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 [thread overview]
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
next prev parent reply other threads:[~2020-05-30 2:03 UTC|newest]
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
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).