From: Evgeniy Temirgaleev via Tarantool-patches <tarantool-patches@dev.tarantool.org>
To: "Sergey Bronnikov" <sergeyb@tarantool.org>
Cc: tarantool-patches@dev.tarantool.org,
"Sergey Bronnikov" <estetus@gmail.com>
Subject: Re: [Tarantool-patches] [PATCH luajit] Prevent sanitizer warnings for lj_tab_new*() and table.new().
Date: Mon, 15 Jun 2026 18:20:22 +0300 [thread overview]
Message-ID: <1781536822.832174736@f756.i.mail.ru> (raw)
In-Reply-To: <8855aae6-92fa-4fdd-80c8-541dd347fecf@tarantool.org>
[-- Attachment #1: Type: text/plain, Size: 4388 bytes --]
Hi, Sergey! Thanks for the patch!
LGTM after fixing a typo:
-test:plan(table_sizes * 2)
+test:plan(#table_sizes * 2)
--
Best regards,
Evgeniy Temirgaleev
>
> From: Sergey Bronnikov via Tarantool-patches <tarantool-patches@dev.tarantool.org
> >
> To: Sergey Kaplun <skaplun@tarantool.org>, Sergey Bronnikov <estetus@gmail.com
> >
> Cc: tarantool-patches@dev.tarantool.org
> Date: Wednesday, June 10, 2026 4:21 PM +03:00
>
>
> Hi, Sergey,
>
>
>
> thanks for the review. Fixes applied and force-pushed.
>
>
>
> Sergey
>
> On 6/9/26 15:03, Sergey Kaplun via Tarantool-patches wrote:
>
>> Hi, Sergey!
>> Thanks for the patch!
>> LGTM, after fixing a few nits below.
>>
>> On
>> 09.06.26, Sergey Bronnikov wrote:
>>
>>> From: Mike Pall <mike>
>>>
>>> Reported by Sergey Bronnikov.
>>>
>>> (cherry picked from
>>> commit 8f421c81ec6aaae0bcd80e01f4353de200afbbc5)
>>>
>>> The Undefined Behaviour
>>> Sanitizer [1] produce a warning because
>>>
>>
>> Typo: s/produce/produces/
>
>
>
> Fixed.
>
>
>
>
>
>
>>
>>> the function `lua_createtable()` takes signed integer arguments,
>>> but the
>>> `lj_tab_new_ah()` was not properly validating or converting
>>> these signed
>>> values before using them in unsigned arithmetic.
>>>
>>> The fix changes the
>>> signature of `lj_tab_new_ah()` to accept
>>> uint32_t directly, and adjusts
>>> `lua_createtable()` to cast the
>>> incoming signed int values to uint32_t
>>> before passing them.
>>>
>>> [1]: https://clang.llvm.org/docs/UndefinedBehaviorSanitizer.html
>>>
>>> Sergey
>>> Bronnikov:
>>> * added the description and the test for the problem
>>>
>>> Part of
>>> tarantool/tarantool#12480
>>> ---
>>> Branch: https://github.com/tarantool/luajit/tree/ligurio/lj-1458-ub-lj_tab_new
>>>
>>>
>>> Related issues:
>>> - https://github.com/tarantool/tarantool/issues/12480
>>> - https://github.com/LuaJIT/LuaJIT/issues/1458
>>>
>>>
>>> src/lj_api.c
>>> | 2 +-
>>> src/lj_tab.c | 4 +--
>>>
>>> src/lj_tab.h | 2 +-
>>>
>>> .../lj-1458-ub-table.new.test.lua | 30 +++++++++++++++++++
>>> 4
>>> files changed, 34 insertions(+), 4 deletions(-)
>>> create mode 100644
>>> test/tarantool-tests/lj-1458-ub-table.new.test.lua
>>>
>>>
>>
>> <snipped>
>>
>>
>>> diff --git a/test/tarantool-tests/lj-1458-ub-table.new.test.lua
>>> b/test/tarantool-tests/lj-1458-ub-table.new.test.lua
>>> new file mode 100644
>>> index 00000000..d0cf6ff5
>>> --- /dev/null
>>> +++
>>> b/test/tarantool-tests/lj-1458-ub-table.new.test.lua
>>>
>>
>> Typo: s/table.new/table-new/
>
> Fixed.
>>
>>> @@ -0,0 +1,30 @@
>>> +local tap = require('tap')
>>> +
>>> +-- The test file to
>>> demonstrate UBSan warning for `table.new()`
>>> +-- with a minimal and maximum
>>> array and hash parts values.
>>> +-- See also: https://github.com/LuaJIT/LuaJIT/issues/1458.
>>> +local test =
>>> tap.test('lj-1458-ub-table-new')
>>> +
>>> +test:plan(8)
>>>
>>
>> Lets use table_sizes * 2 here.
>>
>>
>
>
>
> Updated.
>
>
>
>
>
>
>
>
>
>
>
>
>
>>
>>> +
>>> +local table_new = require('table.new')
>>> +
>>> +local INT_MAX = 2 ^ 31 - 1
>>> +local INT_MIN = -2 ^ 31
>>> +
>>> +local table_sizes = {
>>> + { 0, INT_MIN },
>>> + {
>>> 0, INT_MAX },
>>> + { INT_MIN, 0 },
>>> + { INT_MAX, 0 },
>>> +}
>>> +
>>> +for _, case in
>>> ipairs(table_sizes) do
>>> + local apart, hpart = unpack(case)
>>> + local ok,
>>> err = pcall(table_new, apart, hpart)
>>> + local message = ('table.new(%d,
>>> %d)'):format(apart, hpart)
>>> + test:is(ok, false, message .. ' is failed')
>>>
>>
>> Typo? s/failed/OK/
>
>
>
>
>
>
>
> Right, fixed:
>
>
>
> --- a/test/tarantool-tests/lj-1458-ub-table-new.test.lua
> +++ b/test/tarantool-tests/lj-1458-ub-table-new.test.lua
> @@ -23,7 +23,7 @@ for _, case in ipairs(table_sizes) do
> local apart, hpart = unpack(case)
> local ok, err = pcall(table_new, apart, hpart)
> local message = ('table.new(%d, %d)'):format(apart, hpart)
> - test:is(ok, false, message .. ' is failed')
> + test:is(ok, false, message .. ' is ok')
> test:ok(err:match('table overflow'), message .. ' correct error message')
> end
>
>
>
>>
>>> + test:ok(err:match('table overflow'), message .. ' correct error
>>> message')
>>> +end
>>> +
>>> +test:done(true)
>>> --
>>> 2.43.0
>>>
>>>
>>
>>
>
>
>
[-- Attachment #2: Type: text/html, Size: 7088 bytes --]
prev parent reply other threads:[~2026-06-15 15:20 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-09 10:55 Sergey Bronnikov via Tarantool-patches
2026-06-09 12:03 ` Sergey Kaplun via Tarantool-patches
2026-06-10 11:03 ` Sergey Bronnikov via Tarantool-patches
2026-06-10 13:42 ` Sergey Kaplun via Tarantool-patches
2026-06-15 15:20 ` Evgeniy Temirgaleev via Tarantool-patches [this message]
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=1781536822.832174736@f756.i.mail.ru \
--to=tarantool-patches@dev.tarantool.org \
--cc=e.temirgaleev@tarantool.org \
--cc=estetus@gmail.com \
--cc=sergeyb@tarantool.org \
--subject='Re: [Tarantool-patches] [PATCH luajit] Prevent sanitizer warnings for lj_tab_new*() and table.new().' \
/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