linux-kernel - Re: [PATCH] mm/vmalloc: Fix regression caused by needless vmalloc_sync

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-next>] [day] [month] [year] [list]

Message-ID: <83a39694-373b-da16-3be1-22108ace0107@linux.alibaba.com>
Date:   Thu, 14 Nov 2019 22:40:22 +0800
From:   Shile Zhang <shile.zhang@...ux.alibaba.com>
To:     Michal Hocko <mhocko@...nel.org>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Joerg Roedel <jroedel@...e.de>, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mm/vmalloc: Fix regression caused by needless
 vmalloc_sync_all()



On 2019/11/14 21:56, Michal Hocko wrote:
> On Wed 13-11-19 17:55:30, Shile Zhang wrote:
>> vmalloc_sync_all() was put in the common path in
>> __purge_vmap_area_lazy(), for one sync issue only happened on X86_32
>> with PTI enabled. It is needless for X86_64, which caused a big regression
>> in UnixBench Shell8 testing on X86_64.
>> Similar regression also reported by 0-day kernel test robot in reaim
>> benchmarking:
>> https://lists.01.org/hyperkitty/list/lkp@lists.01.org/thread/4D3JPPHBNOSPFK2KEPC6KGKS6J25AIDB/
>>
>> Fix it by adding more conditions.
> This doesn't really explain a lot. Could you explain why the syncing
> should be limited only to PAE and !SHARED_KERNEL_PMD?

Thanks for your review!
Sorry for that, I'm not clear about the original issue the patch
3f8fd02b1bf1 ("mm/vmalloc: Sync unmappings in __purge_vmap_area_lazy()")
fixed.

I just get that limitation info from the commit log as following:
"
This is only needed x86-32 with !SHARED_KERNEL_PMD, which is the case on 
a PAE kernel with PTI enabled. On affected systems the missing sync 
causes old mappings to persist in some page-tables, causing data 
corruption and other undefined behavior.
"
https://patchwork.kernel.org/cover/11050511/

Hi, Joerg
Could you please help to recall the original issue you encountered before?
Thanks!

>
>> Fixes: 3f8fd02b1bf1 ("mm/vmalloc: Sync unmappings in __purge_vmap_area_lazy()")
>>
>> Signed-off-by: Shile Zhang <shile.zhang@...ux.alibaba.com>
>> ---
>>   mm/vmalloc.c | 8 +++++++-
>>   1 file changed, 7 insertions(+), 1 deletion(-)
>>
>> diff --git a/mm/vmalloc.c b/mm/vmalloc.c
>> index a3c70e275f4e..7b9fc7966da6 100644
>> --- a/mm/vmalloc.c
>> +++ b/mm/vmalloc.c
>> @@ -1255,11 +1255,17 @@ static bool __purge_vmap_area_lazy(unsigned long start, unsigned long end)
>>   	if (unlikely(valist == NULL))
>>   		return false;
>>   
>> +#if defined(CONFIG_X86_32) && defined(CONFIG_X86_PAE)
>>   	/*
>>   	 * First make sure the mappings are removed from all page-tables
>>   	 * before they are freed.
>> +	 *
>> +	 * This is only needed on x86-32 with !SHARED_KERNEL_PMD, which is
>> +	 * the case on a PAE kernel with PTI enabled.
>>   	 */
>> -	vmalloc_sync_all();
>> +	if (!SHARED_KERNEL_PMD && boot_cpu_has(X86_FEATURE_PTI))
>> +		vmalloc_sync_all();
>> +#endif
>>   
>>   	/*
>>   	 * TODO: to calculate a flush range without looping.
>> -- 
>> 2.24.0.rc2
>>