linux_dsm_epyc7002/arch/mips/lib/dump_tlb.c
Paul Burton c8790d657b
MIPS: MemoryMapID (MMID) Support
Introduce support for using MemoryMapIDs (MMIDs) as an alternative to
Address Space IDs (ASIDs). The major difference between the two is that
MMIDs are global - ie. an MMID uniquely identifies an address space
across all coherent CPUs. In contrast ASIDs are non-global per-CPU IDs,
wherein each address space is allocated a separate ASID for each CPU
upon which it is used. This global namespace allows a new GINVT
instruction be used to globally invalidate TLB entries associated with a
particular MMID across all coherent CPUs in the system, removing the
need for IPIs to invalidate entries with separate ASIDs on each CPU.

The allocation scheme used here is largely borrowed from arm64 (see
arch/arm64/mm/context.c). In essence we maintain a bitmap to track
available MMIDs, and MMIDs in active use at the time of a rollover to a
new MMID version are preserved in the new version. The allocation scheme
requires efficient 64 bit atomics in order to perform reasonably, so
this support depends upon CONFIG_GENERIC_ATOMIC64=n (ie. currently it
will only be included in MIPS64 kernels).

The first, and currently only, available CPU with support for MMIDs is
the MIPS I6500. This CPU supports 16 bit MMIDs, and so for now we cap
our MMIDs to 16 bits wide in order to prevent the bitmap growing to
absurd sizes if any future CPU does implement 32 bit MMIDs as the
architecture manuals suggest is recommended.

When MMIDs are in use we also make use of GINVT instruction which is
available due to the global nature of MMIDs. By executing a sequence of
GINVT & SYNC 0x14 instructions we can avoid the overhead of an IPI to
each remote CPU in many cases. One complication is that GINVT will
invalidate wired entries (in all cases apart from type 0, which targets
the entire TLB). In order to avoid GINVT invalidating any wired TLB
entries we set up, we make sure to create those entries using a reserved
MMID (0) that we never associate with any address space.

Also of note is that KVM will require further work in order to support
MMIDs & GINVT, since KVM is involved in allocating IDs for guests & in
configuring the MMU. That work is not part of this patch, so for now
when MMIDs are in use KVM is disabled.

Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org
2019-02-04 10:56:41 -08:00

201 lines
5.4 KiB
C

// SPDX-License-Identifier: GPL-2.0
/*
* Dump R4x00 TLB for debugging purposes.
*
* Copyright (C) 1994, 1995 by Waldorf Electronics, written by Ralf Baechle.
* Copyright (C) 1999 by Silicon Graphics, Inc.
*/
#include <linux/kernel.h>
#include <linux/mm.h>
#include <asm/hazards.h>
#include <asm/mipsregs.h>
#include <asm/mmu_context.h>
#include <asm/page.h>
#include <asm/pgtable.h>
#include <asm/tlbdebug.h>
void dump_tlb_regs(void)
{
const int field = 2 * sizeof(unsigned long);
pr_info("Index : %0x\n", read_c0_index());
pr_info("PageMask : %0x\n", read_c0_pagemask());
if (cpu_has_guestid)
pr_info("GuestCtl1: %0x\n", read_c0_guestctl1());
pr_info("EntryHi : %0*lx\n", field, read_c0_entryhi());
pr_info("EntryLo0 : %0*lx\n", field, read_c0_entrylo0());
pr_info("EntryLo1 : %0*lx\n", field, read_c0_entrylo1());
pr_info("Wired : %0x\n", read_c0_wired());
switch (current_cpu_type()) {
case CPU_R10000:
case CPU_R12000:
case CPU_R14000:
case CPU_R16000:
pr_info("FrameMask: %0x\n", read_c0_framemask());
break;
}
if (cpu_has_small_pages || cpu_has_rixi || cpu_has_xpa)
pr_info("PageGrain: %0x\n", read_c0_pagegrain());
if (cpu_has_htw) {
pr_info("PWField : %0*lx\n", field, read_c0_pwfield());
pr_info("PWSize : %0*lx\n", field, read_c0_pwsize());
pr_info("PWCtl : %0x\n", read_c0_pwctl());
}
}
static inline const char *msk2str(unsigned int mask)
{
switch (mask) {
case PM_4K: return "4kb";
case PM_16K: return "16kb";
case PM_64K: return "64kb";
case PM_256K: return "256kb";
#ifdef CONFIG_CPU_CAVIUM_OCTEON
case PM_8K: return "8kb";
case PM_32K: return "32kb";
case PM_128K: return "128kb";
case PM_512K: return "512kb";
case PM_2M: return "2Mb";
case PM_8M: return "8Mb";
case PM_32M: return "32Mb";
#endif
#ifndef CONFIG_CPU_VR41XX
case PM_1M: return "1Mb";
case PM_4M: return "4Mb";
case PM_16M: return "16Mb";
case PM_64M: return "64Mb";
case PM_256M: return "256Mb";
case PM_1G: return "1Gb";
#endif
}
return "";
}
static void dump_tlb(int first, int last)
{
unsigned long s_entryhi, entryhi, asid, mmid;
unsigned long long entrylo0, entrylo1, pa;
unsigned int s_index, s_pagemask, s_guestctl1 = 0;
unsigned int pagemask, guestctl1 = 0, c0, c1, i;
unsigned long asidmask = cpu_asid_mask(&current_cpu_data);
int asidwidth = DIV_ROUND_UP(ilog2(asidmask) + 1, 4);
unsigned long uninitialized_var(s_mmid);
#ifdef CONFIG_32BIT
bool xpa = cpu_has_xpa && (read_c0_pagegrain() & PG_ELPA);
int pwidth = xpa ? 11 : 8;
int vwidth = 8;
#else
bool xpa = false;
int pwidth = 11;
int vwidth = 11;
#endif
s_pagemask = read_c0_pagemask();
s_entryhi = read_c0_entryhi();
s_index = read_c0_index();
if (cpu_has_mmid)
asid = s_mmid = read_c0_memorymapid();
else
asid = s_entryhi & asidmask;
if (cpu_has_guestid)
s_guestctl1 = read_c0_guestctl1();
for (i = first; i <= last; i++) {
write_c0_index(i);
mtc0_tlbr_hazard();
tlb_read();
tlb_read_hazard();
pagemask = read_c0_pagemask();
entryhi = read_c0_entryhi();
entrylo0 = read_c0_entrylo0();
entrylo1 = read_c0_entrylo1();
if (cpu_has_mmid)
mmid = read_c0_memorymapid();
else
mmid = entryhi & asidmask;
if (cpu_has_guestid)
guestctl1 = read_c0_guestctl1();
/* EHINV bit marks entire entry as invalid */
if (cpu_has_tlbinv && entryhi & MIPS_ENTRYHI_EHINV)
continue;
/*
* Prior to tlbinv, unused entries have a virtual address of
* CKSEG0.
*/
if ((entryhi & ~0x1ffffUL) == CKSEG0)
continue;
/*
* ASID takes effect in absence of G (global) bit.
* We check both G bits, even though architecturally they should
* match one another, because some revisions of the SB1 core may
* leave only a single G bit set after a machine check exception
* due to duplicate TLB entry.
*/
if (!((entrylo0 | entrylo1) & ENTRYLO_G) && (mmid != asid))
continue;
/*
* Only print entries in use
*/
printk("Index: %2d pgmask=%s ", i, msk2str(pagemask));
c0 = (entrylo0 & ENTRYLO_C) >> ENTRYLO_C_SHIFT;
c1 = (entrylo1 & ENTRYLO_C) >> ENTRYLO_C_SHIFT;
pr_cont("va=%0*lx asid=%0*lx",
vwidth, (entryhi & ~0x1fffUL),
asidwidth, mmid);
if (cpu_has_guestid)
pr_cont(" gid=%02lx",
(guestctl1 & MIPS_GCTL1_RID)
>> MIPS_GCTL1_RID_SHIFT);
/* RI/XI are in awkward places, so mask them off separately */
pa = entrylo0 & ~(MIPS_ENTRYLO_RI | MIPS_ENTRYLO_XI);
if (xpa)
pa |= (unsigned long long)readx_c0_entrylo0() << 30;
pa = (pa << 6) & PAGE_MASK;
pr_cont("\n\t[");
if (cpu_has_rixi)
pr_cont("ri=%d xi=%d ",
(entrylo0 & MIPS_ENTRYLO_RI) ? 1 : 0,
(entrylo0 & MIPS_ENTRYLO_XI) ? 1 : 0);
pr_cont("pa=%0*llx c=%d d=%d v=%d g=%d] [",
pwidth, pa, c0,
(entrylo0 & ENTRYLO_D) ? 1 : 0,
(entrylo0 & ENTRYLO_V) ? 1 : 0,
(entrylo0 & ENTRYLO_G) ? 1 : 0);
/* RI/XI are in awkward places, so mask them off separately */
pa = entrylo1 & ~(MIPS_ENTRYLO_RI | MIPS_ENTRYLO_XI);
if (xpa)
pa |= (unsigned long long)readx_c0_entrylo1() << 30;
pa = (pa << 6) & PAGE_MASK;
if (cpu_has_rixi)
pr_cont("ri=%d xi=%d ",
(entrylo1 & MIPS_ENTRYLO_RI) ? 1 : 0,
(entrylo1 & MIPS_ENTRYLO_XI) ? 1 : 0);
pr_cont("pa=%0*llx c=%d d=%d v=%d g=%d]\n",
pwidth, pa, c1,
(entrylo1 & ENTRYLO_D) ? 1 : 0,
(entrylo1 & ENTRYLO_V) ? 1 : 0,
(entrylo1 & ENTRYLO_G) ? 1 : 0);
}
printk("\n");
write_c0_entryhi(s_entryhi);
write_c0_index(s_index);
write_c0_pagemask(s_pagemask);
if (cpu_has_guestid)
write_c0_guestctl1(s_guestctl1);
}
void dump_tlb_all(void)
{
dump_tlb(0, current_cpu_data.tlbsize - 1);
}