checker,transformer: add always true false branch detect

[kbkpbot]

feature request by #23404
This will warn on always true cond, such as a == a, and always false cond, such as 1==2.

Fix transformer support ast.CharLiteral expr, support general var such as struct.f != struct.f

[kbkpbot]

CI fail is because in vtcc/src/i386-asm.v, there are too many stmts like:

        ASMInstr{
                sym:        u16(Tcc_token.tok_asm_shldw)
                opcode:     (u16((if (((4005) & 65280) == 3840) {
                        ((((4005) >> 8) & ~255) | ((4005) & 255))
                } else {
                        (4005)
                })))
                instr_type: u16(((8 | 4096) | ((0) << 13) | (if (((4005) & 65280) == 3840) { 256 } else { 0 })))
                nb_ops:     3
                op_type:    [u8(opt_cl), opt_regw, opt_ea | opt_regw]!
        },

and if (((4005) & 65280) == 3840) is always a constant expr and will eval to true/false.
This will cause the total notice number exceed the limit:

builder error: too many errors/warnings/notices

[spytheman]

What is the performance impact of the additional checks?
Can you please run v run .github/workflows/compare_pr_to_master.v ?

[kbkpbot]

What is the performance impact of the additional checks? Can you please run v run .github/workflows/compare_pr_to_master.v ?

It need more optimization. :)

 >  1           base            114.4ms ± σ:    0.6ms,  113.6ms… 115.1ms `./vold -no-parallel -o ohw.exe examples/hello_world.v`
    2     +7.6%   1.08x slower  123.1ms ± σ:    0.5ms,  122.4ms… 123.6ms `./vnew -no-parallel -o nhw.exe examples/hello_world.v`
>>>>>> size("          nhw.exe") - size("          ohw.exe") =     240697 -     240697 =          0
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 1/3, took: 13079.494 ms
 >  1           base            536.1ms ± σ:    1.5ms,  534.0ms… 537.2ms `./vold -check-syntax           cmd/v`
    2     +1.0%   1.01x slower  541.3ms ± σ:    2.4ms,  538.7ms… 544.5ms `./vnew -check-syntax           cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 2/3, took: 13119.916 ms
 >  1           base            537.4ms ± σ:    1.7ms,  535.0ms… 539.0ms `./vold -check-syntax           cmd/v`
    2     +1.8%   1.02x slower  547.0ms ± σ:    1.5ms,  545.9ms… 549.1ms `./vnew -check-syntax           cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 3/3, took: 13135.491 ms
    1     -0.1%   1.00x ~same~  542.6ms ± σ:    2.7ms,  538.8ms… 545.1ms `./vnew -check-syntax           cmd/v`
 >  2           base            542.9ms ± σ:    0.9ms,  541.6ms… 543.8ms `./vold -check-syntax           cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 1/3, took: 22293.612 ms
 >  1           base            878.6ms ± σ:    2.3ms,  875.9ms… 881.4ms `./vold -check                  cmd/v`
    2     +9.2%   1.09x slower  959.3ms ± σ:   12.2ms,  942.2ms… 969.6ms `./vnew -check                  cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 2/3, took: 22410.240 ms
 >  1           base            885.1ms ± σ:    2.7ms,  881.6ms… 888.3ms `./vold -check                  cmd/v`
    2     +9.4%   1.09x slower  968.0ms ± σ:    2.6ms,  964.4ms… 970.4ms `./vnew -check                  cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 3/3, took: 22377.652 ms
 >  1           base            882.5ms ± σ:    1.8ms,  880.6ms… 885.0ms `./vold -check                  cmd/v`
    2     +9.7%   1.10x slower  967.8ms ± σ:    0.7ms,  966.9ms… 968.6ms `./vnew -check                  cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 1/3, took: 34632.271 ms
 >  1           base           1381.1ms ± σ:    1.8ms, 1378.8ms…1383.2ms `./vold -no-parallel -o ov.c    cmd/v`
    2     +7.6%   1.08x slower 1485.8ms ± σ:    2.7ms, 1482.2ms…1488.7ms `./vnew -no-parallel -o nv.c    cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 2/3, took: 34627.267 ms
 >  1           base           1378.3ms ± σ:    3.3ms, 1373.7ms…1381.2ms `./vold -no-parallel -o ov.c    cmd/v`
    2     +7.5%   1.08x slower 1481.9ms ± σ:    5.1ms, 1474.6ms…1486.2ms `./vnew -no-parallel -o nv.c    cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 3/3, took: 34602.570 ms
 >  1           base           1382.0ms ± σ:    2.6ms, 1378.7ms…1385.1ms `./vold -no-parallel -o ov.c    cmd/v`
    2     +7.5%   1.08x slower 1485.9ms ± σ:    3.2ms, 1481.4ms…1489.0ms `./vnew -no-parallel -o nv.c    cmd/v`
>>>>>> size("             nv.c") - size("             ov.c") =    6778937 -    6779002 =        -65
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 1/3, took: 40363.023 ms
 >  1           base           1616.9ms ± σ:    7.6ms, 1606.9ms…1625.4ms `./vold -no-parallel -o ov.exe  cmd/v`
    2     +6.8%   1.07x slower 1726.1ms ± σ:    2.7ms, 1722.4ms…1728.9ms `./vnew -no-parallel -o nv.exe  cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 2/3, took: 40266.581 ms
 >  1           base           1617.5ms ± σ:    7.7ms, 1606.7ms…1623.7ms `./vold -no-parallel -o ov.exe  cmd/v`
    2     +5.9%   1.06x slower 1713.2ms ± σ:   10.6ms, 1698.3ms…1722.6ms `./vnew -no-parallel -o nv.exe  cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 3/3, took: 40190.929 ms
 >  1           base           1616.3ms ± σ:    1.6ms, 1614.0ms…1617.8ms `./vold -no-parallel -o ov.exe  cmd/v`
    2     +6.3%   1.06x slower 1718.7ms ± σ:    1.6ms, 1716.5ms…1720.1ms `./vnew -no-parallel -o nv.exe  cmd/v`

[kbkpbot]

Is this acceptable ?

Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 3/3, took: 2784.369 ms
 >  1           base            113.6ms ± σ:    0.3ms,  113.2ms… 113.9ms `./vold -no-parallel -o ohw.exe examples/hello_world.v`
    2     +2.4%   1.02x slower  116.3ms ± σ:    0.1ms,  116.2ms… 116.4ms `./vnew -no-parallel -o nhw.exe examples/hello_world.v`
>>>>>> size("          nhw.exe") - size("          ohw.exe") =     240697 -     240697 =          0
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 1/3, took: 13052.584 ms
 >  1           base            538.7ms ± σ:    2.1ms,  535.7ms… 540.6ms `./vold -check-syntax           cmd/v`
    2     +0.3%   1.00x slower  540.1ms ± σ:    0.8ms,  539.0ms… 540.8ms `./vnew -check-syntax           cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 2/3, took: 13080.751 ms
 >  1           base            540.9ms ± σ:    5.3ms,  533.4ms… 544.9ms `./vold -check-syntax           cmd/v`
    2     +0.1%   1.00x ~same~  541.5ms ± σ:    0.4ms,  540.9ms… 541.9ms `./vnew -check-syntax           cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 3/3, took: 13073.723 ms
 >  1           base            537.6ms ± σ:    5.5ms,  531.2ms… 544.5ms `./vold -check-syntax           cmd/v`
    2     +0.5%   1.01x slower  540.6ms ± σ:    2.1ms,  537.7ms… 542.1ms `./vnew -check-syntax           cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 1/3, took: 21559.154 ms
 >  1           base            879.6ms ± σ:    3.4ms,  875.6ms… 883.9ms `./vold -check                  cmd/v`
    2     +2.8%   1.03x slower  904.2ms ± σ:    0.4ms,  903.8ms… 904.7ms `./vnew -check                  cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 2/3, took: 21655.860 ms
 >  1           base            886.1ms ± σ:    1.7ms,  884.7ms… 888.5ms `./vold -check                  cmd/v`
    2     +2.4%   1.02x slower  907.6ms ± σ:    0.2ms,  907.4ms… 907.9ms `./vnew -check                  cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 3/3, took: 21585.384 ms
 >  1           base            880.4ms ± σ:    2.1ms,  878.4ms… 883.4ms `./vold -check                  cmd/v`
    2     +1.8%   1.02x slower  896.7ms ± σ:    7.3ms,  886.4ms… 902.1ms `./vnew -check                  cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 1/3, took: 33666.340 ms
 >  1           base           1379.6ms ± σ:    2.2ms, 1376.7ms…1381.9ms `./vold -no-parallel -o ov.c    cmd/v`
    2     +1.9%   1.02x slower 1406.5ms ± σ:    2.5ms, 1402.9ms…1408.8ms `./vnew -no-parallel -o nv.c    cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 2/3, took: 33742.275 ms
 >  1           base           1378.8ms ± σ:    0.9ms, 1378.0ms…1380.0ms `./vold -no-parallel -o ov.c    cmd/v`
    2     +2.5%   1.03x slower 1413.4ms ± σ:    5.6ms, 1406.4ms…1420.1ms `./vnew -no-parallel -o nv.c    cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 3/3, took: 33671.153 ms
 >  1           base           1381.3ms ± σ:    1.7ms, 1379.3ms…1383.5ms `./vold -no-parallel -o ov.c    cmd/v`
    2     +1.5%   1.02x slower 1402.1ms ± σ:    3.7ms, 1396.9ms…1405.7ms `./vnew -no-parallel -o nv.c    cmd/v`
>>>>>> size("             nv.c") - size("             ov.c") =    6778808 -    6778873 =        -65
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 1/3, took: 39562.733 ms
 >  1           base           1619.1ms ± σ:    2.3ms, 1617.0ms…1622.2ms `./vold -no-parallel -o ov.exe  cmd/v`
    2     +1.8%   1.02x slower 1648.9ms ± σ:    5.8ms, 1640.7ms…1654.0ms `./vnew -no-parallel -o nv.exe  cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 2/3, took: 39361.583 ms
 >  1           base           1620.9ms ± σ:    2.8ms, 1618.7ms…1624.8ms `./vold -no-parallel -o ov.exe  cmd/v`
    2     +1.3%   1.01x slower 1641.8ms ± σ:    5.9ms, 1633.4ms…1646.1ms `./vnew -no-parallel -o nv.exe  cmd/v`
Summary after 1 series x 10 runs (%s are relative to first command, or `base`), discard maxs:  7, repeat: 3/3, took: 39381.921 ms
 >  1           base           1617.8ms ± σ:    1.3ms, 1616.6ms…1619.6ms `./vold -no-parallel -o ov.exe  cmd/v`
    2     +1.7%   1.02x slower 1645.5ms ± σ:    2.2ms, 1642.8ms…1648.2ms `./vnew -no-parallel -o nv.exe  cmd/v`

[spytheman]

Is this acceptable ?

Yes, 1-2% hit is a lot more palatable.
Thank you.

[spytheman]

Excellent work.
Thank you @kbkpbot 🙇🏻‍♂️ .