From e338125b8a886923ba8367207c144764dc352584 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 19 Jul 2008 09:33:21 +0200 Subject: nohz: adjust tick_nohz_stop_sched_tick() call of s390 as well Signed-off-by: Thomas Gleixner --- arch/s390/kernel/process.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index 85defd01d293..9839767d0842 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -142,7 +142,7 @@ static void default_idle(void) void cpu_idle(void) { for (;;) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) default_idle(); tick_nohz_restart_sched_tick(); -- cgit v1.2.3 From 4a0b2b4dbe1335b8b9886ba3dc85a145d5d938ed Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 1 Jul 2008 18:48:41 +0200 Subject: sysdev: Pass the attribute to the low level sysdev show/store function This allow to dynamically generate attributes and share show/store functions between attributes. Right now most attributes are generated by special macros and lots of duplicated code. With the attribute passed it's instead possible to attach some data to the attribute and then use that in shared low level functions to do different things. I need this for the dynamically generated bank attributes in the x86 machine check code, but it'll allow some further cleanups. I converted all users in tree to the new show/store prototype. It's a single huge patch to avoid unbisectable sections. Runtime tested: x86-32, x86-64 Compiled only: ia64, powerpc Not compile tested/only grep converted: sh, arm, avr32 Signed-off-by: Andi Kleen Signed-off-by: Greg Kroah-Hartman --- arch/s390/kernel/smp.c | 36 ++++++++++++++++++++++++------------ arch/s390/kernel/time.c | 35 ++++++++++++++++++++++++----------- 2 files changed, 48 insertions(+), 23 deletions(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index b6781030cfbd..b795b3e24afd 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -864,7 +864,8 @@ int setup_profiling_timer(unsigned int multiplier) } #ifdef CONFIG_HOTPLUG_CPU -static ssize_t cpu_configure_show(struct sys_device *dev, char *buf) +static ssize_t cpu_configure_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { ssize_t count; @@ -874,8 +875,9 @@ static ssize_t cpu_configure_show(struct sys_device *dev, char *buf) return count; } -static ssize_t cpu_configure_store(struct sys_device *dev, const char *buf, - size_t count) +static ssize_t cpu_configure_store(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t count) { int cpu = dev->id; int val, rc; @@ -922,7 +924,8 @@ out: static SYSDEV_ATTR(configure, 0644, cpu_configure_show, cpu_configure_store); #endif /* CONFIG_HOTPLUG_CPU */ -static ssize_t cpu_polarization_show(struct sys_device *dev, char *buf) +static ssize_t cpu_polarization_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { int cpu = dev->id; ssize_t count; @@ -950,7 +953,8 @@ static ssize_t cpu_polarization_show(struct sys_device *dev, char *buf) } static SYSDEV_ATTR(polarization, 0444, cpu_polarization_show, NULL); -static ssize_t show_cpu_address(struct sys_device *dev, char *buf) +static ssize_t show_cpu_address(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { return sprintf(buf, "%d\n", __cpu_logical_map[dev->id]); } @@ -970,7 +974,8 @@ static struct attribute_group cpu_common_attr_group = { .attrs = cpu_common_attrs, }; -static ssize_t show_capability(struct sys_device *dev, char *buf) +static ssize_t show_capability(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { unsigned int capability; int rc; @@ -982,7 +987,8 @@ static ssize_t show_capability(struct sys_device *dev, char *buf) } static SYSDEV_ATTR(capability, 0444, show_capability, NULL); -static ssize_t show_idle_count(struct sys_device *dev, char *buf) +static ssize_t show_idle_count(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct s390_idle_data *idle; unsigned long long idle_count; @@ -995,7 +1001,8 @@ static ssize_t show_idle_count(struct sys_device *dev, char *buf) } static SYSDEV_ATTR(idle_count, 0444, show_idle_count, NULL); -static ssize_t show_idle_time(struct sys_device *dev, char *buf) +static ssize_t show_idle_time(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct s390_idle_data *idle; unsigned long long new_time; @@ -1112,7 +1119,9 @@ out: return rc; } -static ssize_t __ref rescan_store(struct sys_device *dev, const char *buf, +static ssize_t __ref rescan_store(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t count) { int rc; @@ -1123,7 +1132,9 @@ static ssize_t __ref rescan_store(struct sys_device *dev, const char *buf, static SYSDEV_ATTR(rescan, 0200, NULL, rescan_store); #endif /* CONFIG_HOTPLUG_CPU */ -static ssize_t dispatching_show(struct sys_device *dev, char *buf) +static ssize_t dispatching_show(struct sys_device *dev, + struct sysdev_attribute *attr, + char *buf) { ssize_t count; @@ -1133,8 +1144,9 @@ static ssize_t dispatching_show(struct sys_device *dev, char *buf) return count; } -static ssize_t dispatching_store(struct sys_device *dev, const char *buf, - size_t count) +static ssize_t dispatching_store(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t count) { int val, rc; char delim; diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index f2cede3947b2..ab70d9bd9261 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -1100,7 +1100,9 @@ static inline struct etr_aib *etr_aib_from_dev(struct sys_device *dev) return etr_port1_online ? &etr_port1 : NULL; } -static ssize_t etr_online_show(struct sys_device *dev, char *buf) +static ssize_t etr_online_show(struct sys_device *dev, + struct sysdev_attribute *attr, + char *buf) { unsigned int online; @@ -1109,7 +1111,8 @@ static ssize_t etr_online_show(struct sys_device *dev, char *buf) } static ssize_t etr_online_store(struct sys_device *dev, - const char *buf, size_t count) + struct sysdev_attribute *attr, + const char *buf, size_t count) { unsigned int value; @@ -1136,7 +1139,9 @@ static ssize_t etr_online_store(struct sys_device *dev, static SYSDEV_ATTR(online, 0600, etr_online_show, etr_online_store); -static ssize_t etr_stepping_control_show(struct sys_device *dev, char *buf) +static ssize_t etr_stepping_control_show(struct sys_device *dev, + struct sysdev_attribute *attr, + char *buf) { return sprintf(buf, "%i\n", (dev == &etr_port0_dev) ? etr_eacr.e0 : etr_eacr.e1); @@ -1144,7 +1149,8 @@ static ssize_t etr_stepping_control_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(stepping_control, 0400, etr_stepping_control_show, NULL); -static ssize_t etr_mode_code_show(struct sys_device *dev, char *buf) +static ssize_t etr_mode_code_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { if (!etr_port0_online && !etr_port1_online) /* Status word is not uptodate if both ports are offline. */ @@ -1155,7 +1161,8 @@ static ssize_t etr_mode_code_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(state_code, 0400, etr_mode_code_show, NULL); -static ssize_t etr_untuned_show(struct sys_device *dev, char *buf) +static ssize_t etr_untuned_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1166,7 +1173,8 @@ static ssize_t etr_untuned_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(untuned, 0400, etr_untuned_show, NULL); -static ssize_t etr_network_id_show(struct sys_device *dev, char *buf) +static ssize_t etr_network_id_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1177,7 +1185,8 @@ static ssize_t etr_network_id_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(network, 0400, etr_network_id_show, NULL); -static ssize_t etr_id_show(struct sys_device *dev, char *buf) +static ssize_t etr_id_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1188,7 +1197,8 @@ static ssize_t etr_id_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(id, 0400, etr_id_show, NULL); -static ssize_t etr_port_number_show(struct sys_device *dev, char *buf) +static ssize_t etr_port_number_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1199,7 +1209,8 @@ static ssize_t etr_port_number_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(port, 0400, etr_port_number_show, NULL); -static ssize_t etr_coupled_show(struct sys_device *dev, char *buf) +static ssize_t etr_coupled_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1210,7 +1221,8 @@ static ssize_t etr_coupled_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(coupled, 0400, etr_coupled_show, NULL); -static ssize_t etr_local_time_show(struct sys_device *dev, char *buf) +static ssize_t etr_local_time_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1221,7 +1233,8 @@ static ssize_t etr_local_time_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(local_time, 0400, etr_local_time_show, NULL); -static ssize_t etr_utc_offset_show(struct sys_device *dev, char *buf) +static ssize_t etr_utc_offset_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); -- cgit v1.2.3 From faeba830b086bc9e58748869054e994cb09693cd Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Fri, 20 Jun 2008 15:24:18 +0200 Subject: s390: use virtio_console for KVM on s390 This patch enables virtio_console as the default console on kvm for s390. We currently use the same notify hack as lguest for early console output. I will try to address this for lguest and s390 later. Signed-off-by: Christian Borntraeger Signed-off-by: Rusty Russell --- arch/s390/kernel/setup.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index b358e18273b0..62122bad1e33 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -54,6 +54,7 @@ #include #include #include +#include long psw_kernel_bits = (PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_PRIMARY | PSW_MASK_MCHECK | PSW_DEFAULT_KEY); @@ -766,7 +767,8 @@ setup_arch(char **cmdline_p) printk("We are running under VM (64 bit mode)\n"); else if (MACHINE_IS_KVM) { printk("We are running under KVM (64 bit mode)\n"); - add_preferred_console("ttyS", 1, NULL); + add_preferred_console("hvc", 0, NULL); + s390_virtio_console_init(); } else printk("We are running native (64 bit mode)\n"); #endif /* CONFIG_64BIT */ -- cgit v1.2.3 From ef53d9c5e4da147ecaa43c44c5e5945eb83970a2 Mon Sep 17 00:00:00 2001 From: Srinivasa D S Date: Fri, 25 Jul 2008 01:46:04 -0700 Subject: kprobes: improve kretprobe scalability with hashed locking Currently list of kretprobe instances are stored in kretprobe object (as used_instances,free_instances) and in kretprobe hash table. We have one global kretprobe lock to serialise the access to these lists. This causes only one kretprobe handler to execute at a time. Hence affects system performance, particularly on SMP systems and when return probe is set on lot of functions (like on all systemcalls). Solution proposed here gives fine-grain locks that performs better on SMP system compared to present kretprobe implementation. Solution: 1) Instead of having one global lock to protect kretprobe instances present in kretprobe object and kretprobe hash table. We will have two locks, one lock for protecting kretprobe hash table and another lock for kretporbe object. 2) We hold lock present in kretprobe object while we modify kretprobe instance in kretprobe object and we hold per-hash-list lock while modifying kretprobe instances present in that hash list. To prevent deadlock, we never grab a per-hash-list lock while holding a kretprobe lock. 3) We can remove used_instances from struct kretprobe, as we can track used instances of kretprobe instances using kretprobe hash table. Time duration for kernel compilation ("make -j 8") on a 8-way ppc64 system with return probes set on all systemcalls looks like this. cacheline non-cacheline Un-patched kernel aligned patch aligned patch =============================================================================== real 9m46.784s 9m54.412s 10m2.450s user 40m5.715s 40m7.142s 40m4.273s sys 2m57.754s 2m58.583s 3m17.430s =========================================================== Time duration for kernel compilation ("make -j 8) on the same system, when kernel is not probed. ========================= real 9m26.389s user 40m8.775s sys 2m7.283s ========================= Signed-off-by: Srinivasa DS Signed-off-by: Jim Keniston Acked-by: Ananth N Mavinakayanahalli Cc: Anil S Keshavamurthy Cc: David S. Miller Cc: Masami Hiramatsu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/kprobes.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index 288ad490a6dd..4f82e5b5f879 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -270,7 +270,6 @@ static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs, __ctl_store(kcb->kprobe_saved_ctl, 9, 11); } -/* Called with kretprobe_lock held */ void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, struct pt_regs *regs) { @@ -377,8 +376,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p, unsigned long trampoline_address = (unsigned long)&kretprobe_trampoline; INIT_HLIST_HEAD(&empty_rp); - spin_lock_irqsave(&kretprobe_lock, flags); - head = kretprobe_inst_table_head(current); + kretprobe_hash_lock(current, &head, &flags); /* * It is possible to have multiple instances associated with a given @@ -417,7 +415,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p, regs->psw.addr = orig_ret_address | PSW_ADDR_AMODE; reset_current_kprobe(); - spin_unlock_irqrestore(&kretprobe_lock, flags); + kretprobe_hash_unlock(current, &flags); preempt_enable_no_resched(); hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) { -- cgit v1.2.3 From 69b895fd13d73aebf62b75502eb6513d43057ba3 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 25 Jul 2008 01:47:51 -0700 Subject: S390 topology: don't use kthread() for arch_reinit_sched_domains() Now that it is safe to use get_online_cpus() we can revert [S390] cpu topology: Fix possible deadlock. commit: fd781fa25c9e9c6fd1599df060b05e7c4ad724e5 and call arch_reinit_sched_domains() directly from topology_work_fn(). Signed-off-by: Oleg Nesterov Cc: Gautham R Shenoy Tested-by: Heiko Carstens Cc: Max Krasnyansky Cc: Paul Jackson Cc: Paul Menage Cc: Peter Zijlstra Cc: Vegard Nossum Cc: Martin Schwidefsky Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/topology.c | 14 +------------- 1 file changed, 1 insertion(+), 13 deletions(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index 212d618b0095..632b13e10053 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -9,7 +9,6 @@ #include #include #include -#include #include #include #include @@ -230,20 +229,9 @@ void arch_update_cpu_topology(void) } } -static int topology_kthread(void *data) -{ - arch_reinit_sched_domains(); - return 0; -} - static void topology_work_fn(struct work_struct *work) { - /* We can't call arch_reinit_sched_domains() from a multi-threaded - * workqueue context since it may deadlock in case of cpu hotplug. - * So we have to create a kernel thread in order to call - * arch_reinit_sched_domains(). - */ - kthread_run(topology_kthread, NULL, "topology_update"); + arch_reinit_sched_domains(); } void topology_schedule_update(void) -- cgit v1.2.3 From 9b1a4d38373a5581a4e01032a3ccdd94cd93477b Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Mon, 28 Jul 2008 12:16:30 -0500 Subject: stop_machine: Wean existing callers off stop_machine_run() Signed-off-by: Rusty Russell --- arch/s390/kernel/kprobes.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index 4f82e5b5f879..569079ec4ff0 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -197,7 +197,7 @@ void __kprobes arch_arm_kprobe(struct kprobe *p) args.new = BREAKPOINT_INSTRUCTION; kcb->kprobe_status = KPROBE_SWAP_INST; - stop_machine_run(swap_instruction, &args, NR_CPUS); + stop_machine(swap_instruction, &args, NULL); kcb->kprobe_status = status; } @@ -212,7 +212,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) args.new = p->opcode; kcb->kprobe_status = KPROBE_SWAP_INST; - stop_machine_run(swap_instruction, &args, NR_CPUS); + stop_machine(swap_instruction, &args, NULL); kcb->kprobe_status = status; } @@ -331,7 +331,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) * No kprobe at this address. The fault has not been * caused by a kprobe breakpoint. The race of breakpoint * vs. kprobe remove does not exist because on s390 we - * use stop_machine_run to arm/disarm the breakpoints. + * use stop_machine to arm/disarm the breakpoints. */ goto no_kprobe; -- cgit v1.2.3 From 3a95e8eb34f595a0144adb6e5513d456319bd8a5 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Fri, 1 Aug 2008 16:39:10 +0200 Subject: [S390] ipl: Reboot from alternate device does not work when booting from file During startup we check if diag308 works using diag 308 subcode 6, which stores the actual ipl information. This fails with rc = 0x102, if the system has been ipled from the HMC using load from CD or load from file. In the case of rc = 0x102 we have to assume that diag 308 is working, since it still can be used to ipl from an alternative device. Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/ipl.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 54b2779b5e2f..2dcf590faba6 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -1705,7 +1705,10 @@ void __init setup_ipl(void) void __init ipl_update_parameters(void) { - if (diag308(DIAG308_STORE, &ipl_block) == DIAG308_RC_OK) + int rc; + + rc = diag308(DIAG308_STORE, &ipl_block); + if ((rc == DIAG308_RC_OK) || (rc == DIAG308_RC_NOCONFIG)) diag308_set_works = 1; } -- cgit v1.2.3 From 519620cc3d723d41522191ebd150fba4a3790296 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 1 Aug 2008 16:39:15 +0200 Subject: [S390] Wire up new syscalls. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/compat_wrapper.S | 37 +++++++++++++++++++++++++++++++++++++ arch/s390/kernel/syscalls.S | 6 ++++++ 2 files changed, 43 insertions(+) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index d003a6e16afb..328a20e880b5 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -1732,3 +1732,40 @@ compat_sys_timerfd_gettime_wrapper: lgfr %r2,%r2 # int llgtr %r3,%r3 # struct compat_itimerspec * jg compat_sys_timerfd_gettime + + .globl compat_sys_signalfd4_wrapper +compat_sys_signalfd4_wrapper: + lgfr %r2,%r2 # int + llgtr %r3,%r3 # compat_sigset_t * + llgfr %r4,%r4 # compat_size_t + lgfr %r5,%r5 # int + jg compat_sys_signalfd4 + + .globl sys_eventfd2_wrapper +sys_eventfd2_wrapper: + llgfr %r2,%r2 # unsigned int + lgfr %r3,%r3 # int + jg sys_eventfd2 + + .globl sys_inotify_init1_wrapper +sys_inotify_init1_wrapper: + lgfr %r2,%r2 # int + jg sys_inotify_init1 + + .globl sys_pipe2_wrapper +sys_pipe2_wrapper: + llgtr %r2,%r2 # u32 * + lgfr %r3,%r3 # int + jg sys_pipe2 # branch to system call + + .globl sys_dup3_wrapper +sys_dup3_wrapper: + llgfr %r2,%r2 # unsigned int + llgfr %r3,%r3 # unsigned int + lgfr %r4,%r4 # int + jg sys_dup3 # branch to system call + + .globl sys_epoll_create1_wrapper +sys_epoll_create1_wrapper: + lgfr %r2,%r2 # int + jg sys_epoll_create1 # branch to system call diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index c87ec687d4c6..c66d35e55142 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -330,3 +330,9 @@ SYSCALL(sys_eventfd,sys_eventfd,sys_eventfd_wrapper) SYSCALL(sys_timerfd_create,sys_timerfd_create,sys_timerfd_create_wrapper) SYSCALL(sys_timerfd_settime,sys_timerfd_settime,compat_sys_timerfd_settime_wrapper) /* 320 */ SYSCALL(sys_timerfd_gettime,sys_timerfd_gettime,compat_sys_timerfd_gettime_wrapper) +SYSCALL(sys_signalfd4,sys_signalfd4,compat_sys_signalfd4_wrapper) +SYSCALL(sys_eventfd2,sys_eventfd2,sys_eventfd2_wrapper) +SYSCALL(sys_inotify_init1,sys_inotify_init1,sys_inotify_init1_wrapper) +SYSCALL(sys_pipe2,sys_pipe2,sys_pipe2_wrapper) /* 325 */ +SYSCALL(sys_dup3,sys_dup3,sys_dup3_wrapper) +SYSCALL(sys_epoll_create1,sys_epoll_create1,sys_epoll_create1_wrapper) -- cgit v1.2.3 From d918fe2bd72830dfbe8ca2bb30d49930d1356e6d Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 1 Aug 2008 16:39:18 +0200 Subject: [S390] Remove diag 0x260 call from memory detection. The result of the diag 0x260 call is not always what one would expect. So just remove it. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/mem_detect.c | 23 ----------------------- 1 file changed, 23 deletions(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/mem_detect.c b/arch/s390/kernel/mem_detect.c index 18ed7abe16c5..9872999c66d1 100644 --- a/arch/s390/kernel/mem_detect.c +++ b/arch/s390/kernel/mem_detect.c @@ -9,27 +9,6 @@ #include #include -static int memory_fast_detect(struct mem_chunk *chunk) -{ - unsigned long val0 = 0; - unsigned long val1 = 0xc; - int rc = -EOPNOTSUPP; - - if (ipl_flags & IPL_NSS_VALID) - return -EOPNOTSUPP; - asm volatile( - " diag %1,%2,0x260\n" - "0: lhi %0,0\n" - "1:\n" - EX_TABLE(0b,1b) - : "+d" (rc), "+d" (val0), "+d" (val1) : : "cc"); - - if (rc || val0 != val1) - return -EOPNOTSUPP; - chunk->size = val0 + 1; - return 0; -} - static inline int tprot(unsigned long addr) { int rc = -EFAULT; @@ -84,8 +63,6 @@ void detect_memory_layout(struct mem_chunk chunk[]) unsigned long flags, cr0; memset(chunk, 0, MEMORY_CHUNKS * sizeof(struct mem_chunk)); - if (memory_fast_detect(&chunk[0]) == 0) - return; /* Disable IRQs, DAT and low address protection so tprot does the * right thing and we don't get scheduled away with low address * protection disabled. -- cgit v1.2.3 From 8f8470032570988af2eeff520ca01a32fd908b2b Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 1 Aug 2008 16:39:19 +0200 Subject: [S390] stp: fix section mismatch warning. Fix these two (false positive) warnings by adding an __init annoation: WARNING: vmlinux.o(.text+0x7e6a): Section mismatch in reference from the function stp_reset() to the function .init.text:__alloc_bootmem() The function stp_reset() references the function __init __alloc_bootmem(). This is often because stp_reset lacks a __init annotation or the annotation of __alloc_bootmem is wrong. WARNING: vmlinux.o(.text+0x7ece): Section mismatch in reference from the function stp_reset() to the function .init.text:free_bootmem() The function stp_reset() references the function __init free_bootmem(). This is often because stp_reset lacks a __init annotation or the annotation of free_bootmem is wrong. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/time.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index ab70d9bd9261..ca114fe46ffb 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -1348,7 +1348,7 @@ early_param("stp", early_parse_stp); /* * Reset STP attachment. */ -static void stp_reset(void) +static void __init stp_reset(void) { int rc; -- cgit v1.2.3 From 3e972394f9ee1511e5fdbcf50ac7d79e9cf563bd Mon Sep 17 00:00:00 2001 From: Josef 'Jeff' Sipek Date: Thu, 21 Aug 2008 19:46:31 +0200 Subject: [S390] Fix uninitialized spinlock use Ever since commit 43ca5c3a1cefdaa09231d64485b8f676118bf1e0 ([S390] Convert monitor calls to function calls.), the kernel refused to IPL with spinlock debugging enabled. BUG: spinlock bad magic on CPU#0, swapper/0 lock: 00000000003a4668, .magic: 00000000, .owner: /-1, .owner_cpu: 0 CPU: 0 Not tainted 2.6.25 #1 Process swapper (pid: 0, task: 000000000034f958, ksp: 0000000000377d60) 0000000000377ab8 0000000000352628 0000000000377d60 0000000000377d60 0000000000016af4 00000000fffff7b5 0000000000377d60 0000000000000000 0000000000000000 0000000000377a18 0000000000000009 0000000000377a18 0000000000377a78 000000000023c920 0000000000016af4 0000000000377a18 0000000000000005 0000000000000000 0000000000377b58 0000000000377ab8 Call Trace: ([<0000000000016a60>] show_trace+0xdc/0x108) [<0000000000016b4e>] show_stack+0xc2/0xfc [<0000000000016c9a>] dump_stack+0xb2/0xc0 [<0000000000172dd4>] Signed-off-by: Josef 'Jeff' Sipek Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/process.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index 9839767d0842..3e2c05cb6a87 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -75,7 +75,9 @@ unsigned long thread_saved_pc(struct task_struct *tsk) return sf->gprs[8]; } -DEFINE_PER_CPU(struct s390_idle_data, s390_idle); +DEFINE_PER_CPU(struct s390_idle_data, s390_idle) = { + .lock = __SPIN_LOCK_UNLOCKED(s390_idle.lock) +}; static int s390_idle_enter(void) { -- cgit v1.2.3 From 8853e505a15f31fe608cb50e038ba2f794eed923 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 21 Aug 2008 19:46:32 +0200 Subject: [S390] Remove unneeded spinlock initialization. Remove the now unneeded s390_idle.lock spinlock initialization after Josef Sipek did it the right way in arch/s390/kernel/process.c. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/smp.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index b795b3e24afd..00b9b4dec5eb 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -610,7 +610,6 @@ static void __init smp_create_idle(unsigned int cpu) if (IS_ERR(p)) panic("failed fork for CPU %u: %li", cpu, PTR_ERR(p)); current_set[cpu] = p; - spin_lock_init(&(&per_cpu(s390_idle, cpu))->lock); } static int __cpuinit smp_alloc_lowcore(int cpu) @@ -845,7 +844,6 @@ void __init smp_prepare_boot_cpu(void) current_set[0] = current; smp_cpu_state[0] = CPU_STATE_CONFIGURED; smp_cpu_polarization[0] = POLARIZATION_UNKNWN; - spin_lock_init(&(&__get_cpu_var(s390_idle))->lock); } void __init smp_cpus_done(unsigned int max_cpus) -- cgit v1.2.3 From 5453c1a575df6232f84fad1502b2d559909265cc Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 25 Aug 2008 18:13:26 +0200 Subject: [S390] Fix linker script. 6360b1fbb4a939efd34fc770c2ebd927c55506e0 ("move BUG_TABLE into RODATA") causes this build bug (binutils 2.18.50.0.8.20080709, gcc 4.3.1): AS .tmp_kallsyms1.o LD .tmp_vmlinux2 KSYM .tmp_kallsyms2.S s390x-4.3.1-nm: .tmp_vmlinux2: File truncated No valid symbol. make: *** [.tmp_kallsyms2.S] Error 1 So fix this. Cc: Jan Beulich Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/vmlinux.lds.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index 76c1e60c92f3..607bd67a18ce 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -55,7 +55,7 @@ SECTIONS __start___ex_table = .; *(__ex_table) __stop___ex_table = .; - } + } :data .data : { /* Data */ DATA_DATA -- cgit v1.2.3 From e545a6140b698b2494daf0b32107bdcc5e901390 Mon Sep 17 00:00:00 2001 From: Manfred Spraul Date: Sun, 7 Sep 2008 16:57:22 +0200 Subject: kernel/cpu.c: create a CPU_STARTING cpu_chain notifier Right now, there is no notifier that is called on a new cpu, before the new cpu begins processing interrupts/softirqs. Various kernel function would need that notification, e.g. kvm works around by calling smp_call_function_single(), rcu polls cpu_online_map. The patch adds a CPU_STARTING notification. It also adds a helper function that sends the message to all cpu_chain handlers. Tested on x86-64. All other archs are untested. Especially on sparc, I'm not sure if I got it right. Signed-off-by: Manfred Spraul Signed-off-by: Ingo Molnar --- arch/s390/kernel/smp.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 00b9b4dec5eb..9e8b1f9b8f4d 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -585,6 +585,8 @@ int __cpuinit start_secondary(void *cpuvoid) /* Enable pfault pseudo page faults on this cpu. */ pfault_init(); + /* call cpu notifiers */ + notify_cpu_starting(smp_processor_id()); /* Mark this cpu as online */ spin_lock(&call_lock); cpu_set(smp_processor_id(), cpu_online_map); -- cgit v1.2.3 From 3d6e48f43340343d97839eadb1ab7b6a3ea98797 Mon Sep 17 00:00:00 2001 From: Jarod Wilson Date: Tue, 9 Sep 2008 12:38:56 +0200 Subject: [S390] CVE-2008-1514: prevent ptrace padding area read/write in 31-bit mode When running a 31-bit ptrace, on either an s390 or s390x kernel, reads and writes into a padding area in struct user_regs_struct32 will result in a kernel panic. This is also known as CVE-2008-1514. Test case available here: http://sources.redhat.com/cgi-bin/cvsweb.cgi/~checkout~/tests/ptrace-tests/tests/user-area-padding.c?cvsroot=systemtap Steps to reproduce: 1) wget the above 2) gcc -o user-area-padding-31bit user-area-padding.c -Wall -ggdb2 -D_GNU_SOURCE -m31 3) ./user-area-padding-31bit Test status ----------- Without patch, both s390 and s390x kernels panic. With patch, the test case, as well as the gdb testsuite, pass without incident, padding area reads returning zero, writes ignored. Nb: original version returned -EINVAL on write attempts, which broke the gdb test and made the test case slightly unhappy, Jan Kratochvil suggested the change to return 0 on write attempts. Signed-off-by: Jarod Wilson Tested-by: Jan Kratochvil Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/compat_ptrace.h | 1 + arch/s390/kernel/ptrace.c | 28 ++++++++++++++++++++++++++++ 2 files changed, 29 insertions(+) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/compat_ptrace.h b/arch/s390/kernel/compat_ptrace.h index cde81fa64f89..a2be3a978d5c 100644 --- a/arch/s390/kernel/compat_ptrace.h +++ b/arch/s390/kernel/compat_ptrace.h @@ -42,6 +42,7 @@ struct user_regs_struct32 u32 gprs[NUM_GPRS]; u32 acrs[NUM_ACRS]; u32 orig_gpr2; + /* nb: there's a 4-byte hole here */ s390_fp_regs fp_regs; /* * These per registers are in here so that gdb can modify them diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 2815bfe348a6..c8b08289eb87 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -170,6 +170,13 @@ static unsigned long __peek_user(struct task_struct *child, addr_t addr) */ tmp = (addr_t) task_pt_regs(child)->orig_gpr2; + } else if (addr < (addr_t) &dummy->regs.fp_regs) { + /* + * prevent reads of padding hole between + * orig_gpr2 and fp_regs on s390. + */ + tmp = 0; + } else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) { /* * floating point regs. are stored in the thread structure @@ -270,6 +277,13 @@ static int __poke_user(struct task_struct *child, addr_t addr, addr_t data) */ task_pt_regs(child)->orig_gpr2 = data; + } else if (addr < (addr_t) &dummy->regs.fp_regs) { + /* + * prevent writes of padding hole between + * orig_gpr2 and fp_regs on s390. + */ + return 0; + } else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) { /* * floating point regs. are stored in the thread structure @@ -428,6 +442,13 @@ static u32 __peek_user_compat(struct task_struct *child, addr_t addr) */ tmp = *(__u32*)((addr_t) &task_pt_regs(child)->orig_gpr2 + 4); + } else if (addr < (addr_t) &dummy32->regs.fp_regs) { + /* + * prevent reads of padding hole between + * orig_gpr2 and fp_regs on s390. + */ + tmp = 0; + } else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) { /* * floating point regs. are stored in the thread structure @@ -514,6 +535,13 @@ static int __poke_user_compat(struct task_struct *child, */ *(__u32*)((addr_t) &task_pt_regs(child)->orig_gpr2 + 4) = tmp; + } else if (addr < (addr_t) &dummy32->regs.fp_regs) { + /* + * prevent writess of padding hole between + * orig_gpr2 and fp_regs on s390. + */ + return 0; + } else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) { /* * floating point regs. are stored in the thread structure -- cgit v1.2.3 From d3d238c7744d08c36a114a59cb537d4c0c6c9a86 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 3 Oct 2008 21:54:59 +0200 Subject: [S390] nohz: Fix __udelay. This fixes a regression that came with 934b2857cc576ae53c92a66e63fce7ddcfa74691 ("[S390] nohz/sclp: disable timer on synchronous waits."). If udelay() gets called from a disabled context it sets the clock comparator to a value where it expects the next interrupt. When the interrupt happens the clock comparator gets not reset and therefore the interrupt condition doesn't get cleared. The result is an endless timer interrupt loop. In addition this patch fixes also the following: rcutorture reveals that our __udelay implementation is still buggy, since it might schedule tasklets, but prevents their execution: NOHZ: local_softirq_pending 42 NOHZ: local_softirq_pending 02 NOHZ: local_softirq_pending 142 NOHZ: local_softirq_pending 02 To fix this we make sure that only the clock comparator interrupt is enabled when the enabled wait psw is loaded. Also no code gets called anymore which might schedule tasklets. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/time.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index ca114fe46ffb..06acb1a18bbc 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -169,6 +169,8 @@ void init_cpu_timer(void) static void clock_comparator_interrupt(__u16 code) { + if (S390_lowcore.clock_comparator == -1ULL) + set_clock_comparator(S390_lowcore.clock_comparator); } static void etr_timing_alert(struct etr_irq_parm *); -- cgit v1.2.3 From d86730bb9597b02bff59a3a5a01c0094d71a265f Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 10 Oct 2008 21:33:19 +0200 Subject: [S390] s390: use sys_pause for 31bit pause entry point sys32_pause is a useless copy of the generic sys_pause. (and it's certainly not there for old sparc32 binaries..) Signed-off-by: Christoph Hellwig Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/compat_linux.c | 8 -------- arch/s390/kernel/compat_linux.h | 1 - arch/s390/kernel/compat_wrapper.S | 2 -- arch/s390/kernel/syscalls.S | 2 +- 4 files changed, 1 insertion(+), 12 deletions(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index d7f22226fc4e..98e246dc0233 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -608,14 +608,6 @@ asmlinkage long sys32_settimeofday(struct compat_timeval __user *tv, struct time return do_sys_settimeofday(tv ? &kts : NULL, tz ? &ktz : NULL); } -/* These are here just in case some old sparc32 binary calls it. */ -asmlinkage long sys32_pause(void) -{ - current->state = TASK_INTERRUPTIBLE; - schedule(); - return -ERESTARTNOHAND; -} - asmlinkage long sys32_pread64(unsigned int fd, char __user *ubuf, size_t count, u32 poshi, u32 poslo) { diff --git a/arch/s390/kernel/compat_linux.h b/arch/s390/kernel/compat_linux.h index 20723a062017..05f8516366ab 100644 --- a/arch/s390/kernel/compat_linux.h +++ b/arch/s390/kernel/compat_linux.h @@ -206,7 +206,6 @@ long sys32_gettimeofday(struct compat_timeval __user *tv, struct timezone __user *tz); long sys32_settimeofday(struct compat_timeval __user *tv, struct timezone __user *tz); -long sys32_pause(void); long sys32_pread64(unsigned int fd, char __user *ubuf, size_t count, u32 poshi, u32 poslo); long sys32_pwrite64(unsigned int fd, const char __user *ubuf, diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index 328a20e880b5..ee51ca9e23b5 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -128,8 +128,6 @@ sys32_alarm_wrapper: llgfr %r2,%r2 # unsigned int jg sys_alarm # branch to system call -#sys32_pause_wrapper # void - .globl compat_sys_utime_wrapper compat_sys_utime_wrapper: llgtr %r2,%r2 # char * diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index c66d35e55142..3ae303914b42 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -37,7 +37,7 @@ SYSCALL(sys_stime,sys_ni_syscall,sys32_stime_wrapper) /* 25 old stime syscall * SYSCALL(sys_ptrace,sys_ptrace,sys32_ptrace_wrapper) SYSCALL(sys_alarm,sys_alarm,sys32_alarm_wrapper) NI_SYSCALL /* old fstat syscall */ -SYSCALL(sys_pause,sys_pause,sys32_pause) +SYSCALL(sys_pause,sys_pause,sys_pause) SYSCALL(sys_utime,sys_utime,compat_sys_utime_wrapper) /* 30 */ NI_SYSCALL /* old stty syscall */ NI_SYSCALL /* old gtty syscall */ -- cgit v1.2.3 From 753c4dd6a2fa2af81f5d809d610d29f2d9dd9bc1 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 10 Oct 2008 21:33:20 +0200 Subject: [S390] ptrace changes * System call parameter and result access functions * Add tracehook calls * Split syscall_trace into two functions do_syscall_trace_enter and do_syscall_trace_exit Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/entry.S | 50 +++++++++++++++++++++++++++++-------- arch/s390/kernel/entry64.S | 42 +++++++++++++++++++++++-------- arch/s390/kernel/ptrace.c | 61 +++++++++++++++++++++++++--------------------- arch/s390/kernel/signal.c | 13 ++++++++++ 4 files changed, 118 insertions(+), 48 deletions(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 708cf9cf9a35..ed500ef799b7 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -49,9 +49,9 @@ SP_ILC = STACK_FRAME_OVERHEAD + __PT_ILC SP_TRAP = STACK_FRAME_OVERHEAD + __PT_TRAP SP_SIZE = STACK_FRAME_OVERHEAD + __PT_SIZE -_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | \ +_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ _TIF_MCCK_PENDING | _TIF_RESTART_SVC | _TIF_SINGLE_STEP ) -_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | \ +_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ _TIF_MCCK_PENDING) STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER @@ -318,6 +318,8 @@ sysc_work: bo BASED(sysc_reschedule) tm __TI_flags+3(%r9),_TIF_SIGPENDING bnz BASED(sysc_sigpending) + tm __TI_flags+3(%r9),_TIF_NOTIFY_RESUME + bnz BASED(sysc_notify_resume) tm __TI_flags+3(%r9),_TIF_RESTART_SVC bo BASED(sysc_restart) tm __TI_flags+3(%r9),_TIF_SINGLE_STEP @@ -355,6 +357,16 @@ sysc_sigpending: bo BASED(sysc_singlestep) b BASED(sysc_work_loop) +# +# _TIF_NOTIFY_RESUME is set, call do_notify_resume +# +sysc_notify_resume: + la %r2,SP_PTREGS(%r15) # load pt_regs + l %r1,BASED(.Ldo_notify_resume) + la %r14,BASED(sysc_work_loop) + br %r1 # call do_notify_resume + + # # _TIF_RESTART_SVC is set, set up registers and restart svc # @@ -378,20 +390,21 @@ sysc_singlestep: br %r1 # branch to do_single_step # -# call trace before and after sys_call +# call tracehook_report_syscall_entry/tracehook_report_syscall_exit before +# and after the system call # sysc_tracesys: - l %r1,BASED(.Ltrace) + l %r1,BASED(.Ltrace_entry) la %r2,SP_PTREGS(%r15) # load pt_regs la %r3,0 srl %r7,2 st %r7,SP_R2(%r15) basr %r14,%r1 - clc SP_R2(4,%r15),BASED(.Lnr_syscalls) + cl %r2,BASED(.Lnr_syscalls) bnl BASED(sysc_tracenogo) l %r8,BASED(.Lsysc_table) - l %r7,SP_R2(%r15) # strace might have changed the - sll %r7,2 # system call + lr %r7,%r2 + sll %r7,2 # *4 l %r8,0(%r7,%r8) sysc_tracego: lm %r3,%r6,SP_R3(%r15) @@ -401,9 +414,8 @@ sysc_tracego: sysc_tracenogo: tm __TI_flags+3(%r9),(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT) bz BASED(sysc_return) - l %r1,BASED(.Ltrace) + l %r1,BASED(.Ltrace_exit) la %r2,SP_PTREGS(%r15) # load pt_regs - la %r3,1 la %r14,BASED(sysc_return) br %r1 @@ -666,6 +678,8 @@ io_work_loop: bo BASED(io_reschedule) tm __TI_flags+3(%r9),_TIF_SIGPENDING bnz BASED(io_sigpending) + tm __TI_flags+3(%r9),_TIF_NOTIFY_RESUME + bnz BASED(io_notify_resume) b BASED(io_restore) io_work_done: @@ -704,6 +718,19 @@ io_sigpending: TRACE_IRQS_OFF b BASED(io_work_loop) +# +# _TIF_SIGPENDING is set, call do_signal +# +io_notify_resume: + TRACE_IRQS_ON + stosm __SF_EMPTY(%r15),0x03 # reenable interrupts + la %r2,SP_PTREGS(%r15) # load pt_regs + l %r1,BASED(.Ldo_notify_resume) + basr %r14,%r1 # call do_signal + stnsm __SF_EMPTY(%r15),0xfc # disable I/O and ext. interrupts + TRACE_IRQS_OFF + b BASED(io_work_loop) + /* * External interrupt handler routine */ @@ -1070,6 +1097,8 @@ cleanup_io_leave_insn: .Ldo_IRQ: .long do_IRQ .Ldo_extint: .long do_extint .Ldo_signal: .long do_signal +.Ldo_notify_resume: + .long do_notify_resume .Lhandle_per: .long do_single_step .Ldo_execve: .long do_execve .Lexecve_tail: .long execve_tail @@ -1079,7 +1108,8 @@ cleanup_io_leave_insn: .Lpreempt_schedule_irq: .long preempt_schedule_irq #endif -.Ltrace: .long syscall_trace +.Ltrace_entry: .long do_syscall_trace_enter +.Ltrace_exit: .long do_syscall_trace_exit .Lschedtail: .long schedule_tail .Lsysc_table: .long sys_call_table #ifdef CONFIG_TRACE_IRQFLAGS diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index fee10177dbfc..d7ce150453f2 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -52,9 +52,9 @@ SP_SIZE = STACK_FRAME_OVERHEAD + __PT_SIZE STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER STACK_SIZE = 1 << STACK_SHIFT -_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | \ +_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ _TIF_MCCK_PENDING | _TIF_RESTART_SVC | _TIF_SINGLE_STEP ) -_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | \ +_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ _TIF_MCCK_PENDING) #define BASED(name) name-system_call(%r13) @@ -310,6 +310,8 @@ sysc_work: jo sysc_reschedule tm __TI_flags+7(%r9),_TIF_SIGPENDING jnz sysc_sigpending + tm __TI_flags+7(%r9),_TIF_NOTIFY_RESUME + jnz sysc_notify_resume tm __TI_flags+7(%r9),_TIF_RESTART_SVC jo sysc_restart tm __TI_flags+7(%r9),_TIF_SINGLE_STEP @@ -344,6 +346,14 @@ sysc_sigpending: jo sysc_singlestep j sysc_work_loop +# +# _TIF_NOTIFY_RESUME is set, call do_notify_resume +# +sysc_notify_resume: + la %r2,SP_PTREGS(%r15) # load pt_regs + larl %r14,sysc_work_loop + jg do_notify_resume # call do_notify_resume + # # _TIF_RESTART_SVC is set, set up registers and restart svc # @@ -367,20 +377,19 @@ sysc_singlestep: jg do_single_step # branch to do_sigtrap # -# call syscall_trace before and after system call -# special linkage: %r12 contains the return address for trace_svc +# call tracehook_report_syscall_entry/tracehook_report_syscall_exit before +# and after the system call # sysc_tracesys: la %r2,SP_PTREGS(%r15) # load pt_regs la %r3,0 srl %r7,2 stg %r7,SP_R2(%r15) - brasl %r14,syscall_trace + brasl %r14,do_syscall_trace_enter lghi %r0,NR_syscalls - clg %r0,SP_R2(%r15) + clgr %r0,%r2 jnh sysc_tracenogo - lg %r7,SP_R2(%r15) # strace might have changed the - sll %r7,2 # system call + slag %r7,%r2,2 # *4 lgf %r8,0(%r7,%r10) sysc_tracego: lmg %r3,%r6,SP_R3(%r15) @@ -391,9 +400,8 @@ sysc_tracenogo: tm __TI_flags+7(%r9),(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT) jz sysc_return la %r2,SP_PTREGS(%r15) # load pt_regs - la %r3,1 larl %r14,sysc_return # return point is sysc_return - jg syscall_trace + jg do_syscall_trace_exit # # a new process exits the kernel with ret_from_fork @@ -672,6 +680,8 @@ io_work_loop: jo io_reschedule tm __TI_flags+7(%r9),_TIF_SIGPENDING jnz io_sigpending + tm __TI_flags+7(%r9),_TIF_NOTIFY_RESUME + jnz io_notify_resume j io_restore io_work_done: @@ -712,6 +722,18 @@ io_sigpending: TRACE_IRQS_OFF j io_work_loop +# +# _TIF_NOTIFY_RESUME or is set, call do_notify_resume +# +io_notify_resume: + TRACE_IRQS_ON + stosm __SF_EMPTY(%r15),0x03 # reenable interrupts + la %r2,SP_PTREGS(%r15) # load pt_regs + brasl %r14,do_notify_resume # call do_notify_resume + stnsm __SF_EMPTY(%r15),0xfc # disable I/O and ext. interrupts + TRACE_IRQS_OFF + j io_work_loop + /* * External interrupt handler routine */ diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index c8b08289eb87..1f31be1ecc4b 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include @@ -639,40 +640,44 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, } #endif -asmlinkage void -syscall_trace(struct pt_regs *regs, int entryexit) +asmlinkage long do_syscall_trace_enter(struct pt_regs *regs) { - if (unlikely(current->audit_context) && entryexit) - audit_syscall_exit(AUDITSC_RESULT(regs->gprs[2]), regs->gprs[2]); - - if (!test_thread_flag(TIF_SYSCALL_TRACE)) - goto out; - if (!(current->ptrace & PT_PTRACED)) - goto out; - ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD) - ? 0x80 : 0)); + long ret; /* - * If the debuffer has set an invalid system call number, - * we prepare to skip the system call restart handling. + * The sysc_tracesys code in entry.S stored the system + * call number to gprs[2]. */ - if (!entryexit && regs->gprs[2] >= NR_syscalls) + ret = regs->gprs[2]; + if (test_thread_flag(TIF_SYSCALL_TRACE) && + (tracehook_report_syscall_entry(regs) || + regs->gprs[2] >= NR_syscalls)) { + /* + * Tracing decided this syscall should not happen or the + * debugger stored an invalid system call number. Skip + * the system call and the system call restart handling. + */ regs->trap = -1; - - /* - * this isn't the same as continuing with a signal, but it will do - * for normal use. strace only continues with a signal if the - * stopping signal is not SIGTRAP. -brl - */ - if (current->exit_code) { - send_sig(current->exit_code, current, 1); - current->exit_code = 0; + ret = -1; } - out: - if (unlikely(current->audit_context) && !entryexit) - audit_syscall_entry(test_thread_flag(TIF_31BIT)?AUDIT_ARCH_S390:AUDIT_ARCH_S390X, - regs->gprs[2], regs->orig_gpr2, regs->gprs[3], - regs->gprs[4], regs->gprs[5]); + + if (unlikely(current->audit_context)) + audit_syscall_entry(test_thread_flag(TIF_31BIT) ? + AUDIT_ARCH_S390 : AUDIT_ARCH_S390X, + regs->gprs[2], regs->orig_gpr2, + regs->gprs[3], regs->gprs[4], + regs->gprs[5]); + return ret; +} + +asmlinkage void do_syscall_trace_exit(struct pt_regs *regs) +{ + if (unlikely(current->audit_context)) + audit_syscall_exit(AUDITSC_RESULT(regs->gprs[2]), + regs->gprs[2]); + + if (test_thread_flag(TIF_SYSCALL_TRACE)) + tracehook_report_syscall_exit(regs, 0); } /* diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c index b97682040215..4f7fc3059a8e 100644 --- a/arch/s390/kernel/signal.c +++ b/arch/s390/kernel/signal.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -507,6 +508,12 @@ void do_signal(struct pt_regs *regs) */ if (current->thread.per_info.single_step) set_thread_flag(TIF_SINGLE_STEP); + + /* + * Let tracing know that we've done the handler setup. + */ + tracehook_signal_handler(signr, &info, &ka, regs, + test_thread_flag(TIF_SINGLE_STEP)); } return; } @@ -526,3 +533,9 @@ void do_signal(struct pt_regs *regs) set_thread_flag(TIF_RESTART_SVC); } } + +void do_notify_resume(struct pt_regs *regs) +{ + clear_thread_flag(TIF_NOTIFY_RESUME); + tracehook_notify_resume(regs); +} -- cgit v1.2.3 From 4a672cfa3a7fcbc6f2adc558f34148be1096c561 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 10 Oct 2008 21:33:29 +0200 Subject: [S390] fix initialization of stp chsc_sstpc returns -EIO on error and 0 on success but stp_reset checks against 1 instead of 0. chsc_sstpc used to return 1 on success, one call location has not been updated .. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/time.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 06acb1a18bbc..b94e9e3b694a 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -1356,7 +1356,7 @@ static void __init stp_reset(void) stp_page = alloc_bootmem_pages(PAGE_SIZE); rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000); - if (rc == 1) + if (rc == 0) set_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags); else if (stp_online) { printk(KERN_WARNING "Running on non STP capable machine.\n"); -- cgit v1.2.3 From f7a5000f7a8924e9c5fad1801616601d6dc65a17 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 15 Oct 2008 22:02:05 -0700 Subject: compat: move cp_compat_stat to common code struct stat / compat_stat is the same on all architectures, so cp_compat_stat should be, too. Turns out it is, except that various architectures have slightly and some high2lowuid/high2lowgid or the direct assignment instead of the SET_UID/SET_GID that expands to the correct one anyway. This patch replaces the arch-specific cp_compat_stat implementations with a common one based on the x86-64 one. Signed-off-by: Christoph Hellwig Acked-by: David S. Miller [ sparc bits ] Acked-by: Kyle McMartin [ parisc bits ] Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/compat_linux.c | 35 ----------------------------------- 1 file changed, 35 deletions(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index 98e246dc0233..9b471d785ec1 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -362,41 +362,6 @@ asmlinkage long sys32_ftruncate64(unsigned int fd, unsigned long high, unsigned return sys_ftruncate(fd, (high << 32) | low); } -int cp_compat_stat(struct kstat *stat, struct compat_stat __user *statbuf) -{ - compat_ino_t ino; - int err; - - if (!old_valid_dev(stat->dev) || !old_valid_dev(stat->rdev)) - return -EOVERFLOW; - - ino = stat->ino; - if (sizeof(ino) < sizeof(stat->ino) && ino != stat->ino) - return -EOVERFLOW; - - err = put_user(old_encode_dev(stat->dev), &statbuf->st_dev); - err |= put_user(stat->ino, &statbuf->st_ino); - err |= put_user(stat->mode, &statbuf->st_mode); - err |= put_user(stat->nlink, &statbuf->st_nlink); - err |= put_user(high2lowuid(stat->uid), &statbuf->st_uid); - err |= put_user(high2lowgid(stat->gid), &statbuf->st_gid); - err |= put_user(old_encode_dev(stat->rdev), &statbuf->st_rdev); - err |= put_user(stat->size, &statbuf->st_size); - err |= put_user(stat->atime.tv_sec, &statbuf->st_atime); - err |= put_user(stat->atime.tv_nsec, &statbuf->st_atime_nsec); - err |= put_user(stat->mtime.tv_sec, &statbuf->st_mtime); - err |= put_user(stat->mtime.tv_nsec, &statbuf->st_mtime_nsec); - err |= put_user(stat->ctime.tv_sec, &statbuf->st_ctime); - err |= put_user(stat->ctime.tv_nsec, &statbuf->st_ctime_nsec); - err |= put_user(stat->blksize, &statbuf->st_blksize); - err |= put_user(stat->blocks, &statbuf->st_blocks); -/* fixme - err |= put_user(0, &statbuf->__unused4[0]); - err |= put_user(0, &statbuf->__unused4[1]); -*/ - return err; -} - asmlinkage long sys32_sched_rr_get_interval(compat_pid_t pid, struct compat_timespec __user *interval) { -- cgit v1.2.3 From b418da16dd44810e5d5a22bba377cca80512a524 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 15 Oct 2008 22:02:06 -0700 Subject: compat: generic compat get/settimeofday Nothing arch specific in get/settimeofday. The details of the timeval conversion varied a little from arch to arch, but all with the same results. Also add an extern declaration for sys_tz to linux/time.h because externs in .c files are fowned upon. I'll kill the externs in various other files in a sparate patch. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Christoph Hellwig Acked-by: David S. Miller [ sparc bits ] Cc: "Luck, Tony" Cc: Ralf Baechle Acked-by: Kyle McMartin Cc: Matthew Wilcox Cc: Grant Grundler Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: Ingo Molnar Cc: Thomas Gleixner Cc: "H. Peter Anvin" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/compat_linux.c | 67 --------------------------------------- arch/s390/kernel/compat_linux.h | 4 --- arch/s390/kernel/compat_wrapper.S | 12 +++---- arch/s390/kernel/syscalls.S | 4 +-- 4 files changed, 8 insertions(+), 79 deletions(-) (limited to 'arch/s390/kernel') diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index 9b471d785ec1..4646382af34f 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -279,22 +279,6 @@ asmlinkage long sys32_getegid16(void) return high2lowgid(current->egid); } -/* 32-bit timeval and related flotsam. */ - -static inline long get_tv32(struct timeval *o, struct compat_timeval __user *i) -{ - return (!access_ok(VERIFY_READ, o, sizeof(*o)) || - (__get_user(o->tv_sec, &i->tv_sec) || - __get_user(o->tv_usec, &i->tv_usec))); -} - -static inline long put_tv32(struct compat_timeval __user *o, struct timeval *i) -{ - return (!access_ok(VERIFY_WRITE, o, sizeof(*o)) || - (__put_user(i->tv_sec, &o->tv_sec) || - __put_user(i->tv_usec, &o->tv_usec))); -} - /* * sys32_ipc() is the de-multiplexer for the SysV IPC calls in 32bit emulation. * @@ -522,57 +506,6 @@ sys32_delete_module(const char __user *name_user, unsigned int flags) #endif /* CONFIG_MODULES */ -/* Translations due to time_t size differences. Which affects all - sorts of things, like timeval and itimerval. */ - -extern struct timezone sys_tz; - -asmlinkage long sys32_gettimeofday(struct compat_timeval __user *tv, struct timezone __user *tz) -{ - if (tv) { - struct timeval ktv; - do_gettimeofday(&ktv); - if (put_tv32(tv, &ktv)) - return -EFAULT; - } - if (tz) { - if (copy_to_user(tz, &sys_tz, sizeof(sys_tz))) - return -EFAULT; - } - return 0; -} - -static inline long get_ts32(struct timespec *o, struct compat_timeval __user *i) -{ - long usec; - - if (!access_ok(VERIFY_READ, i, sizeof(*i))) - return -EFAULT; - if (__get_user(o->tv_sec, &i->tv_sec)) - return -EFAULT; - if (__get_user(usec, &i->tv_usec)) - return -EFAULT; - o->tv_nsec = usec * 1000; - return 0; -} - -asmlinkage long sys32_settimeofday(struct compat_timeval __user *tv, struct timezone __user *tz) -{ - struct timespec kts; - struct timezone ktz; - - if (tv) { - if (get_ts32(&kts, tv)) - return -EFAULT; - } - if (tz) { - if (copy_from_user(&ktz, tz, sizeof(ktz))) - return -EFAULT; - } - - return do_sys_settimeofday(tv ? &kts : NULL, tz ? &ktz : NULL); -} - asmlinkage long sys32_pread64(unsigned int fd, char __user *ubuf, size_t count, u32 poshi, u32 poslo) { diff --git a/arch/s390/kernel/compat_linux.h b/arch/s390/kernel/compat_linux.h index 05f8516366ab..836a28842900 100644 --- a/arch/s390/kernel/compat_linux.h +++ b/arch/s390/kernel/compat_linux.h @@ -202,10 +202,6 @@ long sys32_execve(void); long sys32_init_module(void __user *umod, unsigned long len, const char __user *uargs); long sys32_delete_module(const char __user *name_user, unsigned int flags); -long sys32_gettimeofday(struct compat_timeval __user *tv, - struct timezone __user *tz); -long sys32_settimeofday(struct compat_timeval __user *tv, - struct timezone __user *tz); long sys32_pread64(unsigned int fd, char __user *ubuf, size_t count, u32 poshi, u32 poslo); long sys32_pwrite64(unsigned int fd, const char __user *ubuf, diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index ee51ca9e23b5..fc2c97197a53 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -332,17 +332,17 @@ compat_sys_getrusage_wrapper: llgtr %r3,%r3 # struct rusage_emu31 * jg compat_sys_getrusage # branch to system call - .globl sys32_gettimeofday_wrapper -sys32_gettimeofday_wrapper: + .globl compat_sys_gettimeofday_wrapper +compat_sys_gettimeofday_wrapper: llgtr %r2,%r2 # struct timeval_emu31 * llgtr %r3,%r3 # struct timezone * - jg sys32_gettimeofday # branch to system call + jg compat_sys_gettimeofday # branch to system call - .globl sys32_settimeofday_wrapper -sys32_settimeofday_wrapper: + .globl compat_sys_settimeofday_wrapper +compat_sys_settimeofday_wrapper: llgtr %r2,%r2 # struct timeval_emu31 * llgtr %r3,%r3 # struct timezone * - jg sys32_settimeofday # branch to system call + jg compat_sys_settimeofday # branch to system call .globl sys32_getgroups16_wrapper sys32_getgroups16_wrapper: diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index 3ae303914b42..2d61787949d5 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -86,8 +86,8 @@ SYSCALL(sys_sethostname,sys_sethostname,sys32_sethostname_wrapper) SYSCALL(sys_setrlimit,sys_setrlimit,compat_sys_setrlimit_wrapper) /* 75 */ SYSCALL(sys_old_getrlimit,sys_getrlimit,compat_sys_old_getrlimit_wrapper) SYSCALL(sys_getrusage,sys_getrusage,compat_sys_getrusage_wrapper) -SYSCALL(sys_gettimeofday,sys_gettimeofday,sys32_gettimeofday_wrapper) -SYSCALL(sys_settimeofday,sys_settimeofday,sys32_settimeofday_wrapper) +SYSCALL(sys_gettimeofday,sys_gettimeofday,compat_sys_gettimeofday_wrapper) +SYSCALL(sys_settimeofday,sys_settimeofday,compat_sys_settimeofday_wrapper) SYSCALL(sys_getgroups16,sys_ni_syscall,sys32_getgroups16_wrapper) /* 80 old getgroups16 syscall */ SYSCALL(sys_setgroups16,sys_ni_syscall,sys32_setgroups16_wrapper) /* old setgroups16 syscall */ NI_SYSCALL /* old select syscall */ -- cgit v1.2.3