:41 ` [PATCH v6 09/13] ARC: Linux Startup and Dynamic Loading Vineet Gupta
2020-04-23  1:41 ` [PATCH v6 10/13] ARC: ABI lists Vineet Gupta
2020-04-23  1:41 ` [PATCH v6 11/13] ARC: Build Infrastructure Vineet Gupta
2020-06-03 19:58   ` Adhemerval Zanella
2020-06-04 15:25     ` Vineet Gupta
2020-06-04 17:05       ` Adhemerval Zanella
2020-06-08  4:18         ` Vineet Gupta
2020-04-23  1:41 ` [PATCH v6 12/13] build-many-glibcs.py: Enable ARC builds Vineet Gupta
2020-04-23  1:41 ` [PATCH v6 13/13] Documentation for ARC port Vineet Gupta
2020-05-04 21:21 ` [PATCH v6 00/13] glibc port to ARC processors Vineet Gupta
2020-05-15  0:45   ` Vineet Gupta
2020-05-27  1:49     ` Vineet Gupta
     [not found]       ` <d7f1176c-87c6-90c6-161c-4705a47837ea@linaro.org>
2020-05-27 18:38         ` Vineet Gupta

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=5024a581-4018-14df-b75c-17aeaf360e46@synopsys.com \
    --to=vineet.gupta1@synopsys.com \
    --cc=adhemerval.zanella@linaro.org \
    --cc=libc-alpha@sourceware.org \
    --cc=linux-snps-arc@lists.infradead.org \
    /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

Linux SNPS ARC Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-snps-arc/0 linux-snps-arc/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-snps-arc linux-snps-arc/ https://lore.kernel.org/linux-snps-arc \
		linux-snps-arc@lists.infradead.org
	public-inbox-index linux-snps-arc

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.infradead.lists.linux-snps-arc


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git