diff options
| -rw-r--r-- | hex.e | 72 | ||||
| -rw-r--r-- | transform.e | 219 |
2 files changed, 205 insertions, 86 deletions
diff --git a/hex.e b/hex.e index fd283ca..d1c4e31 100644 --- a/hex.e +++ b/hex.e @@ -21,88 +21,106 @@ ~ : This is the start routine, the first thing that runs when the ELF loads. ~ : indent current-offset L!' cold-start - ~ The basic registers preserved across syscalls are rbx, rsp, rbp. - ~ To avoid redundant moves, we store the buffer pointer in rbx just once, - ~ and keep it there. We've made sure our load origin fits in 32 bits, so we - ~ can use imm32 for that. We're going to want to do an indirect load from - ~ it, so we can't use rbp for this. + ~ : The basic registers preserved across syscalls are rbx, rsp, rbp. + ~ : To avoid redundant moves, we store the buffer pointer in rbx just once, + ~ : and keep it there. We've made sure our load origin fits in 32 bits, so + ~ : we can use imm32 for that. We're going to want to do an indirect load + ~ : from it, so we can't use rbp for this. L@' buffer L@' origin + :rbx mov-reg64-imm32 + ~ : blank-line current-offset L!' input-loop-start L@' read-byte call-rel-imm32-from-here + ~ : blank-line - ~ If the length is 0, we got EOF. If it's less than zero, we got a read - ~ error. Either way, we exit. This is a signed comparison, as it needs to - ~ be. + ~ : If the length is 0, we got EOF. If it's less than zero, we got a read + ~ : error. Either way, we exit. This is a signed comparison, as it needs to + ~ : be. 0 :rax cmp-reg64-imm8 L@' exit :cc-equal jmp-cc-rel-imm8-from-here L@' read-error :cc-less jmp-cc-rel-imm8-from-here + ~ : blank-line - ~ Now that the length is handled, retrieve the input byte. + ~ : Now that the length is handled, retrieve the input byte. :rbx :rax mov-reg64-indirect-reg64 + ~ : blank-line - ~ If it's space or linefeed, skip it (go back to the loop start). + ~ : If it's space or linefeed, skip it (go back to the loop start). 0x20 :rax cmp-reg64-imm8 ~ ASCII space L@' input-loop-start :cc-equal jmp-cc-rel-imm8-from-here 0x0a :rax cmp-reg64-imm8 ~ ASCII linefeed L@' input-loop-start :cc-equal jmp-cc-rel-imm8-from-here - ~ If it's a comment, skip the whole thing. + ~ : If it's a comment, skip the whole thing. 0x7e :rax cmp-reg64-imm8 ~ ASCII tilde L@' skip-comment :cc-equal jmp-cc-rel-imm8-from-here + ~ : blank-line - ~ Decode the value, or exit with an error. + ~ : Decode the value, or exit with an error. L@' decode-nibble call-rel-imm32-from-here + ~ : blank-line - ~ We use rbp as a place to stash the high nibble. + ~ : We use rbp as a place to stash the high nibble. :rax :rbp mov-reg64-reg64 4 :rbp rol-reg64-imm8 + ~ : blank-line - ~ Now we read another byte. + ~ : Now we read another byte. L@' read-byte call-rel-imm32-from-here + ~ : blank-line - ~ Handle the length. A second hex digit is required here. + ~ : Handle the length. A second hex digit is required here. 0 :rax cmp-reg64-imm8 L@' unexpected-eof :cc-equal jmp-cc-rel-imm8-from-here L@' read-error :cc-less jmp-cc-rel-imm8-from-here + ~ : blank-line - ~ Now that the length is handled, retrieve the input byte. + ~ : Now that the length is handled, retrieve the input byte. :rbx :rax mov-reg64-indirect-reg64 + ~ : blank-line - ~ Decode the value, or exit with an error. + ~ : Decode the value, or exit with an error. L@' decode-nibble call-rel-imm32-from-here + ~ : blank-line - ~ We OR in the low nibble. + ~ : We OR in the low nibble. :rax :rbp or-reg64-reg64 + ~ : blank-line - ~ Output the byte. We reuse the buffer as a place to store it. + ~ : Output the byte. We reuse the buffer as a place to store it. :rbp :rbx mov-indirect-reg64-reg64 :rbx :rsi mov-reg64-reg64 ~ buffer pointer 1 :rdx mov-reg64-imm32 ~ buffer length 1 :rax mov-reg64-imm32 ~ syscall number for sys-write 1 :rdi mov-reg64-imm32 ~ file descriptor 1 is stdout syscall + ~ : blank-line - ~ Back to the start of the loop. + ~ : Back to the start of the loop. L@' input-loop-start jmp-rel-imm8-from-here + ~ : blank-line current-offset L!' skip-comment - ~ Read a byte for the comment. + ~ : Read a byte for the comment. L@' read-byte call-rel-imm32-from-here + ~ : blank-line - ~ Handle the length. We're allowed to end in a comment. + ~ : Handle the length. We're allowed to end in a comment. 0 :rax cmp-reg64-imm8 L@' exit :cc-equal jmp-cc-rel-imm8-from-here L@' read-error :cc-less jmp-cc-rel-imm8-from-here + ~ : blank-line - ~ Now that the length is handled, retrieve the input byte. + ~ : Now that the length is handled, retrieve the input byte. :rbx :rax mov-reg64-indirect-reg64 + ~ : blank-line - ~ If it's linefeed, the comment is over. + ~ : If it's linefeed, the comment is over. 0x0a :rax cmp-reg64-imm8 ~ ASCII linefeed L@' input-loop-start :cc-equal jmp-cc-rel-imm8-from-here + ~ : blank-line - ~ We're still in the comment, keep handling it. + ~ : We're still in the comment, keep handling it. L@' skip-comment jmp-rel-imm8-from-here ~ : deindent ; @@ -119,11 +137,11 @@ ~ : This is the routine named "read-byte". ~ : indent current-offset L!' read-byte - ~ We use self-xor as a concise way to set registers to zero. + ~ : We use self-xor as a concise way to set registers to zero. :rax :rax xor-reg64-reg64 ~ syscall number for sys-read :rdi :rdi xor-reg64-reg64 ~ file descriptor 0 is stdin :rbx :rsi mov-reg64-reg64 ~ buffer pointer - ~ We read one byte at a time, because it makes the loop structure simple. + ~ : We read one byte at a time, because it makes the loop structure simple. 1 :rdx mov-reg64-imm32 ~ buffer length syscall ret diff --git a/transform.e b/transform.e index 7560752..ea93ec3 100644 --- a/transform.e +++ b/transform.e @@ -2738,6 +2738,7 @@ allocate-transformation-state s" transformation-state" variable : hex-output-metadata-entry-type-string-literal 5 ; : hex-output-metadata-entry-type-raw-string-literal 6 ; : hex-output-metadata-entry-type-alignment 7 ; +: hex-output-metadata-entry-type-provide-substring-hex64 8 ; ~ Initialize the contents of the output metadata to all zeroes. This is ~ called from hex-transform at its top level, at the very start, to make sure @@ -2982,6 +2983,16 @@ allocate-transformation-state s" transformation-state" variable s" swap" find entry-to-execution-token , ; +: is-comment-entry + hex-output-metadata-entry-type @ + dup hex-output-metadata-entry-type-line-comment = + swap hex-output-metadata-entry-type-suffix-comment = + || ; + +: is-provide-substring-entry + hex-output-metadata-entry-type @ + hex-output-metadata-entry-type-provide-substring-hex64 = ; + : is-fresh-line@ transformation-state transformation-state-output-metadata @ hex-output-metadata-is-fresh-line @ ; @@ -3018,66 +3029,134 @@ allocate-transformation-state s" transformation-state" variable ~ Any time output is performed, all existing entries are post-processed. ~ Therefore, the post-processing is required to be idempotent, and is. : postprocess-metadata-entries - transformation-state transformation-state-output-metadata @ - hex-output-metadata-first-entry - { dup @ } { - ~ (entry pointer) - - ~ Is the entry a suffix comment with a non-zero data length? - dup hex-output-metadata-entry-type @ - hex-output-metadata-entry-type-suffix-comment = - over hex-output-metadata-entry-data-length @ 0 != && { - ~ (entry pointer) - dup dup - - ~ Find all the remaining entries, and the final terminator, and slide - ~ them forward by the length of one entry. - { dup @ } { hex-output-metadata-next-entry } while 8 + - ~ (entry pointer, entry pointer, entry array end pointer) - swap - - ~ (entry pointer, entry array tail length) - swap dup dup hex-output-metadata-next-entry 4 roll - ~ (entry pointer, entry pointer, adjusted entry pointer, - ~ entry array tail length) - memmove - ~ (entry pointer) - - ~ We have a feature which lets a suffix comment be created with a - ~ negative length, which means it covers bytes before the current output - ~ point rather than after it. This logic here is the implementation of - ~ that feature. - dup hex-output-metadata-entry-data-start @ dup - 2 pick hex-output-metadata-entry-data-length @ + - 2dup min 3unroll max - ~ (entry pointer, low end of data, high end of data) - - ~ Now we have the original entry at the original location, and a copy - ~ of it at the next location. We modify the original one in-place to - ~ turn it into a fresh-line entry at the low end of the data... - 2 pick hex-output-metadata-entry-type - hex-output-metadata-entry-type-fresh-line swap ! - - 2 pick hex-output-metadata-entry-data-start 3roll swap ! - 2 pick hex-output-metadata-entry-data-length 0 swap ! - ~ (entry pointer, high end of data) - - ~ ... then we modify the next entry in-place to set its location to - ~ the end of the data instead of the beginning, and its length to zero. - over hex-output-metadata-next-entry - ~ (entry pointer, high end of data, next entry pointer) - - dup hex-output-metadata-entry-data-start 3roll swap ! - hex-output-metadata-entry-data-length 0 swap ! - ~ (entry pointer) - - ~ This rule won't fire again on the fresh-line entry because it's of a - ~ different type, and it won't fire on the modified suffix-comment entry - ~ because its data length is zero. There's no other rules either, so our - ~ idempotence requirement is satisfied. - } if + 1 { } { + 0 + ~ (did anything this iteration) + + transformation-state transformation-state-output-metadata @ + hex-output-metadata-first-entry + { dup @ } { + ~ (did anything, entry pointer) + + ~ Is the entry a suffix comment with a non-zero data length? + dup hex-output-metadata-entry-type @ + hex-output-metadata-entry-type-suffix-comment = + over hex-output-metadata-entry-data-length @ 0 != && { + ~ Split it into two entries so as not to rely on data length for + ~ formatting. + ~ (did anything, entry pointer) + + ~ Find all the remaining entries, and the final terminator, and slide + ~ them forward by the length of one entry. + dup dup + { dup @ } { hex-output-metadata-next-entry } while 8 + + ~ (..., entry pointer, entry pointer, entry array end pointer) + swap - + ~ (..., entry pointer, entry array tail length) + swap dup dup hex-output-metadata-next-entry 4 roll + ~ (..., entry pointer, entry pointer, adjusted entry pointer, + ~ entry array tail length) + memmove + ~ (..., entry pointer) + + ~ We have a feature which lets a suffix comment be created with a + ~ negative length, which means it covers bytes before the current output + ~ point rather than after it. This logic here is the implementation of + ~ that feature. + dup hex-output-metadata-entry-data-start @ dup + 2 pick hex-output-metadata-entry-data-length @ + + 2dup min 3unroll max + ~ (..., entry pointer, low end of data, high end of data) + + ~ Now we have the original entry at the original location, and a copy + ~ of it at the next location. We modify the original one in-place to + ~ turn it into a fresh-line entry at the low end of the data... + 2 pick hex-output-metadata-entry-type + hex-output-metadata-entry-type-fresh-line swap ! + ~ (..., entry pointer, low end of data, high end of data) + + 2 pick hex-output-metadata-entry-data-start 3roll swap ! + ~ (..., entry pointer, high end of data) + over hex-output-metadata-entry-data-length 0 swap ! + ~ (..., entry pointer, high end of data) + + ~ ... then we modify the next entry in-place to set its location to + ~ the end of the data instead of the beginning, and its length to zero. + over hex-output-metadata-next-entry + ~ (..., entry pointer, high end of data, next entry pointer) + + dup hex-output-metadata-entry-data-start 3roll swap ! + ~ (..., entry pointer, next entry pointer) + hex-output-metadata-entry-data-length 0 swap ! + ~ (..., entry pointer) + + ~ This rule won't fire again on the fresh-line entry because it's of a + ~ a different type, and it won't fire on the modified suffix-comment + ~ entry because its data length is zero. + ~ + ~ It's possible the bubble-sort rule will fire on either of them. + + ~ (did anything, entry pointer) + swap drop 1 swap + } if + + ~ Do this entry and the next entry both exist, and they're out of + ~ order based on their data starts? + dup hex-output-metadata-next-entry dup @ + { hex-output-metadata-entry-data-start @ + over hex-output-metadata-entry-data-start @ > } + { drop 0 } if-else + { + ~ Swap the order of the two entries. Since this whole thing is in a + ~ loop, this functions as a bubble sort. + ~ (did anything, entry pointer) + + ~ Copy the current entry to a scratch area. + dup dup dup hex-output-metadata-next-entry swap - + ~ (..., entry pointer, entry pointer, entry length) + swap-transform-variables here @ swap-transform-variables swap + ~ (..., entry pointer, entry pointer, scratch pointer, entry length) + memcopy + ~ (..., entry pointer) + + ~ Copy the next entry over the current entry. + dup dup hex-output-metadata-next-entry dup + ~ (..., entry pointer, entry pointer, next entry pointer, + ~ next entry pointer) + 3roll - + ~ (..., entry pointer, next entry pointer, entry length) + 2 pick swap + ~ (..., entry pointer, + ~ next entry pointer, entry pointer, entry length) + memcopy + ~ (entry pointer) + + ~ Copy the scratch area over the next entry. + dup hex-output-metadata-next-entry dup + ~ (..., entry pointer, next entry pointer, next entry pointer) + 2 pick - + ~ (..., entry pointer, next entry pointer, entry length) + swap-transform-variables here @ swap-transform-variables 3unroll + ~ (..., entry pointer, + ~ scratch pointer, next entry pointer, entry length) + memcopy + ~ (..., entry pointer) + + ~ This rule won't fire again on the same pair of entries because now + ~ they're in the correct order. It may fire again next iteration, + ~ but only until everything is sorted. + ~ + ~ This rule does nothing to alter whether the suffix-comment rule + ~ will fire on either of the entries it swapped, so there's no + ~ possibility of an infinite loop through that one. + + ~ (did anything, entry pointer) + swap drop 1 swap + } if - hex-output-metadata-next-entry - } while drop ; + hex-output-metadata-next-entry + } while drop + } while ; ~ This "replacement" is a little different from an alternate: When the code ~ under transformation attempts to compile its own version of sys-write, it @@ -3188,6 +3267,19 @@ allocate-transformation-state s" transformation-state" variable 0 current-column! } if + dup hex-output-metadata-entry-type @ + hex-output-metadata-entry-type-provide-substring-hex64 = { + fresh-line + indentation-depth@ dup indent advance-current-column + + ." ~ hex64 value 0x" + dup hex-output-metadata-entry-string @ .hex64 + + newline + 1 is-fresh-line! + 0 current-column! + } if + } if hex-output-metadata-next-entry } while @@ -4013,6 +4105,15 @@ allocate-transformation-state s" transformation-state" variable exit } if + dup s" provide-substring-hex64" stringcmp 0 = { + ~ Create a new "provide substring hex64" entry. Notice that we + ~ intentionally read one stack level deeper than anything of ours, which + ~ will be a value provided by the code the magic comment is embedded in. + drop 2 pick hex-output-metadata-entry-type-provide-substring-hex64 swap + add-hex-output-metadata-entry + exit + } if + ~ If it has a nonzero length, create a new suffix-comment entry. over 0 != { hex-output-metadata-entry-type-suffix-comment swap |