Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <6fe031a6-246f-9759-6bba-8a7cf498bab8@gmail.com>
Date: Thu, 11 Jun 2020 20:04:52 +0200
From: Mikhail Morfikov <mmorfikov@...il.com>
To: lkrg-users@...ts.openwall.com
Subject: Re: RE: ISRA optimized functions

On 11/06/2020 19:59, Adam Zabrocki wrote:
> On Thu, Jun 11, 2020 at 07:29:33PM +0200, Mikhail Morfikov wrote:
>> On 11/06/2020 18:44, Adam Zabrocki wrote:
>>> Hi,
>>>
>>> On Wed, Jun 10, 2020 at 09:18:44AM +0200, Mikhail Morfikov wrote:
>>>> I just tested and:
>>>>
>>>> # cat /proc/version
>>>> Linux version 5.7.1-amd64 (morfik@...fikownia) (gcc version 10.1.0 (Debian 10.1.0-3), GNU ld (GNU Binutils for Debian) 2.34) #3 SMP PREEMPT Wed Jun 10 07:21:29 CEST 2020
>>>>
>>>> # dkms status
>>>> lkrg, 0.7+git20200609, 5.7.1-amd64, x86_64: installed
>>>>
>>>> # modprobe -v p_lkrg
>>>> insmod /lib/modules/5.7.1-amd64/updates/dkms/p_lkrg.ko
>>>>
>>>> # lsmod
>>>> Module                  Size  Used by
>>>> p_lkrg                225280  0
>>>>
>>>> In the syslog I have the following messages now:
>>>>
>>>> kernel: p_lkrg: loading out-of-tree module taints kernel.
>>>> kernel: [p_lkrg] Loading LKRG...
>>>> kernel: [p_lkrg] System does NOT support SMAP. LKRG can't enforce SMAP validation :(
>>>> kernel: Freezing user space processes ... (elapsed 0.031 seconds) done.
>>>> kernel: OOM killer disabled.
>>>> kernel: [p_lkrg] 8/23 UMH paths were whitelisted...
>>>> kernel: [p_lkrg] [kretprobe] register_kretprobe() for <lookup_fast> failed! [err=-22]
>>>> kernel: [p_lkrg] Trying to find ISRA name for <lookup_fast>
>>>> kernel: [p_lkrg] ISRA version not found!
>>>> kernel: [p_lkrg] LKRG won't enforce pCFI validation on 'lookup_fast'
>>>> kernel: [p_lkrg] LKRG initialized successfully!
>>>> kernel: OOM killer enabled.
>>>> kernel: Restarting tasks ... done.
>>>>
>>>> So what to do with this *lookup_fast* ?
>>>>
>>>
>>> It is not critical hook,  that's why LKRG continue initialization. If you have 
>>> set log_level=3 then you should see an extra message printing such information.
>>> However, can you please run the following command and provide the result?
>>>
>>> # cat /proc/kallsyms|grep lookup_fast
>>>
>>> I'm just curious
>>
>> Here it is:
>>
>> # cat /proc/kallsyms|grep lookup_fast
>> ffffffffa16f05d0 t lookup_fast.constprop.0
> 
> You compiler enforced constprop optimization on that function. You can manually 
> replace the name of lookup_fast to lookup_fast.constprop.0 and it will work 
> fine. This function is ISRA safe and should CONSTPROP safe as well.
> 
> 
>> ffffffffa2202580 t nft_hash_lookup_fast
>> ffffffffc019a026 t p_pcfi_lookup_fast_entry.cold        [p_lkrg]
>> ffffffffc01b3520 d p_pcfi_lookup_fast_kretprobe [p_lkrg]
>> ffffffffc0191db0 t p_pcfi_lookup_fast_ret       [p_lkrg]
>> ffffffffc0191fa0 t p_install_pcfi_lookup_fast_hook      [p_lkrg]
>> ffffffffc0191dc0 t p_pcfi_lookup_fast_entry     [p_lkrg]
>> ffffffffc0191fd0 t p_uninstall_pcfi_lookup_fast_hook    [p_lkrg]
>> ffffffffc01bad74 b p_pcfi_lookup_fast_kretprobe_state   [p_lkrg]
>>
> 
> 
> 
> 
I didn't change anything -- it's the default gcc10 config in Debian Sid. 
I just moved from gcc9->gcc10.



Download attachment "signature.asc" of type "application/pgp-signature" (229 bytes)

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.