feature (syscalls): implemented syscall arguments and return value

kernel/inc/platform/syscall.h

@@ -25,19 +25,54 @@
 #define NOX_SYSCALL_H
 
 #include "platform/cpu.h"
+#include "utils/status.h"
 
 typedef enum {
-    SYSCALLS_MISC   = 0x00,
+    SYSCALLS_FILES          = 0x00,
-    SYSCALLS_FILE   = 0x01,
+    SYSCALLS_MEMORY         = 0x01,
-    SYSCALLS_PROC   = 0x02,
+    SYSCALLS_PROC           = 0x02,
-    SYSCALLS_KERNEL = 0xFF,
+    SYSCALLS_RUNTIME_LINKER = 0x02,
+    SYSCALLS_COMPATABILITY  = 0x02,
+
+    SYSCALLS_KERNEL         = 0xFF
 }syscall_group_E;
 
 typedef enum {
-    SYSCALL_KERNEL_SCHEDULER_START  = 0xFF00
+    SYSCALL_FILES_OPEN                  = 0x0101,
+    SYSCALL_FILES_CLOSE                 = 0x0102,
+    SYSCALL_FILES_READ                  = 0x0103,
+    SYSCALL_FILES_WRITE                 = 0x0104,
+    SYSCALL_FILES_DELETE                = 0x0105,
+    SYSCALL_FILES_LIST                  = 0x0106,
+    SYSCALL_FILES_FOLLOW                = 0x0107,   // Warning: nx_follow will probably be kicked out of the ABI
+    SYSCALL_FILES_INFO                  = 0x0108,
+
+    SYSCALL_MEMORY_ALLOC                = 0x0201,
+    SYSCALL_MEMORY_FREE                 = 0x0202,
+    SYSCALL_MEMORY_LABEL                = 0x0203,
+    SYSCALL_MEMORY_RANGE                = 0x0204,
+    SYSCALL_MEMORY_ACCESS               = 0x0205,
+
+    SYSCALL_PROCESS_CREATE              = 0x0301,
+    SYSCALL_PROCESS_ENVFILE             = 0x0302,
+    SYSCALL_PROCESS_START               = 0x0303,
+    SYSCALL_PROCESS_SIGNAL              = 0x0304,
+
+    SYSCALL_RUNTIME_LINKER_OPEN         = 0x0401,
+    SYSCALL_RUNTIME_LINKER_CLOSE        = 0x0402,
+    SYSCALL_RUNTIME_LINKER_LOAD_SYMBOL  = 0x0403,
+    SYSCALL_RUNTIME_LINKER_STATUS       = 0x0404,
+    SYSCALL_RUNTIME_LINKER_STANDARD_MOD = 0x0405,
+
+    SYSCALL_COMPATABILITY_ABI_TYPE      = 0x0501,
+    SYSCALL_COMPATABILITY_ABI_VERSION   = 0x0502,
+    SYSCALL_COMPATABILITY_ACTION        = 0x0503,
+
+    SYSCALL_KERNEL_SCHEDULER_START      = 0xFF00,
+    SYSCALL_KERNEL_PANIC                = 0xFF01
 } syscall_E;
 
-extern void syscall_perform(syscall_E id);
+extern status_E syscall_perform(syscall_E id, uint64_t arg1, uint64_t arg2, uint64_t arg3, uint64_t arg4);
 
 cpu_state_T* syscall_handle(cpu_state_T* state);
 

kernel/inc/utils/status.h

@@ -0,0 +1,38 @@
+/*
+ * Copyright 2023 Antifallobst <antifallobst@systemausfall.org>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy of
+ * this software and associated documentation files (the “Software”), to deal in
+ * the Software without restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
+ * Software, and to permit persons to whom the Software is furnished to do so,
+ * subject to the following conditions:
+ * 
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ * 
+ * THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef NOX_STATUS_H
+#define NOX_STATUS_H
+
+#include "string.h"
+
+typedef enum {
+    STATUS_SUCCESS,
+
+    STATUS_PERMISSION_DENIED,
+    STATUS_RESOURCE_NOT_AVAILABLE
+} status_E;
+
+extern string_t g_status_code_strings[3];
+
+#endif //NOX_STATUS_H

kernel/src/kmain.c

@@ -35,7 +35,7 @@
 #include "drivers/fs/vfs.h"
 #include "proc/scheduler.h"
 
-#include "utils/io.h"
+#include "platform/syscall.h"
 
 void limine_terminal_print(boot_info_T* boot_info, string_t string) {
     boot_info->terminal->write(boot_info->terminal->terminals[0], string, string_length(string));
@@ -72,23 +72,21 @@ void kmain(boot_info_T boot_info) {
 
     log(LOG_INFO, "!=====[ Kernel Initialized ]=====!\n");
 
-    page_map_dump_info(g_kernel_page_map);
+    status_E status = syscall_perform(SYSCALL_FILES_OPEN, (uint64_t)"/initrd/test.txt", 0, 0, 0);
 
-    vfs_node_T* node = vfs_resolve_path(&g_root_fs, "/initrd/test.elf");
+    log(LOG_DEBUG, "syscall returned status: %s", g_status_code_strings[status]);
-    void* buffer = memory_allocate(node->size);
-    vfs_file_read(node, 0, node->size, buffer);
 
-    elf_executable_T* exec = elf_executable_create(buffer);
+//    vfs_node_T* node = vfs_resolve_path(&g_root_fs, "/initrd/test.elf");
-
+//    void* buffer = memory_allocate(node->size);
-    pid_t   process = process_spawn(PROCESS_KERNEL, "test 1", exec, buffer);
+//    vfs_file_read(node, 0, node->size, buffer);
-    void* func = (void*)symbol_resolve_from_name(scheduler_get_process(process)->executable->symbols,
+//
-                                          scheduler_get_process(process)->executable->num_symbols,
+//    elf_executable_T* exec = elf_executable_create(buffer);
-                                          "_start")->address + MEM_REGION_PROCESS_EXEC;
+//
-    thread_start(thread_spawn(process, func));
+//    pid_t   process = process_spawn(PROCESS_KERNEL, "test 1", exec, buffer);
-
+//    void* func = (void*)symbol_resolve_from_name(scheduler_get_process(process)->executable->symbols,
-    while (1){
+//                                          scheduler_get_process(process)->executable->num_symbols,
-        io_out_byte(0x3F8, 'B');
+//                                          "_start")->address + MEM_REGION_PROCESS_EXEC;
-    }
+//    thread_start(thread_spawn(process, func));
 
 //    scheduler_dump_info(scheduler_get_process(PROCESS_KERNEL), 0);
 

kernel/src/platform/syscall.c

@@ -24,30 +24,53 @@
 #include "platform/syscall.h"
 #include "utils/logger.h"
 #include "proc/scheduler.h"
+#include "drivers/fs/vfs.h"
 
 cpu_state_T* syscall_handle(cpu_state_T* state) {
     cpu_state_T*    return_state = state;
     syscall_group_E group_id     = (state->rax & 0xFF00) >> 8;
 
-//    if (group_id == SYSCALLS_KERNEL &&
+    if (group_id == SYSCALLS_KERNEL &&
-//        scheduler_is_initialized() &&
+        scheduler_is_initialized() &&
-//        scheduler_get_current_process() != PROC_KERNEL)
+        scheduler_get_current_process()->id != PROCESS_KERNEL)
-//    {
+    {
-//        log(LOG_WARNING, "non kernel process[%d:%d] tried to perform a kernel syscall (permission denied)", scheduler_get_current_process(), scheduler_get_current_thread());
+        log(LOG_WARNING, "non kernel process[%d:%d] tried to perform a kernel syscall (permission denied)", scheduler_get_current_process(), scheduler_get_current_thread());
-//        return return_state;
+        return return_state;
-//    }
+    }
+
+    // read arguments
+    uint64_t arg1 = state->rdi;
+    uint64_t arg2 = state->rsi;
+    uint64_t arg3 = state->rdx;
+    uint64_t arg4 = state->rcx;
+    uint64_t ret  = STATUS_SUCCESS;
 
     switch (state->rax) {
+        case SYSCALL_FILES_OPEN: {
+            vfs_node_T* node = vfs_resolve_path(&g_root_fs, (string_t)arg1);
+            if (node == NULL) {
+                ret = STATUS_RESOURCE_NOT_AVAILABLE;
+                break;
+            }
+
+            log(LOG_DEBUG, "resolved file '%s' -> 0x%x", arg1, node);
+
+            break;
+        }
+
         case SYSCALL_KERNEL_SCHEDULER_START: {
             return_state = scheduler_start(state);
             break;
         }
 
         default: {
-            log(LOG_WARNING, "Unhandled syscall: 0x%xw    RAX[0x%x]", state->rax, state->rax);
+            log(LOG_WARNING, "unhandled syscall: 0x%xw", state->rax);
             break;
         }
     }
 
+    // apply return value
+    state->rax = ret;
+
     return return_state;
 }

kernel/src/platform/x86/syscall.asm

@@ -20,7 +20,18 @@
 ; OTHER DEALINGS IN THE SOFTWARE.
 
 syscall_perform:
+    ; set the interrupt ID
     mov rax, rdi
+
+    ; set the arguments
+    mov rdi, rsi
+    mov rsi, rdx
+    mov rdx, rcx
+    mov rcx, r8
+
     int 0x80
+
+    ; the return value is already set in RAX
+
     ret
 GLOBAL syscall_perform

kernel/src/proc/scheduler.c

@@ -78,7 +78,7 @@ void scheduler_init(boot_info_T* boot_info) {
 
     process_kernel_spawn(elf_executable_create(boot_info->kernel_file->address));
 
-    syscall_perform(SYSCALL_KERNEL_SCHEDULER_START);
+    syscall_perform(SYSCALL_KERNEL_SCHEDULER_START, 0, 0, 0, 0);
 }
 
 cpu_state_T* scheduler_start(cpu_state_T* state) {

kernel/src/utils/status.c

@@ -0,0 +1,31 @@
+/*
+ * Copyright 2023 Antifallobst <antifallobst@systemausfall.org>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy of
+ * this software and associated documentation files (the “Software”), to deal in
+ * the Software without restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
+ * Software, and to permit persons to whom the Software is furnished to do so,
+ * subject to the following conditions:
+ * 
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ * 
+ * THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include "utils/status.h"
+
+string_t g_status_code_strings[3] = {
+        "Success",
+
+        "Permission Denied",
+        "Resource Not Available"
+};