>>> On 16.12.14 at 21:28, <andrew.coop...@citrix.com> wrote:
> On 16/12/14 19:33, Mihai Donțu wrote:
>> +static bool_t __xmem_pool_check_locked(const char *file, int line,
>> + const struct xmem_pool *pool)
>> +{
>> + unsigned int i;
>> + static bool_t once = 1;
>
> What is this static doing? Surely corruption corruption in one pool has
> no effect on corruption in a separate pool (other than the usual side
> effects of general memory corruption, which tend to be bad).
>
> It looks as if it wants to be an extra field in an xmem_pool.
Question is whether logging more than the first corruption ever is
really all that useful.
>> +bool_t __xmem_pool_check(const char *file, int line, struct xmem_pool *pool)
>> +{
>> + return __xmem_pool_check_unlocked(file, line, pool ?: xenpool);
>
> Why should a NULL pool be tolerated here? This is debug code only, so
> we can require and trust that we are called appropriately.
xenpool is not and should not be visible to code outside this file.
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
