c hello world works

README.adoc

@@ -239,8 +239,6 @@ Source: link:bios_hello_world.S[]
 
 ==== C hello world
 
-TODO get working.
-
 Same output as <<bios-hello-world>>, but written in C:
 
 ....
@@ -250,7 +248,13 @@ cd c_hello_world
 
 Source: link:c_hello_world[]
 
-Single stage, so still limited to 512 bytes of code + data!
+But keep in mind the following limitations and difficulties:
+
+* single stage, so still limited to 512 bytes of code + data!
+* use use GCC's `-m` which does not produce "real" 16 bit code, but rather 32-bit code with `0x66` and `0x67` prefixes: https://wiki.osdev.org/X86-64_Instruction_Encoding#Legacy_Prefixes
+* setting up the initial state and the linker script is much harder and error prone than with assembly
+
+Therefore, for most applications, you will just want to use <<multiboot>> instead, which overcomes all of those problems.
 
 To disassemble the generated C code, try:
 
@@ -258,11 +262,20 @@ To disassemble the generated C code, try:
 objdump -D -m i8086 main.elf
 ....
 
-TODO: I see several `eax` references still, why?
+but note that it still contains references to 32-bit references, e.g.:
+
+....
+00007c17 <main>:
+    7c17:       66 55                   push   %ebp
+    7c19:       66 89 e5                mov    %esp,%ebp
+    7c1c:       66 83 ec 10             sub    $0x10,%esp
+....
+
+This is because those instructions are modified by the prefix `0x66`, which makes them behave like 32-bit.
 
 === No linker script
 
-Print `hello world` without using an explitic linker script:
+Print `hello world` without using an explicit linker script:
 
 ....
 make -C no-linker-script run

c_hello_world/clean

@@ -0,0 +1,2 @@
+#!/usr/bin/env bash
+rm -f *.o *.elf *.img

c_hello_world/entry.S

@@ -2,5 +2,12 @@
 .text
 .global mystart
 mystart:
-    mov $__stack_top, %sp
+    ljmp $0, $.setcs
+.setcs:
+    xor %ax, %ax
+    mov %ax, %ds
+    mov %ax, %es
+    mov %ax, %ss
+    mov $__stack_top, %esp
+    cld
     call main

c_hello_world/linker.ld

@@ -8,7 +8,9 @@ SECTIONS
     *(.data)
     *(.rodata)
     __bss_start = .;
+    /* COMMON vs BSS: https://stackoverflow.com/questions/16835716/bss-vs-common-what-goes-where */
     *(.bss)
+    *(COMMON)
     __bss_end = .;
   }
   /* https://stackoverflow.com/questions/53584666/why-does-gnu-ld-include-a-section-that-does-not-appear-in-the-linker-script */

c_hello_world/main.c

@@ -1,18 +1,11 @@
 void main(void) {
     int i;
-    char s[] = "hello world";
+    char s[] = {'h', 'e', 'l', 'l', 'o', ' ', 'w', 'o', 'r', 'l', 'd'};
-    /* Add a bunch of debug prints to see if the control loop is correct. It is. */
-    __asm__ ("mov $0x0E40, %ax; int $0x10");
     for (i = 0; i < sizeof(s); ++i) {
-        __asm__ ("mov $0x0E41, %ax; int $0x10");
         __asm__ (
-            "mov %0, %%ax; int $0x10"
+            "int $0x10" : : "a" ((0x0e << 8) | s[i])
-            :
-            : "m" (s[i])
-            : "%ax"
         );
     }
-    __asm__ ("mov $0x0E40, %ax; int $0x10");
     while (1) {
         __asm__ ("hlt");
     };

c_hello_world/run

@@ -1,7 +1,7 @@
 #!/usr/bin/env bash
 set -eux
 as -ggdb3 --32 -o entry.o entry.S
-gcc -c -ggdb3 -m16 -nostartfiles -nostdlib -o main.o -std=c99 main.c
+gcc -c -ggdb3 -m16 -ffreestanding -fno-PIE -nostartfiles -nostdlib -o main.o -std=c99 main.c
 ld -m elf_i386 -o main.elf -T linker.ld entry.o main.o
 objcopy -O binary main.elf main.img
 qemu-system-x86_64 -drive file=main.img,format=raw