Implement file path sanitation and relevant unit tests (Issue #137)

The logic is ported from Golang to C.

Reference code: 
- https://cs.opensource.google/go/go/+/refs/tags/go1.21.4:src/path/path.go;l=70
- https://cs.opensource.google/go/go/+/refs/tags/go1.21.4:src/path/path_test.go

Close #137

.gitignore

@@ -15,6 +15,7 @@ build/mini-gdbstub
 build/softfloat
 build/cache/
 build/map/
+build/path/
 *.o
 *.o.d
 tests/**/*.elf

mk/tests.mk

@@ -6,13 +6,20 @@ MAP_TEST_SRCDIR := tests/map
 MAP_TEST_OUTDIR:= build/map
 MAP_TEST_TARGET := $(MAP_TEST_OUTDIR)/test-map
 
+PATH_TEST_SRCDIR := tests/path
+PATH_TEST_OUTDIR := build/path
+PATH_TEST_TARGET := $(PATH_TEST_OUTDIR)/test-path
+
 CACHE_TEST_OBJS := \
 	test-cache.o
 
 MAP_TEST_OBJS := \
 	test-map.o \
 	mt19937.o
 
+PATH_TEST_OBJS := \
+	test-path.o 
+
 CACHE_TEST_OBJS := $(addprefix $(CACHE_TEST_OUTDIR)/, $(CACHE_TEST_OBJS)) \
 		   $(OUT)/cache.o $(OUT)/mpool.o
 OBJS += $(CACHE_TEST_OBJS)
@@ -23,6 +30,11 @@ MAP_TEST_OBJS := $(addprefix $(MAP_TEST_OUTDIR)/, $(MAP_TEST_OBJS)) \
 OBJS += $(MAP_TEST_OBJS)
 deps += $(MAP_TEST_OBJS:%.o=%.o.d)
 
+PATH_TEST_OBJS := $(addprefix $(PATH_TEST_OUTDIR)/, $(PATH_TEST_OBJS)) \
+		   $(OUT)/utils.o 
+OBJS += $(PATH_TEST_OBJS)
+deps += $(PATH_TEST_OBJS:%.o=%.o.d)
+
 # Check adaptive replacement cache policy is enabled or not, default is LFU
 ifeq ($(ENABLE_ARC), 1)
 CACHE_TEST_ACTIONS := \
@@ -43,8 +55,9 @@ endif
 
 CACHE_TEST_OUT = $(addprefix $(CACHE_TEST_OUTDIR)/, $(CACHE_TEST_ACTIONS:%=%.out))
 MAP_TEST_OUT = $(MAP_TEST_TARGET).out
+PATH_TEST_OUT = $(PATH_TEST_TARGET).out
 
-tests : run-test-cache run-test-map
+tests : run-test-cache run-test-map run-test-path
 
 run-test-cache: $(CACHE_TEST_OUT)
 	$(Q)$(foreach e,$(CACHE_TEST_ACTIONS),\
@@ -66,6 +79,15 @@ run-test-map: $(MAP_TEST_OUT)
 	$(PRINTF) "Failed.\n"; \
 	fi;
 
+run-test-path: $(PATH_TEST_OUT)
+	$(Q)$(PATH_TEST_TARGET)
+	$(VECHO) "Running test-path ... "; \
+	if [ $$? -eq 0 ]; then \
+	$(call notice, [OK]); \
+	else \
+	$(PRINTF) "Failed.\n"; \
+	fi;
+
 $(CACHE_TEST_OUT): $(CACHE_TEST_TARGET)
 	$(Q)$(foreach e,$(CACHE_TEST_ACTIONS),\
 	    $(CACHE_TEST_TARGET) $(CACHE_TEST_SRCDIR)/$(e).in > $(CACHE_TEST_OUTDIR)/$(e).out; \
@@ -91,3 +113,15 @@ $(MAP_TEST_OUTDIR)/%.o: $(MAP_TEST_SRCDIR)/%.c
 	$(VECHO) "  CC\t$@\n"
 	$(Q)mkdir -p $(dir $@)
 	$(Q)$(CC) -o $@ $(CFLAGS) -I./src -c -MMD -MF $@.d $<
+
+$(PATH_TEST_OUT): $(PATH_TEST_TARGET)
+	$(Q)touch $@
+
+$(PATH_TEST_TARGET): $(PATH_TEST_OBJS)
+	$(VECHO) "  CC\t$@\n"
+	$(Q)$(CC) $^ -o $@
+
+$(PATH_TEST_OUTDIR)/%.o: $(PATH_TEST_SRCDIR)/%.c
+	$(VECHO) "  CC\t$@\n"
+	$(Q)mkdir -p $(dir $@)
+	$(Q)$(CC) -o $@ $(CFLAGS) -I./src -c -MMD -MF $@.d $<

mk/tools.mk

@@ -7,6 +7,7 @@ HIST_OBJS := \
 	elf.o \
 	decode.o \
 	mpool.o \
+	utils.o \
 	rv_histogram.o
 
 HIST_OBJS := $(addprefix $(OUT)/, $(HIST_OBJS))

src/elf.c

@@ -8,6 +8,7 @@
 
 #include "elf.h"
 #include "io.h"
+#include "utils.h"
 
 #if defined(_WIN32)
 /* fallback to standard I/O text stream */
@@ -290,16 +291,23 @@ bool elf_load(elf_t *e, riscv_t *rv, memory_t *mem)
     return true;
 }
 
-bool elf_open(elf_t *e, const char *path)
+bool elf_open(elf_t *e, const char *input)
 {
     /* free previous memory */
     if (e->raw_data)
         release(e);
 
+    char *path = sanitize_path(input);
+    if (!path) {
+        return false;
+    }
+
 #if defined(USE_MMAP)
     int fd = open(path, O_RDONLY);
-    if (fd < 0)
+    if (fd < 0) {
+        free(path);
         return false;
+    }
 
     /* get file size */
     struct stat st;
@@ -312,21 +320,25 @@ bool elf_open(elf_t *e, const char *path)
     e->raw_data = mmap(0, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
     if (e->raw_data == MAP_FAILED) {
         release(e);
+        free(path);
         return false;
     }
     close(fd);
 
 #else  /* fallback to standard I/O text stream */
     FILE *f = fopen(path, "rb");
-    if (!f)
+    if (!f) {
+        free(path);
         return false;
+    }
 
     /* get file size */
     fseek(f, 0, SEEK_END);
     e->raw_size = ftell(f);
     fseek(f, 0, SEEK_SET);
     if (e->raw_size == 0) {
         fclose(f);
+        free(path);
         return false;
     }
 
@@ -339,6 +351,7 @@ bool elf_open(elf_t *e, const char *path)
     fclose(f);
     if (r != e->raw_size) {
         release(e);
+        free(path);
         return false;
     }
 #endif /* USE_MMAP */
@@ -349,9 +362,11 @@ bool elf_open(elf_t *e, const char *path)
     /* check it is a valid ELF file */
     if (!is_valid(e)) {
         release(e);
+        free(path);
         return false;
     }
 
+    free(path);
     return true;
 }
 

src/main.c

@@ -10,6 +10,7 @@
 
 #include "elf.h"
 #include "state.h"
+#include "utils.h"
 
 /* enable program trace mode */
 static bool opt_trace = false;

src/utils.c

@@ -3,7 +3,10 @@
  * "LICENSE" for information on usage and redistribution of this file.
  */
 
+#include <stdbool.h>
 #include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
 #include <sys/time.h>
 #include <time.h>
 
@@ -21,6 +24,8 @@
 #endif
 #endif
 
+#define MAX_PATH_LEN 1024
+
 static void get_time_info(int32_t *tv_sec, int32_t *tv_usec)
 {
 #if defined(HAVE_POSIX_TIMER)
@@ -67,3 +72,97 @@ void rv_clock_gettime(struct timespec *tp)
     tp->tv_sec = tv_sec;
     tp->tv_nsec = tv_usec / 1000; /* Transfer to microseconds */
 }
+
+char *sanitize_path(const char *input)
+{
+    size_t n = strnlen(input, MAX_PATH_LEN);
+
+    char *ret = calloc(n + 1, sizeof(char));
+    if (!ret) {
+        return NULL;
+    }
+
+    /* After sanitization, the new path will only be shorter than the original
+     * one. Thus, we can reuse the space */
+    if (n == 0) {
+        ret[0] = '.';
+        return ret;
+    }
+
+    bool is_root = (input[0] == '/');
+
+    /*
+     * Invariants:
+     * reading from path; r is index of next byte to process -> path[r]
+     * writing to buf; w is index of next byte to write -> ret[strlen(ret)]
+     * dotdot is index in buf where .. must stop, either because:
+     *  a) it is the leading slash
+     *  b) it is a leading ../../.. prefix.
+     */
+    size_t w = 0;
+    size_t r = 0;
+    size_t dotdot = 0;
+    if (is_root) {
+        ret[w] = '/';
+        w++;
+        r = 1;
+        dotdot = 1;
+    }
+
+    while (r < n) {
+        if (input[r] == '/') {
+            /*  empty path element */
+            r++;
+        } else if (input[r] == '.' && (r + 1 == n || input[r + 1] == '/')) {
+            /* . element */
+            r++;
+        } else if (input[r] == '.' && input[r + 1] == '.' &&
+                   (r + 2 == n || input[r + 2] == '/')) {
+            /* .. element: remove to last / */
+            r += 2;
+
+            if (w > dotdot) {
+                /* can backtrack */
+                w--;
+                while (w > dotdot && ret[w] != '/') {
+                    w--;
+                }
+            } else if (!is_root) {
+                /* cannot backtrack, but not is_root, so append .. element. */
+                if (w > 0) {
+                    ret[w] = '/';
+                    w++;
+                }
+                ret[w] = '.';
+                w++;
+                ret[w] = '.';
+                w++;
+                dotdot = w;
+            }
+        } else {
+            /* real path element, add slash if needed */
+            if ((is_root && w != 1) || (!is_root && w != 0)) {
+                ret[w] = '/';
+                w++;
+            }
+
+            /* copy element */
+            for (; r < n && input[r] != '/'; r++) {
+                ret[w] = input[r];
+                w++;
+            }
+        }
+    }
+
+    /* Turn empty string into "." */
+    if (w == 0) {
+        ret[w] = '.';
+        w++;
+    }
+
+    /* starting from w till the end, we should mark it as \0 since that part of
+     * the buffer is not used */
+    memset(ret + w, '\0', sizeof(n + 1 - w));
+
+    return ret;
+}

src/utils.h

@@ -21,3 +21,29 @@ void rv_clock_gettime(struct timespec *tp);
         /* 0x61C88647 is 32-bit golden ratio */                         \
         return (val * 0x61C88647 >> (32 - size_bits)) & ((size) - (1)); \
     }
+
+/*
+ * Reference:
+ * https://cs.opensource.google/go/go/+/refs/tags/go1.21.4:src/path/path.go;l=51
+ *
+ * sanitize_path returns the shortest path name equivalent to path
+ * by purely lexical processing. It applies the following rules
+ * iteratively until no further processing can be done:
+ *
+ *  1. Replace multiple slashes with a single slash.
+ *  2. Eliminate each . path name element (the current directory).
+ *  3. Eliminate each inner .. path name element (the parent directory)
+ *     along with the non-.. element that precedes it.
+ *  4. Eliminate .. elements that begin a rooted path:
+ *     that is, replace "/.." by "/" at the beginning of a path.
+ *
+ * The returned path ends in a slash only if it is the root "/".
+ *
+ * If the result of this process is an empty string, Clean
+ * returns the string ".".
+ *
+ * See also Rob Pike, “Lexical File Names in Plan 9 or
+ * Getting Dot-Dot Right,”
+ * https://9p.io/sys/doc/lexnames.html
+ */
+char *sanitize_path(const char *input);