kernel_optimize_test/arch/s390/kernel/nmi.c

486 lines
12 KiB
C
Raw Permalink Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* Machine check handler
*
* Copyright IBM Corp. 2000, 2009
* Author(s): Ingo Adlung <adlung@de.ibm.com>,
* Martin Schwidefsky <schwidefsky@de.ibm.com>,
* Cornelia Huck <cornelia.huck@de.ibm.com>,
* Heiko Carstens <heiko.carstens@de.ibm.com>,
*/
#include <linux/kernel_stat.h>
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/hardirq.h>
#include <linux/log2.h>
#include <linux/kprobes.h>
headers: untangle kmemleak.h from mm.h Currently <linux/slab.h> #includes <linux/kmemleak.h> for no obvious reason. It looks like it's only a convenience, so remove kmemleak.h from slab.h and add <linux/kmemleak.h> to any users of kmemleak_* that don't already #include it. Also remove <linux/kmemleak.h> from source files that do not use it. This is tested on i386 allmodconfig and x86_64 allmodconfig. It would be good to run it through the 0day bot for other $ARCHes. I have neither the horsepower nor the storage space for the other $ARCHes. Update: This patch has been extensively build-tested by both the 0day bot & kisskb/ozlabs build farms. Both of them reported 2 build failures for which patches are included here (in v2). [ slab.h is the second most used header file after module.h; kernel.h is right there with slab.h. There could be some minor error in the counting due to some #includes having comments after them and I didn't combine all of those. ] [akpm@linux-foundation.org: security/keys/big_key.c needs vmalloc.h, per sfr] Link: http://lkml.kernel.org/r/e4309f98-3749-93e1-4bb7-d9501a39d015@infradead.org Link: http://kisskb.ellerman.id.au/kisskb/head/13396/ Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Reviewed-by: Ingo Molnar <mingo@kernel.org> Reported-by: Michael Ellerman <mpe@ellerman.id.au> [2 build failures] Reported-by: Fengguang Wu <fengguang.wu@intel.com> [2 build failures] Reviewed-by: Andrew Morton <akpm@linux-foundation.org> Cc: Wei Yongjun <weiyongjun1@huawei.com> Cc: Luis R. Rodriguez <mcgrof@kernel.org> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Mimi Zohar <zohar@linux.vnet.ibm.com> Cc: John Johansen <john.johansen@canonical.com> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-04-06 07:25:34 +08:00
#include <linux/kmemleak.h>
#include <linux/time.h>
#include <linux/module.h>
#include <linux/sched/signal.h>
#include <linux/export.h>
#include <asm/lowcore.h>
#include <asm/smp.h>
#include <asm/stp.h>
#include <asm/cputime.h>
#include <asm/nmi.h>
#include <asm/crw.h>
#include <asm/switch_to.h>
s390/nmi: fix vector register corruption If a machine check happens, the machine has the vector facility installed and the extended save area exists, the cpu will save vector register contents into the extended save area. This is regardless of control register 0 contents, which enables and disables the vector facility during runtime. On each machine check we should validate the vector registers. The current code however tries to validate the registers only if the running task is using vector registers in user space. However even the current code is broken and causes vector register corruption on machine checks, if user space uses them: the prefix area contains a pointer (absolute address) to the machine check extended save area. In order to save some space the save area was put into an unused area of the second prefix page. When validating vector register contents the code uses the absolute address of the extended save area, which is wrong. Due to prefixing the vector instructions will then access contents using absolute addresses instead of real addresses, where the machine stored the contents. If the above would work there is still the problem that register validition would only happen if user space uses vector registers. If kernel space uses them also, this may also lead to vector register content corruption: if the kernel makes use of vector instructions, but the current running user space context does not, the machine check handler will validate floating point registers instead of vector registers. Given the fact that writing to a floating point register may change the upper halve of the corresponding vector register, we also experience vector register corruption in this case. Fix all of these issues, and always validate vector registers on each machine check, if the machine has the vector facility installed and the extended save area is defined. Cc: <stable@vger.kernel.org> # 4.1+ Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-07-07 14:40:49 +08:00
#include <asm/ctl_reg.h>
#include <asm/asm-offsets.h>
#include <linux/kvm_host.h>
struct mcck_struct {
unsigned int kill_task : 1;
unsigned int channel_report : 1;
unsigned int warning : 1;
unsigned int stp_queue : 1;
unsigned long mcck_code;
};
static DEFINE_PER_CPU(struct mcck_struct, cpu_mcck);
static struct kmem_cache *mcesa_cache;
static unsigned long mcesa_origin_lc;
static inline int nmi_needs_mcesa(void)
{
return MACHINE_HAS_VX || MACHINE_HAS_GS;
}
static inline unsigned long nmi_get_mcesa_size(void)
{
if (MACHINE_HAS_GS)
return MCESA_MAX_SIZE;
return MCESA_MIN_SIZE;
}
/*
* The initial machine check extended save area for the boot CPU.
* It will be replaced by nmi_init() with an allocated structure.
* The structure is required for machine check happening early in
* the boot process.
*/
static struct mcesa boot_mcesa __initdata __aligned(MCESA_MAX_SIZE);
void __init nmi_alloc_boot_cpu(struct lowcore *lc)
{
if (!nmi_needs_mcesa())
return;
lc->mcesad = (unsigned long) &boot_mcesa;
if (MACHINE_HAS_GS)
lc->mcesad |= ilog2(MCESA_MAX_SIZE);
}
static int __init nmi_init(void)
{
unsigned long origin, cr0, size;
if (!nmi_needs_mcesa())
return 0;
size = nmi_get_mcesa_size();
if (size > MCESA_MIN_SIZE)
mcesa_origin_lc = ilog2(size);
/* create slab cache for the machine-check-extended-save-areas */
mcesa_cache = kmem_cache_create("nmi_save_areas", size, size, 0, NULL);
if (!mcesa_cache)
panic("Couldn't create nmi save area cache");
origin = (unsigned long) kmem_cache_alloc(mcesa_cache, GFP_KERNEL);
if (!origin)
panic("Couldn't allocate nmi save area");
/* The pointer is stored with mcesa_bits ORed in */
kmemleak_not_leak((void *) origin);
__ctl_store(cr0, 0, 0);
__ctl_clear_bit(0, 28); /* disable lowcore protection */
/* Replace boot_mcesa on the boot CPU */
S390_lowcore.mcesad = origin | mcesa_origin_lc;
__ctl_load(cr0, 0, 0);
return 0;
}
early_initcall(nmi_init);
int nmi_alloc_per_cpu(struct lowcore *lc)
{
unsigned long origin;
if (!nmi_needs_mcesa())
return 0;
origin = (unsigned long) kmem_cache_alloc(mcesa_cache, GFP_KERNEL);
if (!origin)
return -ENOMEM;
/* The pointer is stored with mcesa_bits ORed in */
kmemleak_not_leak((void *) origin);
lc->mcesad = origin | mcesa_origin_lc;
return 0;
}
void nmi_free_per_cpu(struct lowcore *lc)
{
if (!nmi_needs_mcesa())
return;
kmem_cache_free(mcesa_cache, (void *)(lc->mcesad & MCESA_ORIGIN_MASK));
}
static notrace void s390_handle_damage(void)
{
smp_emergency_stop();
disabled_wait();
while (1);
}
NOKPROBE_SYMBOL(s390_handle_damage);
/*
* Main machine check handler function. Will be called with interrupts enabled
* or disabled and machine checks enabled or disabled.
*/
void s390_handle_mcck(void)
{
unsigned long flags;
struct mcck_struct mcck;
/*
* Disable machine checks and get the current state of accumulated
* machine checks. Afterwards delete the old state and enable machine
* checks again.
*/
local_irq_save(flags);
local_mcck_disable();
mcck = *this_cpu_ptr(&cpu_mcck);
memset(this_cpu_ptr(&cpu_mcck), 0, sizeof(mcck));
local_mcck_enable();
local_irq_restore(flags);
if (mcck.channel_report)
crw_handle_channel_report();
/*
* A warning may remain for a prolonged period on the bare iron.
* (actually until the machine is powered off, or the problem is gone)
* So we just stop listening for the WARNING MCH and avoid continuously
* being interrupted. One caveat is however, that we must do this per
* processor and cannot use the smp version of ctl_clear_bit().
* On VM we only get one interrupt per virtally presented machinecheck.
* Though one suffices, we may get one interrupt per (virtual) cpu.
*/
if (mcck.warning) { /* WARNING pending ? */
static int mchchk_wng_posted = 0;
/* Use single cpu clear, as we cannot handle smp here. */
__ctl_clear_bit(14, 24); /* Disable WARNING MCH */
if (xchg(&mchchk_wng_posted, 1) == 0)
kill_cad_pid(SIGPWR, 1);
}
if (mcck.stp_queue)
stp_queue_work();
if (mcck.kill_task) {
local_irq_enable();
printk(KERN_EMERG "mcck: Terminating task because of machine "
"malfunction (code 0x%016lx).\n", mcck.mcck_code);
printk(KERN_EMERG "mcck: task: %s, pid: %d.\n",
current->comm, current->pid);
do_exit(SIGSEGV);
}
}
EXPORT_SYMBOL_GPL(s390_handle_mcck);
/*
* returns 0 if all required registers are available
* returns 1 otherwise
*/
static int notrace s390_check_registers(union mci mci, int umode)
{
union ctlreg2 cr2;
int kill_task;
kill_task = 0;
if (!mci.gr) {
/*
* General purpose registers couldn't be restored and have
* unknown contents. Stop system or terminate process.
*/
if (!umode)
s390_handle_damage();
kill_task = 1;
}
/* Check control registers */
if (!mci.cr) {
/*
* Control registers have unknown contents.
* Can't recover and therefore stopping machine.
*/
s390_handle_damage();
}
if (!mci.fp) {
/*
* Floating point registers can't be restored. If the
* kernel currently uses floating point registers the
* system is stopped. If the process has its floating
* pointer registers loaded it is terminated.
*/
if (S390_lowcore.fpu_flags & KERNEL_VXR_V0V7)
s390_handle_damage();
if (!test_cpu_flag(CIF_FPU))
kill_task = 1;
}
if (!mci.fc) {
/*
* Floating point control register can't be restored.
* If the kernel currently uses the floating pointer
* registers and needs the FPC register the system is
* stopped. If the process has its floating pointer
* registers loaded it is terminated.
*/
if (S390_lowcore.fpu_flags & KERNEL_FPC)
s390_handle_damage();
if (!test_cpu_flag(CIF_FPU))
kill_task = 1;
}
if (MACHINE_HAS_VX) {
if (!mci.vr) {
/*
* Vector registers can't be restored. If the kernel
* currently uses vector registers the system is
* stopped. If the process has its vector registers
* loaded it is terminated.
*/
if (S390_lowcore.fpu_flags & KERNEL_VXR)
s390_handle_damage();
if (!test_cpu_flag(CIF_FPU))
kill_task = 1;
}
}
/* Check if access registers are valid */
if (!mci.ar) {
/*
* Access registers have unknown contents.
* Terminating task.
*/
kill_task = 1;
}
/* Check guarded storage registers */
cr2.val = S390_lowcore.cregs_save_area[2];
if (cr2.gse) {
if (!mci.gs) {
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
/*
* Guarded storage register can't be restored and
* the current processes uses guarded storage.
* It has to be terminated.
*/
kill_task = 1;
}
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
}
/* Check if old PSW is valid */
if (!mci.wp) {
/*
* Can't tell if we come from user or kernel mode
* -> stopping machine.
*/
s390_handle_damage();
}
/* Check for invalid kernel instruction address */
if (!mci.ia && !umode) {
/*
* The instruction address got lost while running
* in the kernel -> stopping machine.
*/
s390_handle_damage();
}
if (!mci.ms || !mci.pm || !mci.ia)
kill_task = 1;
return kill_task;
}
NOKPROBE_SYMBOL(s390_check_registers);
/*
* Backup the guest's machine check info to its description block
*/
static void notrace s390_backup_mcck_info(struct pt_regs *regs)
{
struct mcck_volatile_info *mcck_backup;
struct sie_page *sie_page;
/* r14 contains the sie block, which was set in sie64a */
struct kvm_s390_sie_block *sie_block =
(struct kvm_s390_sie_block *) regs->gprs[14];
if (sie_block == NULL)
/* Something's seriously wrong, stop system. */
s390_handle_damage();
sie_page = container_of(sie_block, struct sie_page, sie_block);
mcck_backup = &sie_page->mcck_info;
mcck_backup->mcic = S390_lowcore.mcck_interruption_code &
~(MCCK_CODE_CP | MCCK_CODE_EXT_DAMAGE);
mcck_backup->ext_damage_code = S390_lowcore.external_damage_code;
mcck_backup->failing_storage_address
= S390_lowcore.failing_storage_address;
}
NOKPROBE_SYMBOL(s390_backup_mcck_info);
#define MAX_IPD_COUNT 29
#define MAX_IPD_TIME (5 * 60 * USEC_PER_SEC) /* 5 minutes */
#define ED_STP_ISLAND 6 /* External damage STP island check */
#define ED_STP_SYNC 7 /* External damage STP sync check */
#define MCCK_CODE_NO_GUEST (MCCK_CODE_CP | MCCK_CODE_EXT_DAMAGE)
/*
* machine check handler.
*/
int notrace s390_do_machine_check(struct pt_regs *regs)
{
static int ipd_count;
static DEFINE_SPINLOCK(ipd_lock);
static unsigned long long last_ipd;
struct mcck_struct *mcck;
unsigned long long tmp;
union mci mci;
unsigned long mcck_dam_code;
int mcck_pending = 0;
nmi_enter();
inc_irq_stat(NMI_NMI);
mci.val = S390_lowcore.mcck_interruption_code;
s390: Replace __get_cpu_var uses __get_cpu_var() is used for multiple purposes in the kernel source. One of them is address calculation via the form &__get_cpu_var(x). This calculates the address for the instance of the percpu variable of the current processor based on an offset. Other use cases are for storing and retrieving data from the current processors percpu area. __get_cpu_var() can be used as an lvalue when writing data or on the right side of an assignment. __get_cpu_var() is defined as : #define __get_cpu_var(var) (*this_cpu_ptr(&(var))) __get_cpu_var() always only does an address determination. However, store and retrieve operations could use a segment prefix (or global register on other platforms) to avoid the address calculation. this_cpu_write() and this_cpu_read() can directly take an offset into a percpu area and use optimized assembly code to read and write per cpu variables. This patch converts __get_cpu_var into either an explicit address calculation using this_cpu_ptr() or into a use of this_cpu operations that use the offset. Thereby address calculations are avoided and less registers are used when code is generated. At the end of the patch set all uses of __get_cpu_var have been removed so the macro is removed too. The patch set includes passes over all arches as well. Once these operations are used throughout then specialized macros can be defined in non -x86 arches as well in order to optimize per cpu access by f.e. using a global register that may be set to the per cpu base. Transformations done to __get_cpu_var() 1. Determine the address of the percpu instance of the current processor. DEFINE_PER_CPU(int, y); int *x = &__get_cpu_var(y); Converts to int *x = this_cpu_ptr(&y); 2. Same as #1 but this time an array structure is involved. DEFINE_PER_CPU(int, y[20]); int *x = __get_cpu_var(y); Converts to int *x = this_cpu_ptr(y); 3. Retrieve the content of the current processors instance of a per cpu variable. DEFINE_PER_CPU(int, y); int x = __get_cpu_var(y) Converts to int x = __this_cpu_read(y); 4. Retrieve the content of a percpu struct DEFINE_PER_CPU(struct mystruct, y); struct mystruct x = __get_cpu_var(y); Converts to memcpy(&x, this_cpu_ptr(&y), sizeof(x)); 5. Assignment to a per cpu variable DEFINE_PER_CPU(int, y) __get_cpu_var(y) = x; Converts to this_cpu_write(y, x); 6. Increment/Decrement etc of a per cpu variable DEFINE_PER_CPU(int, y); __get_cpu_var(y)++ Converts to this_cpu_inc(y) Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> CC: linux390@de.ibm.com Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Christoph Lameter <cl@linux.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2014-08-18 01:30:45 +08:00
mcck = this_cpu_ptr(&cpu_mcck);
if (mci.sd) {
/* System damage -> stopping machine */
s390_handle_damage();
}
/*
* Reinject the instruction processing damages' machine checks
* including Delayed Access Exception into the guest
* instead of damaging the host if they happen in the guest.
*/
if (mci.pd && !test_cpu_flag(CIF_MCCK_GUEST)) {
if (mci.b) {
/* Processing backup -> verify if we can survive this */
u64 z_mcic, o_mcic, t_mcic;
z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<29);
o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 |
1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 |
1ULL<<30 | 1ULL<<21 | 1ULL<<20 | 1ULL<<17 |
1ULL<<16);
t_mcic = mci.val;
if (((t_mcic & z_mcic) != 0) ||
((t_mcic & o_mcic) != o_mcic)) {
s390_handle_damage();
}
/*
* Nullifying exigent condition, therefore we might
* retry this instruction.
*/
spin_lock(&ipd_lock);
tmp = get_tod_clock();
if (((tmp - last_ipd) >> 12) < MAX_IPD_TIME)
ipd_count++;
else
ipd_count = 1;
last_ipd = tmp;
if (ipd_count == MAX_IPD_COUNT)
s390_handle_damage();
spin_unlock(&ipd_lock);
} else {
/* Processing damage -> stopping machine */
s390_handle_damage();
}
}
if (s390_check_registers(mci, user_mode(regs))) {
/*
* Couldn't restore all register contents for the
* user space process -> mark task for termination.
*/
mcck->kill_task = 1;
mcck->mcck_code = mci.val;
mcck_pending = 1;
}
/*
* Backup the machine check's info if it happens when the guest
* is running.
*/
if (test_cpu_flag(CIF_MCCK_GUEST))
s390_backup_mcck_info(regs);
if (mci.cd) {
/* Timing facility damage */
s390_handle_damage();
}
if (mci.ed && mci.ec) {
/* External damage */
if (S390_lowcore.external_damage_code & (1U << ED_STP_SYNC))
mcck->stp_queue |= stp_sync_check();
if (S390_lowcore.external_damage_code & (1U << ED_STP_ISLAND))
mcck->stp_queue |= stp_island_check();
mcck_pending = 1;
}
/*
* Reinject storage related machine checks into the guest if they
* happen when the guest is running.
*/
if (!test_cpu_flag(CIF_MCCK_GUEST)) {
if (mci.se)
/* Storage error uncorrected */
s390_handle_damage();
if (mci.ke)
/* Storage key-error uncorrected */
s390_handle_damage();
if (mci.ds && mci.fa)
/* Storage degradation */
s390_handle_damage();
}
if (mci.cp) {
/* Channel report word pending */
mcck->channel_report = 1;
mcck_pending = 1;
}
if (mci.w) {
/* Warning pending */
mcck->warning = 1;
mcck_pending = 1;
}
/*
* If there are only Channel Report Pending and External Damage
* machine checks, they will not be reinjected into the guest
* because they refer to host conditions only.
*/
mcck_dam_code = (mci.val & MCIC_SUBCLASS_MASK);
if (test_cpu_flag(CIF_MCCK_GUEST) &&
(mcck_dam_code & MCCK_CODE_NO_GUEST) != mcck_dam_code) {
/* Set exit reason code for host's later handling */
*((long *)(regs->gprs[15] + __SF_SIE_REASON)) = -EINTR;
}
clear_cpu_flag(CIF_MCCK_GUEST);
if (user_mode(regs) && mcck_pending) {
nmi_exit();
return 1;
}
if (mcck_pending)
schedule_mcck_handler();
nmi_exit();
return 0;
}
NOKPROBE_SYMBOL(s390_do_machine_check);
static int __init machine_check_init(void)
{
ctl_set_bit(14, 25); /* enable external damage MCH */
ctl_set_bit(14, 27); /* enable system recovery MCH */
ctl_set_bit(14, 24); /* enable warning MCH */
return 0;
}
early_initcall(machine_check_init);