Re: [PATCH v3] ipc,shm: disable shmmax and shmall by default

[Michael Kerrisk (man-pages)]

On Fri, Apr 18, 2014 at 11:26 AM, Manfred Spraul
<manfred@xxxxxxxxxxxxxxxx> wrote:
> Hi Davidlohr,
>
>
> On 04/18/2014 03:25 AM, Davidlohr Bueso wrote:
>>
>> So a value of 0 bytes or pages, for shmmax and shmall, respectively,
>> implies unlimited memory, as opposed to disabling sysv shared memory.
>
> That might be a second risk:
> Right now, a sysadmin can prevent sysv memory allocations with
>
>     # sysctl kernel.shmall=0
>
> After your patch is applied, this line allows unlimited allocations.

Good point. I wonder if some folk may get bitten by this complete
reversal the semantics of shmall==0.

> Obviously my patch has the opposite problem: 64-bit wrap-arounds.

I know you alluded to a case in another thread, but I couldn't quite
work out from the mail you referred to whether this was really the
problem. (And I assume those folks were forced to fix their set-up
scripts anyway.) So, it's not clear to me whether this is a real
problem. (And your patch does not worsen things from the current
situation, right?)

Cheers,

Michael



>> --- a/include/uapi/linux/shm.h
>> +++ b/include/uapi/linux/shm.h
>> @@ -9,14 +9,14 @@
>>     /*
>>    * SHMMAX, SHMMNI and SHMALL are upper limits are defaults which can
>> - * be increased by sysctl
>> + * be modified by sysctl. By default, disable SHMMAX and SHMALL with
>> + * 0 bytes, thus allowing processes to have unlimited shared memory.
>>    */
>> -
>> -#define SHMMAX 0x2000000                /* max shared seg size (bytes) */
>> +#define SHMMAX 0                        /* max shared seg size (bytes) */
>>   #define SHMMIN 1                       /* min shared seg size (bytes) */
>>   #define SHMMNI 4096                    /* max num of segs system wide */
>>   #ifndef __KERNEL__
>> -#define SHMALL (SHMMAX/getpagesize()*(SHMMNI/16))
>> +#define SHMALL 0
>>   #endif
>>   #define SHMSEG SHMMNI                  /* max shared segs per process */
>>
>
> The "#ifndef __KERNEL__" is not required:
> As there is no reference to PAGE_SIZE anymore, one definition for SHMALL is
> sufficient.
>
>
> --
>     Manfred



-- 
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Linux/UNIX System Programming Training: http://man7.org/training/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/