[ARM] 3881/4: xscale: clean up cp0/cp1 handling

XScale cores either have a DSP coprocessor (which contains a single
40 bit accumulator register), or an iWMMXt coprocessor (which contains
eight 64 bit registers.)

Because of the small amount of state in the DSP coprocessor, access to
the DSP coprocessor (CP0) is always enabled, and DSP context switching
is done unconditionally on every task switch.  Access to the iWMMXt
coprocessor (CP0/CP1) is enabled only when an iWMMXt instruction is
first issued, and iWMMXt context switching is done lazily.

CONFIG_IWMMXT is supposed to mean 'the cpu we will be running on will
have iWMMXt support', but boards are supposed to select this config
symbol by hand, and at least one pxa27x board doesn't get this right,
so on that board, proc-xscale.S will incorrectly assume that we have a
DSP coprocessor, enable CP0 on boot, and we will then only save the
first iWMMXt register (wR0) on context switches, which is Bad.

This patch redefines CONFIG_IWMMXT as 'the cpu we will be running on
might have iWMMXt support, and we will enable iWMMXt context switching
if it does.'  This means that with this patch, running a CONFIG_IWMMXT=n
kernel on an iWMMXt-capable CPU will no longer potentially corrupt iWMMXt
state over context switches, and running a CONFIG_IWMMXT=y kernel on a
non-iWMMXt capable CPU will still do DSP context save/restore.

These changes should make iWMMXt work on PXA3xx, and as a side effect,
enable proper acc0 save/restore on non-iWMMXt capable xsc3 cores such
as IOP13xx and IXP23xx (which will not have CONFIG_CPU_XSCALE defined),
as well as setting and using HWCAP_IWMMXT properly.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Acked-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

1 files changed, 16 insertions, 32 deletions

diff --git a/include/asm-arm/elf.h b/include/asm-arm/elf.h
index f3929307a56b..642382d2c9f0 100644
--- a/include/asm-arm/elf.h
+++ b/include/asm-arm/elf.h
@@ -114,40 +114,24 @@ extern char elf_platform[];
    have no such handler.  */
 #define ELF_PLAT_INIT(_r, load_addr)	(_r)->ARM_r0 = 0
 
-#ifndef CONFIG_IWMMXT
-
-/* Old NetWinder binaries were compiled in such a way that the iBCS
-   heuristic always trips on them.  Until these binaries become uncommon
-   enough not to care, don't trust the `ibcs' flag here.  In any case
-   there is no other ELF system currently supported by iBCS.
-   @@ Could print a warning message to encourage users to upgrade.  */
-#define SET_PERSONALITY(ex,ibcs2) \
-	set_personality(((ex).e_flags & EF_ARM_APCS26 ? PER_LINUX : PER_LINUX_32BIT))
-
-#else
-
 /*
- * All iWMMXt capable CPUs don't support 26-bit mode.  Yet they can run
- * legacy binaries which used to contain FPA11 floating point instructions
- * that have always been emulated by the kernel.  PFA11 and iWMMXt overlap
- * on coprocessor 1 space though.  We therefore must decide if given task
- * is allowed to use CP 0 and 1 for iWMMXt, or if they should be blocked
- * at all times for the prefetch exception handler to catch FPA11 opcodes
- * and emulate them.  The best indication to discriminate those two cases
- * is the SOFT_FLOAT flag in the ELF header.
+ * Since the FPA coprocessor uses CP1 and CP2, and iWMMXt uses CP0
+ * and CP1, we only enable access to the iWMMXt coprocessor if the
+ * binary is EABI or softfloat (and thus, guaranteed not to use
+ * FPA instructions.)
  */
-
-#define SET_PERSONALITY(ex,ibcs2) \
-do { \
-	set_personality(PER_LINUX_32BIT); \
-	if (((ex).e_flags & EF_ARM_EABI_MASK) || \
-	    ((ex).e_flags & EF_ARM_SOFT_FLOAT)) \
-		set_thread_flag(TIF_USING_IWMMXT); \
-	else \
-		clear_thread_flag(TIF_USING_IWMMXT); \
-} while (0)
-
-#endif
+#define SET_PERSONALITY(ex, ibcs2)					\
+	do {								\
+		if ((ex).e_flags & EF_ARM_APCS26) {			\
+			set_personality(PER_LINUX);			\
+		} else {						\
+			set_personality(PER_LINUX_32BIT);		\
+			if (elf_hwcap & HWCAP_IWMMXT && (ex).e_flags & (EF_ARM_EABI_MASK | EF_ARM_SOFT_FLOAT)) \
+				set_thread_flag(TIF_USING_IWMMXT);	\
+			else						\
+				clear_thread_flag(TIF_USING_IWMMXT);	\
+		}							\
+	} while (0)
 
 #endif