mirror of
https://git.proxmox.com/git/mirror_ubuntu-kernels.git
synced 2025-11-23 15:09:36 +00:00
On arm64, ARCH_DMA_MINALIGN is 128, larger than the cache line size on
most of the current platforms (typically 64). Define
ARCH_KMALLOC_MINALIGN to 8 (the default for architectures without their
own ARCH_DMA_MINALIGN) and override dma_get_cache_alignment() to return
cache_line_size(), probed at run-time. The kmalloc() caches will be
limited to the cache line size. This will allow the additional
kmalloc-{64,192} caches on most arm64 platforms.
Link: https://lkml.kernel.org/r/20230612153201.554742-12-catalin.marinas@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Tested-by: Isaac J. Manjarres <isaacmanjarres@google.com>
Cc: Will Deacon <will@kernel.org>
Cc: Alasdair Kergon <agk@redhat.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Jerry Snitselaar <jsnitsel@redhat.com>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Jonathan Cameron <jic23@kernel.org>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Cc: Lars-Peter Clausen <lars@metafoo.de>
Cc: Logan Gunthorpe <logang@deltatee.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Mark Brown <broonie@kernel.org>
Cc: Mike Snitzer <snitzer@kernel.org>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Saravana Kannan <saravanak@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
130 lines
3.4 KiB
C
130 lines
3.4 KiB
C
/* SPDX-License-Identifier: GPL-2.0-only */
|
|
/*
|
|
* Copyright (C) 2012 ARM Ltd.
|
|
*/
|
|
#ifndef __ASM_CACHE_H
|
|
#define __ASM_CACHE_H
|
|
|
|
#define L1_CACHE_SHIFT (6)
|
|
#define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT)
|
|
|
|
#define CLIDR_LOUU_SHIFT 27
|
|
#define CLIDR_LOC_SHIFT 24
|
|
#define CLIDR_LOUIS_SHIFT 21
|
|
|
|
#define CLIDR_LOUU(clidr) (((clidr) >> CLIDR_LOUU_SHIFT) & 0x7)
|
|
#define CLIDR_LOC(clidr) (((clidr) >> CLIDR_LOC_SHIFT) & 0x7)
|
|
#define CLIDR_LOUIS(clidr) (((clidr) >> CLIDR_LOUIS_SHIFT) & 0x7)
|
|
|
|
/* Ctypen, bits[3(n - 1) + 2 : 3(n - 1)], for n = 1 to 7 */
|
|
#define CLIDR_CTYPE_SHIFT(level) (3 * (level - 1))
|
|
#define CLIDR_CTYPE_MASK(level) (7 << CLIDR_CTYPE_SHIFT(level))
|
|
#define CLIDR_CTYPE(clidr, level) \
|
|
(((clidr) & CLIDR_CTYPE_MASK(level)) >> CLIDR_CTYPE_SHIFT(level))
|
|
|
|
/* Ttypen, bits [2(n - 1) + 34 : 2(n - 1) + 33], for n = 1 to 7 */
|
|
#define CLIDR_TTYPE_SHIFT(level) (2 * ((level) - 1) + CLIDR_EL1_Ttypen_SHIFT)
|
|
|
|
/*
|
|
* Memory returned by kmalloc() may be used for DMA, so we must make
|
|
* sure that all such allocations are cache aligned. Otherwise,
|
|
* unrelated code may cause parts of the buffer to be read into the
|
|
* cache before the transfer is done, causing old data to be seen by
|
|
* the CPU.
|
|
*/
|
|
#define ARCH_DMA_MINALIGN (128)
|
|
#define ARCH_KMALLOC_MINALIGN (8)
|
|
|
|
#ifndef __ASSEMBLY__
|
|
|
|
#include <linux/bitops.h>
|
|
#include <linux/kasan-enabled.h>
|
|
|
|
#include <asm/cputype.h>
|
|
#include <asm/mte-def.h>
|
|
#include <asm/sysreg.h>
|
|
|
|
#ifdef CONFIG_KASAN_SW_TAGS
|
|
#define ARCH_SLAB_MINALIGN (1ULL << KASAN_SHADOW_SCALE_SHIFT)
|
|
#elif defined(CONFIG_KASAN_HW_TAGS)
|
|
static inline unsigned int arch_slab_minalign(void)
|
|
{
|
|
return kasan_hw_tags_enabled() ? MTE_GRANULE_SIZE :
|
|
__alignof__(unsigned long long);
|
|
}
|
|
#define arch_slab_minalign() arch_slab_minalign()
|
|
#endif
|
|
|
|
#define CTR_L1IP(ctr) SYS_FIELD_GET(CTR_EL0, L1Ip, ctr)
|
|
|
|
#define ICACHEF_ALIASING 0
|
|
#define ICACHEF_VPIPT 1
|
|
extern unsigned long __icache_flags;
|
|
|
|
/*
|
|
* Whilst the D-side always behaves as PIPT on AArch64, aliasing is
|
|
* permitted in the I-cache.
|
|
*/
|
|
static inline int icache_is_aliasing(void)
|
|
{
|
|
return test_bit(ICACHEF_ALIASING, &__icache_flags);
|
|
}
|
|
|
|
static __always_inline int icache_is_vpipt(void)
|
|
{
|
|
return test_bit(ICACHEF_VPIPT, &__icache_flags);
|
|
}
|
|
|
|
static inline u32 cache_type_cwg(void)
|
|
{
|
|
return SYS_FIELD_GET(CTR_EL0, CWG, read_cpuid_cachetype());
|
|
}
|
|
|
|
#define __read_mostly __section(".data..read_mostly")
|
|
|
|
static inline int cache_line_size_of_cpu(void)
|
|
{
|
|
u32 cwg = cache_type_cwg();
|
|
|
|
return cwg ? 4 << cwg : ARCH_DMA_MINALIGN;
|
|
}
|
|
|
|
int cache_line_size(void);
|
|
|
|
#define dma_get_cache_alignment cache_line_size
|
|
|
|
/*
|
|
* Read the effective value of CTR_EL0.
|
|
*
|
|
* According to ARM ARM for ARMv8-A (ARM DDI 0487C.a),
|
|
* section D10.2.33 "CTR_EL0, Cache Type Register" :
|
|
*
|
|
* CTR_EL0.IDC reports the data cache clean requirements for
|
|
* instruction to data coherence.
|
|
*
|
|
* 0 - dcache clean to PoU is required unless :
|
|
* (CLIDR_EL1.LoC == 0) || (CLIDR_EL1.LoUIS == 0 && CLIDR_EL1.LoUU == 0)
|
|
* 1 - dcache clean to PoU is not required for i-to-d coherence.
|
|
*
|
|
* This routine provides the CTR_EL0 with the IDC field updated to the
|
|
* effective state.
|
|
*/
|
|
static inline u32 __attribute_const__ read_cpuid_effective_cachetype(void)
|
|
{
|
|
u32 ctr = read_cpuid_cachetype();
|
|
|
|
if (!(ctr & BIT(CTR_EL0_IDC_SHIFT))) {
|
|
u64 clidr = read_sysreg(clidr_el1);
|
|
|
|
if (CLIDR_LOC(clidr) == 0 ||
|
|
(CLIDR_LOUIS(clidr) == 0 && CLIDR_LOUU(clidr) == 0))
|
|
ctr |= BIT(CTR_EL0_IDC_SHIFT);
|
|
}
|
|
|
|
return ctr;
|
|
}
|
|
|
|
#endif /* __ASSEMBLY__ */
|
|
|
|
#endif
|