arch/tile: finish enabling support for TILE-Gx 64-bit chip
This support was partially present in the existing code (look for
"__tilegx__" ifdefs) but with this change you can build a working
kernel using the TILE-Gx toolchain and ARCH=tilegx.
Most of these files are new, generally adding a foo_64.c file
where previously there was just a foo_32.c file.
The ARCH=tilegx directive redirects to arch/tile, not arch/tilegx,
using the existing SRCARCH mechanism in the top-level Makefile.
Changes to existing files:
- <asm/bitops.h> and <asm/bitops_32.h> changed to factor the
include of <asm-generic/bitops/non-atomic.h> in the common header.
- <asm/compat.h> and arch/tile/kernel/compat.c changed to remove
the "const" markers I had put on compat_sys_execve() when trying
to match some recent similar changes to the non-compat execve.
It turns out the compat version wasn't "upgraded" to use const.
- <asm/opcode-tile_64.h> and <asm/opcode_constants_64.h> were
previously included accidentally, with the 32-bit contents. Now
they have the proper 64-bit contents.
Finally, I had to hack the existing hacky drivers/input/input-compat.h
to add yet another "#ifdef" for INPUT_COMPAT_TEST (same as x86_64).
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
Acked-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> [drivers/input]
diff --git a/arch/tile/lib/spinlock_64.c b/arch/tile/lib/spinlock_64.c
new file mode 100644
index 0000000..d6fb958
--- /dev/null
+++ b/arch/tile/lib/spinlock_64.c
@@ -0,0 +1,104 @@
+/*
+ * Copyright 2011 Tilera Corporation. All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, version 2.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT. See the GNU General Public License for
+ * more details.
+ */
+
+#include <linux/spinlock.h>
+#include <linux/module.h>
+#include <asm/processor.h>
+
+#include "spinlock_common.h"
+
+/*
+ * Read the spinlock value without allocating in our cache and without
+ * causing an invalidation to another cpu with a copy of the cacheline.
+ * This is important when we are spinning waiting for the lock.
+ */
+static inline u32 arch_spin_read_noalloc(void *lock)
+{
+ return atomic_cmpxchg((atomic_t *)lock, -1, -1);
+}
+
+/*
+ * Wait until the high bits (current) match my ticket.
+ * If we notice the overflow bit set on entry, we clear it.
+ */
+void arch_spin_lock_slow(arch_spinlock_t *lock, u32 my_ticket)
+{
+ if (unlikely(my_ticket & __ARCH_SPIN_NEXT_OVERFLOW)) {
+ __insn_fetchand4(&lock->lock, ~__ARCH_SPIN_NEXT_OVERFLOW);
+ my_ticket &= ~__ARCH_SPIN_NEXT_OVERFLOW;
+ }
+
+ for (;;) {
+ u32 val = arch_spin_read_noalloc(lock);
+ u32 delta = my_ticket - arch_spin_current(val);
+ if (delta == 0)
+ return;
+ relax((128 / CYCLES_PER_RELAX_LOOP) * delta);
+ }
+}
+EXPORT_SYMBOL(arch_spin_lock_slow);
+
+/*
+ * Check the lock to see if it is plausible, and try to get it with cmpxchg().
+ */
+int arch_spin_trylock(arch_spinlock_t *lock)
+{
+ u32 val = arch_spin_read_noalloc(lock);
+ if (unlikely(arch_spin_current(val) != arch_spin_next(val)))
+ return 0;
+ return cmpxchg(&lock->lock, val, (val + 1) & ~__ARCH_SPIN_NEXT_OVERFLOW)
+ == val;
+}
+EXPORT_SYMBOL(arch_spin_trylock);
+
+void arch_spin_unlock_wait(arch_spinlock_t *lock)
+{
+ u32 iterations = 0;
+ while (arch_spin_is_locked(lock))
+ delay_backoff(iterations++);
+}
+EXPORT_SYMBOL(arch_spin_unlock_wait);
+
+/*
+ * If the read lock fails due to a writer, we retry periodically
+ * until the value is positive and we write our incremented reader count.
+ */
+void __read_lock_failed(arch_rwlock_t *rw)
+{
+ u32 val;
+ int iterations = 0;
+ do {
+ delay_backoff(iterations++);
+ val = __insn_fetchaddgez4(&rw->lock, 1);
+ } while (unlikely(arch_write_val_locked(val)));
+}
+EXPORT_SYMBOL(__read_lock_failed);
+
+/*
+ * If we failed because there were readers, clear the "writer" bit
+ * so we don't block additional readers. Otherwise, there was another
+ * writer anyway, so our "fetchor" made no difference. Then wait,
+ * issuing periodic fetchor instructions, till we get the lock.
+ */
+void __write_lock_failed(arch_rwlock_t *rw, u32 val)
+{
+ int iterations = 0;
+ do {
+ if (!arch_write_val_locked(val))
+ val = __insn_fetchand4(&rw->lock, ~__WRITE_LOCK_BIT);
+ delay_backoff(iterations++);
+ val = __insn_fetchor4(&rw->lock, __WRITE_LOCK_BIT);
+ } while (val != 0);
+}
+EXPORT_SYMBOL(__write_lock_failed);