linux-kernel - Re: [RFC PATCH 5/5] RISC-V: Support cpu hotplug.

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: <20180821075452.GC28079@infradead.org>
Date:   Tue, 21 Aug 2018 00:54:52 -0700
From:   Christoph Hellwig <hch@...radead.org>
To:     Atish Patra <atish.patra@....com>
Cc:     palmer@...ive.com, linux-riscv@...ts.infradead.org,
        mark.rutland@....com, anup@...infault.org, hch@...radead.org,
        tglx@...utronix.de, Damien.LeMoal@....com,
        linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH 5/5] RISC-V: Support cpu hotplug.

>  	if (!err) {
> +
> +#ifdef CONFIG_HOTPLUG_CPU
> +		arch_send_call_function_single_ipi(cpu);
> +#endif

Please just provide a stub version of arch_send_call_function_single_ipi
for the !CONFIG_HOTPLUG_CPU case instead of the ifdef here.

> +#ifdef CONFIG_HOTPLUG_CPU
> +int can_hotplug_cpu(void)

This should be a bool.

> +{
> +	if (cpu_ops.cpu_die)
> +		return 1;
> +	else
> +		return 0;
> +}

	return cpu_ops.cpu_die != NULL;

> +void default_cpu_die(unsigned int cpu)
> +{
> +	int sipval, sieval, scauseval;
> +
> +	/* clear all pending flags */
> +	csr_write(sip, 0);
> +	/* clear any previous scause data */
> +	csr_write(scause, 0);
> +
> +	do {
> +		wait_for_interrupt();
> +		sipval = csr_read(sip);
> +		sieval = csr_read(sie);
> +		scauseval = csr_read(scause);
> +	/* only break if wfi returns for an enabled interrupt */
> +	} while ((sipval & sieval) == 0 &&
> +		 scauseval != INTERRUPT_CAUSE_SOFTWARE);
> +
> +	boot_sec_cpu();
> +}

I suspect all of this except for the boot_sec_cpu() should go into
a helper in irq.c.  Also as-is this probably doesn't work as scauseval
will have INTERRUPT_CAUSE_FLAG set, making the comparism never true.