LoongArch: Adjust symbol addressing for CONFIG_AS_HAS_EXPLICIT_RELOCS

If explicit relocation hints is used by the toolchain, -Wa,-mla-*
options will be useless for C code.  Only use them for
!CONFIG_AS_HAS_EXPLICIT_RELOCS.

Replace "la" with "la.pcrel" in head.S to keep the semantic consistent
with new and old toolchains for the low level startup code.

For per-CPU variables, the "address" of the symbol is actually an
offset from $r21.  The value is nearing the loading address of main
kernel image, but far from the address of modules.  Use model("extreme")
attibute to tell the compiler that a  a PC-relative addressing with
32-bit offset is not sufficient for local per-CPU variables.

The behavior with different assemblers and compilers are summarized in
the following table:

AS has            CC has
explicit reloc    explicit reloc *  Behavior
==============================================================
No                No                Use la.* macros.
                                    No change from Linux 6.0.
--------------------------------------------------------------
No                Yes               Disable explicit reloc.
                                    No change from Linux 6.0.
--------------------------------------------------------------
Yes               No                Not supported.
--------------------------------------------------------------
Yes               Yes               Use explicit relocs.
                                    No -Wa,-mla* options.
==============================================================
*: We assume CC must have model attribute if it has explicit reloc.
   Both features are added in GCC 13 development cycle, so any GCC
   release >= 13 should be OK.  Using early GCC 13 development snapshots
   may produce modules with unsupported relocations.

Link: https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=f09482a
Link: https://gcc.gnu.org/r13-1834
Link: https://gcc.gnu.org/r13-2199
Signed-off-by: Xi Ruoyao <[email protected]>

arch/loongarch/Makefile

@@ -40,10 +40,27 @@ endif
 
 cflags-y			+= -G0 -pipe -msoft-float
 LDFLAGS_vmlinux			+= -G0 -static -n -nostdlib
+
+# When the assembler supports explicit relocation hint, we must use it.
+# GCC may have -mexplicit-relocs off by default if it was built with an old
+# assembler, so we force it via an option.
+#
+# When the assembler does not supports explicit relocation hint, we can't use
+# it.  Disable it if the compiler supports it.
+#
+# If you've seen "unknown reloc hint" message building the kernel and you are
+# now wondering why "-mexplicit-relocs" is not wrapped with cc-option: the
+# combination of a "new" assembler and "old" compiler is not supported.  Either
+# upgrade the compiler or downgrade the assembler.
+ifdef CONFIG_AS_HAS_EXPLICIT_RELOCS
+cflags-y			+= -mexplicit-relocs
+else
+cflags-y			+= $(call cc-option,-mno-explicit-relocs)
 KBUILD_AFLAGS_KERNEL		+= -Wa,-mla-global-with-pcrel
 KBUILD_CFLAGS_KERNEL		+= -Wa,-mla-global-with-pcrel
 KBUILD_AFLAGS_MODULE		+= -Wa,-mla-global-with-abs
 KBUILD_CFLAGS_MODULE		+= -fplt -Wa,-mla-global-with-abs,-mla-local-with-abs
+endif
 
 cflags-y += -ffreestanding
 cflags-y += $(call cc-option, -mno-check-zero-division)

arch/loongarch/include/asm/percpu.h

@@ -8,6 +8,14 @@
 #include <asm/cmpxchg.h>
 #include <asm/loongarch.h>
 
+#if defined(MODULE) && defined(CONFIG_AS_HAS_EXPLICIT_RELOCS)
+/* The "address" (in fact, offset from $r21) of a per-CPU variable is close
+ * to the load address of main kernel image, but far from where the modules are
+ * loaded.  Tell the compiler this fact.
+ */
+# define PER_CPU_ATTRIBUTES __attribute__((model("extreme")))
+#endif
+
 /* Use r21 for fast access */
 register unsigned long __my_cpu_offset __asm__("$r21");
 

arch/loongarch/kernel/head.S

@@ -55,25 +55,25 @@ SYM_CODE_START(kernel_entry)			# kernel entry point
 	li.w		t0, 0x00		# FPE=0, SXE=0, ASXE=0, BTE=0
 	csrwr		t0, LOONGARCH_CSR_EUEN
 
-	la		t0, __bss_start		# clear .bss
+	la.pcrel	t0, __bss_start		# clear .bss
 	st.d		zero, t0, 0
-	la		t1, __bss_stop - LONGSIZE
+	la.pcrel	t1, __bss_stop - LONGSIZE
 1:
 	addi.d		t0, t0, LONGSIZE
 	st.d		zero, t0, 0
 	bne		t0, t1, 1b
 
-	la		t0, fw_arg0
+	la.pcrel	t0, fw_arg0
 	st.d		a0, t0, 0		# firmware arguments
-	la		t0, fw_arg1
+	la.pcrel	t0, fw_arg1
 	st.d		a1, t0, 0
 
 	/* KSave3 used for percpu base, initialized as 0 */
 	csrwr		zero, PERCPU_BASE_KS
 	/* GPR21 used for percpu base (runtime), initialized as 0 */
 	move		u0, zero
 
-	la		tp, init_thread_union
+	la.pcrel	tp, init_thread_union
 	/* Set the SP after an empty pt_regs.  */
 	PTR_LI		sp, (_THREAD_SIZE - 32 - PT_SIZE)
 	PTR_ADD		sp, sp, tp