linux-kernel - Re: [PATCH 6/6] x86/signal: add SA_{X32,IA32}_ABI sa

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Sat, 4 Jun 2016 18:57:17 +0300
From:	Dmitry Safonov <0x7f454c46@...il.com>
To:	Andy Lutomirski <luto@...capital.net>
Cc:	Dmitry Safonov <dsafonov@...tuozzo.com>,
	"H. Peter Anvin" <hpa@...or.com>, khorenko@...tuozzo.com,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Cyrill Gorcunov <gorcunov@...nvz.org>,
	Oleg Nesterov <oleg@...hat.com>, xemul@...tuozzo.com,
	X86 ML <x86@...nel.org>, Ingo Molnar <mingo@...hat.com>
Subject: Re: [PATCH 6/6] x86/signal: add SA_{X32,IA32}_ABI sa_flags

2016-06-04 8:08 GMT+03:00 Andy Lutomirski <luto@...capital.net>:
> On Jun 1, 2016 6:13 AM, "Dmitry Safonov" <dsafonov@...tuozzo.com> wrote:

>> --- a/arch/x86/include/uapi/asm/signal.h
>> +++ b/arch/x86/include/uapi/asm/signal.h
>> @@ -70,6 +70,8 @@ typedef unsigned long sigset_t;
>>   * SA_RESETHAND clears the handler when the signal is delivered.
>>   * SA_NOCLDWAIT flag on SIGCHLD to inhibit zombies.
>>   * SA_NODEFER prevents the current signal from being masked in the handler.
>> + * SA_IA32_ABI/SA_X32_ABI indicates ABI for a signal frame,
>> + *   if neither is set, the kernel will set them according to a syscall ABI
>
> I would prefer if these weren't in the UAPI header.  If you want a
> foreign signal frame type, do a foreign syscall.  This also means that
> you should strip these flags if a user sets them directly.

Oh, thanks Andy!
I remember you have said it to the previous RFC.
So I did it -- please, check sigaction_compat_abi().
It seems like, I forget to correct this comment accordingly (will do to v2).
Shouldn't I mark in uapi header that those flag bits are reserved,
defining them there?

-- 
Regards,
Safonov Dmitry.