linux-stable/arch/arm64/include/asm/mmu_context.h

275 lines
7.1 KiB
C
Raw Normal View History

/* SPDX-License-Identifier: GPL-2.0-only */
/*
* Based on arch/arm/include/asm/mmu_context.h
*
* Copyright (C) 1996 Russell King.
* Copyright (C) 2012 ARM Ltd.
*/
#ifndef __ASM_MMU_CONTEXT_H
#define __ASM_MMU_CONTEXT_H
arm64: Work around Falkor erratum 1003 The Qualcomm Datacenter Technologies Falkor v1 CPU may allocate TLB entries using an incorrect ASID when TTBRx_EL1 is being updated. When the erratum is triggered, page table entries using the new translation table base address (BADDR) will be allocated into the TLB using the old ASID. All circumstances leading to the incorrect ASID being cached in the TLB arise when software writes TTBRx_EL1[ASID] and TTBRx_EL1[BADDR], a memory operation is in the process of performing a translation using the specific TTBRx_EL1 being written, and the memory operation uses a translation table descriptor designated as non-global. EL2 and EL3 code changing the EL1&0 ASID is not subject to this erratum because hardware is prohibited from performing translations from an out-of-context translation regime. Consider the following pseudo code. write new BADDR and ASID values to TTBRx_EL1 Replacing the above sequence with the one below will ensure that no TLB entries with an incorrect ASID are used by software. write reserved value to TTBRx_EL1[ASID] ISB write new value to TTBRx_EL1[BADDR] ISB write new value to TTBRx_EL1[ASID] ISB When the above sequence is used, page table entries using the new BADDR value may still be incorrectly allocated into the TLB using the reserved ASID. Yet this will not reduce functionality, since TLB entries incorrectly tagged with the reserved ASID will never be hit by a later instruction. Based on work by Shanker Donthineni <shankerd@codeaurora.org> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Christopher Covington <cov@codeaurora.org> Signed-off-by: Will Deacon <will.deacon@arm.com>
2017-02-08 20:08:37 +00:00
#ifndef __ASSEMBLY__
#include <linux/compiler.h>
#include <linux/sched.h>
#include <linux/sched/hotplug.h>
#include <linux/mm_types.h>
mm: reorder includes after introduction of linux/pgtable.h The replacement of <asm/pgrable.h> with <linux/pgtable.h> made the include of the latter in the middle of asm includes. Fix this up with the aid of the below script and manual adjustments here and there. import sys import re if len(sys.argv) is not 3: print "USAGE: %s <file> <header>" % (sys.argv[0]) sys.exit(1) hdr_to_move="#include <linux/%s>" % sys.argv[2] moved = False in_hdrs = False with open(sys.argv[1], "r") as f: lines = f.readlines() for _line in lines: line = _line.rstrip(' ') if line == hdr_to_move: continue if line.startswith("#include <linux/"): in_hdrs = True elif not moved and in_hdrs: moved = True print hdr_to_move print line Signed-off-by: Mike Rapoport <rppt@linux.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Cain <bcain@codeaurora.org> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Chris Zankel <chris@zankel.net> Cc: "David S. Miller" <davem@davemloft.net> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Greentime Hu <green.hu@gmail.com> Cc: Greg Ungerer <gerg@linux-m68k.org> Cc: Guan Xuetao <gxt@pku.edu.cn> Cc: Guo Ren <guoren@kernel.org> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Helge Deller <deller@gmx.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: Ley Foon Tan <ley.foon.tan@intel.com> Cc: Mark Salter <msalter@redhat.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Matt Turner <mattst88@gmail.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Nick Hu <nickhu@andestech.com> Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Richard Weinberger <richard@nod.at> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Stafford Horne <shorne@gmail.com> Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tony Luck <tony.luck@intel.com> Cc: Vincent Chen <deanbo422@gmail.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Will Deacon <will@kernel.org> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Link: http://lkml.kernel.org/r/20200514170327.31389-4-rppt@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-06-09 04:32:42 +00:00
#include <linux/pgtable.h>
#include <asm/cacheflush.h>
#include <asm/cpufeature.h>
#include <asm/daifflags.h>
#include <asm/proc-fns.h>
#include <asm-generic/mm_hooks.h>
#include <asm/cputype.h>
#include <asm/sysreg.h>
#include <asm/tlbflush.h>
extern bool rodata_full;
static inline void contextidr_thread_switch(struct task_struct *next)
{
if (!IS_ENABLED(CONFIG_PID_IN_CONTEXTIDR))
return;
write_sysreg(task_pid_nr(next), contextidr_el1);
isb();
}
/*
* Set TTBR0 to reserved_pg_dir. No translations will be possible via TTBR0.
*/
static inline void cpu_set_reserved_ttbr0_nosync(void)
{
unsigned long ttbr = phys_to_ttbr(__pa_symbol(reserved_pg_dir));
write_sysreg(ttbr, ttbr0_el1);
}
static inline void cpu_set_reserved_ttbr0(void)
{
cpu_set_reserved_ttbr0_nosync();
isb();
}
void cpu_do_switch_mm(phys_addr_t pgd_phys, struct mm_struct *mm);
static inline void cpu_switch_mm(pgd_t *pgd, struct mm_struct *mm)
{
BUG_ON(pgd == swapper_pg_dir);
cpu_do_switch_mm(virt_to_phys(pgd),mm);
}
/*
* TCR.T0SZ value to use when the ID map is active.
*/
#define idmap_t0sz TCR_T0SZ(IDMAP_VA_BITS)
/*
* Ensure TCR.T0SZ is set to the provided value.
*/
static inline void __cpu_set_tcr_t0sz(unsigned long t0sz)
{
unsigned long tcr = read_sysreg(tcr_el1);
if ((tcr & TCR_T0SZ_MASK) >> TCR_T0SZ_OFFSET == t0sz)
return;
tcr &= ~TCR_T0SZ_MASK;
tcr |= t0sz << TCR_T0SZ_OFFSET;
write_sysreg(tcr, tcr_el1);
isb();
}
#define cpu_set_default_tcr_t0sz() __cpu_set_tcr_t0sz(TCR_T0SZ(vabits_actual))
#define cpu_set_idmap_tcr_t0sz() __cpu_set_tcr_t0sz(idmap_t0sz)
/*
* Remove the idmap from TTBR0_EL1 and install the pgd of the active mm.
*
* The idmap lives in the same VA range as userspace, but uses global entries
* and may use a different TCR_EL1.T0SZ. To avoid issues resulting from
* speculative TLB fetches, we must temporarily install the reserved page
* tables while we invalidate the TLBs and set up the correct TCR_EL1.T0SZ.
*
* If current is a not a user task, the mm covers the TTBR1_EL1 page tables,
* which should not be installed in TTBR0_EL1. In this case we can leave the
* reserved page tables in place.
*/
static inline void cpu_uninstall_idmap(void)
{
struct mm_struct *mm = current->active_mm;
cpu_set_reserved_ttbr0();
local_flush_tlb_all();
cpu_set_default_tcr_t0sz();
if (mm != &init_mm && !system_uses_ttbr0_pan())
cpu_switch_mm(mm->pgd, mm);
}
static inline void cpu_install_idmap(void)
{
cpu_set_reserved_ttbr0();
local_flush_tlb_all();
cpu_set_idmap_tcr_t0sz();
cpu_switch_mm(lm_alias(idmap_pg_dir), &init_mm);
}
/*
* Load our new page tables. A strict BBM approach requires that we ensure that
* TLBs are free of any entries that may overlap with the global mappings we are
* about to install.
*
* For a real hibernate/resume/kexec cycle TTBR0 currently points to a zero
* page, but TLBs may contain stale ASID-tagged entries (e.g. for EFI runtime
* services), while for a userspace-driven test_resume cycle it points to
* userspace page tables (and we must point it at a zero page ourselves).
*
* We change T0SZ as part of installing the idmap. This is undone by
* cpu_uninstall_idmap() in __cpu_suspend_exit().
*/
static inline void cpu_install_ttbr0(phys_addr_t ttbr0, unsigned long t0sz)
{
cpu_set_reserved_ttbr0();
local_flush_tlb_all();
__cpu_set_tcr_t0sz(t0sz);
/* avoid cpu_switch_mm() and its SW-PAN and CNP interactions */
write_sysreg(ttbr0, ttbr0_el1);
isb();
}
void __cpu_replace_ttbr1(pgd_t *pgdp, bool cnp);
arm64: mm: add code to safely replace TTBR1_EL1 If page tables are modified without suitable TLB maintenance, the ARM architecture permits multiple TLB entries to be allocated for the same VA. When this occurs, it is permitted that TLB conflict aborts are raised in response to synchronous data/instruction accesses, and/or and amalgamation of the TLB entries may be used as a result of a TLB lookup. The presence of conflicting TLB entries may result in a variety of behaviours detrimental to the system (e.g. erroneous physical addresses may be used by I-cache fetches and/or page table walks). Some of these cases may result in unexpected changes of hardware state, and/or result in the (asynchronous) delivery of SError. To avoid these issues, we must avoid situations where conflicting entries may be allocated into TLBs. For user and module mappings we can follow a strict break-before-make approach, but this cannot work for modifications to the swapper page tables that cover the kernel text and data. Instead, this patch adds code which is intended to be executed from the idmap, which can safely unmap the swapper page tables as it only requires the idmap to be active. This enables us to uninstall the active TTBR1_EL1 entry, invalidate TLBs, then install a new TTBR1_EL1 entry without potentially unmapping code or data required for the sequence. This avoids the risk of conflict, but requires that updates are staged in a copy of the swapper page tables prior to being installed. Signed-off-by: Mark Rutland <mark.rutland@arm.com> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Tested-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Tested-by: Jeremy Linton <jeremy.linton@arm.com> Cc: Laura Abbott <labbott@fedoraproject.org> Cc: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2016-01-25 11:45:01 +00:00
arm64: Avoid cpus_have_const_cap() for ARM64_HAS_CNP In system_supports_cnp() we use cpus_have_const_cap() to check for ARM64_HAS_CNP, but this is only necessary so that the cpu_enable_cnp() callback can run prior to alternatives being patched, and otherwise this is not necessary and alternative_has_cap_*() would be preferable. For historical reasons, cpus_have_const_cap() is more complicated than it needs to be. Before cpucaps are finalized, it will perform a bitmap test of the system_cpucaps bitmap, and once cpucaps are finalized it will use an alternative branch. This used to be necessary to handle some race conditions in the window between cpucap detection and the subsequent patching of alternatives and static branches, where different branches could be out-of-sync with one another (or w.r.t. alternative sequences). Now that we use alternative branches instead of static branches, these are all patched atomically w.r.t. one another, and there are only a handful of cases that need special care in the window between cpucap detection and alternative patching. Due to the above, it would be nice to remove cpus_have_const_cap(), and migrate callers over to alternative_has_cap_*(), cpus_have_final_cap(), or cpus_have_cap() depending on when their requirements. This will remove redundant instructions and improve code generation, and will make it easier to determine how each callsite will behave before, during, and after alternative patching. The cpu_enable_cnp() callback is run immediately after the ARM64_HAS_CNP cpucap is detected system-wide under setup_system_capabilities(), prior to alternatives being patched. During this window cpu_enable_cnp() uses cpu_replace_ttbr1() to set the CNP bit for the swapper_pg_dir in TTBR1. No other users of the ARM64_HAS_CNP cpucap need the up-to-date value during this window: * As KVM isn't initialized yet, kvm_get_vttbr() isn't reachable. * As cpuidle isn't initialized yet, __cpu_suspend_exit() isn't reachable. * At this point all CPUs are using the swapper_pg_dir with a reserved ASID in TTBR1, and the idmap_pg_dir in TTBR0, so neither check_and_switch_context() nor cpu_do_switch_mm() need to do anything special. This patch replaces the use of cpus_have_const_cap() with alternative_has_cap_unlikely(), which will avoid generating code to test the system_cpucaps bitmap and should be better for all subsequent calls at runtime. To allow cpu_enable_cnp() to function prior to alternatives being patched, cpu_replace_ttbr1() is split into cpu_replace_ttbr1() and cpu_enable_swapper_cnp(), with the former only used for early TTBR1 replacement, and the latter used by both cpu_enable_cnp() and __cpu_suspend_exit(). Signed-off-by: Mark Rutland <mark.rutland@arm.com> Cc: Ard Biesheuvel <ardb@kernel.org> Cc: Suzuki K Poulose <suzuki.poulose@arm.com> Cc: Vladimir Murzin <vladimir.murzin@arm.com> Cc: Will Deacon <will@kernel.org> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2023-10-16 10:24:41 +00:00
static inline void cpu_enable_swapper_cnp(void)
{
__cpu_replace_ttbr1(lm_alias(swapper_pg_dir), true);
arm64: Avoid cpus_have_const_cap() for ARM64_HAS_CNP In system_supports_cnp() we use cpus_have_const_cap() to check for ARM64_HAS_CNP, but this is only necessary so that the cpu_enable_cnp() callback can run prior to alternatives being patched, and otherwise this is not necessary and alternative_has_cap_*() would be preferable. For historical reasons, cpus_have_const_cap() is more complicated than it needs to be. Before cpucaps are finalized, it will perform a bitmap test of the system_cpucaps bitmap, and once cpucaps are finalized it will use an alternative branch. This used to be necessary to handle some race conditions in the window between cpucap detection and the subsequent patching of alternatives and static branches, where different branches could be out-of-sync with one another (or w.r.t. alternative sequences). Now that we use alternative branches instead of static branches, these are all patched atomically w.r.t. one another, and there are only a handful of cases that need special care in the window between cpucap detection and alternative patching. Due to the above, it would be nice to remove cpus_have_const_cap(), and migrate callers over to alternative_has_cap_*(), cpus_have_final_cap(), or cpus_have_cap() depending on when their requirements. This will remove redundant instructions and improve code generation, and will make it easier to determine how each callsite will behave before, during, and after alternative patching. The cpu_enable_cnp() callback is run immediately after the ARM64_HAS_CNP cpucap is detected system-wide under setup_system_capabilities(), prior to alternatives being patched. During this window cpu_enable_cnp() uses cpu_replace_ttbr1() to set the CNP bit for the swapper_pg_dir in TTBR1. No other users of the ARM64_HAS_CNP cpucap need the up-to-date value during this window: * As KVM isn't initialized yet, kvm_get_vttbr() isn't reachable. * As cpuidle isn't initialized yet, __cpu_suspend_exit() isn't reachable. * At this point all CPUs are using the swapper_pg_dir with a reserved ASID in TTBR1, and the idmap_pg_dir in TTBR0, so neither check_and_switch_context() nor cpu_do_switch_mm() need to do anything special. This patch replaces the use of cpus_have_const_cap() with alternative_has_cap_unlikely(), which will avoid generating code to test the system_cpucaps bitmap and should be better for all subsequent calls at runtime. To allow cpu_enable_cnp() to function prior to alternatives being patched, cpu_replace_ttbr1() is split into cpu_replace_ttbr1() and cpu_enable_swapper_cnp(), with the former only used for early TTBR1 replacement, and the latter used by both cpu_enable_cnp() and __cpu_suspend_exit(). Signed-off-by: Mark Rutland <mark.rutland@arm.com> Cc: Ard Biesheuvel <ardb@kernel.org> Cc: Suzuki K Poulose <suzuki.poulose@arm.com> Cc: Vladimir Murzin <vladimir.murzin@arm.com> Cc: Will Deacon <will@kernel.org> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2023-10-16 10:24:41 +00:00
}
static inline void cpu_replace_ttbr1(pgd_t *pgdp)
arm64: Avoid cpus_have_const_cap() for ARM64_HAS_CNP In system_supports_cnp() we use cpus_have_const_cap() to check for ARM64_HAS_CNP, but this is only necessary so that the cpu_enable_cnp() callback can run prior to alternatives being patched, and otherwise this is not necessary and alternative_has_cap_*() would be preferable. For historical reasons, cpus_have_const_cap() is more complicated than it needs to be. Before cpucaps are finalized, it will perform a bitmap test of the system_cpucaps bitmap, and once cpucaps are finalized it will use an alternative branch. This used to be necessary to handle some race conditions in the window between cpucap detection and the subsequent patching of alternatives and static branches, where different branches could be out-of-sync with one another (or w.r.t. alternative sequences). Now that we use alternative branches instead of static branches, these are all patched atomically w.r.t. one another, and there are only a handful of cases that need special care in the window between cpucap detection and alternative patching. Due to the above, it would be nice to remove cpus_have_const_cap(), and migrate callers over to alternative_has_cap_*(), cpus_have_final_cap(), or cpus_have_cap() depending on when their requirements. This will remove redundant instructions and improve code generation, and will make it easier to determine how each callsite will behave before, during, and after alternative patching. The cpu_enable_cnp() callback is run immediately after the ARM64_HAS_CNP cpucap is detected system-wide under setup_system_capabilities(), prior to alternatives being patched. During this window cpu_enable_cnp() uses cpu_replace_ttbr1() to set the CNP bit for the swapper_pg_dir in TTBR1. No other users of the ARM64_HAS_CNP cpucap need the up-to-date value during this window: * As KVM isn't initialized yet, kvm_get_vttbr() isn't reachable. * As cpuidle isn't initialized yet, __cpu_suspend_exit() isn't reachable. * At this point all CPUs are using the swapper_pg_dir with a reserved ASID in TTBR1, and the idmap_pg_dir in TTBR0, so neither check_and_switch_context() nor cpu_do_switch_mm() need to do anything special. This patch replaces the use of cpus_have_const_cap() with alternative_has_cap_unlikely(), which will avoid generating code to test the system_cpucaps bitmap and should be better for all subsequent calls at runtime. To allow cpu_enable_cnp() to function prior to alternatives being patched, cpu_replace_ttbr1() is split into cpu_replace_ttbr1() and cpu_enable_swapper_cnp(), with the former only used for early TTBR1 replacement, and the latter used by both cpu_enable_cnp() and __cpu_suspend_exit(). Signed-off-by: Mark Rutland <mark.rutland@arm.com> Cc: Ard Biesheuvel <ardb@kernel.org> Cc: Suzuki K Poulose <suzuki.poulose@arm.com> Cc: Vladimir Murzin <vladimir.murzin@arm.com> Cc: Will Deacon <will@kernel.org> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2023-10-16 10:24:41 +00:00
{
/*
* Only for early TTBR1 replacement before cpucaps are finalized and
* before we've decided whether to use CNP.
*/
WARN_ON(system_capabilities_finalized());
__cpu_replace_ttbr1(pgdp, false);
arm64: Avoid cpus_have_const_cap() for ARM64_HAS_CNP In system_supports_cnp() we use cpus_have_const_cap() to check for ARM64_HAS_CNP, but this is only necessary so that the cpu_enable_cnp() callback can run prior to alternatives being patched, and otherwise this is not necessary and alternative_has_cap_*() would be preferable. For historical reasons, cpus_have_const_cap() is more complicated than it needs to be. Before cpucaps are finalized, it will perform a bitmap test of the system_cpucaps bitmap, and once cpucaps are finalized it will use an alternative branch. This used to be necessary to handle some race conditions in the window between cpucap detection and the subsequent patching of alternatives and static branches, where different branches could be out-of-sync with one another (or w.r.t. alternative sequences). Now that we use alternative branches instead of static branches, these are all patched atomically w.r.t. one another, and there are only a handful of cases that need special care in the window between cpucap detection and alternative patching. Due to the above, it would be nice to remove cpus_have_const_cap(), and migrate callers over to alternative_has_cap_*(), cpus_have_final_cap(), or cpus_have_cap() depending on when their requirements. This will remove redundant instructions and improve code generation, and will make it easier to determine how each callsite will behave before, during, and after alternative patching. The cpu_enable_cnp() callback is run immediately after the ARM64_HAS_CNP cpucap is detected system-wide under setup_system_capabilities(), prior to alternatives being patched. During this window cpu_enable_cnp() uses cpu_replace_ttbr1() to set the CNP bit for the swapper_pg_dir in TTBR1. No other users of the ARM64_HAS_CNP cpucap need the up-to-date value during this window: * As KVM isn't initialized yet, kvm_get_vttbr() isn't reachable. * As cpuidle isn't initialized yet, __cpu_suspend_exit() isn't reachable. * At this point all CPUs are using the swapper_pg_dir with a reserved ASID in TTBR1, and the idmap_pg_dir in TTBR0, so neither check_and_switch_context() nor cpu_do_switch_mm() need to do anything special. This patch replaces the use of cpus_have_const_cap() with alternative_has_cap_unlikely(), which will avoid generating code to test the system_cpucaps bitmap and should be better for all subsequent calls at runtime. To allow cpu_enable_cnp() to function prior to alternatives being patched, cpu_replace_ttbr1() is split into cpu_replace_ttbr1() and cpu_enable_swapper_cnp(), with the former only used for early TTBR1 replacement, and the latter used by both cpu_enable_cnp() and __cpu_suspend_exit(). Signed-off-by: Mark Rutland <mark.rutland@arm.com> Cc: Ard Biesheuvel <ardb@kernel.org> Cc: Suzuki K Poulose <suzuki.poulose@arm.com> Cc: Vladimir Murzin <vladimir.murzin@arm.com> Cc: Will Deacon <will@kernel.org> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2023-10-16 10:24:41 +00:00
}
/*
* It would be nice to return ASIDs back to the allocator, but unfortunately
* that introduces a race with a generation rollover where we could erroneously
* free an ASID allocated in a future generation. We could workaround this by
* freeing the ASID from the context of the dying mm (e.g. in arch_exit_mmap),
* but we'd then need to make sure that we didn't dirty any TLBs afterwards.
* Setting a reserved TTBR0 or EPD0 would work, but it all gets ugly when you
* take CPU migration into account.
*/
arm64/mm: save memory access in check_and_switch_context() fast switch path On arm64, smp_processor_id() reads a per-cpu `cpu_number` variable, using the per-cpu offset stored in the tpidr_el1 system register. In some cases we generate a per-cpu address with a sequence like: cpu_ptr = &per_cpu(ptr, smp_processor_id()); Which potentially incurs a cache miss for both `cpu_number` and the in-memory `__per_cpu_offset` array. This can be written more optimally as: cpu_ptr = this_cpu_ptr(ptr); Which only needs the offset from tpidr_el1, and does not need to load from memory. The following two test cases show a small performance improvement measured on a 46-cpus qualcomm machine with 5.8.0-rc4 kernel. Test 1: (about 0.3% improvement) #cat b.sh make clean && make all -j138 #perf stat --repeat 10 --null --sync sh b.sh - before this patch Performance counter stats for 'sh b.sh' (10 runs): 298.62 +- 1.86 seconds time elapsed ( +- 0.62% ) - after this patch Performance counter stats for 'sh b.sh' (10 runs): 297.734 +- 0.954 seconds time elapsed ( +- 0.32% ) Test 2: (about 1.69% improvement) 'perf stat -r 10 perf bench sched messaging' Then sum the total time of 'sched/messaging' by manual. - before this patch total 0.707 sec for 10 times - after this patch totol 0.695 sec for 10 times Signed-off-by: Pingfan Liu <kernelfans@gmail.com> Acked-by: Mark Rutland <mark.rutland@arm.com> Cc: Will Deacon <will@kernel.org> Cc: Steve Capper <steve.capper@arm.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Vladimir Murzin <vladimir.murzin@arm.com> Cc: Jean-Philippe Brucker <jean-philippe@linaro.org> Link: https://lore.kernel.org/r/1594389852-19949-1-git-send-email-kernelfans@gmail.com Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2020-07-10 14:04:12 +00:00
void check_and_switch_context(struct mm_struct *mm);
#define init_new_context(tsk, mm) init_new_context(tsk, mm)
arm64: mm: Pin down ASIDs for sharing mm with devices To enable address space sharing with the IOMMU, introduce arm64_mm_context_get() and arm64_mm_context_put(), that pin down a context and ensure that it will keep its ASID after a rollover. Export the symbols to let the modular SMMUv3 driver use them. Pinning is necessary because a device constantly needs a valid ASID, unlike tasks that only require one when running. Without pinning, we would need to notify the IOMMU when we're about to use a new ASID for a task, and it would get complicated when a new task is assigned a shared ASID. Consider the following scenario with no ASID pinned: 1. Task t1 is running on CPUx with shared ASID (gen=1, asid=1) 2. Task t2 is scheduled on CPUx, gets ASID (1, 2) 3. Task tn is scheduled on CPUy, a rollover occurs, tn gets ASID (2, 1) We would now have to immediately generate a new ASID for t1, notify the IOMMU, and finally enable task tn. We are holding the lock during all that time, since we can't afford having another CPU trigger a rollover. The IOMMU issues invalidation commands that can take tens of milliseconds. It gets needlessly complicated. All we wanted to do was schedule task tn, that has no business with the IOMMU. By letting the IOMMU pin tasks when needed, we avoid stalling the slow path, and let the pinning fail when we're out of shareable ASIDs. After a rollover, the allocator expects at least one ASID to be available in addition to the reserved ones (one per CPU). So (NR_ASIDS - NR_CPUS - 1) is the maximum number of ASIDs that can be shared with the IOMMU. Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Link: https://lore.kernel.org/r/20200918101852.582559-5-jean-philippe@linaro.org Signed-off-by: Will Deacon <will@kernel.org>
2020-09-18 10:18:44 +00:00
static inline int
init_new_context(struct task_struct *tsk, struct mm_struct *mm)
{
atomic64_set(&mm->context.id, 0);
refcount_set(&mm->context.pinned, 0);
return 0;
}
#ifdef CONFIG_ARM64_SW_TTBR0_PAN
static inline void update_saved_ttbr0(struct task_struct *tsk,
struct mm_struct *mm)
{
u64 ttbr;
if (!system_uses_ttbr0_pan())
return;
if (mm == &init_mm)
ttbr = phys_to_ttbr(__pa_symbol(reserved_pg_dir));
else
ttbr = phys_to_ttbr(virt_to_phys(mm->pgd)) | ASID(mm) << 48;
arm64: kpti: Fix the interaction between ASID switching and software PAN With ARM64_SW_TTBR0_PAN enabled, the exception entry code checks the active ASID to decide whether user access was enabled (non-zero ASID) when the exception was taken. On return from exception, if user access was previously disabled, it re-instates TTBR0_EL1 from the per-thread saved value (updated in switch_mm() or efi_set_pgd()). Commit 7655abb95386 ("arm64: mm: Move ASID from TTBR0 to TTBR1") makes a TTBR0_EL1 + ASID switching non-atomic. Subsequently, commit 27a921e75711 ("arm64: mm: Fix and re-enable ARM64_SW_TTBR0_PAN") changes the __uaccess_ttbr0_disable() function and asm macro to first write the reserved TTBR0_EL1 followed by the ASID=0 update in TTBR1_EL1. If an exception occurs between these two, the exception return code will re-instate a valid TTBR0_EL1. Similar scenario can happen in cpu_switch_mm() between setting the reserved TTBR0_EL1 and the ASID update in cpu_do_switch_mm(). This patch reverts the entry.S check for ASID == 0 to TTBR0_EL1 and disables the interrupts around the TTBR0_EL1 and ASID switching code in __uaccess_ttbr0_disable(). It also ensures that, when returning from the EFI runtime services, efi_set_pgd() doesn't leave a non-zero ASID in TTBR1_EL1 by using uaccess_ttbr0_{enable,disable}. The accesses to current_thread_info()->ttbr0 are updated to use READ_ONCE/WRITE_ONCE. As a safety measure, __uaccess_ttbr0_enable() always masks out any existing non-zero ASID TTBR1_EL1 before writing in the new ASID. Fixes: 27a921e75711 ("arm64: mm: Fix and re-enable ARM64_SW_TTBR0_PAN") Acked-by: Will Deacon <will.deacon@arm.com> Reported-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Tested-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Reviewed-by: James Morse <james.morse@arm.com> Tested-by: James Morse <james.morse@arm.com> Co-developed-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2018-01-10 13:18:30 +00:00
WRITE_ONCE(task_thread_info(tsk)->ttbr0, ttbr);
}
#else
static inline void update_saved_ttbr0(struct task_struct *tsk,
struct mm_struct *mm)
{
}
#endif
#define enter_lazy_tlb enter_lazy_tlb
static inline void
enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
{
/*
* We don't actually care about the ttbr0 mapping, so point it at the
* zero page.
*/
update_saved_ttbr0(tsk, &init_mm);
}
static inline void __switch_mm(struct mm_struct *next)
{
/*
* init_mm.pgd does not contain any user mappings and it is always
* active for kernel addresses in TTBR1. Just set the reserved TTBR0.
*/
if (next == &init_mm) {
cpu_set_reserved_ttbr0();
return;
}
arm64/mm: save memory access in check_and_switch_context() fast switch path On arm64, smp_processor_id() reads a per-cpu `cpu_number` variable, using the per-cpu offset stored in the tpidr_el1 system register. In some cases we generate a per-cpu address with a sequence like: cpu_ptr = &per_cpu(ptr, smp_processor_id()); Which potentially incurs a cache miss for both `cpu_number` and the in-memory `__per_cpu_offset` array. This can be written more optimally as: cpu_ptr = this_cpu_ptr(ptr); Which only needs the offset from tpidr_el1, and does not need to load from memory. The following two test cases show a small performance improvement measured on a 46-cpus qualcomm machine with 5.8.0-rc4 kernel. Test 1: (about 0.3% improvement) #cat b.sh make clean && make all -j138 #perf stat --repeat 10 --null --sync sh b.sh - before this patch Performance counter stats for 'sh b.sh' (10 runs): 298.62 +- 1.86 seconds time elapsed ( +- 0.62% ) - after this patch Performance counter stats for 'sh b.sh' (10 runs): 297.734 +- 0.954 seconds time elapsed ( +- 0.32% ) Test 2: (about 1.69% improvement) 'perf stat -r 10 perf bench sched messaging' Then sum the total time of 'sched/messaging' by manual. - before this patch total 0.707 sec for 10 times - after this patch totol 0.695 sec for 10 times Signed-off-by: Pingfan Liu <kernelfans@gmail.com> Acked-by: Mark Rutland <mark.rutland@arm.com> Cc: Will Deacon <will@kernel.org> Cc: Steve Capper <steve.capper@arm.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Vladimir Murzin <vladimir.murzin@arm.com> Cc: Jean-Philippe Brucker <jean-philippe@linaro.org> Link: https://lore.kernel.org/r/1594389852-19949-1-git-send-email-kernelfans@gmail.com Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2020-07-10 14:04:12 +00:00
check_and_switch_context(next);
}
static inline void
switch_mm(struct mm_struct *prev, struct mm_struct *next,
struct task_struct *tsk)
{
if (prev != next)
__switch_mm(next);
/*
* Update the saved TTBR0_EL1 of the scheduled-in task as the previous
* value may have not been initialised yet (activate_mm caller) or the
* ASID has changed since the last run (following the context switch
* of another thread of the same process).
*/
update_saved_ttbr0(tsk, next);
}
static inline const struct cpumask *
task_cpu_possible_mask(struct task_struct *p)
{
if (!static_branch_unlikely(&arm64_mismatched_32bit_el0))
return cpu_possible_mask;
if (!is_compat_thread(task_thread_info(p)))
return cpu_possible_mask;
return system_32bit_el0_cpumask();
}
#define task_cpu_possible_mask task_cpu_possible_mask
void verify_cpu_asid_bits(void);
arm64: kpti: Fix the interaction between ASID switching and software PAN With ARM64_SW_TTBR0_PAN enabled, the exception entry code checks the active ASID to decide whether user access was enabled (non-zero ASID) when the exception was taken. On return from exception, if user access was previously disabled, it re-instates TTBR0_EL1 from the per-thread saved value (updated in switch_mm() or efi_set_pgd()). Commit 7655abb95386 ("arm64: mm: Move ASID from TTBR0 to TTBR1") makes a TTBR0_EL1 + ASID switching non-atomic. Subsequently, commit 27a921e75711 ("arm64: mm: Fix and re-enable ARM64_SW_TTBR0_PAN") changes the __uaccess_ttbr0_disable() function and asm macro to first write the reserved TTBR0_EL1 followed by the ASID=0 update in TTBR1_EL1. If an exception occurs between these two, the exception return code will re-instate a valid TTBR0_EL1. Similar scenario can happen in cpu_switch_mm() between setting the reserved TTBR0_EL1 and the ASID update in cpu_do_switch_mm(). This patch reverts the entry.S check for ASID == 0 to TTBR0_EL1 and disables the interrupts around the TTBR0_EL1 and ASID switching code in __uaccess_ttbr0_disable(). It also ensures that, when returning from the EFI runtime services, efi_set_pgd() doesn't leave a non-zero ASID in TTBR1_EL1 by using uaccess_ttbr0_{enable,disable}. The accesses to current_thread_info()->ttbr0 are updated to use READ_ONCE/WRITE_ONCE. As a safety measure, __uaccess_ttbr0_enable() always masks out any existing non-zero ASID TTBR1_EL1 before writing in the new ASID. Fixes: 27a921e75711 ("arm64: mm: Fix and re-enable ARM64_SW_TTBR0_PAN") Acked-by: Will Deacon <will.deacon@arm.com> Reported-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Tested-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Reviewed-by: James Morse <james.morse@arm.com> Tested-by: James Morse <james.morse@arm.com> Co-developed-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2018-01-10 13:18:30 +00:00
void post_ttbr_update_workaround(void);
arm64: mm: Pin down ASIDs for sharing mm with devices To enable address space sharing with the IOMMU, introduce arm64_mm_context_get() and arm64_mm_context_put(), that pin down a context and ensure that it will keep its ASID after a rollover. Export the symbols to let the modular SMMUv3 driver use them. Pinning is necessary because a device constantly needs a valid ASID, unlike tasks that only require one when running. Without pinning, we would need to notify the IOMMU when we're about to use a new ASID for a task, and it would get complicated when a new task is assigned a shared ASID. Consider the following scenario with no ASID pinned: 1. Task t1 is running on CPUx with shared ASID (gen=1, asid=1) 2. Task t2 is scheduled on CPUx, gets ASID (1, 2) 3. Task tn is scheduled on CPUy, a rollover occurs, tn gets ASID (2, 1) We would now have to immediately generate a new ASID for t1, notify the IOMMU, and finally enable task tn. We are holding the lock during all that time, since we can't afford having another CPU trigger a rollover. The IOMMU issues invalidation commands that can take tens of milliseconds. It gets needlessly complicated. All we wanted to do was schedule task tn, that has no business with the IOMMU. By letting the IOMMU pin tasks when needed, we avoid stalling the slow path, and let the pinning fail when we're out of shareable ASIDs. After a rollover, the allocator expects at least one ASID to be available in addition to the reserved ones (one per CPU). So (NR_ASIDS - NR_CPUS - 1) is the maximum number of ASIDs that can be shared with the IOMMU. Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Link: https://lore.kernel.org/r/20200918101852.582559-5-jean-philippe@linaro.org Signed-off-by: Will Deacon <will@kernel.org>
2020-09-18 10:18:44 +00:00
unsigned long arm64_mm_context_get(struct mm_struct *mm);
void arm64_mm_context_put(struct mm_struct *mm);
#define mm_untag_mask mm_untag_mask
static inline unsigned long mm_untag_mask(struct mm_struct *mm)
{
return -1UL >> 8;
}
#include <asm-generic/mmu_context.h>
arm64: Work around Falkor erratum 1003 The Qualcomm Datacenter Technologies Falkor v1 CPU may allocate TLB entries using an incorrect ASID when TTBRx_EL1 is being updated. When the erratum is triggered, page table entries using the new translation table base address (BADDR) will be allocated into the TLB using the old ASID. All circumstances leading to the incorrect ASID being cached in the TLB arise when software writes TTBRx_EL1[ASID] and TTBRx_EL1[BADDR], a memory operation is in the process of performing a translation using the specific TTBRx_EL1 being written, and the memory operation uses a translation table descriptor designated as non-global. EL2 and EL3 code changing the EL1&0 ASID is not subject to this erratum because hardware is prohibited from performing translations from an out-of-context translation regime. Consider the following pseudo code. write new BADDR and ASID values to TTBRx_EL1 Replacing the above sequence with the one below will ensure that no TLB entries with an incorrect ASID are used by software. write reserved value to TTBRx_EL1[ASID] ISB write new value to TTBRx_EL1[BADDR] ISB write new value to TTBRx_EL1[ASID] ISB When the above sequence is used, page table entries using the new BADDR value may still be incorrectly allocated into the TLB using the reserved ASID. Yet this will not reduce functionality, since TLB entries incorrectly tagged with the reserved ASID will never be hit by a later instruction. Based on work by Shanker Donthineni <shankerd@codeaurora.org> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Christopher Covington <cov@codeaurora.org> Signed-off-by: Will Deacon <will.deacon@arm.com>
2017-02-08 20:08:37 +00:00
#endif /* !__ASSEMBLY__ */
#endif /* !__ASM_MMU_CONTEXT_H */