List:Commits« Previous MessageNext Message »
From:Dmitry Lenev Date:October 5 2010 9:31am
Subject:Re: bzr commit into mysql-5.5-runtime branch (jon.hauglid:3152)
Bug#57002
View as plain text  
Hi Jon Olav!

* Jon Olav Hauglid <jon.hauglid@stripped> [10/10/05 13:07]:
> On 10/05/2010 10:32 AM, Dmitry Lenev wrote:
>> I have discussed this bug with Kostja.
>>
>> He thinks that solution for this MERGE-specific problem should be
>> localized to MERGE-specific code rather than change code handling
>> general case.
>
> In general, I agree with this line of thinking.
> Especially if not checking MDL lock type when finding
> the TABLE instance to use, will always remain a MERGE-specific issue.
>
>> === modified file 'storage/myisammrg/ha_myisammrg.cc'
>> --- storage/myisammrg/ha_myisammrg.cc   2010-09-08 08:25:37 +0000
>> +++ storage/myisammrg/ha_myisammrg.cc   2010-10-05 08:05:35 +0000
>> @@ -478,8 +478,10 @@ int ha_myisammrg::add_children_list(void
>>       /* Set the expected table version, to not cause spurious re-prepare.
>>   * */
>>       child_l->set_table_ref_id(mrg_child_def->get_child_table_ref_type(),
>>                                 mrg_child_def->get_child_def_version());
>> -    /* Use the same metadata lock type for children. */
>> -    child_l->mdl_request.set_type(parent_l->mdl_request.type);
>> +
>> +    if (! thd->locked_tables_mode&&
>> +        parent_l->mdl_request.type == MDL_SHARED_NO_WRITE)
>> +      child_l->mdl_request.set_type(MDL_SHARED_NO_WRITE);
>>       /* Link TABLE_LIST object into the children list. */
>>       if (this->children_last_l)
>>         child_l->prev_global= this->children_last_l;
>>
>> What do you think about this approach?
>
> My concern is that this will make MERGE tables conceptually harder to
> understand by adding a special case.

Well... MERGE tables are already somewhat convoluted.
If it is a price which has to be paid for keeping other
code clean I would be glad to pay it.

> I also think that the code for selecting a correct TABLE instance
> could use a slight refactoring along the lines you mentioned.
> But this is maybe not the time and place for it?

I suggest to wait until Alik pushes his temporary tables refactoring
patch. This patch should open the way for so much more refactorings
that we are likely to see things better once it is pushed.

So I suggest to follow Kostja's approach at this point.

-- 
Dmitry Lenev, Software Developer
Oracle Development SPB/MySQL, www.mysql.com

Are you MySQL certified?  http://www.mysql.com/certification
Thread
bzr commit into mysql-5.5-runtime branch (jon.hauglid:3152) Bug#57002Jon Olav Hauglid4 Oct
  • Re: bzr commit into mysql-5.5-runtime branch (jon.hauglid:3152)Bug#57002Dmitry Lenev5 Oct
    • Re: bzr commit into mysql-5.5-runtime branch (jon.hauglid:3152)Bug#57002Dmitry Lenev5 Oct
      • Re: bzr commit into mysql-5.5-runtime branch (jon.hauglid:3152) Bug#57002Jon Olav Hauglid5 Oct
        • Re: bzr commit into mysql-5.5-runtime branch (jon.hauglid:3152)Bug#57002Dmitry Lenev5 Oct