|
@@ -30,6 +30,191 @@
|
|
|
*/
|
|
|
SdFile::SdFile(const char* path, uint8_t oflag) : SdBaseFile(path, oflag) {
|
|
|
}
|
|
|
+
|
|
|
+bool SdFile::openFilteredGcode(SdBaseFile* dirFile, const char* path){
|
|
|
+ if( open(dirFile, path, O_READ) ){
|
|
|
+ // compute the block to start with
|
|
|
+ if( ! gfComputeNextFileBlock() )
|
|
|
+ return false;
|
|
|
+ gfReset();
|
|
|
+ return true;
|
|
|
+ } else {
|
|
|
+ return false;
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+bool SdFile::seekSetFilteredGcode(uint32_t pos){
|
|
|
+ if(! seekSet(pos) )return false;
|
|
|
+ if(! gfComputeNextFileBlock() )return false;
|
|
|
+ gfReset();
|
|
|
+ return true;
|
|
|
+}
|
|
|
+
|
|
|
+const uint8_t *SdFile::gfBlockBuffBegin() const {
|
|
|
+ return vol_->cache()->data; // this is constant for the whole time, so it should be fast and sleek
|
|
|
+}
|
|
|
+
|
|
|
+void SdFile::gfReset(){
|
|
|
+ // reset cache read ptr to its begin
|
|
|
+ gfReadPtr = gfBlockBuffBegin() + gfOffset;
|
|
|
+}
|
|
|
+
|
|
|
+// think twice before allowing this to inline - manipulating 4B longs is costly
|
|
|
+// moreover - this function has its parameters in registers only, so no heavy stack usage besides the call/ret
|
|
|
+void __attribute__((noinline)) SdFile::gfUpdateCurrentPosition(uint16_t inc){
|
|
|
+ curPosition_ += inc;
|
|
|
+}
|
|
|
+
|
|
|
+#define find_endl(resultP, startP) \
|
|
|
+__asm__ __volatile__ ( \
|
|
|
+"cycle: \n" \
|
|
|
+"ld r22, Z+ \n" \
|
|
|
+"cpi r22, 0x0A \n" \
|
|
|
+"brne cycle \n" \
|
|
|
+: "=z" (resultP) /* result of the ASM code - in our case the Z register (R30:R31) */ \
|
|
|
+: "z" (startP) /* input of the ASM code - in our case the Z register as well (R30:R31) */ \
|
|
|
+: "r22" /* modifying register R22 - so that the compiler knows */ \
|
|
|
+)
|
|
|
+
|
|
|
+// avoid calling the default heavy-weight read() for just one byte
|
|
|
+int16_t SdFile::readFilteredGcode(){
|
|
|
+ if( ! gfEnsureBlock() ){
|
|
|
+ goto eof_or_fail; // this is unfortunate :( ... other calls are using the cache and we can loose the data block of our gcode file
|
|
|
+ }
|
|
|
+ // assume, we have the 512B block cache filled and terminated with a '\n'
|
|
|
+ {
|
|
|
+ const uint8_t *start = gfReadPtr;
|
|
|
+
|
|
|
+ // It may seem unreasonable to copy the variable into a local one and copy it back at the end of this method,
|
|
|
+ // but there is an important point of view: the compiler is unsure whether it can optimize the reads/writes
|
|
|
+ // to gfReadPtr within this method, because it is a class member variable.
|
|
|
+ // The compiler cannot see, if omitting read/write won't have any incorrect side-effects to the rest of the whole FW.
|
|
|
+ // So this trick explicitly states, that rdPtr is a local variable limited to the scope of this method,
|
|
|
+ // therefore the compiler can omit read/write to it (keep it in registers!) as it sees fit.
|
|
|
+ // And it does! Codesize dropped by 68B!
|
|
|
+ const uint8_t *rdPtr = gfReadPtr;
|
|
|
+
|
|
|
+ // the same applies to gfXBegin, codesize dropped another 100B!
|
|
|
+ const uint8_t *blockBuffBegin = gfBlockBuffBegin();
|
|
|
+
|
|
|
+ uint8_t consecutiveCommentLines = 0;
|
|
|
+ while( *rdPtr == ';' ){
|
|
|
+ for(;;){
|
|
|
+
|
|
|
+ //while( *(++gfReadPtr) != '\n' ); // skip until a newline is found - suboptimal code!
|
|
|
+ // Wondering, why this "nice while cycle" is done in such a weird way using a separate find_endl() function?
|
|
|
+ // Have a look at the ASM code GCC produced!
|
|
|
+
|
|
|
+ // At first - a separate find_endl() makes the compiler understand,
|
|
|
+ // that I don't need to store gfReadPtr every time, I'm only interested in the final address where the '\n' was found
|
|
|
+ // - the cycle can run on CPU registers only without touching memory besides reading the character being compared.
|
|
|
+ // Not only makes the code run considerably faster, but is also 40B shorter!
|
|
|
+ // This was the generated code:
|
|
|
+ //FORCE_INLINE const uint8_t * find_endl(const uint8_t *p){
|
|
|
+ // while( *(++p) != '\n' ); // skip until a newline is found
|
|
|
+ // return p; }
|
|
|
+ // 11c5e: movw r30, r18
|
|
|
+ // 11c60: subi r18, 0xFF ; 255
|
|
|
+ // 11c62: sbci r19, 0xFF ; 255
|
|
|
+ // 11c64: ld r22, Z
|
|
|
+ // 11c66: cpi r22, 0x0A ; 10
|
|
|
+ // 11c68: brne .-12 ; 0x11c5e <get_command()+0x524>
|
|
|
+
|
|
|
+ // Still, even that was suboptimal as the compiler seems not to understand the usage of ld r22, Z+ (the plus is important)
|
|
|
+ // aka automatic increment of the Z register (R30:R31 pair)
|
|
|
+ // There is no other way than pure ASM!
|
|
|
+ find_endl(rdPtr, rdPtr);
|
|
|
+
|
|
|
+ // found a newline, prepare the next block if block cache end reached
|
|
|
+ if( rdPtr - blockBuffBegin > 512 ){
|
|
|
+ // at the end of block cache, fill new data in
|
|
|
+ gfUpdateCurrentPosition( rdPtr - start - 1 );
|
|
|
+ if( ! gfComputeNextFileBlock() )goto eof_or_fail;
|
|
|
+ if( ! gfEnsureBlock() )goto eof_or_fail; // fetch it into RAM
|
|
|
+ rdPtr = start = blockBuffBegin;
|
|
|
+ } else {
|
|
|
+ if(consecutiveCommentLines >= 250){
|
|
|
+ --rdPtr; // unget the already consumed newline
|
|
|
+ goto emit_char;
|
|
|
+ }
|
|
|
+ // peek the next byte - we are inside the block at least at 511th index - still safe
|
|
|
+ if( *rdPtr == ';' ){
|
|
|
+ // consecutive comment
|
|
|
+ ++consecutiveCommentLines;
|
|
|
+ } else {
|
|
|
+ --rdPtr; // unget the already consumed newline
|
|
|
+ goto emit_char;
|
|
|
+ }
|
|
|
+ break; // found the real end of the line even across many blocks
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+emit_char:
|
|
|
+ {
|
|
|
+ gfUpdateCurrentPosition( rdPtr - start + 1 );
|
|
|
+ int16_t rv = *rdPtr++;
|
|
|
+
|
|
|
+ if( curPosition_ >= fileSize_ ){
|
|
|
+ // past the end of file
|
|
|
+ goto eof_or_fail;
|
|
|
+ } else if( rdPtr - blockBuffBegin >= 512 ){
|
|
|
+ // past the end of current bufferred block - prepare the next one...
|
|
|
+ if( ! gfComputeNextFileBlock() )goto eof_or_fail;
|
|
|
+ // don't need to force fetch the block here, it will get loaded on the next call
|
|
|
+ rdPtr = blockBuffBegin;
|
|
|
+ }
|
|
|
+
|
|
|
+ // save the current read ptr for the next run
|
|
|
+ gfReadPtr = rdPtr;
|
|
|
+ return rv;
|
|
|
+ }
|
|
|
+
|
|
|
+}
|
|
|
+
|
|
|
+eof_or_fail:
|
|
|
+ // make the rdptr point to a safe location - end of file
|
|
|
+ gfReadPtr = gfBlockBuffBegin() + 512;
|
|
|
+ return -1;
|
|
|
+}
|
|
|
+
|
|
|
+bool SdFile::gfEnsureBlock(){
|
|
|
+ if ( vol_->cacheRawBlock(gfBlock, SdVolume::CACHE_FOR_READ)){
|
|
|
+ // terminate with a '\n'
|
|
|
+ const uint16_t terminateOfs = fileSize_ - gfOffset;
|
|
|
+ vol_->cache()->data[ terminateOfs < 512 ? terminateOfs : 512 ] = '\n';
|
|
|
+ return true;
|
|
|
+ } else {
|
|
|
+ return false;
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+bool SdFile::gfComputeNextFileBlock() {
|
|
|
+ // error if not open or write only
|
|
|
+ if (!isOpen() || !(flags_ & O_READ)) return false;
|
|
|
+
|
|
|
+ gfOffset = curPosition_ & 0X1FF; // offset in block
|
|
|
+ if (type_ == FAT_FILE_TYPE_ROOT_FIXED) {
|
|
|
+ // SHR by 9 means skip the last byte and shift just 3 bytes by 1
|
|
|
+ // -> should be 8 instructions... and not the horrible loop shifting 4 bytes at once
|
|
|
+ // still need to get some work on this
|
|
|
+ gfBlock = vol_->rootDirStart() + (curPosition_ >> 9);
|
|
|
+ } else {
|
|
|
+ uint8_t blockOfCluster = vol_->blockOfCluster(curPosition_);
|
|
|
+ if (gfOffset == 0 && blockOfCluster == 0) {
|
|
|
+ // start of new cluster
|
|
|
+ if (curPosition_ == 0) {
|
|
|
+ // use first cluster in file
|
|
|
+ curCluster_ = firstCluster_;
|
|
|
+ } else {
|
|
|
+ // get next cluster from FAT
|
|
|
+ if (!vol_->fatGet(curCluster_, &curCluster_)) return false;
|
|
|
+ }
|
|
|
+ }
|
|
|
+ gfBlock = vol_->clusterStartBlock(curCluster_) + blockOfCluster;
|
|
|
+ }
|
|
|
+ return true;
|
|
|
+}
|
|
|
+
|
|
|
//------------------------------------------------------------------------------
|
|
|
/** Write data to an open file.
|
|
|
*
|