From: Cong Wang cong.wang@bytedance.com

NF_HOOK_LIST() uses list_del() to remove skb from the linked list, however, it is not sufficient as skb->next still points to other skb. We should just call skb_list_del_init() to clear skb->next, like the rest places which using skb list.

This has been fixed in upstream by commit ca58fbe06c54 ("netfilter: add and use nf_hook_slow_list()").

Fixes: 9f17dbf04ddf ("netfilter: fix use-after-free in NF_HOOK_LIST") Reported-by: liuzx@knownsec.com Tested-by: liuzx@knownsec.com Cc: Florian Westphal fw@strlen.de Cc: Edward Cree ecree@solarflare.com Cc: stable@vger.kernel.org # between 4.19 and 5.4 Signed-off-by: Cong Wang cong.wang@bytedance.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- include/linux/netfilter.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -300,7 +300,7 @@ NF_HOOK_LIST(uint8_t pf, unsigned int ho

INIT_LIST_HEAD(&sublist); list_for_each_entry_safe(skb, next, head, list) { - list_del(&skb->list); + skb_list_del_init(skb); if (nf_hook(pf, hook, net, sk, skb, in, out, okfn) == 1) list_add_tail(&skb->list, &sublist); }

From: Filipe Manana fdmanana@suse.com

commit 3d05cad3c357a2b749912914356072b38435edfa upstream.

Lockdep reported the following splat when running test btrfs/190 from fstests:

[ 9482.126098] ====================================================== [ 9482.126184] WARNING: possible circular locking dependency detected [ 9482.126281] 5.10.0-rc4-btrfs-next-73 #1 Not tainted [ 9482.126365] ------------------------------------------------------ [ 9482.126456] mount/24187 is trying to acquire lock: [ 9482.126534] ffffa0c869a7dac0 (&fs_info->qgroup_rescan_lock){+.+.}-{3:3}, at: qgroup_rescan_init+0x43/0xf0 [btrfs] [ 9482.126647] but task is already holding lock: [ 9482.126777] ffffa0c892ebd3a0 (btrfs-quota-00){++++}-{3:3}, at: __btrfs_tree_read_lock+0x27/0x120 [btrfs] [ 9482.126886] which lock already depends on the new lock.

[ 9482.127078] the existing dependency chain (in reverse order) is: [ 9482.127213] -> #1 (btrfs-quota-00){++++}-{3:3}: [ 9482.127366] lock_acquire+0xd8/0x490 [ 9482.127436] down_read_nested+0x45/0x220 [ 9482.127528] __btrfs_tree_read_lock+0x27/0x120 [btrfs] [ 9482.127613] btrfs_read_lock_root_node+0x41/0x130 [btrfs] [ 9482.127702] btrfs_search_slot+0x514/0xc30 [btrfs] [ 9482.127788] update_qgroup_status_item+0x72/0x140 [btrfs] [ 9482.127877] btrfs_qgroup_rescan_worker+0xde/0x680 [btrfs] [ 9482.127964] btrfs_work_helper+0xf1/0x600 [btrfs] [ 9482.128039] process_one_work+0x24e/0x5e0 [ 9482.128110] worker_thread+0x50/0x3b0 [ 9482.128181] kthread+0x153/0x170 [ 9482.128256] ret_from_fork+0x22/0x30 [ 9482.128327] -> #0 (&fs_info->qgroup_rescan_lock){+.+.}-{3:3}: [ 9482.128464] check_prev_add+0x91/0xc60 [ 9482.128551] __lock_acquire+0x1740/0x3110 [ 9482.128623] lock_acquire+0xd8/0x490 [ 9482.130029] __mutex_lock+0xa3/0xb30 [ 9482.130590] qgroup_rescan_init+0x43/0xf0 [btrfs] [ 9482.131577] btrfs_read_qgroup_config+0x43a/0x550 [btrfs] [ 9482.132175] open_ctree+0x1228/0x18a0 [btrfs] [ 9482.132756] btrfs_mount_root.cold+0x13/0xed [btrfs] [ 9482.133325] legacy_get_tree+0x30/0x60 [ 9482.133866] vfs_get_tree+0x28/0xe0 [ 9482.134392] fc_mount+0xe/0x40 [ 9482.134908] vfs_kern_mount.part.0+0x71/0x90 [ 9482.135428] btrfs_mount+0x13b/0x3e0 [btrfs] [ 9482.135942] legacy_get_tree+0x30/0x60 [ 9482.136444] vfs_get_tree+0x28/0xe0 [ 9482.136949] path_mount+0x2d7/0xa70 [ 9482.137438] do_mount+0x75/0x90 [ 9482.137923] __x64_sys_mount+0x8e/0xd0 [ 9482.138400] do_syscall_64+0x33/0x80 [ 9482.138873] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 9482.139346] other info that might help us debug this:

[ 9482.140735] Possible unsafe locking scenario:

[ 9482.141594] CPU0 CPU1 [ 9482.142011] ---- ---- [ 9482.142411] lock(btrfs-quota-00); [ 9482.142806] lock(&fs_info->qgroup_rescan_lock); [ 9482.143216] lock(btrfs-quota-00); [ 9482.143629] lock(&fs_info->qgroup_rescan_lock); [ 9482.144056] *** DEADLOCK ***

[ 9482.145242] 2 locks held by mount/24187: [ 9482.145637] #0: ffffa0c8411c40e8 (&type->s_umount_key#44/1){+.+.}-{3:3}, at: alloc_super+0xb9/0x400 [ 9482.146061] #1: ffffa0c892ebd3a0 (btrfs-quota-00){++++}-{3:3}, at: __btrfs_tree_read_lock+0x27/0x120 [btrfs] [ 9482.146509] stack backtrace: [ 9482.147350] CPU: 1 PID: 24187 Comm: mount Not tainted 5.10.0-rc4-btrfs-next-73 #1 [ 9482.147788] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 [ 9482.148709] Call Trace: [ 9482.149169] dump_stack+0x8d/0xb5 [ 9482.149628] check_noncircular+0xff/0x110 [ 9482.150090] check_prev_add+0x91/0xc60 [ 9482.150561] ? kvm_clock_read+0x14/0x30 [ 9482.151017] ? kvm_sched_clock_read+0x5/0x10 [ 9482.151470] __lock_acquire+0x1740/0x3110 [ 9482.151941] ? __btrfs_tree_read_lock+0x27/0x120 [btrfs] [ 9482.152402] lock_acquire+0xd8/0x490 [ 9482.152887] ? qgroup_rescan_init+0x43/0xf0 [btrfs] [ 9482.153354] __mutex_lock+0xa3/0xb30 [ 9482.153826] ? qgroup_rescan_init+0x43/0xf0 [btrfs] [ 9482.154301] ? qgroup_rescan_init+0x43/0xf0 [btrfs] [ 9482.154768] ? qgroup_rescan_init+0x43/0xf0 [btrfs] [ 9482.155226] qgroup_rescan_init+0x43/0xf0 [btrfs] [ 9482.155690] btrfs_read_qgroup_config+0x43a/0x550 [btrfs] [ 9482.156160] open_ctree+0x1228/0x18a0 [btrfs] [ 9482.156643] btrfs_mount_root.cold+0x13/0xed [btrfs] [ 9482.157108] ? rcu_read_lock_sched_held+0x5d/0x90 [ 9482.157567] ? kfree+0x31f/0x3e0 [ 9482.158030] legacy_get_tree+0x30/0x60 [ 9482.158489] vfs_get_tree+0x28/0xe0 [ 9482.158947] fc_mount+0xe/0x40 [ 9482.159403] vfs_kern_mount.part.0+0x71/0x90 [ 9482.159875] btrfs_mount+0x13b/0x3e0 [btrfs] [ 9482.160335] ? rcu_read_lock_sched_held+0x5d/0x90 [ 9482.160805] ? kfree+0x31f/0x3e0 [ 9482.161260] ? legacy_get_tree+0x30/0x60 [ 9482.161714] legacy_get_tree+0x30/0x60 [ 9482.162166] vfs_get_tree+0x28/0xe0 [ 9482.162616] path_mount+0x2d7/0xa70 [ 9482.163070] do_mount+0x75/0x90 [ 9482.163525] __x64_sys_mount+0x8e/0xd0 [ 9482.163986] do_syscall_64+0x33/0x80 [ 9482.164437] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 9482.164902] RIP: 0033:0x7f51e907caaa

This happens because at btrfs_read_qgroup_config() we can call qgroup_rescan_init() while holding a read lock on a quota btree leaf, acquired by the previous call to btrfs_search_slot_for_read(), and qgroup_rescan_init() acquires the mutex qgroup_rescan_lock.

A qgroup rescan worker does the opposite: it acquires the mutex qgroup_rescan_lock, at btrfs_qgroup_rescan_worker(), and then tries to update the qgroup status item in the quota btree through the call to update_qgroup_status_item(). This inversion of locking order between the qgroup_rescan_lock mutex and quota btree locks causes the splat.

Fix this simply by releasing and freeing the path before calling qgroup_rescan_init() at btrfs_read_qgroup_config().

CC: stable@vger.kernel.org # 4.4+ Signed-off-by: Filipe Manana fdmanana@suse.com Reviewed-by: David Sterba dsterba@suse.com Signed-off-by: David Sterba dsterba@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- fs/btrfs/qgroup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -488,13 +488,13 @@ next2: break; } out: + btrfs_free_path(path); fs_info->qgroup_flags |= flags; if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON)) clear_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags); else if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN && ret >= 0) ret = qgroup_rescan_init(fs_info, rescan_progress, 0); - btrfs_free_path(path);

if (ret < 0) { ulist_free(fs_info->qgroup_ulist);

From: Zenghui Yu yuzenghui@huawei.com

commit 23bde34771f1ea92fb5e6682c0d8c04304d34b3b upstream.

It was recently reported that if GICR_TYPER is accessed before the RD base address is set, we'll suffer from the unset @rdreg dereferencing. Oops...

gpa_t last_rdist_typer = rdreg->base + GICR_TYPER + (rdreg->free_index - 1) * KVM_VGIC_V3_REDIST_SIZE;

It's "expected" that users will access registers in the redistributor if the RD has been properly configured (e.g., the RD base address is set). But it hasn't yet been covered by the existing documentation.

Per discussion on the list [1], the reporting of the GICR_TYPER.Last bit for userspace never actually worked. And it's difficult for us to emulate it correctly given that userspace has the flexibility to access it any time. Let's just drop the reporting of the Last bit for userspace for now (userspace should have full knowledge about it anyway) and it at least prevents kernel from panic ;-)

[1] https://lore.kernel.org/kvmarm/c20865a267e44d1e2c0d52ce4e012263@kernel.org/

Fixes: ba7b3f1275fd ("KVM: arm/arm64: Revisit Redistributor TYPER last bit computation") Reported-by: Keqian Zhu zhukeqian1@huawei.com Signed-off-by: Zenghui Yu yuzenghui@huawei.com Signed-off-by: Marc Zyngier maz@kernel.org Reviewed-by: Eric Auger eric.auger@redhat.com Link: https://lore.kernel.org/r/20201117151629.1738-1-yuzenghui@huawei.com Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- virt/kvm/arm/vgic/vgic-mmio-v3.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-)

--- a/virt/kvm/arm/vgic/vgic-mmio-v3.c +++ b/virt/kvm/arm/vgic/vgic-mmio-v3.c @@ -226,6 +226,23 @@ static unsigned long vgic_mmio_read_v3r_ return extract_bytes(value, addr & 7, len); }

+static unsigned long vgic_uaccess_read_v3r_typer(struct kvm_vcpu *vcpu, + gpa_t addr, unsigned int len) +{ + unsigned long mpidr = kvm_vcpu_get_mpidr_aff(vcpu); + int target_vcpu_id = vcpu->vcpu_id; + u64 value; + + value = (u64)(mpidr & GENMASK(23, 0)) << 32; + value |= ((target_vcpu_id & 0xffff) << 8); + + if (vgic_has_its(vcpu->kvm)) + value |= GICR_TYPER_PLPIS; + + /* reporting of the Last bit is not supported for userspace */ + return extract_bytes(value, addr & 7, len); +} + static unsigned long vgic_mmio_read_v3r_iidr(struct kvm_vcpu *vcpu, gpa_t addr, unsigned int len) { @@ -532,8 +549,9 @@ static const struct vgic_register_region REGISTER_DESC_WITH_LENGTH(GICR_IIDR, vgic_mmio_read_v3r_iidr, vgic_mmio_write_wi, 4, VGIC_ACCESS_32bit), - REGISTER_DESC_WITH_LENGTH(GICR_TYPER, - vgic_mmio_read_v3r_typer, vgic_mmio_write_wi, 8, + REGISTER_DESC_WITH_LENGTH_UACCESS(GICR_TYPER, + vgic_mmio_read_v3r_typer, vgic_mmio_write_wi, + vgic_uaccess_read_v3r_typer, vgic_mmio_uaccess_write_wi, 8, VGIC_ACCESS_64bit | VGIC_ACCESS_32bit), REGISTER_DESC_WITH_LENGTH(GICR_WAKER, vgic_mmio_read_raz, vgic_mmio_write_wi, 4,

From: Paolo Bonzini pbonzini@redhat.com

commit 71cc849b7093bb83af966c0e60cb11b7f35cd746 upstream.

kvm_cpu_accept_dm_intr and kvm_vcpu_ready_for_interrupt_injection are a hodge-podge of conditions, hacked together to get something that more or less works. But what is actually needed is much simpler; in both cases the fundamental question is, do we have a place to stash an interrupt if userspace does KVM_INTERRUPT?

In userspace irqchip mode, that is !vcpu->arch.interrupt.injected. Currently kvm_event_needs_reinjection(vcpu) covers it, but it is unnecessarily restrictive.

In split irqchip mode it's a bit more complicated, we need to check kvm_apic_accept_pic_intr(vcpu) (the IRQ window exit is basically an INTACK cycle and thus requires ExtINTs not to be masked) as well as !pending_userspace_extint(vcpu). However, there is no need to check kvm_event_needs_reinjection(vcpu), since split irqchip keeps pending ExtINT state separate from event injection state, and checking kvm_cpu_has_interrupt(vcpu) is wrong too since ExtINT has higher priority than APIC interrupts. In fact the latter fixes a bug: when userspace requests an IRQ window vmexit, an interrupt in the local APIC can cause kvm_cpu_has_interrupt() to be true and thus kvm_vcpu_ready_for_interrupt_injection() to return false. When this happens, vcpu_run does not exit to userspace but the interrupt window vmexits keep occurring. The VM loops without any hope of making progress.

Once we try to fix these with something like

return kvm_arch_interrupt_allowed(vcpu) && - !kvm_cpu_has_interrupt(vcpu) && - !kvm_event_needs_reinjection(vcpu) && - kvm_cpu_accept_dm_intr(vcpu); + (!lapic_in_kernel(vcpu) + ? !vcpu->arch.interrupt.injected + : (kvm_apic_accept_pic_intr(vcpu) + && !pending_userspace_extint(v)));

we realize two things. First, thanks to the previous patch the complex conditional can reuse !kvm_cpu_has_extint(vcpu). Second, the interrupt window request in vcpu_enter_guest()

bool req_int_win = dm_request_for_irq_injection(vcpu) && kvm_cpu_accept_dm_intr(vcpu);

should be kept in sync with kvm_vcpu_ready_for_interrupt_injection(): it is unnecessary to ask the processor for an interrupt window if we would not be able to return to userspace. Therefore, kvm_cpu_accept_dm_intr(vcpu) is basically !kvm_cpu_has_extint(vcpu) ANDed with the existing check for masked ExtINT. It all makes sense:

- we can accept an interrupt from userspace if there is a place to stash it (and, for irqchip split, ExtINTs are not masked). Interrupts from userspace _can_ be accepted even if right now EFLAGS.IF=0.

- in order to tell userspace we will inject its interrupt ("IRQ window open" i.e. kvm_vcpu_ready_for_interrupt_injection), both KVM and the vCPU need to be ready to accept the interrupt.

... and this is what the patch implements.

Reported-by: David Woodhouse dwmw@amazon.co.uk Analyzed-by: David Woodhouse dwmw@amazon.co.uk Cc: stable@vger.kernel.org Signed-off-by: Paolo Bonzini pbonzini@redhat.com Reviewed-by: Nikos Tsironis ntsironis@arrikto.com Reviewed-by: David Woodhouse dwmw@amazon.co.uk Tested-by: David Woodhouse dwmw@amazon.co.uk Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/irq.c | 2 +- arch/x86/kvm/x86.c | 18 ++++++++++-------- 3 files changed, 12 insertions(+), 9 deletions(-)

--- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -1472,6 +1472,7 @@ int kvm_test_age_hva(struct kvm *kvm, un void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte); int kvm_cpu_has_injectable_intr(struct kvm_vcpu *v); int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu); +int kvm_cpu_has_extint(struct kvm_vcpu *v); int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu); int kvm_cpu_get_interrupt(struct kvm_vcpu *v); void kvm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event); --- a/arch/x86/kvm/irq.c +++ b/arch/x86/kvm/irq.c @@ -52,7 +52,7 @@ static int pending_userspace_extint(stru * check if there is pending interrupt from * non-APIC source without intack. */ -static int kvm_cpu_has_extint(struct kvm_vcpu *v) +int kvm_cpu_has_extint(struct kvm_vcpu *v) { /* * FIXME: interrupt.injected represents an interrupt whose --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3351,21 +3351,23 @@ static int kvm_vcpu_ioctl_set_lapic(stru

static int kvm_cpu_accept_dm_intr(struct kvm_vcpu *vcpu) { + /* + * We can accept userspace's request for interrupt injection + * as long as we have a place to store the interrupt number. + * The actual injection will happen when the CPU is able to + * deliver the interrupt. + */ + if (kvm_cpu_has_extint(vcpu)) + return false; + + /* Acknowledging ExtINT does not happen if LINT0 is masked. */ return (!lapic_in_kernel(vcpu) || kvm_apic_accept_pic_intr(vcpu)); }

-/* - * if userspace requested an interrupt window, check that the - * interrupt window is open. - * - * No need to exit to userspace if we already have an interrupt queued. - */ static int kvm_vcpu_ready_for_interrupt_injection(struct kvm_vcpu *vcpu) { return kvm_arch_interrupt_allowed(vcpu) && - !kvm_cpu_has_interrupt(vcpu) && - !kvm_event_needs_reinjection(vcpu) && kvm_cpu_accept_dm_intr(vcpu); }

From: Will Deacon will@kernel.org

commit 07509e10dcc77627f8b6a57381e878fe269958d3 upstream.

pte_accessible() is used by ptep_clear_flush() to figure out whether TLB invalidation is necessary when unmapping pages for reclaim. Although our implementation is correct according to the architecture, returning true only for valid, young ptes in the absence of racing page-table modifications, this is in fact flawed due to lazy invalidation of old ptes in ptep_clear_flush_young() where we elide the expensive DSB instruction for completing the TLB invalidation.

Rather than penalise the aging path, adjust pte_accessible() to return true for any valid pte, even if the access flag is cleared.

Cc: stable@vger.kernel.org Fixes: 76c714be0e5e ("arm64: pgtable: implement pte_accessible()") Reported-by: Yu Zhao yuzhao@google.com Acked-by: Yu Zhao yuzhao@google.com Reviewed-by: Minchan Kim minchan@kernel.org Reviewed-by: Catalin Marinas catalin.marinas@arm.com Link: https://lore.kernel.org/r/20201120143557.6715-2-will@kernel.org Signed-off-by: Will Deacon will@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- arch/arm64/include/asm/pgtable.h | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)

--- a/arch/arm64/include/asm/pgtable.h +++ b/arch/arm64/include/asm/pgtable.h @@ -107,8 +107,6 @@ extern unsigned long empty_zero_page[PAG #define pte_valid(pte) (!!(pte_val(pte) & PTE_VALID)) #define pte_valid_not_user(pte) \ ((pte_val(pte) & (PTE_VALID | PTE_USER)) == PTE_VALID) -#define pte_valid_young(pte) \ - ((pte_val(pte) & (PTE_VALID | PTE_AF)) == (PTE_VALID | PTE_AF)) #define pte_valid_user(pte) \ ((pte_val(pte) & (PTE_VALID | PTE_USER)) == (PTE_VALID | PTE_USER))

@@ -116,9 +114,12 @@ extern unsigned long empty_zero_page[PAG * Could the pte be present in the TLB? We must check mm_tlb_flush_pending * so that we don't erroneously return false for pages that have been * remapped as PROT_NONE but are yet to be flushed from the TLB. + * Note that we can't make any assumptions based on the state of the access + * flag, since ptep_clear_flush_young() elides a DSB when invalidating the + * TLB. */ #define pte_accessible(mm, pte) \ - (mm_tlb_flush_pending(mm) ? pte_present(pte) : pte_valid_young(pte)) + (mm_tlb_flush_pending(mm) ? pte_present(pte) : pte_valid(pte))

/* * p??_access_permitted() is true for valid user mappings (subject to the

From: Lyude Paul lyude@redhat.com

commit de9f8eea5a44b0b756d3d6345af7f8e630a3c8c0 upstream.

Unfortunately, it appears our fix in: commit b5d29843d8ef ("drm/atomic_helper: Allow DPMS On<->Off changes for unregistered connectors")

Which attempted to work around the problems introduced by: commit 4d80273976bf ("drm/atomic_helper: Disallow new modesets on unregistered connectors")

Is still not the right solution, as modesets can still be triggered outside of drm_atomic_set_crtc_for_connector().

So in order to fix this, while still being careful that we don't break modesets that a driver may perform before being registered with userspace, we replace connector->registered with a tristate member, connector->registration_state. This allows us to keep track of whether or not a connector is still initializing and hasn't been exposed to userspace, is currently registered and exposed to userspace, or has been legitimately removed from the system after having once been present.

Using this info, we can prevent userspace from performing new modesets on unregistered connectors while still allowing the driver to perform modesets on unregistered connectors before the driver has finished being registered.

Changes since v1: - Fix WARN_ON() in drm_connector_cleanup() that CI caught with this patchset in igt@drv_module_reload@basic-reload-inject and igt@drv_module_reload@basic-reload by checking if the connector is registered instead of unregistered, as calling drm_connector_cleanup() on a connector that hasn't been registered with userspace yet should stay valid. - Remove unregistered_connector_check(), and just go back to what we were doing before in commit 4d80273976bf ("drm/atomic_helper: Disallow new modesets on unregistered connectors") except replacing READ_ONCE(connector->registered) with drm_connector_is_unregistered(). This gets rid of the behavior of allowing DPMS On<->Off, but that should be fine as it's more consistent with the UAPI we had before - danvet - s/drm_connector_unregistered/drm_connector_is_unregistered/ - danvet - Update documentation, fix some typos.

Fixes: b5d29843d8ef ("drm/atomic_helper: Allow DPMS On<->Off changes for unregistered connectors") Cc: Ville Syrjälä ville.syrjala@linux.intel.com Cc: Daniel Vetter daniel.vetter@ffwll.ch Cc: Rodrigo Vivi rodrigo.vivi@intel.com Cc: stable@vger.kernel.org Cc: David Airlie airlied@linux.ie Signed-off-by: Lyude Paul lyude@redhat.com Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch Link: https://patchwork.freedesktop.org/patch/msgid/20181016203946.9601-1-lyude@re... (cherry picked from commit 39b50c603878f4f8ae541ac4088a805d588abc79) Fixes: e96550956fbc ("drm/atomic_helper: Disallow new modesets on unregistered connectors") Fixes: 34ca26a98ad6 ("drm/atomic_helper: Allow DPMS On<->Off changes for unregistered connectors") Cc: stable@vger.kernel.org Signed-off-by: Joonas Lahtinen joonas.lahtinen@linux.intel.com Cc: Christoph Niedermaier cniedermaier@dh-electronics.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- drivers/gpu/drm/drm_atomic.c | 21 ---------- drivers/gpu/drm/drm_atomic_helper.c | 21 ++++++++++ drivers/gpu/drm/drm_connector.c | 11 +++-- drivers/gpu/drm/i915/intel_dp_mst.c | 8 ++-- include/drm/drm_connector.h | 71 ++++++++++++++++++++++++++++++++++-- 5 files changed, 99 insertions(+), 33 deletions(-)

--- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -1702,27 +1702,6 @@ drm_atomic_set_crtc_for_connector(struct struct drm_connector *connector = conn_state->connector; struct drm_crtc_state *crtc_state;

- /* - * For compatibility with legacy users, we want to make sure that - * we allow DPMS On<->Off modesets on unregistered connectors, since - * legacy modesetting users will not be expecting these to fail. We do - * not however, want to allow legacy users to assign a connector - * that's been unregistered from sysfs to another CRTC, since doing - * this with a now non-existent connector could potentially leave us - * in an invalid state. - * - * Since the connector can be unregistered at any point during an - * atomic check or commit, this is racy. But that's OK: all we care - * about is ensuring that userspace can't use this connector for new - * configurations after it's been notified that the connector is no - * longer present. - */ - if (!READ_ONCE(connector->registered) && crtc) { - DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] is not registered\n", - connector->base.id, connector->name); - return -EINVAL; - } - if (conn_state->crtc == crtc) return 0;

--- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -307,6 +307,26 @@ update_connector_routing(struct drm_atom return 0; }

+ crtc_state = drm_atomic_get_new_crtc_state(state, + new_connector_state->crtc); + /* + * For compatibility with legacy users, we want to make sure that + * we allow DPMS On->Off modesets on unregistered connectors. Modesets + * which would result in anything else must be considered invalid, to + * avoid turning on new displays on dead connectors. + * + * Since the connector can be unregistered at any point during an + * atomic check or commit, this is racy. But that's OK: all we care + * about is ensuring that userspace can't do anything but shut off the + * display on a connector that was destroyed after its been notified, + * not before. + */ + if (drm_connector_is_unregistered(connector) && crtc_state->active) { + DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] is not registered\n", + connector->base.id, connector->name); + return -EINVAL; + } + funcs = connector->helper_private;

if (funcs->atomic_best_encoder) @@ -351,7 +371,6 @@ update_connector_routing(struct drm_atom

set_best_encoder(state, new_connector_state, new_encoder);

- crtc_state = drm_atomic_get_new_crtc_state(state, new_connector_state->crtc); crtc_state->connectors_changed = true;

DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] using [ENCODER:%d:%s] on [CRTC:%d:%s]\n", --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -375,7 +375,8 @@ void drm_connector_cleanup(struct drm_co /* The connector should have been removed from userspace long before * it is finally destroyed. */ - if (WARN_ON(connector->registered)) + if (WARN_ON(connector->registration_state == + DRM_CONNECTOR_REGISTERED)) drm_connector_unregister(connector);

if (connector->tile_group) { @@ -432,7 +433,7 @@ int drm_connector_register(struct drm_co return 0;

mutex_lock(&connector->mutex); - if (connector->registered) + if (connector->registration_state != DRM_CONNECTOR_INITIALIZING) goto unlock;

ret = drm_sysfs_connector_add(connector); @@ -452,7 +453,7 @@ int drm_connector_register(struct drm_co

drm_mode_object_register(connector->dev, &connector->base);

- connector->registered = true; + connector->registration_state = DRM_CONNECTOR_REGISTERED; goto unlock;

err_debugfs: @@ -474,7 +475,7 @@ EXPORT_SYMBOL(drm_connector_register); void drm_connector_unregister(struct drm_connector *connector) { mutex_lock(&connector->mutex); - if (!connector->registered) { + if (connector->registration_state != DRM_CONNECTOR_REGISTERED) { mutex_unlock(&connector->mutex); return; } @@ -485,7 +486,7 @@ void drm_connector_unregister(struct drm drm_sysfs_connector_remove(connector); drm_debugfs_connector_remove(connector);

- connector->registered = false; + connector->registration_state = DRM_CONNECTOR_UNREGISTERED; mutex_unlock(&connector->mutex); } EXPORT_SYMBOL(drm_connector_unregister); --- a/drivers/gpu/drm/i915/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/intel_dp_mst.c @@ -77,7 +77,7 @@ static bool intel_dp_mst_compute_config( pipe_config->pbn = mst_pbn;

/* Zombie connectors can't have VCPI slots */ - if (READ_ONCE(connector->registered)) { + if (!drm_connector_is_unregistered(connector)) { slots = drm_dp_atomic_find_vcpi_slots(state, &intel_dp->mst_mgr, port, @@ -317,7 +317,7 @@ static int intel_dp_mst_get_ddc_modes(st struct edid *edid; int ret;

- if (!READ_ONCE(connector->registered)) + if (drm_connector_is_unregistered(connector)) return intel_connector_update_modes(connector, NULL);

edid = drm_dp_mst_get_edid(connector, &intel_dp->mst_mgr, intel_connector->port); @@ -333,7 +333,7 @@ intel_dp_mst_detect(struct drm_connector struct intel_connector *intel_connector = to_intel_connector(connector); struct intel_dp *intel_dp = intel_connector->mst_port;

- if (!READ_ONCE(connector->registered)) + if (drm_connector_is_unregistered(connector)) return connector_status_disconnected; return drm_dp_mst_detect_port(connector, &intel_dp->mst_mgr, intel_connector->port); @@ -376,7 +376,7 @@ intel_dp_mst_mode_valid(struct drm_conne int bpp = 24; /* MST uses fixed bpp */ int max_rate, mode_rate, max_lanes, max_link_clock;

- if (!READ_ONCE(connector->registered)) + if (drm_connector_is_unregistered(connector)) return MODE_ERROR;

if (mode->flags & DRM_MODE_FLAG_DBLSCAN) --- a/include/drm/drm_connector.h +++ b/include/drm/drm_connector.h @@ -81,6 +81,53 @@ enum drm_connector_status { connector_status_unknown = 3, };

+/** + * enum drm_connector_registration_status - userspace registration status for + * a &drm_connector + * + * This enum is used to track the status of initializing a connector and + * registering it with userspace, so that DRM can prevent bogus modesets on + * connectors that no longer exist. + */ +enum drm_connector_registration_state { + /** + * @DRM_CONNECTOR_INITIALIZING: The connector has just been created, + * but has yet to be exposed to userspace. There should be no + * additional restrictions to how the state of this connector may be + * modified. + */ + DRM_CONNECTOR_INITIALIZING = 0, + + /** + * @DRM_CONNECTOR_REGISTERED: The connector has been fully initialized + * and registered with sysfs, as such it has been exposed to + * userspace. There should be no additional restrictions to how the + * state of this connector may be modified. + */ + DRM_CONNECTOR_REGISTERED = 1, + + /** + * @DRM_CONNECTOR_UNREGISTERED: The connector has either been exposed + * to userspace and has since been unregistered and removed from + * userspace, or the connector was unregistered before it had a chance + * to be exposed to userspace (e.g. still in the + * @DRM_CONNECTOR_INITIALIZING state). When a connector is + * unregistered, there are additional restrictions to how its state + * may be modified: + * + * - An unregistered connector may only have its DPMS changed from + * On->Off. Once DPMS is changed to Off, it may not be switched back + * to On. + * - Modesets are not allowed on unregistered connectors, unless they + * would result in disabling its assigned CRTCs. This means + * disabling a CRTC on an unregistered connector is OK, but enabling + * one is not. + * - Removing a CRTC from an unregistered connector is OK, but new + * CRTCs may never be assigned to an unregistered connector. + */ + DRM_CONNECTOR_UNREGISTERED = 2, +}; + enum subpixel_order { SubPixelUnknown = 0, SubPixelHorizontalRGB, @@ -852,10 +899,12 @@ struct drm_connector { bool ycbcr_420_allowed;

/** - * @registered: Is this connector exposed (registered) with userspace? + * @registration_state: Is this connector initializing, exposed + * (registered) with userspace, or unregistered? + * * Protected by @mutex. */ - bool registered; + enum drm_connector_registration_state registration_state;

/** * @modes: @@ -1165,6 +1214,24 @@ static inline void drm_connector_unrefer drm_connector_put(connector); }

+/** + * drm_connector_is_unregistered - has the connector been unregistered from + * userspace? + * @connector: DRM connector + * + * Checks whether or not @connector has been unregistered from userspace. + * + * Returns: + * True if the connector was unregistered, false if the connector is + * registered or has not yet been registered with userspace. + */ +static inline bool +drm_connector_is_unregistered(struct drm_connector *connector) +{ + return READ_ONCE(connector->registration_state) == + DRM_CONNECTOR_UNREGISTERED; +} + const char *drm_get_connector_status_name(enum drm_connector_status status); const char *drm_get_subpixel_order_name(enum subpixel_order order); const char *drm_get_dpms_name(int val);

From: Frank Yang puilp0502@gmail.com

[ Upstream commit 652f3d00de523a17b0cebe7b90debccf13aa8c31 ]

The Varmilo VA104M Keyboard (04b4:07b1, reported as Varmilo Z104M) exposes media control hotkeys as a USB HID consumer control device, but these keys do not work in the current (5.8-rc1) kernel due to the incorrect HID report descriptor. Fix the problem by modifying the internal HID report descriptor.

More specifically, the keyboard report descriptor specifies the logical boundary as 572~10754 (0x023c ~ 0x2a02) while the usage boundary is specified as 0~10754 (0x00 ~ 0x2a02). This results in an incorrect interpretation of input reports, causing inputs to be ignored. By setting the Logical Minimum to zero, we align the logical boundary with the Usage ID boundary.

Some notes:

* There seem to be multiple variants of the VA104M keyboard. This patch specifically targets 04b4:07b1 variant.

* The device works out-of-the-box on Windows platform with the generic consumer control device driver (hidserv.inf). This suggests that Windows either ignores the Logical Minimum/Logical Maximum or interprets the Usage ID assignment differently from the linux implementation; Maybe there are other devices out there that only works on Windows due to this problem?

Signed-off-by: Frank Yang puilp0502@gmail.com Signed-off-by: Jiri Kosina jkosina@suse.cz Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/hid/hid-cypress.c | 44 ++++++++++++++++++++++++++++++++++----- drivers/hid/hid-ids.h | 2 ++ 2 files changed, 41 insertions(+), 5 deletions(-)

diff --git a/drivers/hid/hid-cypress.c b/drivers/hid/hid-cypress.c index 1689568b597d4..12c5d7c96527a 100644 --- a/drivers/hid/hid-cypress.c +++ b/drivers/hid/hid-cypress.c @@ -26,19 +26,17 @@ #define CP_2WHEEL_MOUSE_HACK 0x02 #define CP_2WHEEL_MOUSE_HACK_ON 0x04

+#define VA_INVAL_LOGICAL_BOUNDARY 0x08 + /* * Some USB barcode readers from cypress have usage min and usage max in * the wrong order */ -static __u8 *cp_report_fixup(struct hid_device *hdev, __u8 *rdesc, +static __u8 *cp_rdesc_fixup(struct hid_device *hdev, __u8 *rdesc, unsigned int *rsize) { - unsigned long quirks = (unsigned long)hid_get_drvdata(hdev); unsigned int i;

- if (!(quirks & CP_RDESC_SWAPPED_MIN_MAX)) - return rdesc; - if (*rsize < 4) return rdesc;

@@ -51,6 +49,40 @@ static __u8 *cp_report_fixup(struct hid_device *hdev, __u8 *rdesc, return rdesc; }

+static __u8 *va_logical_boundary_fixup(struct hid_device *hdev, __u8 *rdesc, + unsigned int *rsize) +{ + /* + * Varmilo VA104M (with VID Cypress and device ID 07B1) incorrectly + * reports Logical Minimum of its Consumer Control device as 572 + * (0x02 0x3c). Fix this by setting its Logical Minimum to zero. + */ + if (*rsize == 25 && + rdesc[0] == 0x05 && rdesc[1] == 0x0c && + rdesc[2] == 0x09 && rdesc[3] == 0x01 && + rdesc[6] == 0x19 && rdesc[7] == 0x00 && + rdesc[11] == 0x16 && rdesc[12] == 0x3c && rdesc[13] == 0x02) { + hid_info(hdev, + "fixing up varmilo VA104M consumer control report descriptor\n"); + rdesc[12] = 0x00; + rdesc[13] = 0x00; + } + return rdesc; +} + +static __u8 *cp_report_fixup(struct hid_device *hdev, __u8 *rdesc, + unsigned int *rsize) +{ + unsigned long quirks = (unsigned long)hid_get_drvdata(hdev); + + if (quirks & CP_RDESC_SWAPPED_MIN_MAX) + rdesc = cp_rdesc_fixup(hdev, rdesc, rsize); + if (quirks & VA_INVAL_LOGICAL_BOUNDARY) + rdesc = va_logical_boundary_fixup(hdev, rdesc, rsize); + + return rdesc; +} + static int cp_input_mapped(struct hid_device *hdev, struct hid_input *hi, struct hid_field *field, struct hid_usage *usage, unsigned long **bit, int *max) @@ -131,6 +163,8 @@ static const struct hid_device_id cp_devices[] = { .driver_data = CP_RDESC_SWAPPED_MIN_MAX }, { HID_USB_DEVICE(USB_VENDOR_ID_CYPRESS, USB_DEVICE_ID_CYPRESS_MOUSE), .driver_data = CP_2WHEEL_MOUSE_HACK }, + { HID_USB_DEVICE(USB_VENDOR_ID_CYPRESS, USB_DEVICE_ID_CYPRESS_VARMILO_VA104M_07B1), + .driver_data = VA_INVAL_LOGICAL_BOUNDARY }, { } }; MODULE_DEVICE_TABLE(hid, cp_devices); diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index e18d796d985f8..e9155f7c8c51c 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -330,6 +330,8 @@ #define USB_DEVICE_ID_CYPRESS_BARCODE_4 0xed81 #define USB_DEVICE_ID_CYPRESS_TRUETOUCH 0xc001

+#define USB_DEVICE_ID_CYPRESS_VARMILO_VA104M_07B1 0X07b1 + #define USB_VENDOR_ID_DATA_MODUL 0x7374 #define USB_VENDOR_ID_DATA_MODUL_EASYMAXTOUCH 0x1201

From: Hans de Goede hdegoede@redhat.com

[ Upstream commit b1884583fcd17d6a1b1bba94bbb5826e6b5c6e17 ]

The i8042 module exports several symbols which may be used by other modules.

Before this commit it would refuse to load (when built as a module itself) on systems without an i8042 controller.

This is a problem specifically for the asus-nb-wmi module. Many Asus laptops support the Asus WMI interface. Some of them have an i8042 controller and need to use i8042_install_filter() to filter some kbd events. Other models do not have an i8042 controller (e.g. they use an USB attached kbd).

Before this commit the asus-nb-wmi driver could not be loaded on Asus models without an i8042 controller, when the i8042 code was built as a module (as Arch Linux does) because the module_init function of the i8042 module would fail with -ENODEV and thus the i8042_install_filter symbol could not be loaded.

This commit fixes this by exiting from module_init with a return code of 0 if no controller is found. It also adds a i8042_present bool to make the module_exit function a no-op in this case and also adds a check for i8042_present to the exported i8042_command function.

The latter i8042_present check should not really be necessary because when builtin that function can already be used on systems without an i8042 controller, but better safe then sorry.

Reported-and-tested-by: Marius Iacob themariusus@gmail.com Signed-off-by: Hans de Goede hdegoede@redhat.com Link: https://lore.kernel.org/r/20201008112628.3979-2-hdegoede@redhat.com Signed-off-by: Dmitry Torokhov dmitry.torokhov@gmail.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/input/serio/i8042.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/input/serio/i8042.c b/drivers/input/serio/i8042.c index 95a78ccbd8470..fef3b4064f187 100644 --- a/drivers/input/serio/i8042.c +++ b/drivers/input/serio/i8042.c @@ -125,6 +125,7 @@ module_param_named(unmask_kbd_data, i8042_unmask_kbd_data, bool, 0600); MODULE_PARM_DESC(unmask_kbd_data, "Unconditional enable (may reveal sensitive data) of normally sanitize-filtered kbd data traffic debug log [pre-condition: i8042.debug=1 enabled]"); #endif

+static bool i8042_present; static bool i8042_bypass_aux_irq_test; static char i8042_kbd_firmware_id[128]; static char i8042_aux_firmware_id[128]; @@ -345,6 +346,9 @@ int i8042_command(unsigned char *param, int command) unsigned long flags; int retval;

+ if (!i8042_present) + return -1; + spin_lock_irqsave(&i8042_lock, flags); retval = __i8042_command(param, command); spin_unlock_irqrestore(&i8042_lock, flags); @@ -1613,12 +1617,15 @@ static int __init i8042_init(void)

err = i8042_platform_init(); if (err) - return err; + return (err == -ENODEV) ? 0 : err;

err = i8042_controller_check(); if (err) goto err_platform_exit;

+ /* Set this before creating the dev to allow i8042_command to work right away */ + i8042_present = true; + pdev = platform_create_bundle(&i8042_driver, i8042_probe, NULL, 0, NULL, 0); if (IS_ERR(pdev)) { err = PTR_ERR(pdev); @@ -1637,6 +1644,9 @@ static int __init i8042_init(void)

static void __exit i8042_exit(void) { + if (!i8042_present) + return; + platform_device_unregister(i8042_platform_device); platform_driver_unregister(&i8042_driver); i8042_platform_exit();

From: Chris Ye lzye@google.com

[ Upstream commit f59ee399de4a8ca4d7d19cdcabb4b63e94867f09 ]

Kernel 5.4 introduces HID_QUIRK_INCREMENT_USAGE_ON_DUPLICATE, devices need to be set explicitly with this flag.

Signed-off-by: Chris Ye lzye@google.com Signed-off-by: Jiri Kosina jkosina@suse.cz Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/hid/hid-ids.h | 4 ++++ drivers/hid/hid-quirks.c | 4 ++++ 2 files changed, 8 insertions(+)

diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index 4d2a1d540a821..ad079aca68898 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -444,6 +444,10 @@ #define USB_VENDOR_ID_FRUCTEL 0x25B6 #define USB_DEVICE_ID_GAMETEL_MT_MODE 0x0002

+#define USB_VENDOR_ID_GAMEVICE 0x27F8 +#define USB_DEVICE_ID_GAMEVICE_GV186 0x0BBE +#define USB_DEVICE_ID_GAMEVICE_KISHI 0x0BBF + #define USB_VENDOR_ID_GAMERON 0x0810 #define USB_DEVICE_ID_GAMERON_DUAL_PSX_ADAPTOR 0x0001 #define USB_DEVICE_ID_GAMERON_DUAL_PCS_ADAPTOR 0x0002 diff --git a/drivers/hid/hid-quirks.c b/drivers/hid/hid-quirks.c index 493e2e7e12de0..10cb42a00fe87 100644 --- a/drivers/hid/hid-quirks.c +++ b/drivers/hid/hid-quirks.c @@ -87,6 +87,10 @@ static const struct hid_device_id hid_quirks[] = { { HID_USB_DEVICE(USB_VENDOR_ID_FUTABA, USB_DEVICE_ID_LED_DISPLAY), HID_QUIRK_NO_INIT_REPORTS }, { HID_USB_DEVICE(USB_VENDOR_ID_GREENASIA, USB_DEVICE_ID_GREENASIA_DUAL_SAT_ADAPTOR), HID_QUIRK_MULTI_INPUT }, { HID_USB_DEVICE(USB_VENDOR_ID_GREENASIA, USB_DEVICE_ID_GREENASIA_DUAL_USB_JOYPAD), HID_QUIRK_MULTI_INPUT }, + { HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_GAMEVICE, USB_DEVICE_ID_GAMEVICE_GV186), + HID_QUIRK_INCREMENT_USAGE_ON_DUPLICATE }, + { HID_USB_DEVICE(USB_VENDOR_ID_GAMEVICE, USB_DEVICE_ID_GAMEVICE_KISHI), + HID_QUIRK_INCREMENT_USAGE_ON_DUPLICATE }, { HID_USB_DEVICE(USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_DRIVING), HID_QUIRK_BADPAD | HID_QUIRK_MULTI_INPUT }, { HID_USB_DEVICE(USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_FIGHTING), HID_QUIRK_BADPAD | HID_QUIRK_MULTI_INPUT }, { HID_USB_DEVICE(USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_FLYING), HID_QUIRK_BADPAD | HID_QUIRK_MULTI_INPUT },

From: Brian Masney bmasney@redhat.com

[ Upstream commit 65cae18882f943215d0505ddc7e70495877308e6 ]

When booting a hyperthreaded system with the kernel parameter 'mitigations=auto,nosmt', the following warning occurs:

WARNING: CPU: 0 PID: 1 at drivers/xen/events/events_base.c:1112 unbind_from_irqhandler+0x4e/0x60 ... Hardware name: Xen HVM domU, BIOS 4.2.amazon 08/24/2006 ... Call Trace: xen_uninit_lock_cpu+0x28/0x62 xen_hvm_cpu_die+0x21/0x30 takedown_cpu+0x9c/0xe0 ? trace_suspend_resume+0x60/0x60 cpuhp_invoke_callback+0x9a/0x530 _cpu_up+0x11a/0x130 cpu_up+0x7e/0xc0 bringup_nonboot_cpus+0x48/0x50 smp_init+0x26/0x79 kernel_init_freeable+0xea/0x229 ? rest_init+0xaa/0xaa kernel_init+0xa/0x106 ret_from_fork+0x35/0x40

The secondary CPUs are not activated with the nosmt mitigations and only the primary thread on each CPU core is used. In this situation, xen_hvm_smp_prepare_cpus(), and more importantly xen_init_lock_cpu(), is not called, so the lock_kicker_irq is not initialized for the secondary CPUs. Let's fix this by exiting early in xen_uninit_lock_cpu() if the irq is not set to avoid the warning from above for each secondary CPU.

Signed-off-by: Brian Masney bmasney@redhat.com Link: https://lore.kernel.org/r/20201107011119.631442-1-bmasney@redhat.com Reviewed-by: Juergen Gross jgross@suse.com Signed-off-by: Boris Ostrovsky boris.ostrovsky@oracle.com Signed-off-by: Sasha Levin sashal@kernel.org --- arch/x86/xen/spinlock.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c index 717b4847b473f..6fffb86a32add 100644 --- a/arch/x86/xen/spinlock.c +++ b/arch/x86/xen/spinlock.c @@ -101,10 +101,20 @@ void xen_init_lock_cpu(int cpu)

void xen_uninit_lock_cpu(int cpu) { + int irq; + if (!xen_pvspin) return;

- unbind_from_irqhandler(per_cpu(lock_kicker_irq, cpu), NULL); + /* + * When booting the kernel with 'mitigations=auto,nosmt', the secondary + * CPUs are not activated, and lock_kicker_irq is not initialized. + */ + irq = per_cpu(lock_kicker_irq, cpu); + if (irq == -1) + return; + + unbind_from_irqhandler(irq, NULL); per_cpu(lock_kicker_irq, cpu) = -1; kfree(per_cpu(irq_name, cpu)); per_cpu(irq_name, cpu) = NULL;

From: Jens Axboe axboe@kernel.dk

[ Upstream commit 8d4c3e76e3be11a64df95ddee52e99092d42fc19 ]

If this is attempted by a kthread, then return -EOPNOTSUPP as we don't currently support that. Once we can get task_pid_ptr() doing the right thing, then this can go away again.

Signed-off-by: Jens Axboe axboe@kernel.dk Signed-off-by: Sasha Levin sashal@kernel.org --- fs/proc/self.c | 7 +++++++ 1 file changed, 7 insertions(+)

diff --git a/fs/proc/self.c b/fs/proc/self.c index cc6d4253399d1..7922edf70ce1a 100644 --- a/fs/proc/self.c +++ b/fs/proc/self.c @@ -16,6 +16,13 @@ static const char *proc_self_get_link(struct dentry *dentry, pid_t tgid = task_tgid_nr_ns(current, ns); char *name;

+ /* + * Not currently supported. Once we can inherit all of struct pid, + * we can allow this. + */ + if (current->flags & PF_KTHREAD) + return ERR_PTR(-EOPNOTSUPP); + if (!tgid) return ERR_PTR(-ENOENT); /* max length of unsigned int in decimal + NULL term */

From: Sugar Zhang sugar.zhang@rock-chips.com

[ Upstream commit e773ca7da8beeca7f17fe4c9d1284a2b66839cc1 ]

Actually, burst size is equal to '1 << desc->rqcfg.brst_size'. we should use burst size, not desc->rqcfg.brst_size.

dma memcpy performance on Rockchip RV1126 @ 1512MHz A7, 1056MHz LPDDR3, 200MHz DMA:

dmatest:

/# echo dma0chan0 > /sys/module/dmatest/parameters/channel /# echo 4194304 > /sys/module/dmatest/parameters/test_buf_size /# echo 8 > /sys/module/dmatest/parameters/iterations /# echo y > /sys/module/dmatest/parameters/norandom /# echo y > /sys/module/dmatest/parameters/verbose /# echo 1 > /sys/module/dmatest/parameters/run

dmatest: dma0chan0-copy0: result #1: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000 dmatest: dma0chan0-copy0: result #2: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000 dmatest: dma0chan0-copy0: result #3: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000 dmatest: dma0chan0-copy0: result #4: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000 dmatest: dma0chan0-copy0: result #5: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000 dmatest: dma0chan0-copy0: result #6: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000 dmatest: dma0chan0-copy0: result #7: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000 dmatest: dma0chan0-copy0: result #8: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000

Before:

dmatest: dma0chan0-copy0: summary 8 tests, 0 failures 48 iops 200338 KB/s (0)

After this patch:

dmatest: dma0chan0-copy0: summary 8 tests, 0 failures 179 iops 734873 KB/s (0)

After this patch and increase dma clk to 400MHz:

dmatest: dma0chan0-copy0: summary 8 tests, 0 failures 259 iops 1062929 KB/s (0)

Signed-off-by: Sugar Zhang sugar.zhang@rock-chips.com Link: https://lore.kernel.org/r/1605326106-55681-1-git-send-email-sugar.zhang@rock... Signed-off-by: Vinod Koul vkoul@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/dma/pl330.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c index c564df713efc3..15b30d2d8f7ed 100644 --- a/drivers/dma/pl330.c +++ b/drivers/dma/pl330.c @@ -2774,7 +2774,7 @@ pl330_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dst, * If burst size is smaller than bus width then make sure we only * transfer one at a time to avoid a burst stradling an MFIFO entry. */ - if (desc->rqcfg.brst_size * 8 < pl330->pcfg.data_bus_width) + if (burst * 8 < pl330->pcfg.data_bus_width) desc->rqcfg.brst_len = 1;

desc->bytes_requested = len;

From: Mike Christie michael.christie@oracle.com

[ Upstream commit f36199355c64a39fe82cfddc7623d827c7e050da ]

Maurizio found a race where the abort and cmd stop paths can race as follows:

1. thread1 runs iscsit_release_commands_from_conn and sets CMD_T_FABRIC_STOP.

2. thread2 runs iscsit_aborted_task and then does __iscsit_free_cmd. It then returns from the aborted_task callout and we finish target_handle_abort and do:

target_handle_abort -> transport_cmd_check_stop_to_fabric -> lio_check_stop_free -> target_put_sess_cmd

The cmd is now freed.

3. thread1 now finishes iscsit_release_commands_from_conn and runs iscsit_free_cmd while accessing a command we just released.

In __target_check_io_state we check for CMD_T_FABRIC_STOP and set the CMD_T_ABORTED if the driver is not cleaning up the cmd because of a session shutdown. However, iscsit_release_commands_from_conn only sets the CMD_T_FABRIC_STOP and does not check to see if the abort path has claimed completion ownership of the command.

This adds a check in iscsit_release_commands_from_conn so only the abort or fabric stop path cleanup the command.

Link: https://lore.kernel.org/r/1605318378-9269-1-git-send-email-michael.christie@... Reported-by: Maurizio Lombardi mlombard@redhat.com Reviewed-by: Maurizio Lombardi mlombard@redhat.com Signed-off-by: Mike Christie michael.christie@oracle.com Signed-off-by: Martin K. Petersen martin.petersen@oracle.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/target/iscsi/iscsi_target.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index 2602b57936d4b..58ccded1be857 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -492,8 +492,7 @@ EXPORT_SYMBOL(iscsit_queue_rsp); void iscsit_aborted_task(struct iscsi_conn *conn, struct iscsi_cmd *cmd) { spin_lock_bh(&conn->cmd_lock); - if (!list_empty(&cmd->i_conn_node) && - !(cmd->se_cmd.transport_state & CMD_T_FABRIC_STOP)) + if (!list_empty(&cmd->i_conn_node)) list_del_init(&cmd->i_conn_node); spin_unlock_bh(&conn->cmd_lock);

@@ -4054,12 +4053,22 @@ static void iscsit_release_commands_from_conn(struct iscsi_conn *conn) spin_lock_bh(&conn->cmd_lock); list_splice_init(&conn->conn_cmd_list, &tmp_list);

- list_for_each_entry(cmd, &tmp_list, i_conn_node) { + list_for_each_entry_safe(cmd, cmd_tmp, &tmp_list, i_conn_node) { struct se_cmd *se_cmd = &cmd->se_cmd;

if (se_cmd->se_tfo != NULL) { spin_lock_irq(&se_cmd->t_state_lock); - se_cmd->transport_state |= CMD_T_FABRIC_STOP; + if (se_cmd->transport_state & CMD_T_ABORTED) { + /* + * LIO's abort path owns the cleanup for this, + * so put it back on the list and let + * aborted_task handle it. + */ + list_move_tail(&cmd->i_conn_node, + &conn->conn_cmd_list); + } else { + se_cmd->transport_state |= CMD_T_FABRIC_STOP; + } spin_unlock_irq(&se_cmd->t_state_lock); } }

From: Laurent Pinchart laurent.pinchart@ideasonboard.com

[ Upstream commit dc293f2106903ab9c24e9cea18c276e32c394c33 ]

When adding __user annotations in commit 2adf5352a34a, the strncpy_from_user() function declaration for the CONFIG_GENERIC_STRNCPY_FROM_USER case was missed. Fix it.

Reported-by: kernel test robot lkp@intel.com Signed-off-by: Laurent Pinchart laurent.pinchart@ideasonboard.com Message-Id: 20200831210937.17938-1-laurent.pinchart@ideasonboard.com Signed-off-by: Max Filippov jcmvbkbc@gmail.com Signed-off-by: Sasha Levin sashal@kernel.org --- arch/xtensa/include/asm/uaccess.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/xtensa/include/asm/uaccess.h b/arch/xtensa/include/asm/uaccess.h index f1158b4c629cf..da4effe270072 100644 --- a/arch/xtensa/include/asm/uaccess.h +++ b/arch/xtensa/include/asm/uaccess.h @@ -291,7 +291,7 @@ strncpy_from_user(char *dst, const char *src, long count) return -EFAULT; } #else -long strncpy_from_user(char *dst, const char *src, long count); +long strncpy_from_user(char *dst, const char __user *src, long count); #endif

/*

From: Marc Zyngier maz@kernel.org

[ Upstream commit eb9c4dd9bdfdebaa13846c16a8c79b5b336066b6 ]

If, for some reason, the xusb PHY fails to probe, it leaves a dangling pointer attached to the platform device structure.

This would normally be harmless, but the Tegra XHCI driver then goes and extract that pointer from the PHY device. Things go downhill from there:

8.752082] [004d554e5145533c] address between user and kernel address ranges [ 8.752085] Internal error: Oops: 96000004 [#1] PREEMPT SMP [ 8.752088] Modules linked in: max77620_regulator(E+) xhci_tegra(E+) sdhci_tegra(E+) xhci_hcd(E) sdhci_pltfm(E) cqhci(E) fixed(E) usbcore(E) scsi_mod(E) sdhci(E) host1x(E+) [ 8.752103] CPU: 4 PID: 158 Comm: systemd-udevd Tainted: G S W E 5.9.0-rc7-00298-gf6337624c4fe #1980 [ 8.752105] Hardware name: NVIDIA Jetson TX2 Developer Kit (DT) [ 8.752108] pstate: 20000005 (nzCv daif -PAN -UAO BTYPE=--) [ 8.752115] pc : kobject_put+0x1c/0x21c [ 8.752120] lr : put_device+0x20/0x30 [ 8.752121] sp : ffffffc012eb3840 [ 8.752122] x29: ffffffc012eb3840 x28: ffffffc010e82638 [ 8.752125] x27: ffffffc008d56440 x26: 0000000000000000 [ 8.752128] x25: ffffff81eb508200 x24: 0000000000000000 [ 8.752130] x23: ffffff81eb538800 x22: 0000000000000000 [ 8.752132] x21: 00000000fffffdfb x20: ffffff81eb538810 [ 8.752134] x19: 3d4d554e51455300 x18: 0000000000000020 [ 8.752136] x17: ffffffc008d00270 x16: ffffffc008d00c94 [ 8.752138] x15: 0000000000000004 x14: ffffff81ebd4ae90 [ 8.752140] x13: 0000000000000000 x12: ffffff81eb86a4e8 [ 8.752142] x11: ffffff81eb86a480 x10: ffffff81eb862fea [ 8.752144] x9 : ffffffc01055fb28 x8 : ffffff81eb86a4a8 [ 8.752146] x7 : 0000000000000001 x6 : 0000000000000001 [ 8.752148] x5 : ffffff81dff8bc38 x4 : 0000000000000000 [ 8.752150] x3 : 0000000000000001 x2 : 0000000000000001 [ 8.752152] x1 : 0000000000000002 x0 : 3d4d554e51455300 [ 8.752155] Call trace: [ 8.752157] kobject_put+0x1c/0x21c [ 8.752160] put_device+0x20/0x30 [ 8.752164] tegra_xusb_padctl_put+0x24/0x3c [ 8.752170] tegra_xusb_probe+0x8b0/0xd10 [xhci_tegra] [ 8.752174] platform_drv_probe+0x60/0xb4 [ 8.752176] really_probe+0xf0/0x504 [ 8.752179] driver_probe_device+0x100/0x170 [ 8.752181] device_driver_attach+0xcc/0xd4 [ 8.752183] __driver_attach+0xb0/0x17c [ 8.752185] bus_for_each_dev+0x7c/0xd4 [ 8.752187] driver_attach+0x30/0x3c [ 8.752189] bus_add_driver+0x154/0x250 [ 8.752191] driver_register+0x84/0x140 [ 8.752193] __platform_driver_register+0x54/0x60 [ 8.752197] tegra_xusb_init+0x40/0x1000 [xhci_tegra] [ 8.752201] do_one_initcall+0x54/0x2d0 [ 8.752205] do_init_module+0x68/0x29c [ 8.752207] load_module+0x2178/0x26c0 [ 8.752209] __do_sys_finit_module+0xb0/0x120 [ 8.752211] __arm64_sys_finit_module+0x2c/0x40 [ 8.752215] el0_svc_common.constprop.0+0x80/0x240 [ 8.752218] do_el0_svc+0x30/0xa0 [ 8.752220] el0_svc+0x18/0x50 [ 8.752223] el0_sync_handler+0x90/0x318 [ 8.752225] el0_sync+0x158/0x180 [ 8.752230] Code: a9bd7bfd 910003fd a90153f3 aa0003f3 (3940f000) [ 8.752232] ---[ end trace 90f6c89d62d85ff5 ]---

Reset the pointer on probe failure fixes the issue.

Fixes: 53d2a715c2403 ("phy: Add Tegra XUSB pad controller support") Signed-off-by: Marc Zyngier maz@kernel.org Link: https://lore.kernel.org/r/20201013095820.311376-1-maz@kernel.org Signed-off-by: Vinod Koul vkoul@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/phy/tegra/xusb.c | 1 + 1 file changed, 1 insertion(+)

diff --git a/drivers/phy/tegra/xusb.c b/drivers/phy/tegra/xusb.c index de1b4ebe4de28..39c01ef57d83c 100644 --- a/drivers/phy/tegra/xusb.c +++ b/drivers/phy/tegra/xusb.c @@ -899,6 +899,7 @@ remove_pads: reset: reset_control_assert(padctl->rst); remove: + platform_set_drvdata(pdev, NULL); soc->ops->remove(padctl); return err; }

From: Marc Kleine-Budde mkl@pengutronix.de

[ Upstream commit 05d5de6ba7dbe490dd413b5ca11d0875bd2bc006 ]

According to the bosch,m_can.yaml bindings the first clock shall be the "hclk", while the second clock "cclk".

This patch fixes the order accordingly.

Fixes: 0adbe832f21a ("ARM: dts: dra76x: Add MCAN node") Cc: Faiz Abbas faiz_abbas@ti.com Cc: Tony Lindgren tony@atomide.com Cc: linux-omap@vger.kernel.org Signed-off-by: Marc Kleine-Budde mkl@pengutronix.de Signed-off-by: Tony Lindgren tony@atomide.com Signed-off-by: Sasha Levin sashal@kernel.org --- arch/arm/boot/dts/dra76x.dtsi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm/boot/dts/dra76x.dtsi b/arch/arm/boot/dts/dra76x.dtsi index 216e1d1a69c7d..473d6721b7887 100644 --- a/arch/arm/boot/dts/dra76x.dtsi +++ b/arch/arm/boot/dts/dra76x.dtsi @@ -35,8 +35,8 @@ interrupts = <GIC_SPI 67 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 68 IRQ_TYPE_LEVEL_HIGH>; interrupt-names = "int0", "int1"; - clocks = <&mcan_clk>, <&l3_iclk_div>; - clock-names = "cclk", "hclk"; + clocks = <&l3_iclk_div>, <&mcan_clk>; + clock-names = "hclk", "cclk"; bosch,mram-cfg = <0x0 0 0 32 0 0 1 1>; }; };

From: Zhang Changzhong zhangchangzhong@huawei.com

[ Upstream commit b5f796b62c98cd8c219c4b788ecb6e1218e648cb ]

Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function.

Fixes: c213eae8d3cd ("bnxt_en: Improve VF/PF link change logic.") Reported-by: Hulk Robot hulkci@huawei.com Signed-off-by: Zhang Changzhong zhangchangzhong@huawei.com Reviewed-by: Edwin Peer edwin.peer@broadcom.com Link: https://lore.kernel.org/r/1605701851-20270-1-git-send-email-zhangchangzhong@... Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 1 + 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 01d28ede1fb20..50732ab2c2bc9 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -9120,6 +9120,7 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) create_singlethread_workqueue("bnxt_pf_wq"); if (!bnxt_pf_wq) { dev_err(&pdev->dev, "Unable to create workqueue.\n"); + rc = -ENOMEM; goto init_err_pci_clean; } }

From: Dexuan Cui decui@microsoft.com

[ Upstream commit 5f1251a48c17b54939d7477305e39679a565382c ]

x86 Hyper-V used to essentially always overwrite the effective cache type of guest memory accesses to WB. This was problematic in cases where there is a physical device assigned to the VM, since that often requires that the VM should have control over cache types. Thus, on newer Hyper-V since 2018, Hyper-V always honors the VM's cache type, but unexpectedly Linux VM users start to complain that Linux VM's VRAM becomes very slow, and it turns out that Linux VM should not map the VRAM uncacheable by ioremap(). Fix this slowness issue by using ioremap_cache().

On ARM64, ioremap_cache() is also required as the host also maps the VRAM cacheable, otherwise VM Connect can't display properly with ioremap() or ioremap_wc().

With this change, the VRAM on new Hyper-V is as fast as regular RAM, so it's no longer necessary to use the hacks we added to mitigate the slowness, i.e. we no longer need to allocate physical memory and use it to back up the VRAM in Generation-1 VM, and we also no longer need to allocate physical memory to back up the framebuffer in a Generation-2 VM and copy the framebuffer to the real VRAM. A further big change will address these for v5.11.

Fixes: 68a2d20b79b1 ("drivers/video: add Hyper-V Synthetic Video Frame Buffer Driver") Tested-by: Boqun Feng boqun.feng@gmail.com Signed-off-by: Dexuan Cui decui@microsoft.com Reviewed-by: Michael Kelley mikelley@microsoft.com Reviewed-by: Haiyang Zhang haiyangz@microsoft.com Link: https://lore.kernel.org/r/20201118000305.24797-1-decui@microsoft.com Signed-off-by: Wei Liu wei.liu@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/video/fbdev/hyperv_fb.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/video/fbdev/hyperv_fb.c b/drivers/video/fbdev/hyperv_fb.c index 403d8cd3e5827..56e70f12c9960 100644 --- a/drivers/video/fbdev/hyperv_fb.c +++ b/drivers/video/fbdev/hyperv_fb.c @@ -712,7 +712,12 @@ static int hvfb_getmem(struct hv_device *hdev, struct fb_info *info) goto err1; }

- fb_virt = ioremap(par->mem->start, screen_fb_size); + /* + * Map the VRAM cacheable for performance. This is also required for + * VM Connect to display properly for ARM64 Linux VM, as the host also + * maps the VRAM cacheable. + */ + fb_virt = ioremap_cache(par->mem->start, screen_fb_size); if (!fb_virt) goto err2;

From: Raju Rangoju rajur@chelsio.com

[ Upstream commit bff453921ae105a8dbbad0ed7dd5f5ce424536e7 ]

SMT entry is allocated only when loopback Source MAC rewriting is requested. Accessing SMT entry for non smac rewrite cases results in kernel panic.

Fix the panic caused by non smac rewrite

Fixes: 937d84205884 ("cxgb4: set up filter action after rewrites") Signed-off-by: Raju Rangoju rajur@chelsio.com Link: https://lore.kernel.org/r/20201118143213.13319-1-rajur@chelsio.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c index a62c96001761b..9160b44c68bbf 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c @@ -626,7 +626,8 @@ int set_filter_wr(struct adapter *adapter, int fidx) FW_FILTER_WR_OVLAN_VLD_V(f->fs.val.ovlan_vld) | FW_FILTER_WR_IVLAN_VLDM_V(f->fs.mask.ivlan_vld) | FW_FILTER_WR_OVLAN_VLDM_V(f->fs.mask.ovlan_vld)); - fwr->smac_sel = f->smt->idx; + if (f->fs.newsmac) + fwr->smac_sel = f->smt->idx; fwr->rx_chan_rx_rpl_iq = htons(FW_FILTER_WR_RX_CHAN_V(0) | FW_FILTER_WR_RX_RPL_IQ_V(adapter->sge.fw_evtq.abs_id));

From: Xiongfeng Wang wangxiongfeng2@huawei.com

[ Upstream commit 6830ff853a5764c75e56750d59d0bbb6b26f1835 ]

We return 'err' in the error branch, but this variable may be set as zero by the above code. Fix it by setting 'err' as a negative value before we goto the error label.

Fixes: 74c2174e7be5 ("IB uverbs: add mthca user CQ support") Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Link: https://lore.kernel.org/r/1605837422-42724-1-git-send-email-wangxiongfeng2@h... Reported-by: Hulk Robot hulkci@huawei.com Signed-off-by: Xiongfeng Wang wangxiongfeng2@huawei.com Signed-off-by: Jason Gunthorpe jgg@nvidia.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/infiniband/hw/mthca/mthca_cq.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/infiniband/hw/mthca/mthca_cq.c b/drivers/infiniband/hw/mthca/mthca_cq.c index a6531ffe29a6f..a5694dec3f2ee 100644 --- a/drivers/infiniband/hw/mthca/mthca_cq.c +++ b/drivers/infiniband/hw/mthca/mthca_cq.c @@ -808,8 +808,10 @@ int mthca_init_cq(struct mthca_dev *dev, int nent, }

mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL); - if (IS_ERR(mailbox)) + if (IS_ERR(mailbox)) { + err = PTR_ERR(mailbox); goto err_out_arm; + }

cq_context = mailbox->buf;

@@ -851,9 +853,9 @@ int mthca_init_cq(struct mthca_dev *dev, int nent, }

spin_lock_irq(&dev->cq_table.lock); - if (mthca_array_set(&dev->cq_table.cq, - cq->cqn & (dev->limits.num_cqs - 1), - cq)) { + err = mthca_array_set(&dev->cq_table.cq, + cq->cqn & (dev->limits.num_cqs - 1), cq); + if (err) { spin_unlock_irq(&dev->cq_table.lock); goto err_out_free_mr; }

From: Shay Agroskin shayagr@amazon.com

[ Upstream commit 09323b3bca95181c0da79daebc8b0603e500f573 ]

The ENA driver uses the readless mechanism, which uses DMA, to find out what the DMA mask is supposed to be.

If DMA is used without setting the dma_mask first, it causes the Intel IOMMU driver to think that ENA is a 32-bit device and therefore disables IOMMU passthrough permanently.

This patch sets the dma_mask to be ENA_MAX_PHYS_ADDR_SIZE_BITS=48 before readless initialization in ena_device_init()->ena_com_mmio_reg_read_request_init(), which is large enough to workaround the intel_iommu issue.

DMA mask is set again to the correct value after it's received from the device after readless is initialized.

The patch also changes the driver to use dma_set_mask_and_coherent() function instead of the two pci_set_dma_mask() and pci_set_consistent_dma_mask() ones. Both methods achieve the same effect.

Fixes: 1738cd3ed342 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)") Signed-off-by: Mike Cui mikecui@amazon.com Signed-off-by: Arthur Kiyanovski akiyano@amazon.com Signed-off-by: Shay Agroskin shayagr@amazon.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/amazon/ena/ena_netdev.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index 3c3222e2dcfcf..9aea4cf19d0c9 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -2433,16 +2433,9 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev, goto err_mmio_read_less; }

- rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(dma_width)); + rc = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(dma_width)); if (rc) { - dev_err(dev, "pci_set_dma_mask failed 0x%x\n", rc); - goto err_mmio_read_less; - } - - rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(dma_width)); - if (rc) { - dev_err(dev, "err_pci_set_consistent_dma_mask failed 0x%x\n", - rc); + dev_err(dev, "dma_set_mask_and_coherent failed %d\n", rc); goto err_mmio_read_less; }

@@ -3183,6 +3176,12 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return rc; }

+ rc = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(ENA_MAX_PHYS_ADDR_SIZE_BITS)); + if (rc) { + dev_err(&pdev->dev, "dma_set_mask_and_coherent failed %d\n", rc); + goto err_disable_device; + } + pci_set_master(pdev);

ena_dev = vzalloc(sizeof(*ena_dev));

From: Lijun Pan ljp@linux.ibm.com

[ Upstream commit 0e435befaea45f7ea58682eecab5e37e05b2ce65 ]

crq->msgs could be NULL if the previous reset did not complete after freeing crq->msgs. Check for NULL before dereferencing them.

Snippet of call trace: ... ibmvnic 30000003 env3 (unregistering): Releasing sub-CRQ ibmvnic 30000003 env3 (unregistering): Releasing CRQ BUG: Kernel NULL pointer dereference on read at 0x00000000 Faulting instruction address: 0xc0000000000c1a30 Oops: Kernel access of bad area, sig: 11 [#1] LE PAGE_SIZE=64K MMU=Hash SMP NR_CPUS=2048 NUMA pSeries Modules linked in: ibmvnic(E-) rpadlpar_io rpaphp xt_CHECKSUM xt_MASQUERADE xt_conntrack ipt_REJECT nf_reject_ipv4 nft_compat nft_counter nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 nf_tables xsk_diag tcp_diag udp_diag tun raw_diag inet_diag unix_diag bridge af_packet_diag netlink_diag stp llc rfkill sunrpc pseries_rng xts vmx_crypto uio_pdrv_genirq uio binfmt_misc ip_tables xfs libcrc32c sd_mod t10_pi sg ibmvscsi ibmveth scsi_transport_srp dm_mirror dm_region_hash dm_log dm_mod [last unloaded: ibmvnic] CPU: 20 PID: 8426 Comm: kworker/20:0 Tainted: G E 5.10.0-rc1+ #12 Workqueue: events __ibmvnic_reset [ibmvnic] NIP: c0000000000c1a30 LR: c008000001b00c18 CTR: 0000000000000400 REGS: c00000000d05b7a0 TRAP: 0380 Tainted: G E (5.10.0-rc1+) MSR: 800000000280b033 <SF,VEC,VSX,EE,FP,ME,IR,DR,RI,LE> CR: 44002480 XER: 20040000 CFAR: c0000000000c19ec IRQMASK: 0 GPR00: 0000000000000400 c00000000d05ba30 c008000001b17c00 0000000000000000 GPR04: 0000000000000000 0000000000000000 0000000000000000 00000000000001e2 GPR08: 000000000001f400 ffffffffffffd950 0000000000000000 c008000001b0b280 GPR12: c0000000000c19c8 c00000001ec72e00 c00000000019a778 c00000002647b440 GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000 GPR20: 0000000000000006 0000000000000001 0000000000000003 0000000000000002 GPR24: 0000000000001000 c008000001b0d570 0000000000000005 c00000007ab5d550 GPR28: c00000007ab5c000 c000000032fcf848 c00000007ab5cc00 c000000032fcf800 NIP [c0000000000c1a30] memset+0x68/0x104 LR [c008000001b00c18] ibmvnic_reset_crq+0x70/0x110 [ibmvnic] Call Trace: [c00000000d05ba30] [0000000000000800] 0x800 (unreliable) [c00000000d05bab0] [c008000001b0a930] do_reset.isra.40+0x224/0x634 [ibmvnic] [c00000000d05bb80] [c008000001b08574] __ibmvnic_reset+0x17c/0x3c0 [ibmvnic] [c00000000d05bc50] [c00000000018d9ac] process_one_work+0x2cc/0x800 [c00000000d05bd20] [c00000000018df58] worker_thread+0x78/0x520 [c00000000d05bdb0] [c00000000019a934] kthread+0x1c4/0x1d0 [c00000000d05be20] [c00000000000d5d0] ret_from_kernel_thread+0x5c/0x6c

Fixes: 032c5e82847a ("Driver for IBM System i/p VNIC protocol") Signed-off-by: Lijun Pan ljp@linux.ibm.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/ibm/ibmvnic.c | 3 +++ 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 2fc8f281c2766..2938ac440fb36 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -4462,6 +4462,9 @@ static int ibmvnic_reset_crq(struct ibmvnic_adapter *adapter) } while (rc == H_BUSY || H_IS_LONG_BUSY(rc));

/* Clean out the queue */ + if (!crq->msgs) + return -EINVAL; + memset(crq->msgs, 0, PAGE_SIZE); crq->cur = 0; crq->active = false;

From: Ard Biesheuvel ardb@kernel.org

[ Upstream commit ff04f3b6f2e27f8ae28a498416af2a8dd5072b43 ]

The memory leak addressed by commit fe5186cf12e3 is a false positive: all allocations are recorded in a linked list, and freed when the filesystem is unmounted. This leads to double frees, and as reported by David, leads to crashes if SLUB is configured to self destruct when double frees occur.

So drop the redundant kfree() again, and instead, mark the offending pointer variable so the allocation is ignored by kmemleak.

Cc: Vamshi K Sthambamkadi vamshi.k.sthambamkadi@gmail.com Fixes: fe5186cf12e3 ("efivarfs: fix memory leak in efivarfs_create()") Reported-by: David Laight David.Laight@aculab.com Signed-off-by: Ard Biesheuvel ardb@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- fs/efivarfs/inode.c | 2 ++ fs/efivarfs/super.c | 1 - 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/fs/efivarfs/inode.c b/fs/efivarfs/inode.c index 8c6ab6c95727e..7f40343b39b05 100644 --- a/fs/efivarfs/inode.c +++ b/fs/efivarfs/inode.c @@ -10,6 +10,7 @@ #include <linux/efi.h> #include <linux/fs.h> #include <linux/ctype.h> +#include <linux/kmemleak.h> #include <linux/slab.h> #include <linux/uuid.h>

@@ -106,6 +107,7 @@ static int efivarfs_create(struct inode *dir, struct dentry *dentry, var->var.VariableName[i] = '\0';

inode->i_private = var; + kmemleak_ignore(var);

err = efivar_entry_add(var, &efivarfs_list); if (err) diff --git a/fs/efivarfs/super.c b/fs/efivarfs/super.c index 7808a26bd33fa..834615f13f3e3 100644 --- a/fs/efivarfs/super.c +++ b/fs/efivarfs/super.c @@ -23,7 +23,6 @@ LIST_HEAD(efivarfs_list); static void efivarfs_evict_inode(struct inode *inode) { clear_inode(inode); - kfree(inode->i_private); }

static const struct super_operations efivarfs_ops = {

From: Benjamin Berg bberg@redhat.com

[ Upstream commit e40cc1b476d60f22628741e53cf3446a29e6e6b9 ]

The lid state may change while the machine is suspended. As such, we may need to re-check the state at wake-up time (at least when waking up from hibernation). Add the appropriate call to the resume handler in order to sync the SW_TABLET_MODE switch state with the hardware state.

Fixes: dda3ec0aa631 ("platform/x86: thinkpad_acpi: Implement tablet mode using GMMS method") BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=210269 Signed-off-by: Benjamin Berg bberg@redhat.com Acked-by: Henrique de Moraes Holschuh hnh@hmh.eng.br Link: https://lore.kernel.org/r/20201123132157.866303-1-benjamin@sipsolutions.net Signed-off-by: Hans de Goede hdegoede@redhat.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/platform/x86/thinkpad_acpi.c | 1 + 1 file changed, 1 insertion(+)

diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c index 98bd8213b0378..8cc01857bc5c0 100644 --- a/drivers/platform/x86/thinkpad_acpi.c +++ b/drivers/platform/x86/thinkpad_acpi.c @@ -4251,6 +4251,7 @@ static void hotkey_resume(void) pr_err("error while attempting to reset the event firmware interface\n");

tpacpi_send_radiosw_update(); + tpacpi_input_send_tabletsw(); hotkey_tablet_mode_notify_change(); hotkey_wakeup_reason_notify_change(); hotkey_wakeup_hotunplug_complete_notify_change();

From: Marc Kleine-Budde mkl@pengutronix.de

[ Upstream commit e3409e4192535fbcc86a84b7a65d9351f46039ec ]

At lest the revision 3.3.0 of the bosch m_can IP core specifies that valid register values for "Nominal Time segment after sample point (NTSEG2)" are from 1 to 127. As the hardware uses a value of one more than the programmed value, mean tseg2_min is 2.

This patch fixes the tseg2_min value accordingly.

Cc: Dan Murphy dmurphy@ti.com Cc: Mario Huettel mario.huettel@gmx.net Acked-by: Sriram Dash sriram.dash@samsung.com Link: https://lore.kernel.org/r/20201124190751.3972238-1-mkl@pengutronix.de Fixes: b03cfc5bb0e1 ("can: m_can: Enable M_CAN version dependent initialization") Signed-off-by: Marc Kleine-Budde mkl@pengutronix.de Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/can/m_can/m_can.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index efaa342600c41..fbb970220c2d7 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -976,7 +976,7 @@ static const struct can_bittiming_const m_can_bittiming_const_31X = { .name = KBUILD_MODNAME, .tseg1_min = 2, /* Time segment 1 = prop_seg + phase_seg1 */ .tseg1_max = 256, - .tseg2_min = 1, /* Time segment 2 = phase_seg2 */ + .tseg2_min = 2, /* Time segment 2 = phase_seg2 */ .tseg2_max = 128, .sjw_max = 128, .brp_min = 1,

From: Alan Stern stern@rowland.harvard.edu

commit f3bc432aa8a7a2bfe9ebb432502be5c5d979d7fe upstream.

Commit 2f964780c03b ("USB: core: replace %p with %pK") used the %pK format specifier for a bunch of __user pointers. But as the 'K' in the specifier indicates, it is meant for kernel pointers. The reason for the %pK specifier is to avoid leaks of kernel addresses, but when the pointer is to an address in userspace the security implications are minimal. In particular, no kernel information is leaked.

This patch changes the __user %pK specifiers (used in a bunch of debugging output lines) to %px, which will always print the actual address with no mangling. (Notably, there is no printk format specifier particularly intended for __user pointers.)

Fixes: 2f964780c03b ("USB: core: replace %p with %pK") CC: Vamsi Krishna Samavedam vskrishn@codeaurora.org CC: stable@vger.kernel.org Signed-off-by: Alan Stern stern@rowland.harvard.edu Link: https://lore.kernel.org/r/20201119170228.GB576844@rowland.harvard.edu Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- drivers/usb/core/devio.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)

--- a/drivers/usb/core/devio.c +++ b/drivers/usb/core/devio.c @@ -463,11 +463,11 @@ static void snoop_urb(struct usb_device

if (userurb) { /* Async */ if (when == SUBMIT) - dev_info(&udev->dev, "userurb %pK, ep%d %s-%s, " + dev_info(&udev->dev, "userurb %px, ep%d %s-%s, " "length %u\n", userurb, ep, t, d, length); else - dev_info(&udev->dev, "userurb %pK, ep%d %s-%s, " + dev_info(&udev->dev, "userurb %px, ep%d %s-%s, " "actual_length %u status %d\n", userurb, ep, t, d, length, timeout_or_status); @@ -1927,7 +1927,7 @@ static int proc_reapurb(struct usb_dev_s if (as) { int retval;

- snoop(&ps->dev->dev, "reap %pK\n", as->userurb); + snoop(&ps->dev->dev, "reap %px\n", as->userurb); retval = processcompl(as, (void __user * __user *)arg); free_async(as); return retval; @@ -1944,7 +1944,7 @@ static int proc_reapurbnonblock(struct u

as = async_getcompleted(ps); if (as) { - snoop(&ps->dev->dev, "reap %pK\n", as->userurb); + snoop(&ps->dev->dev, "reap %px\n", as->userurb); retval = processcompl(as, (void __user * __user *)arg); free_async(as); } else { @@ -2070,7 +2070,7 @@ static int proc_reapurb_compat(struct us if (as) { int retval;

- snoop(&ps->dev->dev, "reap %pK\n", as->userurb); + snoop(&ps->dev->dev, "reap %px\n", as->userurb); retval = processcompl_compat(as, (void __user * __user *)arg); free_async(as); return retval; @@ -2087,7 +2087,7 @@ static int proc_reapurbnonblock_compat(s

as = async_getcompleted(ps); if (as) { - snoop(&ps->dev->dev, "reap %pK\n", as->userurb); + snoop(&ps->dev->dev, "reap %px\n", as->userurb); retval = processcompl_compat(as, (void __user * __user *)arg); free_async(as); } else { @@ -2512,7 +2512,7 @@ static long usbdev_do_ioctl(struct file #endif

case USBDEVFS_DISCARDURB: - snoop(&dev->dev, "%s: DISCARDURB %pK\n", __func__, p); + snoop(&dev->dev, "%s: DISCARDURB %px\n", __func__, p); ret = proc_unlinkurb(ps, p); break;

From: penghao penghao@uniontech.com

commit 9ca57518361418ad5ae7dc38a2128fbf4855e1a2 upstream.

Add a USB_QUIRK_DISCONNECT_SUSPEND quirk for the Lenovo TIO built-in usb-audio. when A630Z going into S3,the system immediately wakeup 7-8 seconds later by usb-audio disconnect interrupt to avoids the issue. eg dmesg: .... [ 626.974091 ] usb 7-1.1: USB disconnect, device number 3 .... .... [ 1774.486691] usb 7-1.1: new full-speed USB device number 5 using xhci_hcd [ 1774.947742] usb 7-1.1: New USB device found, idVendor=17ef, idProduct=a012, bcdDevice= 0.55 [ 1774.956588] usb 7-1.1: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [ 1774.964339] usb 7-1.1: Product: Thinkcentre TIO24Gen3 for USB-audio [ 1774.970999] usb 7-1.1: Manufacturer: Lenovo [ 1774.975447] usb 7-1.1: SerialNumber: 000000000000 [ 1775.048590] usb 7-1.1: 2:1: cannot get freq at ep 0x1 ....... Seeking a better fix, we've tried a lot of things, including: - Check that the device's power/wakeup is disabled - Check that remote wakeup is off at the USB level - All the quirks in drivers/usb/core/quirks.c e.g. USB_QUIRK_RESET_RESUME, USB_QUIRK_RESET, USB_QUIRK_IGNORE_REMOTE_WAKEUP, USB_QUIRK_NO_LPM.

but none of that makes any difference.

There are no errors in the logs showing any suspend/resume-related issues. When the system wakes up due to the modem, log-wise it appears to be a normal resume.

Introduce a quirk to disable the port during suspend when the modem is detected.

Signed-off-by: penghao penghao@uniontech.com Link: https://lore.kernel.org/r/20201118123039.11696-1-penghao@uniontech.com Cc: stable stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- drivers/usb/core/quirks.c | 4 ++++ 1 file changed, 4 insertions(+)

--- a/drivers/usb/core/quirks.c +++ b/drivers/usb/core/quirks.c @@ -421,6 +421,10 @@ static const struct usb_device_id usb_qu { USB_DEVICE(0x1532, 0x0116), .driver_info = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL },

+ /* Lenovo ThinkCenter A630Z TI024Gen3 usb-audio */ + { USB_DEVICE(0x17ef, 0xa012), .driver_info = + USB_QUIRK_DISCONNECT_SUSPEND }, + /* BUILDWIN Photo Frame */ { USB_DEVICE(0x1908, 0x1315), .driver_info = USB_QUIRK_HONOR_BNUMINTERFACES },

From: Anand K Mistry amistry@google.com

commit 33fc379df76b4991e5ae312f07bcd6820811971e upstream.

When spectre_v2_user={seccomp,prctl},ibpb is specified on the command line, IBPB is force-enabled and STIPB is conditionally-enabled (or not available).

However, since

21998a351512 ("x86/speculation: Avoid force-disabling IBPB based on STIBP and enhanced IBRS.")

the spectre_v2_user_ibpb variable is set to SPECTRE_V2_USER_{PRCTL,SECCOMP} instead of SPECTRE_V2_USER_STRICT, which is the actual behaviour. Because the issuing of IBPB relies on the switch_mm_*_ibpb static branches, the mitigations behave as expected.

Since

1978b3a53a74 ("x86/speculation: Allow IBPB to be conditionally enabled on CPUs with always-on STIBP")

this discrepency caused the misreporting of IB speculation via prctl().

On CPUs with STIBP always-on and spectre_v2_user=seccomp,ibpb, prctl(PR_GET_SPECULATION_CTRL) would return PR_SPEC_PRCTL | PR_SPEC_ENABLE instead of PR_SPEC_DISABLE since both IBPB and STIPB are always on. It also allowed prctl(PR_SET_SPECULATION_CTRL) to set the IB speculation mode, even though the flag is ignored.

Similarly, for CPUs without SMT, prctl(PR_GET_SPECULATION_CTRL) should also return PR_SPEC_DISABLE since IBPB is always on and STIBP is not available.

[ bp: Massage commit message. ]

Fixes: 21998a351512 ("x86/speculation: Avoid force-disabling IBPB based on STIBP and enhanced IBRS.") Fixes: 1978b3a53a74 ("x86/speculation: Allow IBPB to be conditionally enabled on CPUs with always-on STIBP") Signed-off-by: Anand K Mistry amistry@google.com Signed-off-by: Borislav Petkov bp@suse.de Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/20201110123349.1.Id0cbf996d2151f4c143c90f9028651a5... Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- arch/x86/kernel/cpu/bugs.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -733,11 +733,13 @@ spectre_v2_user_select_mitigation(enum s if (boot_cpu_has(X86_FEATURE_IBPB)) { setup_force_cpu_cap(X86_FEATURE_USE_IBPB);

+ spectre_v2_user_ibpb = mode; switch (cmd) { case SPECTRE_V2_USER_CMD_FORCE: case SPECTRE_V2_USER_CMD_PRCTL_IBPB: case SPECTRE_V2_USER_CMD_SECCOMP_IBPB: static_branch_enable(&switch_mm_always_ibpb); + spectre_v2_user_ibpb = SPECTRE_V2_USER_STRICT; break; case SPECTRE_V2_USER_CMD_PRCTL: case SPECTRE_V2_USER_CMD_AUTO: @@ -751,8 +753,6 @@ spectre_v2_user_select_mitigation(enum s pr_info("mitigation: Enabling %s Indirect Branch Prediction Barrier\n", static_key_enabled(&switch_mm_always_ibpb) ? "always-on" : "conditional"); - - spectre_v2_user_ibpb = mode; }

/*

From: Xiaochen Shen xiaochen.shen@intel.com

commit 758999246965eeb8b253d47e72f7bfe508804b16 upstream.

On resource group creation via a mkdir an extra kernfs_node reference is obtained by kernfs_get() to ensure that the rdtgroup structure remains accessible for the rdtgroup_kn_unlock() calls where it is removed on deletion. Currently the extra kernfs_node reference count is only dropped by kernfs_put() in rdtgroup_kn_unlock() while the rdtgroup structure is removed in a few other locations that lack the matching reference drop.

In call paths of rmdir and umount, when a control group is removed, kernfs_remove() is called to remove the whole kernfs nodes tree of the control group (including the kernfs nodes trees of all child monitoring groups), and then rdtgroup structure is freed by kfree(). The rdtgroup structures of all child monitoring groups under the control group are freed by kfree() in free_all_child_rdtgrp().

Before calling kfree() to free the rdtgroup structures, the kernfs node of the control group itself as well as the kernfs nodes of all child monitoring groups still take the extra references which will never be dropped to 0 and the kernfs nodes will never be freed. It leads to reference count leak and kernfs_node_cache memory leak.

For example, reference count leak is observed in these two cases: (1) mount -t resctrl resctrl /sys/fs/resctrl mkdir /sys/fs/resctrl/c1 mkdir /sys/fs/resctrl/c1/mon_groups/m1 umount /sys/fs/resctrl

(2) mkdir /sys/fs/resctrl/c1 mkdir /sys/fs/resctrl/c1/mon_groups/m1 rmdir /sys/fs/resctrl/c1

The same reference count leak issue also exists in the error exit paths of mkdir in mkdir_rdt_prepare() and rdtgroup_mkdir_ctrl_mon().

Fix this issue by following changes to make sure the extra kernfs_node reference on rdtgroup is dropped before freeing the rdtgroup structure. (1) Introduce rdtgroup removal helper rdtgroup_remove() to wrap up kernfs_put() and kfree().

(2) Call rdtgroup_remove() in rdtgroup removal path where the rdtgroup structure is about to be freed by kfree().

(3) Call rdtgroup_remove() or kernfs_put() as appropriate in the error exit paths of mkdir where an extra reference is taken by kernfs_get().

Backporting notes:

Since upstream commit fa7d949337cc ("x86/resctrl: Rename and move rdt files to a separate directory"), the file arch/x86/kernel/cpu/intel_rdt_rdtgroup.c has been renamed and moved to arch/x86/kernel/cpu/resctrl/rdtgroup.c. Apply the change against file arch/x86/kernel/cpu/intel_rdt_rdtgroup.c in older stable trees.

Fixes: f3cbeacaa06e ("x86/intel_rdt/cqm: Add rmdir support") Fixes: e02737d5b826 ("x86/intel_rdt: Add tasks files") Fixes: 60cf5e101fd4 ("x86/intel_rdt: Add mkdir to resctrl file system") Reported-by: Willem de Bruijn willemb@google.com Signed-off-by: Xiaochen Shen xiaochen.shen@intel.com Signed-off-by: Borislav Petkov bp@suse.de Reviewed-by: Reinette Chatre reinette.chatre@intel.com Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/1604085088-31707-1-git-send-email-xiaochen.shen@in... Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- arch/x86/kernel/cpu/intel_rdt_rdtgroup.c | 32 ++++++++++++++++++++++++------- 1 file changed, 25 insertions(+), 7 deletions(-)

--- a/arch/x86/kernel/cpu/intel_rdt_rdtgroup.c +++ b/arch/x86/kernel/cpu/intel_rdt_rdtgroup.c @@ -515,6 +515,24 @@ unlock: return ret ?: nbytes; }

+/** + * rdtgroup_remove - the helper to remove resource group safely + * @rdtgrp: resource group to remove + * + * On resource group creation via a mkdir, an extra kernfs_node reference is + * taken to ensure that the rdtgroup structure remains accessible for the + * rdtgroup_kn_unlock() calls where it is removed. + * + * Drop the extra reference here, then free the rdtgroup structure. + * + * Return: void + */ +static void rdtgroup_remove(struct rdtgroup *rdtgrp) +{ + kernfs_put(rdtgrp->kn); + kfree(rdtgrp); +} + struct task_move_callback { struct callback_head work; struct rdtgroup *rdtgrp; @@ -537,7 +555,7 @@ static void move_myself(struct callback_ (rdtgrp->flags & RDT_DELETED)) { current->closid = 0; current->rmid = 0; - kfree(rdtgrp); + rdtgroup_remove(rdtgrp); }

preempt_disable(); @@ -1959,8 +1977,7 @@ void rdtgroup_kn_unlock(struct kernfs_no rdtgrp->mode == RDT_MODE_PSEUDO_LOCKED) rdtgroup_pseudo_lock_remove(rdtgrp); kernfs_unbreak_active_protection(kn); - kernfs_put(rdtgrp->kn); - kfree(rdtgrp); + rdtgroup_remove(rdtgrp); } else { kernfs_unbreak_active_protection(kn); } @@ -2169,7 +2186,7 @@ static void free_all_child_rdtgrp(struct if (atomic_read(&sentry->waitcount) != 0) sentry->flags = RDT_DELETED; else - kfree(sentry); + rdtgroup_remove(sentry); } }

@@ -2211,7 +2228,7 @@ static void rmdir_all_sub(void) if (atomic_read(&rdtgrp->waitcount) != 0) rdtgrp->flags = RDT_DELETED; else - kfree(rdtgrp); + rdtgroup_remove(rdtgrp); } /* Notify online CPUs to update per cpu storage and PQR_ASSOC MSR */ update_closid_rmid(cpu_online_mask, &rdtgroup_default); @@ -2602,7 +2619,7 @@ static int mkdir_rdt_prepare(struct kern * kernfs_remove() will drop the reference count on "kn" which * will free it. But we still need it to stick around for the * rdtgroup_kn_unlock(kn) call. Take one extra reference here, - * which will be dropped inside rdtgroup_kn_unlock(). + * which will be dropped by kernfs_put() in rdtgroup_remove(). */ kernfs_get(kn);

@@ -2643,6 +2660,7 @@ static int mkdir_rdt_prepare(struct kern out_idfree: free_rmid(rdtgrp->mon.rmid); out_destroy: + kernfs_put(rdtgrp->kn); kernfs_remove(rdtgrp->kn); out_free_rgrp: kfree(rdtgrp); @@ -2655,7 +2673,7 @@ static void mkdir_rdt_prepare_clean(stru { kernfs_remove(rgrp->kn); free_rmid(rgrp->mon.rmid); - kfree(rgrp); + rdtgroup_remove(rgrp); }

/*