linux-kernel - Re: [PATCH v11 2/3] tty: hvc: pass DMA capable memory to put

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:   Tue, 26 Oct 2021 14:02:21 +0800
From:   Xianting Tian <xianting.tian@...ux.alibaba.com>
To:     Jiri Slaby <jirislaby@...nel.org>, gregkh@...uxfoundation.org,
        amit@...nel.org, arnd@...db.de, osandov@...com
Cc:     shile.zhang@...ux.alibaba.com, linuxppc-dev@...ts.ozlabs.org,
        virtualization@...ts.linux-foundation.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v11 2/3] tty: hvc: pass DMA capable memory to put_chars()

在 2021/10/26 下午1:10, Jiri Slaby 写道:
> On 15. 10. 21, 4:46, Xianting Tian wrote:
>> @@ -151,9 +142,11 @@ static uint32_t vtermnos[MAX_NR_HVC_CONSOLES] =
>>   static void hvc_console_print(struct console *co, const char *b,
>>                     unsigned count)
>>   {
>> -    char c[N_OUTBUF] __ALIGNED__;
>> +    char *c;
>>       unsigned i = 0, n = 0;
>>       int r, donecr = 0, index = co->index;
>> +    unsigned long flags;
>> +    struct hvc_struct *hp;
>>         /* Console access attempt outside of acceptable console 
>> range. */
>>       if (index >= MAX_NR_HVC_CONSOLES)
>> @@ -163,6 +156,13 @@ static void hvc_console_print(struct console 
>> *co, const char *b,
>>       if (vtermnos[index] == -1)
>>           return;
>>   +    hp = cons_hvcs[index];
>> +    if (!hp)
>> +        return;
>
> You effectively make the console unusable until someone calls 
> hvc_alloc() for this device, correct? This doesn't look right. Neither 
> you describe this change of behaviour in the commit log.

I mentioned such info in the commit log:
'Introduce another array(cons_hvcs[]) for hvc pointers next to the
cons_ops[] and vtermnos[] arrays. With the array, we can easily find
hvc's cons_outbuf and its lock.'

After you pointed it out, I just found what you said make sense, I checked the code hvc_console_print() can support print before hvc_alloc() is called when someone use hvc_instantiate() for an early console discovery method.
I send a patch to fix the issue?  or these serial pathches reverted fisrtly then I resend new version patches? thanks


>
> regards,