* Re: [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization.
2024-01-12 13:26 [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization Maxim Kokryashkin via Tarantool-patches
@ 2024-01-15 15:22 ` Sergey Kaplun via Tarantool-patches
2024-02-02 12:21 ` Maxim Kokryashkin via Tarantool-patches
2024-01-16 8:46 ` Sergey Bronnikov via Tarantool-patches
` (2 subsequent siblings)
3 siblings, 1 reply; 10+ messages in thread
From: Sergey Kaplun via Tarantool-patches @ 2024-01-15 15:22 UTC (permalink / raw)
To: Maxim Kokryashkin; +Cc: tarantool-patches
Hi, Maxim!
Thanks for the patch!
LGTM with a few comments below.
On 12.01.24, Maxim Kokryashkin wrote:
> From: Mike Pall <mike>
>
> Contributed by XmiliaH.
>
> (cherry-picked from commit 91bc6b8ad1f373c1ce9003dc024b2e21fad0e444)
>
> In `lj_record_idx` when `ix->oldv` is the global nilnode and the
> required key is not present in the table, it is possible to pass
> the constant key lookup optimization condition because of the
> `uint32_t` overflow. Because of that, further recording
I suggest clarifying like the following:
| `uint32_t` (`MSize`)
Feel free to ignore.
> incorrectly removes the check for the nilnode, which produces
> wrong results when trace is called for a different table.
Nit: Please mention also how the problem is fixed.
>
> Maxim Kokryashkin:
> * added the description and the test for the problem
>
> Part of tarantool/tarantool#9145
> ---
> Branch: https://github.com/tarantool/luajit/tree/fckxorg/lj-840-fix-hrefk-optimization
> PR: https://github.com/tarantool/tarantool/pull/9591
> Issues: https://github.com/LuaJIT/LuaJIT/issues/840
> https://github.com/tarantool/tarantool/issues/9145
>
> src/lj_record.c | 8 +--
> .../lj-840-fix-hrefk-optimization.test.lua | 58 +++++++++++++++++++
> 2 files changed, 62 insertions(+), 4 deletions(-)
> create mode 100644 test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
>
> diff --git a/src/lj_record.c b/src/lj_record.c
> index a929b8aa..919e7169 100644
> --- a/src/lj_record.c
> +++ b/src/lj_record.c
<snipped>
> diff --git a/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua b/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
Nit: We can also add gc64 prefix for this test like:
<lj-840-gc64-fix-hrefk-optimization.test.lua>
Feel free to ignore.
> new file mode 100644
> index 00000000..a11b91e3
> --- /dev/null
> +++ b/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
> @@ -0,0 +1,58 @@
> +local tap = require('tap')
> +
> +-- Test file to demonstrate incorrect HREFK optimization
> +-- in LuaJIT.
> +
> +local ffi = require('ffi')
> +local test = tap.test('lj-840-fix-hrefk-optimization'):skipcond({
> + ['Test requires GC64 mode enabled'] = not ffi.abi('gc64'),
> + ['Test requires JIT enabled'] = not jit.status(),
> +})
> +test:plan(1)
> +
> +local table_new = require('table.new')
> +
> +-- Size of single hash node in bytes.
> +local NODE_SIZE = 24
> +-- Number of hash nodes to allocate on each iteration
> +-- based on the condition from `rec_idx_key`
Nit: Missed dot at the end of the sentence.
It is more correct to say that we use this is restricted by the IR
format:
`op2` field in the HREFK IR is a slot number and it is 16-bit wide.
65535 == 2^16 - 1; i.e., it is the maximum value that can be stored in a
16-bit field.
> +local HASH_NODES = 65535
> +-- The vector of hash nodes should have a raw size of
> +-- `HASH_NODES * NODE_SIZE`, which is allocated in
> +-- `lj_alloc_malloc` directly with `mmap`. However,
> +-- the LuaJIT allocator adds a bunch of small paddings
> +-- and aligns the required size to LJ_PAGESIZE, which is
> +-- 4096, so the actual allocated size includes alignment.
> +local ALIGNMENT = 4096
Minor: So, maybe name it `LJ_PAGESIZE`?
Feel free to ignore.
> +-- The vector for hash nodes in the table is allocated based on
> +-- `hbits`, so it's actually got a size of 65536 nodes.
> +local SINGLE_ITERATION_ALLOC = (HASH_NODES + 1) * NODE_SIZE + ALIGNMENT + 72
What is the magic number 72?
> +-- We need to overflow the 32-bit distance to the global nilnode,
> +-- so we divide 2^32 by the SINGLE_ITERATION_ALLOC. There are a
> +-- bunch of non-table.new allocations already performed, so one
> +-- iteration is subtracted to account for them.
Why is it crucial to subtract it? What happens without it?
I suppose that the new table will still be huge enough, won't it?
> +local N_ITERATIONS = 0x100000000 / SINGLE_ITERATION_ALLOC - 1
Minor: We can use `(2 ^ 32)` instead of 0x100000000 (it is easier to
read).
Feel free to ignore.
> +-- Prevent anchor table from interfering with target table allocations.
Nit: Comment length is more than 66 symbols.
> +local anchor = table.new(N_ITERATIONS, 0)
> +
> +-- Construct table.
> +for _ = 1, N_ITERATIONS do
> + table.insert(anchor, table_new(0, HASH_NODES))
> +end
> +
> +jit.opt.start('hotloop=1')
> +local function get_n(tab)
> + local x
> + for _ = 1, 4 do
> + x = tab.n
> + end
> + return x
> +end
> +
> +-- Record the trace for the constructed table.
> +get_n(anchor[#anchor])
> +
> +-- Check the result for the table that has the required key.
> +local result = get_n({n=1})
> +test:is(result, 1, 'correct value retrieved')
> +test:done(true)
> --
> 2.43.0
>
--
Best regards,
Sergey Kaplun
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization.
2024-01-15 15:22 ` Sergey Kaplun via Tarantool-patches
@ 2024-02-02 12:21 ` Maxim Kokryashkin via Tarantool-patches
2024-02-05 9:53 ` Sergey Kaplun via Tarantool-patches
0 siblings, 1 reply; 10+ messages in thread
From: Maxim Kokryashkin via Tarantool-patches @ 2024-02-02 12:21 UTC (permalink / raw)
To: Sergey Kaplun; +Cc: Maxim Kokryashkin, tarantool-patches
Hi, Sergey!
Thanks for the review!
Fixed your comments, branch is force-pushed.
Here is the diff with changes:
===
diff --git a/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua b/test/tarantool-tests/lj-840-gc64-fix-hrefk-optimization.test.lua
similarity index 67%
rename from test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
rename to test/tarantool-tests/lj-840-gc64-fix-hrefk-optimization.test.lua
index a11b91e3..49168bc4 100644
--- a/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
+++ b/test/tarantool-tests/lj-840-gc64-fix-hrefk-optimization.test.lua
@@ -4,7 +4,7 @@ local tap = require('tap')
-- in LuaJIT.
local ffi = require('ffi')
-local test = tap.test('lj-840-fix-hrefk-optimization'):skipcond({
+local test = tap.test('lj-840-gc64-fix-hrefk-optimization'):skipcond({
['Test requires GC64 mode enabled'] = not ffi.abi('gc64'),
['Test requires JIT enabled'] = not jit.status(),
})
@@ -12,10 +12,14 @@ test:plan(1)
local table_new = require('table.new')
+local SIZEOF_GCTAB = 64
+-- See `chunk2mem` in lj_alloc.c for details.
+local ALLOC_CHUNK_HDR = 16
+local GCTAB_FOOTPRINT = SIZEOF_GCTAB + ALLOC_CHUNK_HDR
-- Size of single hash node in bytes.
local NODE_SIZE = 24
--- Number of hash nodes to allocate on each iteration
--- based on the condition from `rec_idx_key`
+-- The maximum value that can be stored in a 16-bit `op2`
+-- field in HREFK IR.
local HASH_NODES = 65535
-- The vector of hash nodes should have a raw size of
-- `HASH_NODES * NODE_SIZE`, which is allocated in
@@ -23,16 +27,17 @@ local HASH_NODES = 65535
-- the LuaJIT allocator adds a bunch of small paddings
-- and aligns the required size to LJ_PAGESIZE, which is
-- 4096, so the actual allocated size includes alignment.
-local ALIGNMENT = 4096
+local LJ_PAGESIZE = 4096
-- The vector for hash nodes in the table is allocated based on
-- `hbits`, so it's actually got a size of 65536 nodes.
-local SINGLE_ITERATION_ALLOC = (HASH_NODES + 1) * NODE_SIZE + ALIGNMENT + 72
+local NODE_VECTOR_SIZE = (HASH_NODES + 1) * NODE_SIZE + LJ_PAGESIZE
+local SINGLE_ITERATION_ALLOC = NODE_VECTOR_SIZE + GCTAB_FOOTPRINT
-- We need to overflow the 32-bit distance to the global nilnode,
--- so we divide 2^32 by the SINGLE_ITERATION_ALLOC. There are a
--- bunch of non-table.new allocations already performed, so one
--- iteration is subtracted to account for them.
-local N_ITERATIONS = 0x100000000 / SINGLE_ITERATION_ALLOC - 1
--- Prevent anchor table from interfering with target table allocations.
+-- so we divide 2^32 by the SINGLE_ITERATION_ALLOC and ceil
+-- the result.
+local N_ITERATIONS = math.ceil((2 ^ 32) / SINGLE_ITERATION_ALLOC)
+-- Prevent anchor table from interfering with target
+-- table allocations.
local anchor = table.new(N_ITERATIONS, 0)
-- Construct table.
===
New commit message:
===
LJ_GC64: Fix HREFK optimization.
Contributed by XmiliaH.
(cherry-picked from commit 91bc6b8ad1f373c1ce9003dc024b2e21fad0e444)
In `lj_record_idx` when `ix->oldv` is the global nilnode and the
required key is not present in the table, it is possible to pass
the constant key lookup optimization condition because of the
`uint32_t` (`MSize`) overflow. Because of that, further recording
incorrectly removes the check for the nilnode, which produces
wrong results when trace is called for a different table.
The issue is solved by using `GCSize`, which has a size of
64 bits, instead of `MSize`.
Maxim Kokryashkin:
* added the description and the test for the problem
Part of tarantool/tarantool#9145
===
On Mon, Jan 15, 2024 at 06:22:29PM +0300, Sergey Kaplun via Tarantool-patches wrote:
> Hi, Maxim!
> Thanks for the patch!
> LGTM with a few comments below.
>
> On 12.01.24, Maxim Kokryashkin wrote:
> > From: Mike Pall <mike>
> >
> > Contributed by XmiliaH.
> >
> > (cherry-picked from commit 91bc6b8ad1f373c1ce9003dc024b2e21fad0e444)
> >
> > In `lj_record_idx` when `ix->oldv` is the global nilnode and the
> > required key is not present in the table, it is possible to pass
> > the constant key lookup optimization condition because of the
> > `uint32_t` overflow. Because of that, further recording
>
> I suggest clarifying like the following:
> | `uint32_t` (`MSize`)
> Feel free to ignore.
Fixed.
>
> > incorrectly removes the check for the nilnode, which produces
> > wrong results when trace is called for a different table.
>
> Nit: Please mention also how the problem is fixed.
Fixed.
>
> >
> > Maxim Kokryashkin:
> > * added the description and the test for the problem
> >
> > Part of tarantool/tarantool#9145
> > ---
> > Branch: https://github.com/tarantool/luajit/tree/fckxorg/lj-840-fix-hrefk-optimization
> > PR: https://github.com/tarantool/tarantool/pull/9591
> > Issues: https://github.com/LuaJIT/LuaJIT/issues/840
> > https://github.com/tarantool/tarantool/issues/9145
> >
> > src/lj_record.c | 8 +--
> > .../lj-840-fix-hrefk-optimization.test.lua | 58 +++++++++++++++++++
> > 2 files changed, 62 insertions(+), 4 deletions(-)
> > create mode 100644 test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
> >
> > diff --git a/src/lj_record.c b/src/lj_record.c
> > index a929b8aa..919e7169 100644
> > --- a/src/lj_record.c
> > +++ b/src/lj_record.c
>
> <snipped>
>
> > diff --git a/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua b/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
>
> Nit: We can also add gc64 prefix for this test like:
> <lj-840-gc64-fix-hrefk-optimization.test.lua>
> Feel free to ignore.
Fixed.
>
> > new file mode 100644
> > index 00000000..a11b91e3
> > --- /dev/null
> > +++ b/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
> > @@ -0,0 +1,58 @@
> > +local tap = require('tap')
> > +
> > +-- Test file to demonstrate incorrect HREFK optimization
> > +-- in LuaJIT.
> > +
> > +local ffi = require('ffi')
> > +local test = tap.test('lj-840-fix-hrefk-optimization'):skipcond({
> > + ['Test requires GC64 mode enabled'] = not ffi.abi('gc64'),
> > + ['Test requires JIT enabled'] = not jit.status(),
> > +})
> > +test:plan(1)
> > +
> > +local table_new = require('table.new')
> > +
> > +-- Size of single hash node in bytes.
> > +local NODE_SIZE = 24
> > +-- Number of hash nodes to allocate on each iteration
> > +-- based on the condition from `rec_idx_key`
>
> Nit: Missed dot at the end of the sentence.
>
> It is more correct to say that we use this is restricted by the IR
> format:
> `op2` field in the HREFK IR is a slot number and it is 16-bit wide.
> 65535 == 2^16 - 1; i.e., it is the maximum value that can be stored in a
> 16-bit field.
Fixed.
>
> > +local HASH_NODES = 65535
> > +-- The vector of hash nodes should have a raw size of
> > +-- `HASH_NODES * NODE_SIZE`, which is allocated in
> > +-- `lj_alloc_malloc` directly with `mmap`. However,
> > +-- the LuaJIT allocator adds a bunch of small paddings
> > +-- and aligns the required size to LJ_PAGESIZE, which is
> > +-- 4096, so the actual allocated size includes alignment.
> > +local ALIGNMENT = 4096
>
> Minor: So, maybe name it `LJ_PAGESIZE`?
> Feel free to ignore.
>
> > +-- The vector for hash nodes in the table is allocated based on
> > +-- `hbits`, so it's actually got a size of 65536 nodes.
> > +local SINGLE_ITERATION_ALLOC = (HASH_NODES + 1) * NODE_SIZE + ALIGNMENT + 72
>
> What is the magic number 72?
It's GCtab memory footprint, but I got it wrong and it's actually 80.
Fixed.
>
> > +-- We need to overflow the 32-bit distance to the global nilnode,
> > +-- so we divide 2^32 by the SINGLE_ITERATION_ALLOC. There are a
> > +-- bunch of non-table.new allocations already performed, so one
> > +-- iteration is subtracted to account for them.
>
> Why is it crucial to subtract it? What happens without it?
> I suppose that the new table will still be huge enough, won't it?
>
> > +local N_ITERATIONS = 0x100000000 / SINGLE_ITERATION_ALLOC - 1
I've got a mistake here too. There must be no subtraction, moreover, it
must be a ceil of the result. Fixed.
>
> Minor: We can use `(2 ^ 32)` instead of 0x100000000 (it is easier to
> read).
> Feel free to ignore.
Fixed.
>
> > +-- Prevent anchor table from interfering with target table allocations.
>
> Nit: Comment length is more than 66 symbols.
Fixed.
>
> > +local anchor = table.new(N_ITERATIONS, 0)
> > +
> > +-- Construct table.
> > +for _ = 1, N_ITERATIONS do
> > + table.insert(anchor, table_new(0, HASH_NODES))
> > +end
> > +
> > +jit.opt.start('hotloop=1')
> > +local function get_n(tab)
> > + local x
> > + for _ = 1, 4 do
> > + x = tab.n
> > + end
> > + return x
> > +end
> > +
> > +-- Record the trace for the constructed table.
> > +get_n(anchor[#anchor])
> > +
> > +-- Check the result for the table that has the required key.
> > +local result = get_n({n=1})
> > +test:is(result, 1, 'correct value retrieved')
> > +test:done(true)
> > --
> > 2.43.0
> >
>
> --
> Best regards,
> Sergey Kaplun
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization.
2024-02-02 12:21 ` Maxim Kokryashkin via Tarantool-patches
@ 2024-02-05 9:53 ` Sergey Kaplun via Tarantool-patches
2024-02-06 11:09 ` Maxim Kokryashkin via Tarantool-patches
0 siblings, 1 reply; 10+ messages in thread
From: Sergey Kaplun via Tarantool-patches @ 2024-02-05 9:53 UTC (permalink / raw)
To: Maxim Kokryashkin; +Cc: Maxim Kokryashkin, tarantool-patches
Hi, Maxim!
Thanks for the fixes!
LGTM, except a single nit below.
On 02.02.24, Maxim Kokryashkin wrote:
> Hi, Sergey!
> Thanks for the review!
> Fixed your comments, branch is force-pushed.
> Here is the diff with changes:
<snipped>
> New commit message:
> ===
> LJ_GC64: Fix HREFK optimization.
>
> Contributed by XmiliaH.
>
> (cherry-picked from commit 91bc6b8ad1f373c1ce9003dc024b2e21fad0e444)
>
> In `lj_record_idx` when `ix->oldv` is the global nilnode and the
> required key is not present in the table, it is possible to pass
> the constant key lookup optimization condition because of the
> `uint32_t` (`MSize`) overflow. Because of that, further recording
> incorrectly removes the check for the nilnode, which produces
> wrong results when trace is called for a different table.
>
> The issue is solved by using `GCSize`, which has a size of
Minor: I'd say 'which size is defines by the target system' or something
like that (to mention that it is 32 bit for non-GC64 build).
> 64 bits, instead of `MSize`.
>
> Maxim Kokryashkin:
> * added the description and the test for the problem
>
> Part of tarantool/tarantool#9145
> ===
<snipped>
> >
> > --
> > Best regards,
> > Sergey Kaplun
--
Best regards,
Sergey Kaplun
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization.
2024-02-05 9:53 ` Sergey Kaplun via Tarantool-patches
@ 2024-02-06 11:09 ` Maxim Kokryashkin via Tarantool-patches
0 siblings, 0 replies; 10+ messages in thread
From: Maxim Kokryashkin via Tarantool-patches @ 2024-02-06 11:09 UTC (permalink / raw)
To: Sergey Kaplun; +Cc: Maxim Kokryashkin, tarantool-patches
[-- Attachment #1: Type: text/plain, Size: 1625 bytes --]
Hi, Sergey!
Thanks for the review!
Now this line looks like the following:
| The issue is solved by using `GCSize`, which has a size of
| 64 bits on GC64 builds, instead of `MSize`.
--
Best regards,
Maxim Kokryashkin
>Понедельник, 5 февраля 2024, 12:57 +03:00 от Sergey Kaplun <skaplun@tarantool.org>:
>
>Hi, Maxim!
>Thanks for the fixes!
>LGTM, except a single nit below.
>
>On 02.02.24, Maxim Kokryashkin wrote:
>> Hi, Sergey!
>> Thanks for the review!
>> Fixed your comments, branch is force-pushed.
>> Here is the diff with changes:
>
><snipped>
>
>> New commit message:
>> ===
>> LJ_GC64: Fix HREFK optimization.
>>
>> Contributed by XmiliaH.
>>
>> (cherry-picked from commit 91bc6b8ad1f373c1ce9003dc024b2e21fad0e444)
>>
>> In `lj_record_idx` when `ix->oldv` is the global nilnode and the
>> required key is not present in the table, it is possible to pass
>> the constant key lookup optimization condition because of the
>> `uint32_t` (`MSize`) overflow. Because of that, further recording
>> incorrectly removes the check for the nilnode, which produces
>> wrong results when trace is called for a different table.
>>
>> The issue is solved by using `GCSize`, which has a size of
>
>Minor: I'd say 'which size is defines by the target system' or something
>like that (to mention that it is 32 bit for non-GC64 build).
>
>> 64 bits, instead of `MSize`.
>>
>> Maxim Kokryashkin:
>> * added the description and the test for the problem
>>
>> Part of tarantool/tarantool#9145
>> ===
>
><snipped>
>
>> >
>> > --
>> > Best regards,
>> > Sergey Kaplun
>
>--
>Best regards,
>Sergey Kaplun
[-- Attachment #2: Type: text/html, Size: 2342 bytes --]
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization.
2024-01-12 13:26 [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization Maxim Kokryashkin via Tarantool-patches
2024-01-15 15:22 ` Sergey Kaplun via Tarantool-patches
@ 2024-01-16 8:46 ` Sergey Bronnikov via Tarantool-patches
2024-01-16 12:09 ` Sergey Kaplun via Tarantool-patches
2024-01-16 12:54 ` Sergey Bronnikov via Tarantool-patches
2024-02-15 13:42 ` Igor Munkin via Tarantool-patches
3 siblings, 1 reply; 10+ messages in thread
From: Sergey Bronnikov via Tarantool-patches @ 2024-01-16 8:46 UTC (permalink / raw)
To: Maxim Kokryashkin, tarantool-patches, skaplun
Hi, Max
thanks for the patch!
test is passed after reverting the patch.
On the same build repro from the issue [1] has failed.
1. https://github.com/LuaJIT/LuaJIT/issues/840
Sergey
On 1/12/24 16:26, Maxim Kokryashkin wrote:
> From: Mike Pall <mike>
>
> Contributed by XmiliaH.
>
> (cherry-picked from commit 91bc6b8ad1f373c1ce9003dc024b2e21fad0e444)
>
> In `lj_record_idx` when `ix->oldv` is the global nilnode and the
> required key is not present in the table, it is possible to pass
> the constant key lookup optimization condition because of the
> `uint32_t` overflow. Because of that, further recording
> incorrectly removes the check for the nilnode, which produces
> wrong results when trace is called for a different table.
>
> Maxim Kokryashkin:
> * added the description and the test for the problem
>
> Part of tarantool/tarantool#9145
> ---
> Branch: https://github.com/tarantool/luajit/tree/fckxorg/lj-840-fix-hrefk-optimization
> PR: https://github.com/tarantool/tarantool/pull/9591
> Issues: https://github.com/LuaJIT/LuaJIT/issues/840
> https://github.com/tarantool/tarantool/issues/9145
>
> src/lj_record.c | 8 +--
> .../lj-840-fix-hrefk-optimization.test.lua | 58 +++++++++++++++++++
> 2 files changed, 62 insertions(+), 4 deletions(-)
> create mode 100644 test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
>
> diff --git a/src/lj_record.c b/src/lj_record.c
> index a929b8aa..919e7169 100644
> --- a/src/lj_record.c
> +++ b/src/lj_record.c
> @@ -1374,16 +1374,16 @@ static TRef rec_idx_key(jit_State *J, RecordIndex *ix, IRRef *rbref,
> key = emitir(IRTN(IR_CONV), key, IRCONV_NUM_INT);
> if (tref_isk(key)) {
> /* Optimize lookup of constant hash keys. */
> - MSize hslot = (MSize)((char *)ix->oldv - (char *)&noderef(t->node)[0].val);
> - if (t->hmask > 0 && hslot <= t->hmask*(MSize)sizeof(Node) &&
> - hslot <= 65535*(MSize)sizeof(Node)) {
> + GCSize hslot = (GCSize)((char *)ix->oldv-(char *)&noderef(t->node)[0].val);
> + if (hslot <= t->hmask*(GCSize)sizeof(Node) &&
> + hslot <= 65535*(GCSize)sizeof(Node)) {
> TRef node, kslot, hm;
> *rbref = J->cur.nins; /* Mark possible rollback point. */
> *rbguard = J->guardemit;
> hm = emitir(IRTI(IR_FLOAD), ix->tab, IRFL_TAB_HMASK);
> emitir(IRTGI(IR_EQ), hm, lj_ir_kint(J, (int32_t)t->hmask));
> node = emitir(IRT(IR_FLOAD, IRT_PGC), ix->tab, IRFL_TAB_NODE);
> - kslot = lj_ir_kslot(J, key, hslot / sizeof(Node));
> + kslot = lj_ir_kslot(J, key, (IRRef)(hslot / sizeof(Node)));
> return emitir(IRTG(IR_HREFK, IRT_PGC), node, kslot);
> }
> }
> diff --git a/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua b/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
> new file mode 100644
> index 00000000..a11b91e3
> --- /dev/null
> +++ b/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
> @@ -0,0 +1,58 @@
> +local tap = require('tap')
> +
> +-- Test file to demonstrate incorrect HREFK optimization
> +-- in LuaJIT.
> +
> +local ffi = require('ffi')
> +local test = tap.test('lj-840-fix-hrefk-optimization'):skipcond({
> + ['Test requires GC64 mode enabled'] = not ffi.abi('gc64'),
> + ['Test requires JIT enabled'] = not jit.status(),
> +})
> +test:plan(1)
> +
> +local table_new = require('table.new')
> +
> +-- Size of single hash node in bytes.
> +local NODE_SIZE = 24
> +-- Number of hash nodes to allocate on each iteration
> +-- based on the condition from `rec_idx_key`
> +local HASH_NODES = 65535
> +-- The vector of hash nodes should have a raw size of
> +-- `HASH_NODES * NODE_SIZE`, which is allocated in
> +-- `lj_alloc_malloc` directly with `mmap`. However,
> +-- the LuaJIT allocator adds a bunch of small paddings
> +-- and aligns the required size to LJ_PAGESIZE, which is
> +-- 4096, so the actual allocated size includes alignment.
> +local ALIGNMENT = 4096
> +-- The vector for hash nodes in the table is allocated based on
> +-- `hbits`, so it's actually got a size of 65536 nodes.
> +local SINGLE_ITERATION_ALLOC = (HASH_NODES + 1) * NODE_SIZE + ALIGNMENT + 72
> +-- We need to overflow the 32-bit distance to the global nilnode,
> +-- so we divide 2^32 by the SINGLE_ITERATION_ALLOC. There are a
> +-- bunch of non-table.new allocations already performed, so one
> +-- iteration is subtracted to account for them.
> +local N_ITERATIONS = 0x100000000 / SINGLE_ITERATION_ALLOC - 1
> +-- Prevent anchor table from interfering with target table allocations.
> +local anchor = table.new(N_ITERATIONS, 0)
> +
> +-- Construct table.
> +for _ = 1, N_ITERATIONS do
> + table.insert(anchor, table_new(0, HASH_NODES))
> +end
> +
> +jit.opt.start('hotloop=1')
> +local function get_n(tab)
> + local x
> + for _ = 1, 4 do
> + x = tab.n
> + end
> + return x
> +end
> +
> +-- Record the trace for the constructed table.
> +get_n(anchor[#anchor])
> +
> +-- Check the result for the table that has the required key.
> +local result = get_n({n=1})
> +test:is(result, 1, 'correct value retrieved')
> +test:done(true)
> --
> 2.43.0
>
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization.
2024-01-16 8:46 ` Sergey Bronnikov via Tarantool-patches
@ 2024-01-16 12:09 ` Sergey Kaplun via Tarantool-patches
2024-01-16 12:53 ` Sergey Bronnikov via Tarantool-patches
0 siblings, 1 reply; 10+ messages in thread
From: Sergey Kaplun via Tarantool-patches @ 2024-01-16 12:09 UTC (permalink / raw)
To: Sergey Bronnikov; +Cc: Maxim Kokryashkin, tarantool-patches
Hi, Sergey!
On 16.01.24, Sergey Bronnikov wrote:
> Hi, Max
>
>
> thanks for the patch!
>
> test is passed after reverting the patch.
Do you check the GC64 build?
*NB*: You should use GC64 *without* SYSMALLOC (and hence without ASAN),
since we need to use more predictible LuaJIT's dlmalloc.
>
> On the same build repro from the issue [1] has failed.
>
>
> 1. https://github.com/LuaJIT/LuaJIT/issues/840
>
> Sergey
--
Best regards,
Sergey Kaplun
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization.
2024-01-16 12:09 ` Sergey Kaplun via Tarantool-patches
@ 2024-01-16 12:53 ` Sergey Bronnikov via Tarantool-patches
0 siblings, 0 replies; 10+ messages in thread
From: Sergey Bronnikov via Tarantool-patches @ 2024-01-16 12:53 UTC (permalink / raw)
To: Sergey Kaplun; +Cc: Maxim Kokryashkin, tarantool-patches
On 1/16/24 15:09, Sergey Kaplun wrote:
> Hi, Sergey!
>
> On 16.01.24, Sergey Bronnikov wrote:
>> Hi, Max
>>
>>
>> thanks for the patch!
>>
>> test is passed after reverting the patch.
> Do you check the GC64 build?
> *NB*: You should use GC64 *without* SYSMALLOC (and hence without ASAN),
> since we need to use more predictible LuaJIT's dlmalloc.
Reproduced with:
cmake -S . -B build -DCMAKE_BUILD_TYPE=Debug -DLUA_USE_ASSERT=ON
-DLUA_USE_APICHECK=ON -DLUAJIT_ENABLE_GC64=ON
Sergey, thanks!
>> On the same build repro from the issue [1] has failed.
>>
>>
>> 1. https://github.com/LuaJIT/LuaJIT/issues/840
>>
>> Sergey
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization.
2024-01-12 13:26 [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization Maxim Kokryashkin via Tarantool-patches
2024-01-15 15:22 ` Sergey Kaplun via Tarantool-patches
2024-01-16 8:46 ` Sergey Bronnikov via Tarantool-patches
@ 2024-01-16 12:54 ` Sergey Bronnikov via Tarantool-patches
2024-02-15 13:42 ` Igor Munkin via Tarantool-patches
3 siblings, 0 replies; 10+ messages in thread
From: Sergey Bronnikov via Tarantool-patches @ 2024-01-16 12:54 UTC (permalink / raw)
To: Maxim Kokryashkin, tarantool-patches, skaplun
Max, thanks for the patch! LGTM
On 1/12/24 16:26, Maxim Kokryashkin wrote:
> From: Mike Pall <mike>
>
> Contributed by XmiliaH.
>
> (cherry-picked from commit 91bc6b8ad1f373c1ce9003dc024b2e21fad0e444)
>
> In `lj_record_idx` when `ix->oldv` is the global nilnode and the
> required key is not present in the table, it is possible to pass
> the constant key lookup optimization condition because of the
> `uint32_t` overflow. Because of that, further recording
> incorrectly removes the check for the nilnode, which produces
> wrong results when trace is called for a different table.
>
> Maxim Kokryashkin:
> * added the description and the test for the problem
>
> Part of tarantool/tarantool#9145
> ---
> Branch: https://github.com/tarantool/luajit/tree/fckxorg/lj-840-fix-hrefk-optimization
> PR: https://github.com/tarantool/tarantool/pull/9591
> Issues: https://github.com/LuaJIT/LuaJIT/issues/840
> https://github.com/tarantool/tarantool/issues/9145
>
> src/lj_record.c | 8 +--
> .../lj-840-fix-hrefk-optimization.test.lua | 58 +++++++++++++++++++
> 2 files changed, 62 insertions(+), 4 deletions(-)
> create mode 100644 test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
>
> diff --git a/src/lj_record.c b/src/lj_record.c
> index a929b8aa..919e7169 100644
> --- a/src/lj_record.c
> +++ b/src/lj_record.c
> @@ -1374,16 +1374,16 @@ static TRef rec_idx_key(jit_State *J, RecordIndex *ix, IRRef *rbref,
> key = emitir(IRTN(IR_CONV), key, IRCONV_NUM_INT);
> if (tref_isk(key)) {
> /* Optimize lookup of constant hash keys. */
> - MSize hslot = (MSize)((char *)ix->oldv - (char *)&noderef(t->node)[0].val);
> - if (t->hmask > 0 && hslot <= t->hmask*(MSize)sizeof(Node) &&
> - hslot <= 65535*(MSize)sizeof(Node)) {
> + GCSize hslot = (GCSize)((char *)ix->oldv-(char *)&noderef(t->node)[0].val);
> + if (hslot <= t->hmask*(GCSize)sizeof(Node) &&
> + hslot <= 65535*(GCSize)sizeof(Node)) {
> TRef node, kslot, hm;
> *rbref = J->cur.nins; /* Mark possible rollback point. */
> *rbguard = J->guardemit;
> hm = emitir(IRTI(IR_FLOAD), ix->tab, IRFL_TAB_HMASK);
> emitir(IRTGI(IR_EQ), hm, lj_ir_kint(J, (int32_t)t->hmask));
> node = emitir(IRT(IR_FLOAD, IRT_PGC), ix->tab, IRFL_TAB_NODE);
> - kslot = lj_ir_kslot(J, key, hslot / sizeof(Node));
> + kslot = lj_ir_kslot(J, key, (IRRef)(hslot / sizeof(Node)));
> return emitir(IRTG(IR_HREFK, IRT_PGC), node, kslot);
> }
> }
> diff --git a/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua b/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
> new file mode 100644
> index 00000000..a11b91e3
> --- /dev/null
> +++ b/test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
> @@ -0,0 +1,58 @@
> +local tap = require('tap')
> +
> +-- Test file to demonstrate incorrect HREFK optimization
> +-- in LuaJIT.
> +
> +local ffi = require('ffi')
> +local test = tap.test('lj-840-fix-hrefk-optimization'):skipcond({
> + ['Test requires GC64 mode enabled'] = not ffi.abi('gc64'),
> + ['Test requires JIT enabled'] = not jit.status(),
> +})
> +test:plan(1)
> +
> +local table_new = require('table.new')
> +
> +-- Size of single hash node in bytes.
> +local NODE_SIZE = 24
> +-- Number of hash nodes to allocate on each iteration
> +-- based on the condition from `rec_idx_key`
> +local HASH_NODES = 65535
> +-- The vector of hash nodes should have a raw size of
> +-- `HASH_NODES * NODE_SIZE`, which is allocated in
> +-- `lj_alloc_malloc` directly with `mmap`. However,
> +-- the LuaJIT allocator adds a bunch of small paddings
> +-- and aligns the required size to LJ_PAGESIZE, which is
> +-- 4096, so the actual allocated size includes alignment.
> +local ALIGNMENT = 4096
> +-- The vector for hash nodes in the table is allocated based on
> +-- `hbits`, so it's actually got a size of 65536 nodes.
> +local SINGLE_ITERATION_ALLOC = (HASH_NODES + 1) * NODE_SIZE + ALIGNMENT + 72
> +-- We need to overflow the 32-bit distance to the global nilnode,
> +-- so we divide 2^32 by the SINGLE_ITERATION_ALLOC. There are a
> +-- bunch of non-table.new allocations already performed, so one
> +-- iteration is subtracted to account for them.
> +local N_ITERATIONS = 0x100000000 / SINGLE_ITERATION_ALLOC - 1
> +-- Prevent anchor table from interfering with target table allocations.
> +local anchor = table.new(N_ITERATIONS, 0)
> +
> +-- Construct table.
> +for _ = 1, N_ITERATIONS do
> + table.insert(anchor, table_new(0, HASH_NODES))
> +end
> +
> +jit.opt.start('hotloop=1')
> +local function get_n(tab)
> + local x
> + for _ = 1, 4 do
> + x = tab.n
> + end
> + return x
> +end
> +
> +-- Record the trace for the constructed table.
> +get_n(anchor[#anchor])
> +
> +-- Check the result for the table that has the required key.
> +local result = get_n({n=1})
> +test:is(result, 1, 'correct value retrieved')
> +test:done(true)
> --
> 2.43.0
>
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization.
2024-01-12 13:26 [Tarantool-patches] [PATCH luajit] LJ_GC64: Fix HREFK optimization Maxim Kokryashkin via Tarantool-patches
` (2 preceding siblings ...)
2024-01-16 12:54 ` Sergey Bronnikov via Tarantool-patches
@ 2024-02-15 13:42 ` Igor Munkin via Tarantool-patches
3 siblings, 0 replies; 10+ messages in thread
From: Igor Munkin via Tarantool-patches @ 2024-02-15 13:42 UTC (permalink / raw)
To: Maxim Kokryashkin; +Cc: tarantool-patches
Max,
I've checked the patchset into all long-term branches in
tarantool/luajit and bumped a new version in master, release/3.0 and
release/2.11.
On 12.01.24, Maxim Kokryashkin via Tarantool-patches wrote:
> From: Mike Pall <mike>
>
> Contributed by XmiliaH.
>
> (cherry-picked from commit 91bc6b8ad1f373c1ce9003dc024b2e21fad0e444)
>
> In `lj_record_idx` when `ix->oldv` is the global nilnode and the
> required key is not present in the table, it is possible to pass
> the constant key lookup optimization condition because of the
> `uint32_t` overflow. Because of that, further recording
> incorrectly removes the check for the nilnode, which produces
> wrong results when trace is called for a different table.
>
> Maxim Kokryashkin:
> * added the description and the test for the problem
>
> Part of tarantool/tarantool#9145
> ---
> Branch: https://github.com/tarantool/luajit/tree/fckxorg/lj-840-fix-hrefk-optimization
> PR: https://github.com/tarantool/tarantool/pull/9591
> Issues: https://github.com/LuaJIT/LuaJIT/issues/840
> https://github.com/tarantool/tarantool/issues/9145
>
> src/lj_record.c | 8 +--
> .../lj-840-fix-hrefk-optimization.test.lua | 58 +++++++++++++++++++
> 2 files changed, 62 insertions(+), 4 deletions(-)
> create mode 100644 test/tarantool-tests/lj-840-fix-hrefk-optimization.test.lua
>
<snipped>
> --
> 2.43.0
>
--
Best regards,
IM
^ permalink raw reply [flat|nested] 10+ messages in thread