summaryrefslogtreecommitdiffstats
path: root/arch/s390/kernel
diff options
context:
space:
mode:
Diffstat (limited to 'arch/s390/kernel')
-rw-r--r--arch/s390/kernel/asm-offsets.c1
-rw-r--r--arch/s390/kernel/compat_signal.c32
-rw-r--r--arch/s390/kernel/cpcmd.c13
-rw-r--r--arch/s390/kernel/debug.c9
-rw-r--r--arch/s390/kernel/dumpstack.c2
-rw-r--r--arch/s390/kernel/early.c29
-rw-r--r--arch/s390/kernel/entry.S7
-rw-r--r--arch/s390/kernel/head.S3
-rw-r--r--arch/s390/kernel/head64.S4
-rw-r--r--arch/s390/kernel/irq.c3
-rw-r--r--arch/s390/kernel/perf_cpum_sf.c11
-rw-r--r--arch/s390/kernel/relocate_kernel.S5
-rw-r--r--arch/s390/kernel/setup.c14
-rw-r--r--arch/s390/kernel/smp.c4
-rw-r--r--arch/s390/kernel/suspend.c24
-rw-r--r--arch/s390/kernel/time.c67
-rw-r--r--arch/s390/kernel/topology.c148
-rw-r--r--arch/s390/kernel/vdso.c2
-rw-r--r--arch/s390/kernel/vdso32/vdso32.lds.S4
-rw-r--r--arch/s390/kernel/vdso64/vdso64.lds.S4
20 files changed, 280 insertions, 106 deletions
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index b65c414b6c0e..3d42f91c95fd 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -158,6 +158,7 @@ int main(void)
OFFSET(__LC_LAST_UPDATE_CLOCK, lowcore, last_update_clock);
OFFSET(__LC_INT_CLOCK, lowcore, int_clock);
OFFSET(__LC_MCCK_CLOCK, lowcore, mcck_clock);
+ OFFSET(__LC_BOOT_CLOCK, lowcore, boot_clock);
OFFSET(__LC_CURRENT, lowcore, current_task);
OFFSET(__LC_KERNEL_STACK, lowcore, kernel_stack);
OFFSET(__LC_ASYNC_STACK, lowcore, async_stack);
diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c
index c620049c61f2..f549c4657376 100644
--- a/arch/s390/kernel/compat_signal.c
+++ b/arch/s390/kernel/compat_signal.c
@@ -75,35 +75,34 @@ int copy_siginfo_to_user32(compat_siginfo_t __user *to, const siginfo_t *from)
at the same time. */
err = __put_user(from->si_signo, &to->si_signo);
err |= __put_user(from->si_errno, &to->si_errno);
- err |= __put_user((short)from->si_code, &to->si_code);
+ err |= __put_user(from->si_code, &to->si_code);
if (from->si_code < 0)
err |= __copy_to_user(&to->_sifields._pad, &from->_sifields._pad, SI_PAD_SIZE);
else {
- switch (from->si_code >> 16) {
- case __SI_RT >> 16: /* This is not generated by the kernel as of now. */
- case __SI_MESGQ >> 16:
+ switch (siginfo_layout(from->si_signo, from->si_code)) {
+ case SIL_RT:
err |= __put_user(from->si_int, &to->si_int);
/* fallthrough */
- case __SI_KILL >> 16:
+ case SIL_KILL:
err |= __put_user(from->si_pid, &to->si_pid);
err |= __put_user(from->si_uid, &to->si_uid);
break;
- case __SI_CHLD >> 16:
+ case SIL_CHLD:
err |= __put_user(from->si_pid, &to->si_pid);
err |= __put_user(from->si_uid, &to->si_uid);
err |= __put_user(from->si_utime, &to->si_utime);
err |= __put_user(from->si_stime, &to->si_stime);
err |= __put_user(from->si_status, &to->si_status);
break;
- case __SI_FAULT >> 16:
+ case SIL_FAULT:
err |= __put_user((unsigned long) from->si_addr,
&to->si_addr);
break;
- case __SI_POLL >> 16:
+ case SIL_POLL:
err |= __put_user(from->si_band, &to->si_band);
err |= __put_user(from->si_fd, &to->si_fd);
break;
- case __SI_TIMER >> 16:
+ case SIL_TIMER:
err |= __put_user(from->si_tid, &to->si_tid);
err |= __put_user(from->si_overrun, &to->si_overrun);
err |= __put_user(from->si_int, &to->si_int);
@@ -127,32 +126,31 @@ int copy_siginfo_from_user32(siginfo_t *to, compat_siginfo_t __user *from)
if (to->si_code < 0)
err |= __copy_from_user(&to->_sifields._pad, &from->_sifields._pad, SI_PAD_SIZE);
else {
- switch (to->si_code >> 16) {
- case __SI_RT >> 16: /* This is not generated by the kernel as of now. */
- case __SI_MESGQ >> 16:
+ switch (siginfo_layout(to->si_signo, to->si_code)) {
+ case SIL_RT:
err |= __get_user(to->si_int, &from->si_int);
/* fallthrough */
- case __SI_KILL >> 16:
+ case SIL_KILL:
err |= __get_user(to->si_pid, &from->si_pid);
err |= __get_user(to->si_uid, &from->si_uid);
break;
- case __SI_CHLD >> 16:
+ case SIL_CHLD:
err |= __get_user(to->si_pid, &from->si_pid);
err |= __get_user(to->si_uid, &from->si_uid);
err |= __get_user(to->si_utime, &from->si_utime);
err |= __get_user(to->si_stime, &from->si_stime);
err |= __get_user(to->si_status, &from->si_status);
break;
- case __SI_FAULT >> 16:
+ case SIL_FAULT:
err |= __get_user(tmp, &from->si_addr);
to->si_addr = (void __force __user *)
(u64) (tmp & PSW32_ADDR_INSN);
break;
- case __SI_POLL >> 16:
+ case SIL_POLL:
err |= __get_user(to->si_band, &from->si_band);
err |= __get_user(to->si_fd, &from->si_fd);
break;
- case __SI_TIMER >> 16:
+ case SIL_TIMER:
err |= __get_user(to->si_tid, &from->si_tid);
err |= __get_user(to->si_overrun, &from->si_overrun);
err |= __get_user(to->si_int, &from->si_int);
diff --git a/arch/s390/kernel/cpcmd.c b/arch/s390/kernel/cpcmd.c
index 9f0e4a2785f7..63bc6603e0ed 100644
--- a/arch/s390/kernel/cpcmd.c
+++ b/arch/s390/kernel/cpcmd.c
@@ -14,6 +14,7 @@
#include <linux/spinlock.h>
#include <linux/stddef.h>
#include <linux/string.h>
+#include <linux/mm.h>
#include <asm/diag.h>
#include <asm/ebcdic.h>
#include <asm/cpcmd.h>
@@ -28,9 +29,7 @@ static int diag8_noresponse(int cmdlen)
register unsigned long reg3 asm ("3") = cmdlen;
asm volatile(
- " sam31\n"
" diag %1,%0,0x8\n"
- " sam64\n"
: "+d" (reg3) : "d" (reg2) : "cc");
return reg3;
}
@@ -43,9 +42,7 @@ static int diag8_response(int cmdlen, char *response, int *rlen)
register unsigned long reg5 asm ("5") = *rlen;
asm volatile(
- " sam31\n"
" diag %2,%0,0x8\n"
- " sam64\n"
" brc 8,1f\n"
" agr %1,%4\n"
"1:\n"
@@ -57,7 +54,6 @@ static int diag8_response(int cmdlen, char *response, int *rlen)
/*
* __cpcmd has some restrictions over cpcmd
- * - the response buffer must reside below 2GB (if any)
* - __cpcmd is unlocked and therefore not SMP-safe
*/
int __cpcmd(const char *cmd, char *response, int rlen, int *response_code)
@@ -88,13 +84,12 @@ EXPORT_SYMBOL(__cpcmd);
int cpcmd(const char *cmd, char *response, int rlen, int *response_code)
{
+ unsigned long flags;
char *lowbuf;
int len;
- unsigned long flags;
- if ((virt_to_phys(response) != (unsigned long) response) ||
- (((unsigned long)response + rlen) >> 31)) {
- lowbuf = kmalloc(rlen, GFP_KERNEL | GFP_DMA);
+ if (is_vmalloc_or_module_addr(response)) {
+ lowbuf = kmalloc(rlen, GFP_KERNEL);
if (!lowbuf) {
pr_warn("The cpcmd kernel function failed to allocate a response buffer\n");
return -ENOMEM;
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index 86b3e74f569e..1d9e83c401fc 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -866,7 +866,8 @@ static inline void
debug_finish_entry(debug_info_t * id, debug_entry_t* active, int level,
int exception)
{
- active->id.stck = get_tod_clock_fast() - sched_clock_base_cc;
+ active->id.stck = get_tod_clock_fast() -
+ *(unsigned long long *) &tod_clock_base[1];
active->id.fields.cpuid = smp_processor_id();
active->caller = __builtin_return_address(0);
active->id.fields.exception = exception;
@@ -1455,15 +1456,15 @@ int
debug_dflt_header_fn(debug_info_t * id, struct debug_view *view,
int area, debug_entry_t * entry, char *out_buf)
{
- unsigned long sec, usec;
+ unsigned long base, sec, usec;
char *except_str;
unsigned long caller;
int rc = 0;
unsigned int level;
level = entry->id.fields.level;
- sec = (entry->id.stck >> 12) + (sched_clock_base_cc >> 12);
- sec = sec - (TOD_UNIX_EPOCH >> 12);
+ base = (*(unsigned long *) &tod_clock_base[0]) >> 4;
+ sec = (entry->id.stck >> 12) + base - (TOD_UNIX_EPOCH >> 12);
usec = do_div(sec, USEC_PER_SEC);
if (entry->id.fields.exception)
diff --git a/arch/s390/kernel/dumpstack.c b/arch/s390/kernel/dumpstack.c
index dab78babfab6..2aa545dca4d5 100644
--- a/arch/s390/kernel/dumpstack.c
+++ b/arch/s390/kernel/dumpstack.c
@@ -76,7 +76,7 @@ void dump_trace(dump_trace_func_t func, void *data, struct task_struct *task,
frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
#ifdef CONFIG_CHECK_STACK
sp = __dump_trace(func, data, sp,
- S390_lowcore.panic_stack + frame_size - 4096,
+ S390_lowcore.panic_stack + frame_size - PAGE_SIZE,
S390_lowcore.panic_stack + frame_size);
#endif
sp = __dump_trace(func, data, sp,
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index 5d20182ee8ae..60181caf8e8a 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -53,8 +53,9 @@ static void __init reset_tod_clock(void)
if (set_tod_clock(TOD_UNIX_EPOCH) != 0 || store_tod_clock(&time) != 0)
disabled_wait(0);
- sched_clock_base_cc = TOD_UNIX_EPOCH;
- S390_lowcore.last_update_clock = sched_clock_base_cc;
+ memset(tod_clock_base, 0, 16);
+ *(__u64 *) &tod_clock_base[1] = TOD_UNIX_EPOCH;
+ S390_lowcore.last_update_clock = TOD_UNIX_EPOCH;
}
#ifdef CONFIG_SHARED_KERNEL
@@ -165,8 +166,8 @@ static noinline __init void create_kernel_nss(void)
}
/* re-initialize cputime accounting. */
- sched_clock_base_cc = get_tod_clock();
- S390_lowcore.last_update_clock = sched_clock_base_cc;
+ get_tod_clock_ext(tod_clock_base);
+ S390_lowcore.last_update_clock = *(__u64 *) &tod_clock_base[1];
S390_lowcore.last_update_timer = 0x7fffffffffffffffULL;
S390_lowcore.user_timer = 0;
S390_lowcore.system_timer = 0;
@@ -387,6 +388,12 @@ static __init void detect_machine_facilities(void)
}
if (test_facility(133))
S390_lowcore.machine_flags |= MACHINE_FLAG_GS;
+ if (test_facility(139) && (tod_clock_base[1] & 0x80)) {
+ /* Enabled signed clock comparator comparisons */
+ S390_lowcore.machine_flags |= MACHINE_FLAG_SCC;
+ clock_comparator_max = -1ULL >> 1;
+ __ctl_set_bit(0, 53);
+ }
}
static inline void save_vector_registers(void)
@@ -397,23 +404,11 @@ static inline void save_vector_registers(void)
#endif
}
-static int __init topology_setup(char *str)
-{
- bool enabled;
- int rc;
-
- rc = kstrtobool(str, &enabled);
- if (!rc && !enabled)
- S390_lowcore.machine_flags &= ~MACHINE_FLAG_TOPOLOGY;
- return rc;
-}
-early_param("topology", topology_setup);
-
static int __init disable_vector_extension(char *str)
{
S390_lowcore.machine_flags &= ~MACHINE_FLAG_VX;
__ctl_clear_bit(0, 17);
- return 1;
+ return 0;
}
early_param("novx", disable_vector_extension);
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 21900e1cee9c..d185aa3965bf 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -521,12 +521,15 @@ ENTRY(pgm_check_handler)
tmhh %r8,0x0001 # test problem state bit
jnz 2f # -> fault in user space
#if IS_ENABLED(CONFIG_KVM)
- # cleanup critical section for sie64a
+ # cleanup critical section for program checks in sie64a
lgr %r14,%r9
slg %r14,BASED(.Lsie_critical_start)
clg %r14,BASED(.Lsie_critical_length)
jhe 0f
- brasl %r14,.Lcleanup_sie
+ lg %r14,__SF_EMPTY(%r15) # get control block pointer
+ ni __SIE_PROG0C+3(%r14),0xfe # no longer in SIE
+ lctlg %c1,%c1,__LC_USER_ASCE # load primary asce
+ larl %r9,sie_exit # skip forward to sie_exit
#endif
0: tmhh %r8,0x4000 # PER bit set in old PSW ?
jnz 1f # -> enabled, can't be a double fault
diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S
index eff5b31671d4..8ed753c72d9b 100644
--- a/arch/s390/kernel/head.S
+++ b/arch/s390/kernel/head.S
@@ -302,7 +302,8 @@ ENTRY(startup_kdump)
xc 0xe00(256),0xe00
xc 0xf00(256),0xf00
lctlg %c0,%c15,0x200(%r0) # initialize control registers
- stck __LC_LAST_UPDATE_CLOCK
+ stcke __LC_BOOT_CLOCK
+ mvc __LC_LAST_UPDATE_CLOCK(8),__LC_BOOT_CLOCK+1
spt 6f-.LPG0(%r13)
mvc __LC_LAST_UPDATE_TIMER(8),6f-.LPG0(%r13)
l %r15,.Lstack-.LPG0(%r13)
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index 31c91f24e562..0d8f2a858ced 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -21,8 +21,8 @@ ENTRY(startup_continue)
xc __LC_LPP+1(7,0),__LC_LPP+1 # clear lpp and current_pid
mvi __LC_LPP,0x80 # and set LPP_MAGIC
.insn s,0xb2800000,__LC_LPP # load program parameter
-0: larl %r1,sched_clock_base_cc
- mvc 0(8,%r1),__LC_LAST_UPDATE_CLOCK
+0: larl %r1,tod_clock_base
+ mvc 0(16,%r1),__LC_BOOT_CLOCK
larl %r13,.LPG1 # get base
lctlg %c0,%c15,.Lctl-.LPG1(%r13) # load control registers
lg %r12,.Lparmaddr-.LPG1(%r13) # pointer to parameter area
diff --git a/arch/s390/kernel/irq.c b/arch/s390/kernel/irq.c
index 6dca93b29bed..a2fdff0e730b 100644
--- a/arch/s390/kernel/irq.c
+++ b/arch/s390/kernel/irq.c
@@ -105,7 +105,8 @@ void do_IRQ(struct pt_regs *regs, int irq)
old_regs = set_irq_regs(regs);
irq_enter();
- if (S390_lowcore.int_clock >= S390_lowcore.clock_comparator)
+ if (tod_after_eq(S390_lowcore.int_clock,
+ S390_lowcore.clock_comparator))
/* Serve timer interrupts first. */
clock_comparator_work();
generic_handle_irq(irq);
diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c
index 0c82f7903fc7..7e1e40323b78 100644
--- a/arch/s390/kernel/perf_cpum_sf.c
+++ b/arch/s390/kernel/perf_cpum_sf.c
@@ -823,9 +823,12 @@ static int cpumsf_pmu_event_init(struct perf_event *event)
}
/* Check online status of the CPU to which the event is pinned */
- if ((unsigned int)event->cpu >= nr_cpumask_bits ||
- (event->cpu >= 0 && !cpu_online(event->cpu)))
- return -ENODEV;
+ if (event->cpu >= 0) {
+ if ((unsigned int)event->cpu >= nr_cpumask_bits)
+ return -ENODEV;
+ if (!cpu_online(event->cpu))
+ return -ENODEV;
+ }
/* Force reset of idle/hv excludes regardless of what the
* user requested.
@@ -998,7 +1001,7 @@ static int perf_push_sample(struct perf_event *event, struct sf_raw_sample *sfr)
psw_bits(regs.psw).ia = sfr->basic.ia;
psw_bits(regs.psw).dat = sfr->basic.T;
psw_bits(regs.psw).wait = sfr->basic.W;
- psw_bits(regs.psw).per = sfr->basic.P;
+ psw_bits(regs.psw).pstate = sfr->basic.P;
psw_bits(regs.psw).as = sfr->basic.AS;
/*
diff --git a/arch/s390/kernel/relocate_kernel.S b/arch/s390/kernel/relocate_kernel.S
index cfac28330b03..4bdc65636603 100644
--- a/arch/s390/kernel/relocate_kernel.S
+++ b/arch/s390/kernel/relocate_kernel.S
@@ -7,6 +7,7 @@
*/
#include <linux/linkage.h>
+#include <asm/page.h>
#include <asm/sigp.h>
/*
@@ -55,8 +56,8 @@ ENTRY(relocate_kernel)
.back_pgm:
lmg %r0,%r15,gprregs-.base(%r13)
.top:
- lghi %r7,4096 # load PAGE_SIZE in r7
- lghi %r9,4096 # load PAGE_SIZE in r9
+ lghi %r7,PAGE_SIZE # load PAGE_SIZE in r7
+ lghi %r9,PAGE_SIZE # load PAGE_SIZE in r9
lg %r5,0(%r2) # read another word for indirection page
aghi %r2,8 # increment pointer
tml %r5,0x1 # is it a destination page?
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 3d1d808ea8a9..164a1e16b53e 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -305,7 +305,7 @@ static void __init setup_lowcore(void)
/*
* Setup lowcore for boot cpu
*/
- BUILD_BUG_ON(sizeof(struct lowcore) != LC_PAGES * 4096);
+ BUILD_BUG_ON(sizeof(struct lowcore) != LC_PAGES * PAGE_SIZE);
lc = memblock_virt_alloc_low(sizeof(*lc), sizeof(*lc));
lc->restart_psw.mask = PSW_KERNEL_BITS;
lc->restart_psw.addr = (unsigned long) restart_int_handler;
@@ -323,7 +323,7 @@ static void __init setup_lowcore(void)
lc->io_new_psw.mask = PSW_KERNEL_BITS |
PSW_MASK_DAT | PSW_MASK_MCHECK;
lc->io_new_psw.addr = (unsigned long) io_int_handler;
- lc->clock_comparator = -1ULL;
+ lc->clock_comparator = clock_comparator_max;
lc->kernel_stack = ((unsigned long) &init_thread_union)
+ THREAD_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs);
lc->async_stack = (unsigned long)
@@ -469,10 +469,10 @@ static void __init setup_memory_end(void)
vmalloc_size = VMALLOC_END ?: (128UL << 30) - MODULES_LEN;
tmp = (memory_end ?: max_physmem_end) / PAGE_SIZE;
tmp = tmp * (sizeof(struct page) + PAGE_SIZE);
- if (tmp + vmalloc_size + MODULES_LEN <= (1UL << 42))
- vmax = 1UL << 42; /* 3-level kernel page table */
+ if (tmp + vmalloc_size + MODULES_LEN <= _REGION2_SIZE)
+ vmax = _REGION2_SIZE; /* 3-level kernel page table */
else
- vmax = 1UL << 53; /* 4-level kernel page table */
+ vmax = _REGION1_SIZE; /* 4-level kernel page table */
/* module area is at the end of the kernel address space. */
MODULES_END = vmax;
MODULES_VADDR = MODULES_END - MODULES_LEN;
@@ -818,6 +818,9 @@ static int __init setup_hwcaps(void)
case 0x2965:
strcpy(elf_platform, "z13");
break;
+ case 0x3906:
+ strcpy(elf_platform, "z14");
+ break;
}
/*
@@ -922,6 +925,7 @@ void __init setup_arch(char **cmdline_p)
setup_memory_end();
setup_memory();
dma_contiguous_reserve(memory_end);
+ vmcp_cma_reserve();
check_initrd();
reserve_crashkernel();
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 1020a11a24e5..495ff6959dec 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -293,7 +293,10 @@ static void pcpu_attach_task(struct pcpu *pcpu, struct task_struct *tsk)
lc->lpp = LPP_MAGIC;
lc->current_pid = tsk->pid;
lc->user_timer = tsk->thread.user_timer;
+ lc->guest_timer = tsk->thread.guest_timer;
lc->system_timer = tsk->thread.system_timer;
+ lc->hardirq_timer = tsk->thread.hardirq_timer;
+ lc->softirq_timer = tsk->thread.softirq_timer;
lc->steal_timer = 0;
}
@@ -1181,6 +1184,7 @@ static int __init s390_smp_init(void)
rc = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "s390/smp:online",
smp_cpu_online, smp_cpu_pre_down);
+ rc = rc <= 0 ? rc : 0;
out:
return rc;
}
diff --git a/arch/s390/kernel/suspend.c b/arch/s390/kernel/suspend.c
index 39e2f41b6cf0..c8ea715bfe10 100644
--- a/arch/s390/kernel/suspend.c
+++ b/arch/s390/kernel/suspend.c
@@ -98,10 +98,16 @@ int page_key_alloc(unsigned long pages)
*/
void page_key_read(unsigned long *pfn)
{
+ struct page *page;
unsigned long addr;
-
- addr = (unsigned long) page_address(pfn_to_page(*pfn));
- *(unsigned char *) pfn = (unsigned char) page_get_storage_key(addr);
+ unsigned char key;
+
+ page = pfn_to_page(*pfn);
+ addr = (unsigned long) page_address(page);
+ key = (unsigned char) page_get_storage_key(addr) & 0x7f;
+ if (arch_test_page_nodat(page))
+ key |= 0x80;
+ *(unsigned char *) pfn = key;
}
/*
@@ -126,8 +132,16 @@ void page_key_memorize(unsigned long *pfn)
*/
void page_key_write(void *address)
{
- page_set_storage_key((unsigned long) address,
- page_key_rp->data[page_key_rx], 0);
+ struct page *page;
+ unsigned char key;
+
+ key = page_key_rp->data[page_key_rx];
+ page_set_storage_key((unsigned long) address, key & 0x7f, 0);
+ page = virt_to_page(address);
+ if (key & 0x80)
+ arch_set_page_nodat(page, 0);
+ else
+ arch_set_page_dat(page, 0);
if (++page_key_rx >= PAGE_KEY_DATA_SIZE)
return;
page_key_rp = page_key_rp->next;
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 192efdfac918..5cbd52169348 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -51,8 +51,15 @@
#include <asm/cio.h>
#include "entry.h"
-u64 sched_clock_base_cc = -1; /* Force to data section. */
-EXPORT_SYMBOL_GPL(sched_clock_base_cc);
+unsigned char tod_clock_base[16] __aligned(8) = {
+ /* Force to data section. */
+ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff
+};
+EXPORT_SYMBOL_GPL(tod_clock_base);
+
+u64 clock_comparator_max = -1ULL;
+EXPORT_SYMBOL_GPL(clock_comparator_max);
static DEFINE_PER_CPU(struct clock_event_device, comparators);
@@ -75,7 +82,7 @@ void __init time_early_init(void)
struct ptff_qui qui;
/* Initialize TOD steering parameters */
- tod_steering_end = sched_clock_base_cc;
+ tod_steering_end = *(unsigned long long *) &tod_clock_base[1];
vdso_data->ts_end = tod_steering_end;
if (!test_facility(28))
@@ -111,22 +118,27 @@ unsigned long long monotonic_clock(void)
}
EXPORT_SYMBOL(monotonic_clock);
-static void tod_to_timeval(__u64 todval, struct timespec64 *xt)
+static void ext_to_timespec64(unsigned char *clk, struct timespec64 *xt)
{
- unsigned long long sec;
+ unsigned long long high, low, rem, sec, nsec;
+
+ /* Split extendnd TOD clock to micro-seconds and sub-micro-seconds */
+ high = (*(unsigned long long *) clk) >> 4;
+ low = (*(unsigned long long *)&clk[7]) << 4;
+ /* Calculate seconds and nano-seconds */
+ sec = high;
+ rem = do_div(sec, 1000000);
+ nsec = (((low >> 32) + (rem << 32)) * 1000) >> 32;
- sec = todval >> 12;
- do_div(sec, 1000000);
xt->tv_sec = sec;
- todval -= (sec * 1000000) << 12;
- xt->tv_nsec = ((todval * 1000) >> 12);
+ xt->tv_nsec = nsec;
}
void clock_comparator_work(void)
{
struct clock_event_device *cd;
- S390_lowcore.clock_comparator = -1ULL;
+ S390_lowcore.clock_comparator = clock_comparator_max;
cd = this_cpu_ptr(&comparators);
cd->event_handler(cd);
}
@@ -148,7 +160,7 @@ void init_cpu_timer(void)
struct clock_event_device *cd;
int cpu;
- S390_lowcore.clock_comparator = -1ULL;
+ S390_lowcore.clock_comparator = clock_comparator_max;
set_clock_comparator(S390_lowcore.clock_comparator);
cpu = smp_processor_id();
@@ -179,7 +191,7 @@ static void clock_comparator_interrupt(struct ext_code ext_code,
unsigned long param64)
{
inc_irq_stat(IRQEXT_CLK);
- if (S390_lowcore.clock_comparator == -1ULL)
+ if (S390_lowcore.clock_comparator == clock_comparator_max)
set_clock_comparator(S390_lowcore.clock_comparator);
}
@@ -197,18 +209,28 @@ static void stp_reset(void);
void read_persistent_clock64(struct timespec64 *ts)
{
- __u64 clock;
+ unsigned char clk[STORE_CLOCK_EXT_SIZE];
+ __u64 delta;
- clock = get_tod_clock() - initial_leap_seconds;
- tod_to_timeval(clock - TOD_UNIX_EPOCH, ts);
+ delta = initial_leap_seconds + TOD_UNIX_EPOCH;
+ get_tod_clock_ext(clk);
+ *(__u64 *) &clk[1] -= delta;
+ if (*(__u64 *) &clk[1] > delta)
+ clk[0]--;
+ ext_to_timespec64(clk, ts);
}
void read_boot_clock64(struct timespec64 *ts)
{
- __u64 clock;
+ unsigned char clk[STORE_CLOCK_EXT_SIZE];
+ __u64 delta;
- clock = sched_clock_base_cc - initial_leap_seconds;
- tod_to_timeval(clock - TOD_UNIX_EPOCH, ts);
+ delta = initial_leap_seconds + TOD_UNIX_EPOCH;
+ memcpy(clk, tod_clock_base, 16);
+ *(__u64 *) &clk[1] -= delta;
+ if (*(__u64 *) &clk[1] > delta)
+ clk[0]--;
+ ext_to_timespec64(clk, ts);
}
static u64 read_tod_clock(struct clocksource *cs)
@@ -335,7 +357,7 @@ static unsigned long clock_sync_flags;
* source. If the clock mode is local it will return -EOPNOTSUPP and
* -EAGAIN if the clock is not in sync with the external reference.
*/
-int get_phys_clock(unsigned long long *clock)
+int get_phys_clock(unsigned long *clock)
{
atomic_t *sw_ptr;
unsigned int sw0, sw1;
@@ -406,7 +428,10 @@ static void clock_sync_global(unsigned long long delta)
struct ptff_qto qto;
/* Fixup the monotonic sched clock. */
- sched_clock_base_cc += delta;
+ *(unsigned long long *) &tod_clock_base[1] += delta;
+ if (*(unsigned long long *) &tod_clock_base[1] < delta)
+ /* Epoch overflow */
+ tod_clock_base[0]++;
/* Adjust TOD steering parameters. */
vdso_data->tb_update_count++;
now = get_tod_clock();
@@ -437,7 +462,7 @@ static void clock_sync_global(unsigned long long delta)
static void clock_sync_local(unsigned long long delta)
{
/* Add the delta to the clock comparator. */
- if (S390_lowcore.clock_comparator != -1ULL) {
+ if (S390_lowcore.clock_comparator != clock_comparator_max) {
S390_lowcore.clock_comparator += delta;
set_clock_comparator(S390_lowcore.clock_comparator);
}
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index bb47c92476f0..ed0bdd220e1a 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -8,6 +8,8 @@
#include <linux/workqueue.h>
#include <linux/bootmem.h>
+#include <linux/uaccess.h>
+#include <linux/sysctl.h>
#include <linux/cpuset.h>
#include <linux/device.h>
#include <linux/export.h>
@@ -29,12 +31,20 @@
#define PTF_VERTICAL (1UL)
#define PTF_CHECK (2UL)
+enum {
+ TOPOLOGY_MODE_HW,
+ TOPOLOGY_MODE_SINGLE,
+ TOPOLOGY_MODE_PACKAGE,
+ TOPOLOGY_MODE_UNINITIALIZED
+};
+
struct mask_info {
struct mask_info *next;
unsigned char id;
cpumask_t mask;
};
+static int topology_mode = TOPOLOGY_MODE_UNINITIALIZED;
static void set_topology_timer(void);
static void topology_work_fn(struct work_struct *work);
static struct sysinfo_15_1_x *tl_info;
@@ -59,11 +69,26 @@ static cpumask_t cpu_group_map(struct mask_info *info, unsigned int cpu)
cpumask_t mask;
cpumask_copy(&mask, cpumask_of(cpu));
- if (!MACHINE_HAS_TOPOLOGY)
- return mask;
- for (; info; info = info->next) {
- if (cpumask_test_cpu(cpu, &info->mask))
- return info->mask;
+ switch (topology_mode) {
+ case TOPOLOGY_MODE_HW:
+ while (info) {
+ if (cpumask_test_cpu(cpu, &info->mask)) {
+ mask = info->mask;
+ break;
+ }
+ info = info->next;
+ }
+ if (cpumask_empty(&mask))
+ cpumask_copy(&mask, cpumask_of(cpu));
+ break;
+ case TOPOLOGY_MODE_PACKAGE:
+ cpumask_copy(&mask, cpu_present_mask);
+ break;
+ default:
+ /* fallthrough */
+ case TOPOLOGY_MODE_SINGLE:
+ cpumask_copy(&mask, cpumask_of(cpu));
+ break;
}
return mask;
}
@@ -74,7 +99,7 @@ static cpumask_t cpu_thread_map(unsigned int cpu)
int i;
cpumask_copy(&mask, cpumask_of(cpu));
- if (!MACHINE_HAS_TOPOLOGY)
+ if (topology_mode != TOPOLOGY_MODE_HW)
return mask;
cpu -= cpu % (smp_cpu_mtid + 1);
for (i = 0; i <= smp_cpu_mtid; i++)
@@ -184,10 +209,8 @@ static void topology_update_polarization_simple(void)
{
int cpu;
- mutex_lock(&smp_cpu_state_mutex);
for_each_possible_cpu(cpu)
smp_cpu_set_polarization(cpu, POLARIZATION_HRZ);
- mutex_unlock(&smp_cpu_state_mutex);
}
static int ptf(unsigned long fc)
@@ -223,7 +246,7 @@ int topology_set_cpu_management(int fc)
static void update_cpu_masks(void)
{
struct cpu_topology_s390 *topo;
- int cpu;
+ int cpu, id;
for_each_possible_cpu(cpu) {
topo = &cpu_topology[cpu];
@@ -231,12 +254,13 @@ static void update_cpu_masks(void)
topo->core_mask = cpu_group_map(&socket_info, cpu);
topo->book_mask = cpu_group_map(&book_info, cpu);
topo->drawer_mask = cpu_group_map(&drawer_info, cpu);
- if (!MACHINE_HAS_TOPOLOGY) {
+ if (topology_mode != TOPOLOGY_MODE_HW) {
+ id = topology_mode == TOPOLOGY_MODE_PACKAGE ? 0 : cpu;
topo->thread_id = cpu;
topo->core_id = cpu;
- topo->socket_id = cpu;
- topo->book_id = cpu;
- topo->drawer_id = cpu;
+ topo->socket_id = id;
+ topo->book_id = id;
+ topo->drawer_id = id;
if (cpu_present(cpu))
cpumask_set_cpu(cpu, &cpus_with_topology);
}
@@ -254,6 +278,7 @@ static int __arch_update_cpu_topology(void)
struct sysinfo_15_1_x *info = tl_info;
int rc = 0;
+ mutex_lock(&smp_cpu_state_mutex);
cpumask_clear(&cpus_with_topology);
if (MACHINE_HAS_TOPOLOGY) {
rc = 1;
@@ -263,6 +288,7 @@ static int __arch_update_cpu_topology(void)
update_cpu_masks();
if (!MACHINE_HAS_TOPOLOGY)
topology_update_polarization_simple();
+ mutex_unlock(&smp_cpu_state_mutex);
return rc;
}
@@ -289,6 +315,11 @@ void topology_schedule_update(void)
schedule_work(&topology_work);
}
+static void topology_flush_work(void)
+{
+ flush_work(&topology_work);
+}
+
static void topology_timer_fn(unsigned long ignored)
{
if (ptf(PTF_CHECK))
@@ -459,6 +490,12 @@ void __init topology_init_early(void)
struct sysinfo_15_1_x *info;
set_sched_topology(s390_topology);
+ if (topology_mode == TOPOLOGY_MODE_UNINITIALIZED) {
+ if (MACHINE_HAS_TOPOLOGY)
+ topology_mode = TOPOLOGY_MODE_HW;
+ else
+ topology_mode = TOPOLOGY_MODE_SINGLE;
+ }
if (!MACHINE_HAS_TOPOLOGY)
goto out;
tl_info = memblock_virt_alloc(PAGE_SIZE, PAGE_SIZE);
@@ -474,12 +511,97 @@ out:
__arch_update_cpu_topology();
}
+static inline int topology_get_mode(int enabled)
+{
+ if (!enabled)
+ return TOPOLOGY_MODE_SINGLE;
+ return MACHINE_HAS_TOPOLOGY ? TOPOLOGY_MODE_HW : TOPOLOGY_MODE_PACKAGE;
+}
+
+static inline int topology_is_enabled(void)
+{
+ return topology_mode != TOPOLOGY_MODE_SINGLE;
+}
+
+static int __init topology_setup(char *str)
+{
+ bool enabled;
+ int rc;
+
+ rc = kstrtobool(str, &enabled);
+ if (rc)
+ return rc;
+ topology_mode = topology_get_mode(enabled);
+ return 0;
+}
+early_param("topology", topology_setup);
+
+static int topology_ctl_handler(struct ctl_table *ctl, int write,
+ void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+ unsigned int len;
+ int new_mode;
+ char buf[2];
+
+ if (!*lenp || *ppos) {
+ *lenp = 0;
+ return 0;
+ }
+ if (!write) {
+ strncpy(buf, topology_is_enabled() ? "1\n" : "0\n",
+ ARRAY_SIZE(buf));
+ len = strnlen(buf, ARRAY_SIZE(buf));
+ if (len > *lenp)
+ len = *lenp;
+ if (copy_to_user(buffer, buf, len))
+ return -EFAULT;
+ goto out;
+ }
+ len = *lenp;
+ if (copy_from_user(buf, buffer, len > sizeof(buf) ? sizeof(buf) : len))
+ return -EFAULT;
+ if (buf[0] != '0' && buf[0] != '1')
+ return -EINVAL;
+ mutex_lock(&smp_cpu_state_mutex);
+ new_mode = topology_get_mode(buf[0] == '1');
+ if (topology_mode != new_mode) {
+ topology_mode = new_mode;
+ topology_schedule_update();
+ }
+ mutex_unlock(&smp_cpu_state_mutex);
+ topology_flush_work();
+out:
+ *lenp = len;
+ *ppos += len;
+ return 0;
+}
+
+static struct ctl_table topology_ctl_table[] = {
+ {
+ .procname = "topology",
+ .mode = 0644,
+ .proc_handler = topology_ctl_handler,
+ },
+ { },
+};
+
+static struct ctl_table topology_dir_table[] = {
+ {
+ .procname = "s390",
+ .maxlen = 0,
+ .mode = 0555,
+ .child = topology_ctl_table,
+ },
+ { },
+};
+
static int __init topology_init(void)
{
if (MACHINE_HAS_TOPOLOGY)
set_topology_timer();
else
topology_update_polarization_simple();
+ register_sysctl_table(topology_dir_table);
return device_create_file(cpu_subsys.dev_root, &dev_attr_dispatching);
}
device_initcall(topology_init);
diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c
index b89d19f6f2ab..eacda05b45d7 100644
--- a/arch/s390/kernel/vdso.c
+++ b/arch/s390/kernel/vdso.c
@@ -157,6 +157,8 @@ int vdso_alloc_per_cpu(struct lowcore *lowcore)
page_frame = get_zeroed_page(GFP_KERNEL);
if (!segment_table || !page_table || !page_frame)
goto out;
+ arch_set_page_dat(virt_to_page(segment_table), SEGMENT_ORDER);
+ arch_set_page_dat(virt_to_page(page_table), 0);
/* Initialize per-cpu vdso data page */
vd = (struct vdso_per_cpu_data *) page_frame;
diff --git a/arch/s390/kernel/vdso32/vdso32.lds.S b/arch/s390/kernel/vdso32/vdso32.lds.S
index 8f048c2d6d13..263a7f9eee1e 100644
--- a/arch/s390/kernel/vdso32/vdso32.lds.S
+++ b/arch/s390/kernel/vdso32/vdso32.lds.S
@@ -2,6 +2,8 @@
* This is the infamous ld script for the 32 bits vdso
* library
*/
+
+#include <asm/page.h>
#include <asm/vdso.h>
OUTPUT_FORMAT("elf32-s390", "elf32-s390", "elf32-s390")
@@ -91,7 +93,7 @@ SECTIONS
.debug_ranges 0 : { *(.debug_ranges) }
.gnu.attributes 0 : { KEEP (*(.gnu.attributes)) }
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
PROVIDE(_vdso_data = .);
/DISCARD/ : {
diff --git a/arch/s390/kernel/vdso64/vdso64.lds.S b/arch/s390/kernel/vdso64/vdso64.lds.S
index f35455d497fe..9e3dbbcc1cfc 100644
--- a/arch/s390/kernel/vdso64/vdso64.lds.S
+++ b/arch/s390/kernel/vdso64/vdso64.lds.S
@@ -2,6 +2,8 @@
* This is the infamous ld script for the 64 bits vdso
* library
*/
+
+#include <asm/page.h>
#include <asm/vdso.h>
OUTPUT_FORMAT("elf64-s390", "elf64-s390", "elf64-s390")
@@ -91,7 +93,7 @@ SECTIONS
.debug_ranges 0 : { *(.debug_ranges) }
.gnu.attributes 0 : { KEEP (*(.gnu.attributes)) }
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
PROVIDE(_vdso_data = .);
/DISCARD/ : {
generated by cgit v1.2.3-55-g7522 (git 2.39.0) at 2025-05-05 19:26:35 +0200