[llvm-dev] Problem with clang optimizer?

Sat Sep 25 17:48:19 PDT 2021

I found that 
The problem disappears with -mno-sse4.1
The problem manifests with both Apple Clang from Xcode-13, and LLVM Clang-12 (and not with Xcode-12 or LLVM Clang-11)
I could experiment only on Apple platform, as that’s the only one I have that runs LLVM Clang-12.

--

Regards,

Uri

There are two ways to design a system. One is to make is so simple there are obviously no deficiencies.

The other is to make it so complex there are no obvious deficiencies.

                                                                                                                                     -  C. A. R. Hoare

From: Craig Topper <craig.topper at gmail.com>
Date: Saturday, September 25, 2021 at 12:07
To: Dimitry Andric <dimitry at andric.com>
Cc: Uri Blumenthal <uri at ll.mit.edu>, LLVM-DEV LIST <llvm-dev at lists.llvm.org>
Subject: Re: [llvm-dev] Problem with clang optimizer?

It reproduced for me with -march=nehalem which does not have AVX.

On Sat, Sep 25, 2021 at 2:51 AM Dimitry Andric via llvm-dev <llvm-dev at lists.llvm.org> wrote:

It is only occurring (as far as I can see now) on x86_64, with -mavx enabled. Or with a target CPU that supports AVX. And it is not Apple clang specific.

-Dimitry

On 24 Sep 2021, at 15:30, Blumenthal, Uri - 0553 - MITLL via llvm-dev <llvm-dev at lists.llvm.org> wrote:

I tried to reproduce it on goldbolt with clang 12.0.0 and 12.0.1 but things seem fine when I run it there: https://godbolt.org/z/vrq8j6Kj7.
Can you share your exact clang invocation? Does it only reproduce in some specific environment?

Save the source I posted before into “sha3-reproducer.cxx” file. Let me know if you want it re-posted here.

$ clang++-mp-12 -v

clang version 12.0.1

Target: x86_64-apple-darwin20.6.0

Thread model: posix

InstalledDir: /opt/local/libexec/llvm-12/bin

$ clang++-mp-12 -o s -O3 sha3-reproducer.cxx

$ ./s

Assertion failed: (T[0] == 16394434931424703552u), function main, file sha3-reproducer.cxx, line 103.

Abort trap: 6

$ clang++-mp-12 -o s -O2 sha3-reproducer.cxx

$ ./s

Assertion failed: (T[0] == 16394434931424703552u), function main, file sha3-reproducer.cxx, line 103.

Abort trap: 6

$ clang++-mp-12 -o s -O1 sha3-reproducer.cxx

$ ./s

$ 

Clang-12 is installed via Macports, which is why we invoke the executable as clang++-mp-12.

The same problem manifests in exactly the same way in the Xcode-13 version of Clang (presumably based on LLVM Clang-12). 

I’ll be happy to provide more of specific details, if you let me know what you need.

Also, it generally helps to reduce code bug reports as much as possible; creduce can help with that: https://embed.cs.utah.edu/creduce/using/.

Understood. Unfortunately, the above reproducer is the best we could come up with. An alternative is trying to build the Botan package itself https://github.com/randombit/botan.git.

On Thu, Sep 23, 2021 at 10:14 PM Blumenthal, Uri - 0553 - MITLL via llvm-dev <llvm-dev at lists.llvm.org> wrote:

I’m not sure if this is the correct list, so please direct me to the right one if this bug report shouldn’t go here.

The problem is: invoking clang (v12) with -O2 or better optimization flags generates wrong object code for the following C++. Compiling it with -O1 generates working binary.

=================

#include <cstdint>

#include <cassert>

template<size_t ROT, typename T>

inline constexpr T rotl(T input)

   {

   static_assert(ROT > 0 && ROT < 8*sizeof(T), "Invalid rotation constant");

   return static_cast<T>((input << ROT) | (input >> (8*sizeof(T) - ROT)));

   }

inline void SHA3_round(uint64_t T[25], const uint64_t A[25], uint64_t RC)

   {

   const uint64_t C0 = A[0] ^ A[5] ^ A[10] ^ A[15] ^ A[20];

   const uint64_t C1 = A[1] ^ A[6] ^ A[11] ^ A[16] ^ A[21];

   // the calculation of C2 fails for -O3 or -O2 with clang 12

   // FWIW: it would produce a value that doesn't fit into a _signed_ 64-bit int

   const uint64_t C2 = A[2] ^ A[7] ^ A[12] ^ A[17] ^ A[22];

   const uint64_t C3 = A[3] ^ A[8] ^ A[13] ^ A[18] ^ A[23];

   const uint64_t C4 = A[4] ^ A[9] ^ A[14] ^ A[19] ^ A[24];

   const uint64_t D0 = rotl<1>(C0) ^ C3;

   const uint64_t D1 = rotl<1>(C1) ^ C4;

   const uint64_t D2 = rotl<1>(C2) ^ C0;

   const uint64_t D3 = rotl<1>(C3) ^ C1;

   const uint64_t D4 = rotl<1>(C4) ^ C2;

   const uint64_t B00 =          A[ 0] ^ D1;

   const uint64_t B01 = rotl<44>(A[ 6] ^ D2);

   const uint64_t B02 = rotl<43>(A[12] ^ D3);

   const uint64_t B03 = rotl<21>(A[18] ^ D4);

   const uint64_t B04 = rotl<14>(A[24] ^ D0);

   T[ 0] = B00 ^ (~B01 & B02) ^ RC;

   T[ 1] = B01 ^ (~B02 & B03);

   T[ 2] = B02 ^ (~B03 & B04);

   T[ 3] = B03 ^ (~B04 & B00);

   T[ 4] = B04 ^ (~B00 & B01);

   const uint64_t B05 = rotl<28>(A[ 3] ^ D4);

   const uint64_t B06 = rotl<20>(A[ 9] ^ D0);

   const uint64_t B07 = rotl< 3>(A[10] ^ D1);

   const uint64_t B08 = rotl<45>(A[16] ^ D2);

   const uint64_t B09 = rotl<61>(A[22] ^ D3);

   T[ 5] = B05 ^ (~B06 & B07);

   T[ 6] = B06 ^ (~B07 & B08);

   T[ 7] = B07 ^ (~B08 & B09);

   T[ 8] = B08 ^ (~B09 & B05);

   T[ 9] = B09 ^ (~B05 & B06);

   // --- instructions starting from here can be removed

   //     and the -O3 dicrepancy is still triggered

   const uint64_t B10 = rotl< 1>(A[ 1] ^ D2);

   const uint64_t B11 = rotl< 6>(A[ 7] ^ D3);

   const uint64_t B12 = rotl<25>(A[13] ^ D4);

   const uint64_t B13 = rotl< 8>(A[19] ^ D0);

   const uint64_t B14 = rotl<18>(A[20] ^ D1);

   T[10] = B10 ^ (~B11 & B12);

   T[11] = B11 ^ (~B12 & B13);

   T[12] = B12 ^ (~B13 & B14);

   T[13] = B13 ^ (~B14 & B10);

   T[14] = B14 ^ (~B10 & B11);

   const uint64_t B15 = rotl<27>(A[ 4] ^ D0);

   const uint64_t B16 = rotl<36>(A[ 5] ^ D1);

   const uint64_t B17 = rotl<10>(A[11] ^ D2);

   const uint64_t B18 = rotl<15>(A[17] ^ D3);

   const uint64_t B19 = rotl<56>(A[23] ^ D4);

   T[15] = B15 ^ (~B16 & B17);

   T[16] = B16 ^ (~B17 & B18);

   T[17] = B17 ^ (~B18 & B19);

   T[18] = B18 ^ (~B19 & B15);

   T[19] = B19 ^ (~B15 & B16);

   const uint64_t B20 = rotl<62>(A[ 2] ^ D3);

   const uint64_t B21 = rotl<55>(A[ 8] ^ D4);

   const uint64_t B22 = rotl<39>(A[14] ^ D0);

   const uint64_t B23 = rotl<41>(A[15] ^ D1);

   const uint64_t B24 = rotl< 2>(A[21] ^ D2);

   T[20] = B20 ^ (~B21 & B22);

   T[21] = B21 ^ (~B22 & B23);

   T[22] = B22 ^ (~B23 & B24);

   T[23] = B23 ^ (~B24 & B20);

   T[24] = B24 ^ (~B20 & B21);

   }

int main()

{

    uint64_t T[25];

    uint64_t A[25] = {

        15515230172486u, 9751542238472685244u, 220181482233372672u,

        2303197730119u, 9537012007446913720u, 0u, 14782389640143539577u,

        2305843009213693952u, 1056340403235818873u, 16396894922196123648u,

        13438274300558u, 3440198220943040u, 0u, 3435902021559310u, 64u,

        14313837075027532897u, 32768u, 6880396441885696u, 14320469711924527201u,

        0u, 9814829303127743595u, 18014398509481984u, 14444556046857390455u,

        4611686018427387904u, 18041275058083100u };

    SHA3_round(T, A, 0x0000000000008082);

    assert(T[0]  == 16394434931424703552u);

    assert(T[1]  == 10202638136074191489u);

    assert(T[2]  == 6432602484395933614u);

    assert(T[3]  == 10616058301262943899u);

    assert(T[4]  == 14391824303596635982u);

    assert(T[5]  == 5673590995284149638u);

    assert(T[6]  == 15681872423764765508u);

    assert(T[7]  == 11470206704342013341u);

    assert(T[8]  == 8508807405493883168u);

    assert(T[9]  == 9461805213344568570u);

    assert(T[10] == 8792313850970105187u);

    assert(T[11] == 13508586629627657374u);

    assert(T[12] == 5157283382205130943u);

    assert(T[13] == 375019647457809685u);

    assert(T[14] == 9294608398083155963u);

    assert(T[15] == 16923121173371064314u);

    assert(T[16] == 4737739424553008030u);

    assert(T[17] == 5823987023293412593u);

    assert(T[18] == 13908063749137376267u);

    assert(T[19] == 13781177305593198238u);

    assert(T[20] == 9673833001659673401u);

    assert(T[21] == 17282395057630454440u);

    assert(T[22] == 12906624984756985556u);

    assert(T[23] == 3081478361927354234u);

    assert(T[24] == 93297594635310132u);

    return 0;

}

=================

Your help debugging and fixing this problem is appreciated!

--

Regards,

Uri Blumenthal                              Voice: (781) 981-1638 

Secure Resilient Systems and Technologies   Cell:  (339) 223-5363

MIT Lincoln Laboratory                      

244 Wood Street, Lexington, MA  02420-9108      

Web:     https://www.ll.mit.edu/biographies/uri-blumenthal

Root CA: https://www.ll.mit.edu/llrca2.pem

There are two ways to design a system. One is to make is so simple there are obviously no deficiencies.

The other is to make it so complex there are no obvious deficiencies.

                                                                                                                                     -  C. A. R. Hoare

_______________________________________________
LLVM Developers mailing list
llvm-dev at lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev

-- 

Jakub Kuderski

_______________________________________________
LLVM Developers mailing list
llvm-dev at lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev

_______________________________________________
LLVM Developers mailing list
llvm-dev at lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev

-- 

~Craig

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20210926/af324929/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5249 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20210926/af324929/attachment.bin>