From 316ec154810960052d4586b634156c54d0778f74 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Wed, 15 Apr 2020 15:21:01 +0200 Subject: s390/mm: fix page table upgrade vs 2ndary address mode accesses A page table upgrade in a kernel section that uses secondary address mode will mess up the kernel instructions as follows: Consider the following scenario: two threads are sharing memory. On CPU1 thread 1 does e.g. strnlen_user(). That gets to old_fs = enable_sacf_uaccess(); len = strnlen_user_srst(src, size); and " la %2,0(%1)\n" " la %3,0(%0,%1)\n" " slgr %0,%0\n" " sacf 256\n" "0: srst %3,%2\n" in strnlen_user_srst(). At that point we are in secondary space mode, control register 1 points to kernel page table and instruction fetching happens via c1, rather than usual c13. Interrupts are not disabled, for obvious reasons. On CPU2 thread 2 does MAP_FIXED mmap(), forcing the upgrade of page table from 3-level to e.g. 4-level one. We'd allocated new top-level table, set it up and now we hit this: notify = 1; spin_unlock_bh(&mm->page_table_lock); } if (notify) on_each_cpu(__crst_table_upgrade, mm, 0); OK, we need to actually change over to use of new page table and we need that to happen in all threads that are currently running. Which happens to include the thread 1. IPI is delivered and we have static void __crst_table_upgrade(void *arg) { struct mm_struct *mm = arg; if (current->active_mm == mm) set_user_asce(mm); __tlb_flush_local(); } run on CPU1. That does static inline void set_user_asce(struct mm_struct *mm) { S390_lowcore.user_asce = mm->context.asce; OK, user page table address updated... __ctl_load(S390_lowcore.user_asce, 1, 1); ... and control register 1 set to it. clear_cpu_flag(CIF_ASCE_PRIMARY); } IPI is run in home space mode, so it's fine - insns are fetched using c13, which always points to kernel page table. But as soon as we return from the interrupt, previous PSW is restored, putting CPU1 back into secondary space mode, at which point we no longer get the kernel instructions from the kernel mapping. The fix is to only fixup the control registers that are currently in use for user processes during the page table update. We must also disable interrupts in enable_sacf_uaccess to synchronize the cr and thread.mm_segment updates against the on_each-cpu. Fixes: 0aaba41b58bc ("s390: remove all code using the access register mode") Cc: stable@vger.kernel.org # 4.15+ Reported-by: Al Viro Reviewed-by: Gerald Schaefer References: CVE-2020-11884 Signed-off-by: Christian Borntraeger --- arch/s390/lib/uaccess.c | 4 ++++ arch/s390/mm/pgalloc.c | 16 ++++++++++++++-- 2 files changed, 18 insertions(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/lib/uaccess.c b/arch/s390/lib/uaccess.c index c4f8039a35e8..0267405ab7c6 100644 --- a/arch/s390/lib/uaccess.c +++ b/arch/s390/lib/uaccess.c @@ -64,10 +64,13 @@ mm_segment_t enable_sacf_uaccess(void) { mm_segment_t old_fs; unsigned long asce, cr; + unsigned long flags; old_fs = current->thread.mm_segment; if (old_fs & 1) return old_fs; + /* protect against a concurrent page table upgrade */ + local_irq_save(flags); current->thread.mm_segment |= 1; asce = S390_lowcore.kernel_asce; if (likely(old_fs == USER_DS)) { @@ -83,6 +86,7 @@ mm_segment_t enable_sacf_uaccess(void) __ctl_load(asce, 7, 7); set_cpu_flag(CIF_ASCE_SECONDARY); } + local_irq_restore(flags); return old_fs; } EXPORT_SYMBOL(enable_sacf_uaccess); diff --git a/arch/s390/mm/pgalloc.c b/arch/s390/mm/pgalloc.c index 3dd253f81a77..46071be897ab 100644 --- a/arch/s390/mm/pgalloc.c +++ b/arch/s390/mm/pgalloc.c @@ -70,8 +70,20 @@ static void __crst_table_upgrade(void *arg) { struct mm_struct *mm = arg; - if (current->active_mm == mm) - set_user_asce(mm); + /* we must change all active ASCEs to avoid the creation of new TLBs */ + if (current->active_mm == mm) { + S390_lowcore.user_asce = mm->context.asce; + if (current->thread.mm_segment == USER_DS) { + __ctl_load(S390_lowcore.user_asce, 1, 1); + /* Mark user-ASCE present in CR1 */ + clear_cpu_flag(CIF_ASCE_PRIMARY); + } + if (current->thread.mm_segment == USER_DS_SACF) { + __ctl_load(S390_lowcore.user_asce, 7, 7); + /* enable_sacf_uaccess does all or nothing */ + WARN_ON(!test_cpu_flag(CIF_ASCE_SECONDARY)); + } + } __tlb_flush_local(); } -- cgit v1.2.3 From 8ebf6da9db1b2a20bb86cc1bee2552e894d03308 Mon Sep 17 00:00:00 2001 From: Philipp Rudo Date: Mon, 6 Apr 2020 14:47:48 +0200 Subject: s390/ftrace: fix potential crashes when switching tracers Switching tracers include instruction patching. To prevent that a instruction is patched while it's read the instruction patching is done in stop_machine 'context'. This also means that any function called during stop_machine must not be traced. Thus add 'notrace' to all functions called within stop_machine. Fixes: 1ec2772e0c3c ("s390/diag: add a statistic for diagnose calls") Fixes: 38f2c691a4b3 ("s390: improve wait logic of stop_machine") Fixes: 4ecf0a43e729 ("processor: get rid of cpu_relax_yield") Signed-off-by: Philipp Rudo Signed-off-by: Vasily Gorbik --- arch/s390/kernel/diag.c | 2 +- arch/s390/kernel/smp.c | 4 ++-- arch/s390/kernel/trace.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/diag.c b/arch/s390/kernel/diag.c index 61f2b0412345..ccba63aaeb47 100644 --- a/arch/s390/kernel/diag.c +++ b/arch/s390/kernel/diag.c @@ -133,7 +133,7 @@ void diag_stat_inc(enum diag_stat_enum nr) } EXPORT_SYMBOL(diag_stat_inc); -void diag_stat_inc_norecursion(enum diag_stat_enum nr) +void notrace diag_stat_inc_norecursion(enum diag_stat_enum nr) { this_cpu_inc(diag_stat.counter[nr]); trace_s390_diagnose_norecursion(diag_map[nr].code); diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 7eaabbab2213..10dbb12eb14d 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -403,7 +403,7 @@ int smp_find_processor_id(u16 address) return -1; } -bool arch_vcpu_is_preempted(int cpu) +bool notrace arch_vcpu_is_preempted(int cpu) { if (test_cpu_flag_of(CIF_ENABLED_WAIT, cpu)) return false; @@ -413,7 +413,7 @@ bool arch_vcpu_is_preempted(int cpu) } EXPORT_SYMBOL(arch_vcpu_is_preempted); -void smp_yield_cpu(int cpu) +void notrace smp_yield_cpu(int cpu) { if (!MACHINE_HAS_DIAG9C) return; diff --git a/arch/s390/kernel/trace.c b/arch/s390/kernel/trace.c index 490b52e85014..11a669f3cc93 100644 --- a/arch/s390/kernel/trace.c +++ b/arch/s390/kernel/trace.c @@ -14,7 +14,7 @@ EXPORT_TRACEPOINT_SYMBOL(s390_diagnose); static DEFINE_PER_CPU(unsigned int, diagnose_trace_depth); -void trace_s390_diagnose_norecursion(int diag_nr) +void notrace trace_s390_diagnose_norecursion(int diag_nr) { unsigned long flags; unsigned int *depth; -- cgit v1.2.3 From 86dbf32da150339ca81509fa2eb84c814b55258b Mon Sep 17 00:00:00 2001 From: Niklas Schnelle Date: Thu, 16 Apr 2020 13:44:30 +0200 Subject: s390/pci: do not set affinity for floating irqs with the introduction of CPU directed interrupts the kernel parameter pci=force_floating was introduced to fall back to the previous behavior using floating irqs. However we were still setting the affinity in that case, both in __irq_alloc_descs() and via the irq_set_affinity callback in struct irq_chip. For the former only set the affinity in the directed case. The latter is explicitly set in zpci_directed_irq_init() so we can just leave it unset for the floating case. Fixes: e979ce7bced2 ("s390/pci: provide support for CPU directed interrupts") Co-developed-by: Alexander Schmidt Signed-off-by: Alexander Schmidt Signed-off-by: Niklas Schnelle Signed-off-by: Vasily Gorbik --- arch/s390/pci/pci_irq.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/pci/pci_irq.c b/arch/s390/pci/pci_irq.c index fbe97ab2e228..743f257cf2cb 100644 --- a/arch/s390/pci/pci_irq.c +++ b/arch/s390/pci/pci_irq.c @@ -115,7 +115,6 @@ static struct irq_chip zpci_irq_chip = { .name = "PCI-MSI", .irq_unmask = pci_msi_unmask_irq, .irq_mask = pci_msi_mask_irq, - .irq_set_affinity = zpci_set_irq_affinity, }; static void zpci_handle_cpu_local_irq(bool rescan) @@ -276,7 +275,9 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) rc = -EIO; if (hwirq - bit >= msi_vecs) break; - irq = __irq_alloc_descs(-1, 0, 1, 0, THIS_MODULE, msi->affinity); + irq = __irq_alloc_descs(-1, 0, 1, 0, THIS_MODULE, + (irq_delivery == DIRECTED) ? + msi->affinity : NULL); if (irq < 0) return -ENOMEM; rc = irq_set_msi_desc(irq, msi); -- cgit v1.2.3 From 673deb0beba5d39c2e0bc6536e00b03b6ef59cc0 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Apr 2020 14:01:14 +0200 Subject: s390/protvirt: fix compilation issue The kernel fails to compile with CONFIG_PROTECTED_VIRTUALIZATION_GUEST set but CONFIG_KVM unset. This patch fixes the issue by making the needed variable always available. Link: https://lkml.kernel.org/r/20200423120114.2027410-1-imbrenda@linux.ibm.com Fixes: a0f60f843199 ("s390/protvirt: Add sysfs firmware interface for Ultravisor information") Reported-by: kbuild test robot Reported-by: Philipp Rudo Suggested-by: Philipp Rudo Reviewed-by: David Hildenbrand Reviewed-by: Cornelia Huck Reviewed-by: Vasily Gorbik Acked-by: Christian Borntraeger Signed-off-by: Claudio Imbrenda Signed-off-by: Vasily Gorbik --- arch/s390/boot/uv.c | 2 -- arch/s390/kernel/uv.c | 3 ++- 2 files changed, 2 insertions(+), 3 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/boot/uv.c b/arch/s390/boot/uv.c index 8fde561f1d07..f887a479cdc7 100644 --- a/arch/s390/boot/uv.c +++ b/arch/s390/boot/uv.c @@ -7,9 +7,7 @@ #ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST int __bootdata_preserved(prot_virt_guest); #endif -#if IS_ENABLED(CONFIG_KVM) struct uv_info __bootdata_preserved(uv_info); -#endif void uv_query_info(void) { diff --git a/arch/s390/kernel/uv.c b/arch/s390/kernel/uv.c index c86d654351d1..4c0677fc8904 100644 --- a/arch/s390/kernel/uv.c +++ b/arch/s390/kernel/uv.c @@ -23,10 +23,11 @@ int __bootdata_preserved(prot_virt_guest); #endif +struct uv_info __bootdata_preserved(uv_info); + #if IS_ENABLED(CONFIG_KVM) int prot_virt_host; EXPORT_SYMBOL(prot_virt_host); -struct uv_info __bootdata_preserved(uv_info); EXPORT_SYMBOL(uv_info); static int __init prot_virt_setup(char *val) -- cgit v1.2.3