linux-kernel - Re: [RFC] mm: Generalize notify_page

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]

Message-ID: <20190530110639.GC23461@bombadil.infradead.org>
Date:   Thu, 30 May 2019 04:06:39 -0700
From:   Matthew Wilcox <willy@...radead.org>
To:     Anshuman Khandual <anshuman.khandual@....com>
Cc:     linux-kernel@...r.kernel.org, linux-mm@...ck.org,
        linux-arm-kernel@...ts.infradead.org, linux-ia64@...r.kernel.org,
        linuxppc-dev@...ts.ozlabs.org, linux-s390@...r.kernel.org,
        linux-sh@...r.kernel.org, sparclinux@...r.kernel.org,
        Andrew Morton <akpm@...ux-foundation.org>,
        Michal Hocko <mhocko@...e.com>,
        Mark Rutland <mark.rutland@....com>,
        Christophe Leroy <christophe.leroy@....fr>,
        Stephen Rothwell <sfr@...b.auug.org.au>,
        Andrey Konovalov <andreyknvl@...gle.com>,
        Michael Ellerman <mpe@...erman.id.au>,
        Paul Mackerras <paulus@...ba.org>,
        Russell King <linux@...linux.org.uk>,
        Catalin Marinas <catalin.marinas@....com>,
        Will Deacon <will.deacon@....com>,
        Tony Luck <tony.luck@...el.com>,
        Fenghua Yu <fenghua.yu@...el.com>,
        Martin Schwidefsky <schwidefsky@...ibm.com>,
        Heiko Carstens <heiko.carstens@...ibm.com>,
        Yoshinori Sato <ysato@...rs.sourceforge.jp>,
        "David S. Miller" <davem@...emloft.net>
Subject: Re: [RFC] mm: Generalize notify_page_fault()

On Thu, May 30, 2019 at 11:25:13AM +0530, Anshuman Khandual wrote:
> Similar notify_page_fault() definitions are being used by architectures
> duplicating much of the same code. This attempts to unify them into a
> single implementation, generalize it and then move it to a common place.
> kprobes_built_in() can detect CONFIG_KPROBES, hence notify_page_fault()
> must not be wrapped again within CONFIG_KPROBES. Trap number argument can

This is a funny quirk of the English language.  "must not" means "is not
allowed to be", not "does not have to be".

> @@ -141,6 +142,19 @@ static int __init init_zero_pfn(void)
>  core_initcall(init_zero_pfn);
>  
>  
> +int __kprobes notify_page_fault(struct pt_regs *regs, unsigned int trap)
> +{
> +	int ret = 0;
> +
> +	if (kprobes_built_in() && !user_mode(regs)) {
> +		preempt_disable();
> +		if (kprobe_running() && kprobe_fault_handler(regs, trap))
> +			ret = 1;
> +		preempt_enable();
> +	}
> +	return ret;
> +}
> +
>  #if defined(SPLIT_RSS_COUNTING)

Comparing this to the canonical implementation (ie x86), it looks similar.

static nokprobe_inline int kprobes_fault(struct pt_regs *regs)
{
        if (!kprobes_built_in())
                return 0;
        if (user_mode(regs))
                return 0;
        /*
         * To be potentially processing a kprobe fault and to be allowed to call
         * kprobe_running(), we have to be non-preemptible.
         */
        if (preemptible())
                return 0;
        if (!kprobe_running())
                return 0;
        return kprobe_fault_handler(regs, X86_TRAP_PF);
}

The two handle preemption differently.  Why is x86 wrong and this one
correct?