diff options
author | Kartik Agaram <vc@akkartik.com> | 2021-01-09 18:55:24 -0800 |
---|---|---|
committer | Kartik Agaram <vc@akkartik.com> | 2021-01-09 18:55:24 -0800 |
commit | 57e4978eac1e5deb544b07db77ac382433d35d03 (patch) | |
tree | be0bd25ed1bbd7bc5a7d1d8cda114d904c13f631 /baremetal | |
parent | 0518944e379f343542c872a20d3d5a2aee744297 (diff) | |
download | mu-57e4978eac1e5deb544b07db77ac382433d35d03.tar.gz |
7492
Port some support for unicode to baremetal.
Diffstat (limited to 'baremetal')
-rw-r--r-- | baremetal/107trace.subx | 87 | ||||
-rw-r--r-- | baremetal/108write.subx | 53 | ||||
-rw-r--r-- | baremetal/112read-byte.subx | 33 | ||||
-rw-r--r-- | baremetal/115write-byte.subx | 37 | ||||
-rw-r--r-- | baremetal/120allocate.subx | 52 | ||||
-rw-r--r-- | baremetal/302stack_allocate.subx | 61 | ||||
-rw-r--r-- | baremetal/309stream.subx | 200 | ||||
-rw-r--r-- | baremetal/400.mu | 5 | ||||
-rw-r--r-- | baremetal/401draw-text-rightward.mu | 19 | ||||
-rw-r--r-- | baremetal/403unicode.mu | 193 |
10 files changed, 729 insertions, 11 deletions
diff --git a/baremetal/107trace.subx b/baremetal/107trace.subx new file mode 100644 index 00000000..cdbdffc5 --- /dev/null +++ b/baremetal/107trace.subx @@ -0,0 +1,87 @@ +# instruction effective address register displacement immediate +# . op subop mod rm32 base index scale r32 +# . 1-3 bytes 3 bits 2 bits 3 bits 3 bits 3 bits 2 bits 2 bits 0/1/2/4 bytes 0/1/2/4 bytes + +# 3-argument variant of _append +_append-3: # out: (addr byte), outend: (addr byte), s: (addr array byte) -> num_bytes_appended/eax + # . prologue + 55/push-ebp + 89/copy 3/mod/direct 5/rm32/ebp . . . 4/r32/esp . . # copy esp to ebp + # . save registers + 51/push-ecx + # eax = _append-4(out, outend, &s->data[0], &s->data[s->size]) + # . . push &s->data[s->size] + 8b/copy 1/mod/*+disp8 5/rm32/ebp . . 0/r32/eax 0x10/disp8 . # copy *(ebp+16) to eax + 8b/copy 0/mod/indirect 0/rm32/eax . . . 1/r32/ecx . . # copy *eax to ecx + 8d/copy-address 1/mod/*+disp8 4/rm32/sib 0/base/eax 1/index/ecx . 1/r32/ecx 4/disp8 . # copy eax+ecx+4 to ecx + 51/push-ecx + # . . push &s->data[0] + 8d/copy-address 1/mod/*+disp8 0/rm32/eax . . . 1/r32/ecx 4/disp8 . # copy eax+4 to ecx + 51/push-ecx + # . . push outend + ff 6/subop/push 1/mod/*+disp8 5/rm32/ebp . . . . 0xc/disp8 . # push *(ebp+12) + # . . push out + ff 6/subop/push 1/mod/*+disp8 5/rm32/ebp . . . . 8/disp8 . # push *(ebp+8) + # . . call + e8/call _append-4/disp32 + # . . discard args + 81 0/subop/add 3/mod/direct 4/rm32/esp . . . . . 0x10/imm32 # add to esp +$_append-3:end: + # . restore registers + 59/pop-to-ecx + # . epilogue + 89/copy 3/mod/direct 4/rm32/esp . . . 5/r32/ebp . . # copy ebp to esp + 5d/pop-to-ebp + c3/return + +# 4-argument variant of _append +_append-4: # out: (addr byte), outend: (addr byte), in: (addr byte), inend: (addr byte) -> num_bytes_appended/eax: int + # . prologue + 55/push-ebp + 89/copy 3/mod/direct 5/rm32/ebp . . . 4/r32/esp . . # copy esp to ebp + # . save registers + 51/push-ecx + 52/push-edx + 53/push-ebx + 56/push-esi + 57/push-edi + # num_bytes_appended = 0 + b8/copy-to-eax 0/imm32 + # edi = out + 8b/copy 1/mod/*+disp8 5/rm32/ebp . . . 7/r32/edi 8/disp8 . # copy *(ebp+8) to edi + # edx = outend + 8b/copy 1/mod/*+disp8 5/rm32/ebp . . . 2/r32/edx 0xc/disp8 . # copy *(ebp+12) to edx + # esi = in + 8b/copy 1/mod/*+disp8 5/rm32/ebp . . . 6/r32/esi 0x10/disp8 . # copy *(ebp+16) to esi + # ecx = inend + 8b/copy 1/mod/*+disp8 5/rm32/ebp . . . 1/r32/ecx 0x14/disp8 . # copy *(ebp+20) to ecx +$_append-4:loop: + # if (in >= inend) break + 39/compare 3/mod/direct 6/rm32/esi . . . 1/r32/ecx . . # compare esi with ecx + 73/jump-if-addr>= $_append-4:end/disp8 + # if (out >= outend) abort # just to catch test failures fast + 39/compare 3/mod/direct 7/rm32/edi . . . 2/r32/edx . . # compare edi with edx + 73/jump-if-addr>= $_append-4:end/disp8 # TODO: abort + # *out = *in + 8a/copy-byte 0/mod/indirect 6/rm32/esi . . . 3/r32/BL . . # copy byte at *esi to BL + 88/copy-byte 0/mod/indirect 7/rm32/edi . . . 3/r32/BL . . # copy byte at BL to *edi + # ++num_bytes_appended + 40/increment-eax + # ++in + 46/increment-esi + # ++out + 47/increment-edi + eb/jump $_append-4:loop/disp8 +$_append-4:end: + # . restore registers + 5f/pop-to-edi + 5e/pop-to-esi + 5b/pop-to-ebx + 5a/pop-to-edx + 59/pop-to-ecx + # . epilogue + 89/copy 3/mod/direct 4/rm32/esp . . . 5/r32/ebp . . # copy ebp to esp + 5d/pop-to-ebp + c3/return + +# . . vim:nowrap:textwidth=0 diff --git a/baremetal/108write.subx b/baremetal/108write.subx new file mode 100644 index 00000000..daf1f535 --- /dev/null +++ b/baremetal/108write.subx @@ -0,0 +1,53 @@ +# instruction effective address register displacement immediate +# . op subop mod rm32 base index scale r32 +# . 1-3 bytes 3 bits 2 bits 3 bits 3 bits 3 bits 2 bits 2 bits 0/1/2/4 bytes 0/1/2/4 bytes + +write: # f: (addr stream byte), s: (addr array byte) + # . prologue + 55/push-ebp + 89/copy 3/mod/direct 5/rm32/ebp . . . 4/r32/esp . . # copy esp to ebp + # if (s == 0) return + 81 7/subop/compare 1/mod/*+disp8 5/rm32/ebp . . . . 0xc/disp8 0/imm32 # compare *(ebp+12) + 74/jump-if-= $write:end/disp8 + # TODO: write to file + # otherwise, treat 'f' as a stream to append to + # . save registers + 50/push-eax + 51/push-ecx + 52/push-edx + 53/push-ebx + # ecx = f + 8b/copy 1/mod/*+disp8 5/rm32/ebp . . 1/r32/ecx 8/disp8 . # copy *(ebp+8) to ecx + # edx = f->write + 8b/copy 0/mod/indirect 1/rm32/ecx . . . 2/r32/edx . . # copy *ecx to edx + # ebx = f->size + 8b/copy 1/mod/*+disp8 1/rm32/ecx . . . 3/r32/ebx 8/disp8 . # copy *(ecx+8) to ebx + # eax = _append-3(&f->data[f->write], &f->data[f->size], s) + # . . push s + ff 6/subop/push 1/mod/*+disp8 5/rm32/ebp . . . . 0xc/disp8 . # push *(ebp+12) + # . . push &f->data[f->size] + 8d/copy-address 1/mod/*+disp8 4/rm32/sib 1/base/ecx 3/index/ebx . 3/r32/ebx 0xc/disp8 . # copy ecx+ebx+12 to ebx + 53/push-ebx + # . . push &f->data[f->write] + 8d/copy-address 1/mod/*+disp8 4/rm32/sib 1/base/ecx 2/index/edx . 3/r32/ebx 0xc/disp8 . # copy ecx+edx+12 to ebx + 53/push-ebx + # . . call + e8/call _append-3/disp32 + # . . discard args + 81 0/subop/add 3/mod/direct 4/rm32/esp . . . . . 0xc/imm32 # add to esp + # f->write += eax + 01/add 0/mod/indirect 1/rm32/ecx . . . 0/r32/eax . . # add eax to *ecx + # . restore registers + 5b/pop-to-ebx + 5a/pop-to-edx + 59/pop-to-ecx + 58/pop-to-eax +$write:end: + # . epilogue + 89/copy 3/mod/direct 4/rm32/esp . . . 5/r32/ebp . . # copy ebp to esp + 5d/pop-to-ebp + c3/return + +# TODO: bring in tests once we have check-ints-equal + +# . . vim:nowrap:textwidth=0 diff --git a/baremetal/112read-byte.subx b/baremetal/112read-byte.subx new file mode 100644 index 00000000..e8731044 --- /dev/null +++ b/baremetal/112read-byte.subx @@ -0,0 +1,33 @@ +# TODO: read-byte-buffered + +# Return next byte value in eax, with top 3 bytes cleared. +# Abort on reaching end of stream. +read-byte: # s: (addr stream byte) -> result/eax: byte + # . prologue + 55/push-ebp + 89/copy 3/mod/direct 5/rm32/ebp . . . 4/r32/esp . . # copy esp to ebp + # . save registers + 51/push-ecx + 56/push-esi + # esi = s + 8b/copy 1/mod/*+disp8 5/rm32/ebp . . . 6/r32/esi 8/disp8 . # copy *(ebp+8) to esi + # ecx = s->read + 8b/copy 1/mod/*+disp8 6/rm32/esi . . . 1/r32/ecx 4/disp8 . # copy *(esi+4) to ecx + # if (f->read >= f->write) abort + 3b/compare 0/mod/indirect 6/rm32/esi . . . 1/r32/ecx . . # compare ecx with *esi + 0f 8d/jump-if->= $read-byte:end/disp32 # TODO: abort + # result = f->data[f->read] + 31/xor 3/mod/direct 0/rm32/eax . . . 0/r32/eax . . # clear eax + 8a/copy-byte 1/mod/*+disp8 4/rm32/sib 6/base/esi 1/index/ecx . 0/r32/AL 0xc/disp8 . # copy byte at *(esi+ecx+12) to AL + # ++f->read + ff 0/subop/increment 1/mod/*+disp8 6/rm32/esi . . . . 4/disp8 . # increment *(esi+4) +$read-byte:end: + # . restore registers + 5e/pop-to-esi + 59/pop-to-ecx + # . epilogue + 89/copy 3/mod/direct 4/rm32/esp . . . 5/r32/ebp . . # copy ebp to esp + 5d/pop-to-ebp + c3/return + +# . . vim:nowrap:textwidth=0 diff --git a/baremetal/115write-byte.subx b/baremetal/115write-byte.subx new file mode 100644 index 00000000..32245116 --- /dev/null +++ b/baremetal/115write-byte.subx @@ -0,0 +1,37 @@ +# instruction effective address register displacement immediate +# . op subop mod rm32 base index scale r32 +# . 1-3 bytes 3 bits 2 bits 3 bits 3 bits 3 bits 2 bits 2 bits 0/1/2/4 bytes 0/1/2/4 bytes + +# Write lower byte of 'n' to 'f'. +append-byte: # f: (addr stream byte), n: int + # . prologue + 55/push-ebp + 89/copy 3/mod/direct 5/rm32/ebp . . . 4/r32/esp . . # copy esp to ebp + # . save registers + 51/push-ecx + 57/push-edi + # edi = f + 8b/copy 1/mod/*+disp8 5/rm32/ebp . . . 7/r32/edi 8/disp8 . # copy *(ebp+8) to edi + # ecx = f->write + 8b/copy 0/mod/indirect 7/rm32/edi . . . 1/r32/ecx . . # copy *edi to ecx + # if (f->write >= f->size) abort + 3b/compare 1/mod/*+disp8 7/rm32/edi . . . 1/r32/ecx 8/disp8 . # compare ecx with *(edi+8) + 7d/jump-if->= $append-byte:end/disp8 # TODO: abort +$append-byte:to-stream: + # write to stream + # f->data[f->write] = LSB(n) + 31/xor 3/mod/direct 0/rm32/eax . . . 0/r32/eax . . # clear eax + 8a/copy-byte 1/mod/*+disp8 5/rm32/ebp . . . 0/r32/AL 0xc/disp8 . # copy byte at *(ebp+12) to AL + 88/copy-byte 1/mod/*+disp8 4/rm32/sib 7/base/edi 1/index/ecx . 0/r32/AL 0xc/disp8 . # copy AL to *(edi+ecx+12) + # ++f->write + ff 0/subop/increment 0/mod/indirect 7/rm32/edi . . . . . . # increment *edi +$append-byte:end: + # . restore registers + 5f/pop-to-edi + 59/pop-to-ecx + # . epilogue + 89/copy 3/mod/direct 4/rm32/esp . . . 5/r32/ebp . . # copy ebp to esp + 5d/pop-to-ebp + c3/return + +# . . vim:nowrap:textwidth=0 diff --git a/baremetal/120allocate.subx b/baremetal/120allocate.subx new file mode 100644 index 00000000..fbe998bb --- /dev/null +++ b/baremetal/120allocate.subx @@ -0,0 +1,52 @@ +# instruction effective address register displacement immediate +# . op subop mod rm32 base index scale r32 +# . 1-3 bytes 3 bits 2 bits 3 bits 3 bits 3 bits 2 bits 2 bits 0/1/2/4 bytes 0/1/2/4 bytes + +# Fill a region of memory with zeroes. +zero-out: # start: (addr byte), size: int + # pseudocode: + # curr/esi = start + # i/ecx = 0 + # while true + # if (i >= size) break + # *curr = 0 + # ++curr + # ++i + # + # . prologue + 55/push-ebp + 89/copy 3/mod/direct 5/rm32/ebp . . . 4/r32/esp . . # copy esp to ebp + # . save registers + 50/push-eax + 51/push-ecx + 52/push-edx + 56/push-esi + # curr/esi = start + 8b/copy 1/mod/*+disp8 5/rm32/ebp . . . 6/r32/esi 8/disp8 . # copy *(ebp+8) to esi + # var i/ecx: int = 0 + 31/xor 3/mod/direct 1/rm32/ecx . . . 1/r32/ecx . . # clear ecx + # edx = size + 8b/copy 1/mod/*+disp8 5/rm32/ebp . . . 2/r32/edx 0xc/disp8 . # copy *(ebp+12) to edx +$zero-out:loop: + # if (i >= size) break + 39/compare 3/mod/direct 1/rm32/ecx . . . 2/r32/edx . . # compare ecx with edx + 7d/jump-if->= $zero-out:end/disp8 + # *curr = 0 + c6 0/subop/copy-byte 0/mod/direct 6/rm32/esi . . . . . 0/imm8 # copy byte to *esi + # ++curr + 46/increment-esi + # ++i + 41/increment-ecx + eb/jump $zero-out:loop/disp8 +$zero-out:end: + # . restore registers + 5e/pop-to-esi + 5a/pop-to-edx + 59/pop-to-ecx + 58/pop-to-eax + # . epilogue + 89/copy 3/mod/direct 4/rm32/esp . . . 5/r32/ebp . . # copy ebp to esp + 5d/pop-to-ebp + c3/return + +# . . vim:nowrap:textwidth=0 diff --git a/baremetal/302stack_allocate.subx b/baremetal/302stack_allocate.subx new file mode 100644 index 00000000..cd51d5ff --- /dev/null +++ b/baremetal/302stack_allocate.subx @@ -0,0 +1,61 @@ +# A function which pushes n zeros on the stack. +# Really only intended to be called from code generated by mu.subx (for array +# vars on the stack). + +== code + +#? Entry: +#? # . prologue +#? 89/<- %ebp 4/r32/esp +#? # +#? 68/push 0xfcfdfeff/imm32 +#? b8/copy-to-eax 0x34353637/imm32 +#? $dump-stack0: +#? (push-n-zero-bytes 4) +#? 68/push 0x20/imm32 +#? $dump-stack9: +#? b8/copy-to-eax 1/imm32/exit +#? cd/syscall 0x80/imm8 + +# This is not a regular function, so it won't be idiomatic. +# Registers must be properly restored. +# Registers can be spilled, but that modifies the stack and needs to be +# cleaned up. + +# Overhead: +# 62 + n*6 instructions to push n bytes. +# If we just emitted code to push n zeroes, it would be: +# 5 bytes for 4 zero bytes, so 1.25 bytes per zero. And that's not even +# instructions. +# But on the other hand it would destroy the instruction cache, where this +# approach requires 15 instructions, fixed. + +# n must be positive +push-n-zero-bytes: # n: int +$push-n-zero-bytes:prologue: + 89/<- *Push-n-zero-bytes-ebp 5/r32/ebp # spill ebp without affecting stack + 89/<- %ebp 4/r32/esp +$push-n-zero-bytes:copy-ra: + # -- esp = ebp + 89/<- *Push-n-zero-bytes-eax 0/r32/eax + 8b/-> *esp 0/r32/eax + 2b/subtract *(ebp+4) 4/r32/esp + # -- esp+n = ebp + 89/<- *esp 0/r32/eax + 8b/-> *Push-n-zero-bytes-eax 0/r32/eax +$push-n-zero-bytes:bulk-cleaning: + 89/<- *Push-n-zero-bytes-esp 4/r32/esp + 81 0/subop/add *Push-n-zero-bytes-esp 4/imm32 + 81 0/subop/add *(ebp+4) 4/imm32 + (zero-out *Push-n-zero-bytes-esp *(ebp+4)) # n+4 +$push-n-zero-bytes:epilogue: + 8b/-> *Push-n-zero-bytes-ebp 5/r32/ebp # restore spill + c3/return + +== data +Push-n-zero-bytes-ebp: # (addr int) + 0/imm32 +Push-n-zero-bytes-esp: # (addr int) + 0/imm32 +Push-n-zero-bytes-eax: + 0/imm32 diff --git a/baremetal/309stream.subx b/baremetal/309stream.subx new file mode 100644 index 00000000..720ee0eb --- /dev/null +++ b/baremetal/309stream.subx @@ -0,0 +1,200 @@ +# Some unsafe methods not intended to be used directly in SubX, only through +# Mu after proper type-checking. + +== code + +stream-empty?: # s: (addr stream _) -> result/eax: boolean + # . prologue + 55/push-ebp + 89/<- %ebp 4/r32/esp + # . save registers + 51/push-ecx + 56/push-esi + # result = false + b8/copy-to-eax 0/imm32/false + # esi = s + 8b/-> *(ebp+8) 6/r32/esi + # return s->read >= s->write + 8b/-> *esi 1/r32/ecx + 39/compare-with *(esi+4) 1/r32/ecx + 0f 9d/set-if->= %al +$stream-empty?:end: + # . restore registers + 5e/pop-to-esi + 59/pop-to-ecx + # . epilogue + 89/<- %esp 5/r32/ebp + 5d/pop-to-ebp + c3/return + +stream-full?: # s: (addr stream _) -> result/eax: boolean + # . prologue + 55/push-ebp + 89/<- %ebp 4/r32/esp + # . save registers + 51/push-ecx + 56/push-esi + # result = false + b8/copy-to-eax 0/imm32/false + # esi = s + 8b/-> *(ebp+8) 6/r32/esi + # return s->write >= s->size + 8b/-> *(esi+8) 1/r32/ecx + 39/compare-with *esi 1/r32/ecx + 0f 9d/set-if->= %al +$stream-full?:end: + # . restore registers + 5e/pop-to-esi + 59/pop-to-ecx + # . epilogue + 89/<- %esp 5/r32/ebp + 5d/pop-to-ebp + c3/return + +write-to-stream: # s: (addr stream _), in: (addr byte), n: int + # . prologue + 55/push-ebp + 89/<- %ebp 4/r32/esp + # . save registers + 50/push-eax + 51/push-ecx + 52/push-edx + 53/push-ebx + 57/push-edi + # edi = s + 8b/-> *(ebp+8) 7/r32/edi + # var swrite/edx: int = s->write + 8b/-> *edi 2/r32/edx + # if (swrite + n > s->size) return + 8b/-> *(ebp+0x10) 1/r32/ecx + 01/add-to %ecx 2/r32/edx + 3b/compare 1/r32/ecx *(edi+8) + 0f 8f/jump-if-> $write-to-stream:end/disp32 # TODO: abort + # var out/edx: (addr byte) = s->data + s->write + 8d/copy-address *(edi+edx+0xc) 2/r32/edx + # var outend/ebx: (addr byte) = out + n + 8b/-> *(ebp+0x10) 3/r32/ebx + 8d/copy-address *(edx+ebx) 3/r32/ebx + # eax = in + 8b/-> *(ebp+0xc) 0/r32/eax + # var inend/ecx: (addr byte) = in + n + 8b/-> *(ebp+0x10) 1/r32/ecx + 8d/copy-address *(eax+ecx) 1/r32/ecx + # + (_append-4 %edx %ebx %eax %ecx) # => eax + # s->write += n + 8b/-> *(ebp+0x10) 1/r32/ecx + 01/add-to *edi 1/r32/ecx +$write-to-stream:end: + # . restore registers + 5f/pop-to-edi + 5b/pop-to-ebx + 5a/pop-to-edx + 59/pop-to-ecx + 58/pop-to-eax + # . epilogue + 89/<- %esp 5/r32/ebp + 5d/pop-to-ebp + c3/return + +read-from-stream: # s: (addr stream _), out: (addr byte), n: int + # . prologue + 55/push-ebp + 89/<- %ebp 4/r32/esp + # . save registers + 50/push-eax + 51/push-ecx + 52/push-edx + 53/push-ebx + 56/push-esi + # esi = s + 8b/-> *(ebp+8) 6/r32/esi + # var sread/edx: int = s->read + 8b/-> *(esi+4) 2/r32/edx + # if (sread + n > s->write) return + 8b/-> *(ebp+0x10) 1/r32/ecx + 01/add-to %ecx 2/r32/edx + 3b/compare 1/r32/ecx *esi + 0f 8f/jump-if-> $read-from-stream:end/disp32 # TODO: abort + # var in/edx: (addr byte) = s->data + s->read + 8d/copy-address *(esi+edx+0xc) 2/r32/edx + # var inend/ebx: (addr byte) = in + n + 8b/-> *(ebp+0x10) 3/r32/ebx + 8d/copy-address *(edx+ebx) 3/r32/ebx + # eax = out + 8b/-> *(ebp+0xc) 0/r32/eax + # var outend/ecx: (addr byte) = out + n + 8b/-> *(ebp+0x10) 1/r32/ecx + 8d/copy-address *(eax+ecx) 1/r32/ecx + # + (_append-4 %eax %ecx %edx %ebx) # => eax + # s->read += n + 8b/-> *(ebp+0x10) 1/r32/ecx + 01/add-to *(esi+4) 1/r32/ecx +$read-from-stream:end: + # . restore registers + 5e/pop-to-esi + 5b/pop-to-ebx + 5a/pop-to-edx + 59/pop-to-ecx + 58/pop-to-eax + # . epilogue + 89/<- %esp 5/r32/ebp + 5d/pop-to-ebp + c3/return + +stream-first: # s: (addr stream byte) -> result/eax: byte + # . prologue + 55/push-ebp + 89/<- %ebp 4/r32/esp + # . save registers + 51/push-ecx + 56/push-esi + # result = false + b8/copy-to-eax 0/imm32 + # esi = s + 8b/-> *(ebp+8) 6/r32/esi + # var idx/ecx: int = s->read + 8b/-> *(esi+4) 1/r32/ecx + # if idx >= s->write return 0 + 3b/compare-with 1/r32/ecx *esi + 7d/jump-if->= $stream-first:end/disp8 + # result = s->data[idx] + 8a/byte-> *(esi+ecx+0xc) 0/r32/AL +$stream-first:end: + # . restore registers + 5e/pop-to-esi + 59/pop-to-ecx + # . epilogue + 89/<- %esp 5/r32/ebp + 5d/pop-to-ebp + c3/return + +stream-final: # s: (addr stream byte) -> result/eax: byte + # . prologue + 55/push-ebp + 89/<- %ebp 4/r32/esp + # . save registers + 51/push-ecx + 56/push-esi + # result = false + b8/copy-to-eax 0/imm32 + # esi = s + 8b/-> *(ebp+8) 6/r32/esi + # var max/ecx: int = s->write + 8b/-> *esi 1/r32/ecx + # if s->read >= max return 0 + 39/compare-with *(esi+4) 1/r32/ecx + 7d/jump-if->= $stream-final:end/disp8 + # var idx/ecx: int = max - 1 + 49/decrement-ecx + # result = s->data[idx] + 8a/byte-> *(esi+ecx+0xc) 0/r32/AL +$stream-final:end: + # . restore registers + 5e/pop-to-esi + 59/pop-to-ecx + # . epilogue + 89/<- %esp 5/r32/ebp + 5d/pop-to-ebp + c3/return diff --git a/baremetal/400.mu b/baremetal/400.mu index 016204c5..f25e13a3 100644 --- a/baremetal/400.mu +++ b/baremetal/400.mu @@ -1,3 +1,8 @@ sig pixel screen: (addr screen), x: int, y: int, color: int sig read-key kbd: (addr keyboard) -> _/eax: byte sig draw-grapheme screen: (addr screen), g: grapheme, x: int, y: int, color: int + +sig write f: (addr stream byte), s: (addr array byte) +sig append-byte f: (addr stream byte), n: int +sig read-byte s: (addr stream byte) -> _/eax: byte +sig stream-empty? s: (addr stream _) -> _/eax: boolean diff --git a/baremetal/401draw-text-rightward.mu b/baremetal/401draw-text-rightward.mu index 089c5d5c..61b2d9ca 100644 --- a/baremetal/401draw-text-rightward.mu +++ b/baremetal/401draw-text-rightward.mu @@ -1,16 +1,13 @@ -fn draw-text-rightward screen: (addr screen), _text: (addr array byte), x: int, y: int, color: int { - var text/esi: (addr array byte) <- copy _text - var len/ecx: int <- length text - var i/edx: int <- copy 0 +fn draw-text-rightward screen: (addr screen), text: (addr array byte), x: int, y: int, color: int { + var stream-storage: (stream byte 0x100) + var stream/esi: (addr stream byte) <- address stream-storage + write stream, text { - compare i, len - break-if->= - var g/eax: (addr byte) <- index text, i - var g2/eax: byte <- copy-byte *g - var g3/eax: grapheme <- copy g2 - draw-grapheme screen, g3, x, y, color + var g/eax: grapheme <- read-grapheme stream + compare g, 0xffffffff # end-of-file + break-if-= + draw-grapheme screen, g, x, y, color add-to x, 8 # font-width - i <- increment loop } } diff --git a/baremetal/403unicode.mu b/baremetal/403unicode.mu new file mode 100644 index 00000000..ea45f707 --- /dev/null +++ b/baremetal/403unicode.mu @@ -0,0 +1,193 @@ +# Helpers for Unicode. +# +# Mu has no characters, only code points and graphemes. +# Code points are the indivisible atoms of text streams. +# https://en.wikipedia.org/wiki/Code_point +# Graphemes are the smallest self-contained unit of text. +# Graphemes may consist of multiple code points. +# +# Mu graphemes are always represented in utf-8, and they are required to fit +# in 4 bytes. +# +# Mu doesn't currently support combining code points, or graphemes made of +# multiple code points. One day we will. +# We also don't currently support code points that translate into multiple +# or wide graphemes. (In particular, Tab will never be supported.) + +# transliterated from tb_utf8_unicode_to_char in https://github.com/nsf/termbox +# https://wiki.tcl-lang.org/page/UTF%2D8+bit+by+bit explains the algorithm +# +# The day we want to support combining characters, this function will need to +# take multiple code points. Or something. +fn to-grapheme in: code-point -> _/eax: grapheme { + var c/eax: int <- copy in + var num-trailers/ecx: int <- copy 0 + var first/edx: int <- copy 0 + $to-grapheme:compute-length: { + # single byte: just return it + compare c, 0x7f + { + break-if-> + var g/eax: grapheme <- copy c + return g + } + # 2 bytes + compare c, 0x7ff + { + break-if-> + num-trailers <- copy 1 + first <- copy 0xc0 + break $to-grapheme:compute-length + } + # 3 bytes + compare c, 0xffff + { + break-if-> + num-trailers <- copy 2 + first <- copy 0xe0 + break $to-grapheme:compute-length + } + # 4 bytes + compare c, 0x1fffff + { + break-if-> + num-trailers <- copy 3 + first <- copy 0xf0 + break $to-grapheme:compute-length + } + # more than 4 bytes: unsupported + # TODO: print error message to stderr + compare c, 0x1fffff + { + break-if-> + return 0 + } + } + # emit trailer bytes, 6 bits from 'in', first two bits '10' + var result/edi: grapheme <- copy 0 + { + compare num-trailers, 0 + break-if-<= + var tmp/esi: int <- copy c + tmp <- and 0x3f + tmp <- or 0x80 + result <- shift-left 8 + result <- or tmp + # update loop state + c <- shift-right 6 + num-trailers <- decrement + loop + } + # emit engine + result <- shift-left 8 + result <- or c + result <- or first + # + return result +} + +# TODO: bring in tests once we have check-ints-equal + +# read the next grapheme from a stream of bytes +fn read-grapheme in: (addr stream byte) -> _/eax: grapheme { + # if at eof, return EOF + { + var eof?/eax: boolean <- stream-empty? in + compare eof?, 0 # false + break-if-= + return 0xffffffff + } + var c/eax: byte <- read-byte in + var num-trailers/ecx: int <- copy 0 + $read-grapheme:compute-length: { + # single byte: just return it + compare c, 0xc0 + { + break-if->= + var g/eax: grapheme <- copy c + return g + } + compare c, 0xfe + { + break-if-< + var g/eax: grapheme <- copy c + return g + } + # 2 bytes + compare c, 0xe0 + { + break-if->= + num-trailers <- copy 1 + break $read-grapheme:compute-length + } + # 3 bytes + compare c, 0xf0 + { + break-if->= + num-trailers <- copy 2 + break $read-grapheme:compute-length + } + # 4 bytes + compare c, 0xf8 + { + break-if->= + num-trailers <- copy 3 + break $read-grapheme:compute-length + } + # TODO: print error message + return 0 + } + # prepend trailer bytes + var result/edi: grapheme <- copy c + var num-byte-shifts/edx: int <- copy 1 + { + compare num-trailers, 0 + break-if-<= + var tmp/eax: byte <- read-byte in + var tmp2/eax: int <- copy tmp + tmp2 <- shift-left-bytes tmp2, num-byte-shifts + result <- or tmp2 + # update loop state + num-byte-shifts <- increment + num-trailers <- decrement + loop + } + return result +} + +# needed because available primitives only shift by a literal/constant number of bits +fn shift-left-bytes n: int, k: int -> _/eax: int { + var i/ecx: int <- copy 0 + var result/eax: int <- copy n + { + compare i, k + break-if->= + compare i, 4 # only 4 bytes in 32 bits + break-if->= + result <- shift-left 8 + i <- increment + loop + } + return result +} + +# write a grapheme to a stream of bytes +# this is like write-to-stream, except we skip leading 0 bytes +fn write-grapheme out: (addr stream byte), g: grapheme { +$write-grapheme:body: { + var c/eax: int <- copy g + append-byte out, c # first byte is always written + c <- shift-right 8 + compare c, 0 + break-if-= $write-grapheme:body + append-byte out, c + c <- shift-right 8 + compare c, 0 + break-if-= $write-grapheme:body + append-byte out, c + c <- shift-right 8 + compare c, 0 + break-if-= $write-grapheme:body + append-byte out, c +} +} |