Tarantool development patches archive
 help / color / mirror / Atom feed
From: Vladislav Shpilevoy <v.shpilevoy@tarantool.org>
To: Igor Munkin <imun@tarantool.org>
Cc: tarantool-patches@dev.tarantool.org
Subject: Re: [Tarantool-patches] [PATCH 1/2] fiber: unref fiber.storage via global Lua state
Date: Fri, 13 Dec 2019 00:38:00 +0100	[thread overview]
Message-ID: <175e0c98-8f2d-ff20-4280-63c7a3f90432@tarantool.org> (raw)
In-Reply-To: <20191211235703.GK1214@tarantool.org>

Hi! Thanks for the review!

On 12/12/2019 00:57, Igor Munkin wrote:
> Vlad,
> 
> Thanks for the patch!
> 
> I spent some time knee deep into fibers machinery and totally agree with
> this fix: the most safe approach is "unrefing" storage entity via
> tarantool_L coro, taking into account the possible unref of the fiber's
> one and its consequtive collection.
> 
> However, I don't get why we need to apply these changes, considering
> your follow-up patch. Could you please provide a bit more detailed
> rationale?

My follow up patch is about a different bug and I wanted to keep
these independent fixes separated. That would have been strange,
if in the second commit I had just silently replaced the old L
with tarantool_L.

> 
> On 08.12.19, Vladislav Shpilevoy wrote:
>> Fiber.storage is a table, available from anywhere in the fiber. It
>> is destroyed after fiber function is finished. That provides a
>> reliable fiber-local storage, similar to thread-local in C/C++.
>>
>> But there is a problem that the storage may be created via one
>> struct lua_State, and destroyed via another. Here is an example:
>>
>>     function test_storage()
>>         fiber.self().storage.key = 100
>>     end
>>     box.schema.func.create('test_storage')
>>     _ = fiber.create(function()
>>         box.func.test_storage:call()
>>     end)
>>
>> There are 3 struct lua_State:
>>     tarantool_L - global always alive Lua state;
>>     L1 - stack of the fiber, created by fiber.create();
>>     L2 - stack created by that fiber to execute test_storage().
> 
> Minor: Strictly saying, lua_State is a Lua coroutine, and not just a
> guest stack. Please, consider adjusting the commit message, but feel
> free to ignore this remark.
> 

Agree, here is a new version of this paragraph:

"
    There are 3 struct lua_State:
        tarantool_L - global always alive state;
        L1 - Lua coroutine of the fiber, created by fiber.create();
        L2 - Lua coroutine created by that fiber to execute
             test_storage().
"

  parent reply	other threads:[~2019-12-12 23:38 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-08 19:28 [Tarantool-patches] [PATCH 0/2] Fiber storage leak Vladislav Shpilevoy
2019-12-08 19:28 ` [Tarantool-patches] [PATCH 1/2] fiber: unref fiber.storage via global Lua state Vladislav Shpilevoy
2019-12-11 23:57   ` Igor Munkin
2019-12-12  8:50     ` Konstantin Osipov
2019-12-12 23:38     ` Vladislav Shpilevoy [this message]
2019-12-13 10:44       ` Igor Munkin
2019-12-08 19:28 ` [Tarantool-patches] [PATCH 2/2] fiber: destroy fiber.storage created by iproto Vladislav Shpilevoy
2019-12-09  7:21   ` Konstantin Osipov
2019-12-09 23:31     ` Vladislav Shpilevoy
2019-12-10  8:21       ` Konstantin Osipov
2019-12-10  8:32   ` Konstantin Osipov
2019-12-10 22:59     ` Vladislav Shpilevoy
2019-12-11  7:08       ` Konstantin Osipov
2019-12-11 21:23         ` Vladislav Shpilevoy
2019-12-12  0:00           ` Igor Munkin
2019-12-12 23:37             ` Vladislav Shpilevoy
2019-12-13 13:35               ` Igor Munkin
2019-12-12  8:46           ` Konstantin Osipov
2019-12-13  0:02             ` Vladislav Shpilevoy
2019-12-13  7:58               ` Konstantin Osipov
2019-12-13 23:11                 ` Vladislav Shpilevoy
2019-12-14 12:26                   ` Konstantin Osipov
2019-12-14 12:30                     ` Konstantin Osipov
2019-12-14 12:33                       ` Konstantin Osipov
2019-12-14 16:49                         ` Vladislav Shpilevoy
2019-12-14 20:59                           ` Vladislav Shpilevoy

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=175e0c98-8f2d-ff20-4280-63c7a3f90432@tarantool.org \
    --to=v.shpilevoy@tarantool.org \
    --cc=imun@tarantool.org \
    --cc=tarantool-patches@dev.tarantool.org \
    --subject='Re: [Tarantool-patches] [PATCH 1/2] fiber: unref fiber.storage via global Lua state' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox