linux-kernel - Re: [PATCH v2] x86: Quark: Add if/else to setup

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:	Fri, 26 Sep 2014 17:59:55 -0300
From:	Henrique de Moraes Holschuh <hmh@....eng.br>
To:	Bryan O'Donoghue <pure.logic@...us-software.ie>
Cc:	hpa@...or.com, mingo@...hat.com, tglx@...utronix.de,
	davej@...hat.com, boon.leong.ong@...el.com, x86@...nel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] x86: Quark: Add if/else to setup_arch for Quark TLB
 bug

On Fri, 26 Sep 2014, Bryan O'Donoghue wrote:
> Quark X1000 incorrectly advertises PGE. In later stages of boot
> specifically in early_init_intel we setup_clear_cpu_cap for PGE.
> At this point in time cpu_has_pge() will still be true.
> 
> Use the boot_cpu_data to decide if __flush_tlb_all() or __flush_tlb()
> should be called subsequent to loading CR3

...

>  	load_cr3(swapper_pg_dir);
> -	__flush_tlb_all();
> +
> +	/*
> +	 * Flush the TLB after loading CR3
> +	 *
> +	 * Quark X1000 wrongly advertises PGE. Use boot_cpu_data to
> +	 * to make sure ithe TLB is flushed correctly in the early
> +	 * stage of setup_arch() for Quark X1000.
> +	 * X86_FEATURE_PGE flag is only setup later stage at
> +	 * early_cpu_init();
> +	 */
> +	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
> +	    boot_cpu_data.x86 == 5 && boot_cpu_data.x86_model == 9)
> +		__flush_tlb();
> +	else
> +		__flush_tlb_all();
>  #else
>  	printk(KERN_INFO "Command line: %s\n", boot_command_line);
>  #endif

I'm confused, now.

Wasn't the other patch -- which just added a comment -- the one selected as
a better fix, because there is absolutely no point in calling __flush_tlb()
on Quark X1000 *right after* you just flushed the TLB [on these processors]
by doing a load_cr3() ?

Should this one ([PATCH v2] x86: Quark: Add if/else to setup_arch for Quark
TLB bug) be ignored?  Or should the other one which just adds a comment be
ignored?

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/