linux_dsm_epyc7002/drivers/gpu/drm/i915/i915_memcpy.h
Chris Wilson 6aacb5a3b0 drm/i915: Align start for memcpy_from_wc
The movntqda requires 16-byte alignment for the source pointer. Avoid
falling back to clflush if the source pointer is misaligned by doing the
doing a small uncached memcpy to fixup the alignments.

v2: Turn the unaligned copy into a genuine helper

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191211110437.4082687-5-chris@chris-wilson.co.uk
2019-12-11 22:40:41 +00:00

35 lines
1.2 KiB
C

/* SPDX-License-Identifier: MIT */
/*
* Copyright © 2019 Intel Corporation
*/
#ifndef __I915_MEMCPY_H__
#define __I915_MEMCPY_H__
#include <linux/types.h>
struct drm_i915_private;
void i915_memcpy_init_early(struct drm_i915_private *i915);
bool i915_memcpy_from_wc(void *dst, const void *src, unsigned long len);
void i915_unaligned_memcpy_from_wc(void *dst, void *src, unsigned long len);
/* The movntdqa instructions used for memcpy-from-wc require 16-byte alignment,
* as well as SSE4.1 support. i915_memcpy_from_wc() will report if it cannot
* perform the operation. To check beforehand, pass in the parameters to
* to i915_can_memcpy_from_wc() - since we only care about the low 4 bits,
* you only need to pass in the minor offsets, page-aligned pointers are
* always valid.
*
* For just checking for SSE4.1, in the foreknowledge that the future use
* will be correctly aligned, just use i915_has_memcpy_from_wc().
*/
#define i915_can_memcpy_from_wc(dst, src, len) \
i915_memcpy_from_wc((void *)((unsigned long)(dst) | (unsigned long)(src) | (len)), NULL, 0)
#define i915_has_memcpy_from_wc() \
i915_memcpy_from_wc(NULL, NULL, 0)
#endif /* __I915_MEMCPY_H__ */