mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-05 07:16:56 +07:00
dd7b2f035e
On 64-bit CPUs with no-execute support and non-snooping icache, such as
970 or POWER4, we have a software mechanism to ensure coherency of the
cache (using exec faults when needed).
This was broken due to a logic error when the code was rewritten
from assembly to C, previously the assembly code did:
BEGIN_FTR_SECTION
mr r4,r30
mr r5,r7
bl hash_page_do_lazy_icache
END_FTR_SECTION(CPU_FTR_NOEXECUTE|CPU_FTR_COHERENT_ICACHE, CPU_FTR_NOEXECUTE)
Which tests that:
(cpu_features & (NOEXECUTE | COHERENT_ICACHE)) == NOEXECUTE
Which says that the current cpu does have NOEXECUTE, but does not have
COHERENT_ICACHE.
Fixes: 91f1da9979 ("powerpc/mm: Convert 4k hash insert to C")
Fixes: 89ff725051 ("powerpc/mm: Convert __hash_page_64K to C")
Fixes: a43c0eb836 ("powerpc/mm: Convert 4k insert from asm to C")
Cc: stable@vger.kernel.org # v4.5+
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
[mpe: Change log verbosification]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
127 lines
3.7 KiB
C
127 lines
3.7 KiB
C
/*
|
|
* Copyright IBM Corporation, 2015
|
|
* Author Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
|
|
*
|
|
* This program is free software; you can redistribute it and/or modify it
|
|
* under the terms of version 2 of the GNU Lesser General Public License
|
|
* as published by the Free Software Foundation.
|
|
*
|
|
* This program is distributed in the hope that it would be useful, but
|
|
* WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
|
|
*
|
|
*/
|
|
|
|
#include <linux/mm.h>
|
|
#include <asm/machdep.h>
|
|
#include <asm/mmu.h>
|
|
|
|
int __hash_page_4K(unsigned long ea, unsigned long access, unsigned long vsid,
|
|
pte_t *ptep, unsigned long trap, unsigned long flags,
|
|
int ssize, int subpg_prot)
|
|
{
|
|
unsigned long hpte_group;
|
|
unsigned long rflags, pa;
|
|
unsigned long old_pte, new_pte;
|
|
unsigned long vpn, hash, slot;
|
|
unsigned long shift = mmu_psize_defs[MMU_PAGE_4K].shift;
|
|
|
|
/*
|
|
* atomically mark the linux large page PTE busy and dirty
|
|
*/
|
|
do {
|
|
pte_t pte = READ_ONCE(*ptep);
|
|
|
|
old_pte = pte_val(pte);
|
|
/* If PTE busy, retry the access */
|
|
if (unlikely(old_pte & H_PAGE_BUSY))
|
|
return 0;
|
|
/* If PTE permissions don't match, take page fault */
|
|
if (unlikely(!check_pte_access(access, old_pte)))
|
|
return 1;
|
|
/*
|
|
* Try to lock the PTE, add ACCESSED and DIRTY if it was
|
|
* a write access. Since this is 4K insert of 64K page size
|
|
* also add H_PAGE_COMBO
|
|
*/
|
|
new_pte = old_pte | H_PAGE_BUSY | _PAGE_ACCESSED;
|
|
if (access & _PAGE_WRITE)
|
|
new_pte |= _PAGE_DIRTY;
|
|
} while (!pte_xchg(ptep, __pte(old_pte), __pte(new_pte)));
|
|
|
|
/*
|
|
* PP bits. _PAGE_USER is already PP bit 0x2, so we only
|
|
* need to add in 0x1 if it's a read-only user page
|
|
*/
|
|
rflags = htab_convert_pte_flags(new_pte);
|
|
|
|
if (cpu_has_feature(CPU_FTR_NOEXECUTE) &&
|
|
!cpu_has_feature(CPU_FTR_COHERENT_ICACHE))
|
|
rflags = hash_page_do_lazy_icache(rflags, __pte(old_pte), trap);
|
|
|
|
vpn = hpt_vpn(ea, vsid, ssize);
|
|
if (unlikely(old_pte & H_PAGE_HASHPTE)) {
|
|
/*
|
|
* There MIGHT be an HPTE for this pte
|
|
*/
|
|
hash = hpt_hash(vpn, shift, ssize);
|
|
if (old_pte & H_PAGE_F_SECOND)
|
|
hash = ~hash;
|
|
slot = (hash & htab_hash_mask) * HPTES_PER_GROUP;
|
|
slot += (old_pte & H_PAGE_F_GIX) >> H_PAGE_F_GIX_SHIFT;
|
|
|
|
if (mmu_hash_ops.hpte_updatepp(slot, rflags, vpn, MMU_PAGE_4K,
|
|
MMU_PAGE_4K, ssize, flags) == -1)
|
|
old_pte &= ~_PAGE_HPTEFLAGS;
|
|
}
|
|
|
|
if (likely(!(old_pte & H_PAGE_HASHPTE))) {
|
|
|
|
pa = pte_pfn(__pte(old_pte)) << PAGE_SHIFT;
|
|
hash = hpt_hash(vpn, shift, ssize);
|
|
|
|
repeat:
|
|
hpte_group = ((hash & htab_hash_mask) * HPTES_PER_GROUP) & ~0x7UL;
|
|
|
|
/* Insert into the hash table, primary slot */
|
|
slot = mmu_hash_ops.hpte_insert(hpte_group, vpn, pa, rflags, 0,
|
|
MMU_PAGE_4K, MMU_PAGE_4K, ssize);
|
|
/*
|
|
* Primary is full, try the secondary
|
|
*/
|
|
if (unlikely(slot == -1)) {
|
|
hpte_group = ((~hash & htab_hash_mask) * HPTES_PER_GROUP) & ~0x7UL;
|
|
slot = mmu_hash_ops.hpte_insert(hpte_group, vpn, pa,
|
|
rflags,
|
|
HPTE_V_SECONDARY,
|
|
MMU_PAGE_4K,
|
|
MMU_PAGE_4K, ssize);
|
|
if (slot == -1) {
|
|
if (mftb() & 0x1)
|
|
hpte_group = ((hash & htab_hash_mask) *
|
|
HPTES_PER_GROUP) & ~0x7UL;
|
|
mmu_hash_ops.hpte_remove(hpte_group);
|
|
/*
|
|
* FIXME!! Should be try the group from which we removed ?
|
|
*/
|
|
goto repeat;
|
|
}
|
|
}
|
|
/*
|
|
* Hypervisor failure. Restore old pte and return -1
|
|
* similar to __hash_page_*
|
|
*/
|
|
if (unlikely(slot == -2)) {
|
|
*ptep = __pte(old_pte);
|
|
hash_failure_debug(ea, access, vsid, trap, ssize,
|
|
MMU_PAGE_4K, MMU_PAGE_4K, old_pte);
|
|
return -1;
|
|
}
|
|
new_pte = (new_pte & ~_PAGE_HPTEFLAGS) | H_PAGE_HASHPTE;
|
|
new_pte |= (slot << H_PAGE_F_GIX_SHIFT) &
|
|
(H_PAGE_F_SECOND | H_PAGE_F_GIX);
|
|
}
|
|
*ptep = __pte(new_pte & ~H_PAGE_BUSY);
|
|
return 0;
|
|
}
|