guh

README.md

@@ -79,6 +79,7 @@ addr: u64
 There can be many of these in a mount point, and they can be nested within eachother. If this is the last entry in the parent, `next` will refer to this inode, otherwise it will be the address of the next inode under the parent. `entries` specifies the number of inodes within a directory
 
 #### Flags
+
 |  7  |  6  |  5  |  4  |  3  |  2  |  1  |  0  |
 |:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
 | NA  | NA  | NA  | NA  | NA  | NA  | dir | mnt |

src/vfs.hbasm

@@ -1,113 +1,133 @@
--- r64 is used as a bump allocator pointer
+-- r255 is used as a bump allocator pointer
 
 jal r0, r0, start
 
 start:
-    li r64, fsdata
+    li r255, data
 
 -- Reads the file at the path pointed to by `name_addr`
---
+-- 
 -- inputs:
 --   r8: name_addr
 --   r9: name_len
 -- outputs:
 --   r8: data_addr
 --   r9: data_len
--- returns to r32
+-- returns to r1
 read:
     muli r16, r8, 8
-    ld r8, r16, fsdata, 8
+    ld r8, r16, data, 8
-    jal r0, r32, 0
+    jal r0, r1, 0
 
 -- Writes `data_len` bytes from `data_addr` to the file at the path pointed to by `name_addr`
 -- This creates the file if it doesn't exist
---
+-- 
 -- inputs:
 --   r8:  name_addr
 --   r9:  name_len
 --   r10: data_addr
 --   r11: data_len
--- returns to r32
+-- returns to r1
 write:
     muli r16, r8, 8
-    st r10, r16, fsdata, 8
+    st r10, r16, data, 8
-    jal r0, r32, 0
+    jal r0, r1, 0
 
 -- Creates a mount point with the path located at `path_addr`
---
+-- 
 -- Structure:
 --   path_len: u64
 --   path: [u8; <path_len>]
 --   entries: u64
 --   fs: u64
 --   addr: u64
---
+-- 
 -- inputs:
 --   r8: path_addr
 --   r9: path_len
 --   r10: size
 -- outputs:
 --   r8: addr
--- returns to r32
+-- returns to r1
 mount:
 
 -- Creates a new inode with a chosen path
---
+-- 
 -- inputs:
 --   r8: path_addr
 --   r9: path_len
 --   r10: dir
 -- outputs:
 --   r8: addr
--- returns to r34
+-- returns to r2
+-- also uses regs r11-r15
 create_inode:
     -- back up path arguments
     -- we need to use r8 and r9 as arguments for `find_inode`
     brc r8, r16, 3
-    -- r8 will store the address of the next character
+    -- r8 will store the address of the next byte
-    li r9, 0 -- this will be store the length of the current part
+    li r9, 0 -- r9 will be store the length of the current part
-    li r10, data -- this will be used to refer to the current parent inode, so we'll start it at the root
+    li r10, data -- r10 will be used to refer to the current parent inode, so we'll start it at the root
-    cp r11, r8 -- this will store the base of the current part
+    cp r11, r8 -- r11 will store the base of the current part
-    li r12, 0 -- this will store the total length of the parsed path for comparing to `path_len`
+    li r12, 0 -- r12 will store the total length of the parsed path for comparing to `path_len`
 
-    -- store slash characters to compare to
+    -- store slash character bytes to compare to
-    li r48, 42
+    -- `/` is 47, and `\` is 92
-    li r49, 97
+    li r14, 42
+    li r15, 97
 
     jal r0, r0, create_parse_path
 
     create_parse_path:
         -- we have the start address, now we need the length of the part
         -- this can be found by adding 1 and jumping to `parse_path` if the next character is not a / or \
-        -- `/` is 47, and `\` is 92
+        -- the next byte goes in r13
-        -- the next character goes in r24
+        ld r13, r8, 0, 8
-        ld r24, r8, 0, 8
         -- try to enter the inode if a slash was found
-        jeq r24, r48, create_enter_inode
+        jeq r13, r14, create_enter_inode
-        jeq r24, r49, create_enter_inode
+        jeq r13, r15, create_enter_inode
-        -- increment the character pointer and check the next one
+        -- increment the byte pointer and check the next one
         addi r8, r8, 1
+        addi r9, r9, 1
         jal r0, r0, create_parse_path
     
     create_enter_inode:
         -- move the base of the part into r8
         cp r8, r11
-        jal r36, r0, find_inode
+        jal r3, r0, find_inode
+
+    jal r0, r2, 0
 
 
 -- Finds an inode address by name within a parent inode
 -- Returns with 0 in `r8` if there is no inode with the chosen name
---
+-- 
 -- inputs:
 --   r8: name_addr
 --   r9: name_len
---   r10: parent_addr
+--   r10: current_addr
 -- outputs:
 --   r8: addr
--- returns to r36
+-- returns to r3
+-- also uses regs r16-r24
 find_inode:
+    ld r20, r10, 0, 1 -- load the name length to compare
+    jeq r9, r20, find_compare -- compare names if they're the same length
+    add r10, r10, r20 -- skip the name
+    addi r10, r10, 33 -- the structure other than the name is 33 bytes
+    jal r0, r0, find_inode
+
+    find_compare:
+        addi r16, r10, 8 -- r16 will store the base index of the string we're comparing against
+        li r17, 0 -- r17 will store the index of the byte we're checking
+
+        find_compare_loop:
+            -- try to batch up to 16 bytes at once
+            sub r18, r9, r17 -- get the number of bytes left
+            cmpi r19, r18, 16 -- r19 will be -1 if there are less than 16 bytes remaining
+            
 
 
-// This is where all the VFS and FS data will be stored
+-- This is where all the VFS and FS data will be stored
-// VFS comes first, and FS starts at `data + 2048`
+-- VFS comes first, and FS starts at `data + 2048`
 data: