Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <9869257e-d257-da83-edd6-0c167f915829@de.ibm.com>
Date:   Fri, 25 Feb 2022 14:29:31 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.5.0
Subject: Re: [PATCH v3 1/1] KVM: s390: pv: make use of ultravisor AIV support
Content-Language: en-US
To:     Michael Mueller <mimu@linux.ibm.com>, kvm@vger.kernel.org
Cc:     cohuck@redhat.com, frankja@linux.ibm.com, thuth@redhat.com,
        pasic@linux.ibm.com, david@redhat.com, linux-s390@vger.kernel.org,
        linux-kernel@vger.kernel.org
References: <20220209152217.1793281-1-mimu@linux.ibm.com>
 <20220209152217.1793281-2-mimu@linux.ibm.com>
 <803275d5-58f4-21d9-8020-e56f05737450@de.ibm.com>
 <d58a6a64-a6b6-3238-00d7-573e4177f38c@linux.ibm.com>
From:   Christian Borntraeger <borntraeger@de.ibm.com>
In-Reply-To: <d58a6a64-a6b6-3238-00d7-573e4177f38c@linux.ibm.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk


Am 24.02.22 um 16:47 schrieb Michael Mueller:
> 
> 
> On 22.02.22 09:13, Christian Borntraeger wrote:
>> Am 09.02.22 um 16:22 schrieb Michael Mueller:
>>> This patch enables the ultravisor adapter interruption vitualization
>>> support indicated by UV feature BIT_UV_FEAT_AIV. This allows ISC
>>> interruption injection directly into the GISA IPM for PV kvm guests.
>>>
>>> Hardware that does not support this feature will continue to use the
>>> UV interruption interception method to deliver ISC interruptions to
>>> PV kvm guests. For this purpose, the ECA_AIV bit for all guest cpus
>>> will be cleared and the GISA will be disabled during PV CPU setup.
>>>
>>> In addition a check in __inject_io() has been removed. That reduces the
>>> required instructions for interruption handling for PV and traditional
>>> kvm guests.
>>>
>>> Signed-off-by: Michael Mueller <mimu@linux.ibm.com>
>>
>> The CI said the following with gisa_disable in the calltrace.
>> Will drop from next for now.
> 
> It turns out this is related to kvm_s390_set_tod_clock() which
> is triggered by a kvm-unit-test (sac_PV) and not related directly
> to this patch. Please re-apply.

Done. We need to fix the sck handler instead.

> 
>>
>>     LOCKDEP_CIRCULAR (suite: kvm-unit-tests-kvm, case: -)
>>                  WARNING: possible circular locking dependency detected
>> 5.17.0-20220221.rc5.git1.b8f0356a093a.300.fc35.s390x+debug #1 Not tainted
>>                  ------------------------------------------------------
>>                  qemu-system-s39/161139 is trying to acquire lock:
>>                  0000000280dc0b98 (&kvm->lock){+.+.}-{3:3}, at: kvm_s390_set_tod_clock+0x36/0x220 [kvm]
>>                  but task is already holding lock:
>>                  0000000280f4e4b8 (&vcpu->mutex){+.+.}-{3:3}, at: kvm_vcpu_ioctl+0x9a/0xa40 [kvm]
>>                  which lock already depends on the new lock.
>>                  the existing dependency chain (in reverse order) is:
>>                  -> #1 (&vcpu->mutex){+.+.}-{3:3}:
>>                         __lock_acquire+0x604/0xbd8
>>                         lock_acquire.part.0+0xe2/0x250
>>                         lock_acquire+0xb0/0x200
>>                         __mutex_lock+0x9e/0x8a0
>>                         mutex_lock_nested+0x32/0x40
>>                         kvm_s390_gisa_disable+0xa4/0x130 [kvm]
>>                         kvm_s390_handle_pv+0x718/0x778 [kvm]
>>                         kvm_arch_vm_ioctl+0x4ac/0x5f8 [kvm]
>>                         kvm_vm_ioctl+0x336/0x530 [kvm]
>>                         __s390x_sys_ioctl+0xbe/0x100
>>                         __do_syscall+0x1da/0x208
>>                         system_call+0x82/0xb0
>>                  -> #0 (&kvm->lock){+.+.}-{3:3}:
>>                         check_prev_add+0xe0/0xed8
>>                         validate_chain+0x736/0xb20
>>                         __lock_acquire+0x604/0xbd8
>>                         lock_acquire.part.0+0xe2/0x250
>>                         lock_acquire+0xb0/0x200
>>                         __mutex_lock+0x9e/0x8a0
>>                         mutex_lock_nested+0x32/0x40
>>                         kvm_s390_set_tod_clock+0x36/0x220 [kvm]
>>                         kvm_s390_handle_b2+0x378/0x728 [kvm]
>>                         kvm_handle_sie_intercept+0x13a/0x448 [kvm]
>>                         vcpu_post_run+0x28e/0x560 [kvm]
>>                         __vcpu_run+0x266/0x388 [kvm]
>>                         kvm_arch_vcpu_ioctl_run+0x10a/0x270 [kvm]
>>                         kvm_vcpu_ioctl+0x27c/0xa40 [kvm]
>>                         __s390x_sys_ioctl+0xbe/0x100
>>                         __do_syscall+0x1da/0x208
>>                         system_call+0x82/0xb0
>>                  other info that might help us debug this:
>>                   Possible unsafe locking scenario:
>>                         CPU0                    CPU1
>>                         ----                    ----
>>                    lock(&vcpu->mutex);
>>                                                 lock(&kvm->lock);
>>                                                 lock(&vcpu->mutex);
>>                    lock(&kvm->lock);
>>                   *** DEADLOCK ***
>>                  2 locks held by qemu-system-s39/161139:
>>                   #0: 0000000280f4e4b8 (&vcpu->mutex){+.+.}-{3:3}, at: kvm_vcpu_ioctl+0x9a/0xa40 [kvm]
>>                   #1: 0000000280dc47c8 (&kvm->srcu){....}-{0:0}, at: __vcpu_run+0x1d4/0x388 [kvm]
>>                  stack backtrace:
>>                  CPU: 10 PID: 161139 Comm: qemu-system-s39 Not tainted 5.17.0-20220221.rc5.git1.b8f0356a093a.300.fc35.s390x+debug #1
>>                  Hardware name: IBM 8561 T01 701 (LPAR)
>>                  Call Trace:
>>                   [<00000001da4e89de>] dump_stack_lvl+0x8e/0xc8
>>                   [<00000001d9876c56>] check_noncircular+0x136/0x158
>>                   [<00000001d9877c70>] check_prev_add+0xe0/0xed8
>>                   [<00000001d987919e>] validate_chain+0x736/0xb20
>>                   [<00000001d987b23c>] __lock_acquire+0x604/0xbd8
>>                   [<00000001d987c432>] lock_acquire.part.0+0xe2/0x250
>>                   [<00000001d987c650>] lock_acquire+0xb0/0x200
>>                   [<00000001da4f72ae>] __mutex_lock+0x9e/0x8a0
>>                   [<00000001da4f7ae2>] mutex_lock_nested+0x32/0x40
>>                   [<000003ff8070cd6e>] kvm_s390_set_tod_clock+0x36/0x220 [kvm]
>>                   [<000003ff8071dd68>] kvm_s390_handle_b2+0x378/0x728 [kvm]
>>                   [<000003ff8071146a>] kvm_handle_sie_intercept+0x13a/0x448 [kvm]
>>                   [<000003ff8070dd46>] vcpu_post_run+0x28e/0x560 [kvm]
>>                   [<000003ff8070e27e>] __vcpu_run+0x266/0x388 [kvm]
>>                   [<000003ff8070eba2>] kvm_arch_vcpu_ioctl_run+0x10a/0x270 [kvm]
>>                   [<000003ff806f4044>] kvm_vcpu_ioctl+0x27c/0xa40 [kvm]
>>                   [<00000001d9b47ac6>] __s390x_sys_ioctl+0xbe/0x100
>>                   [<00000001da4ec152>] __do_syscall+0x1da/0x208
>>                   [<00000001da4fec42>] system_call+0x82/0xb0
>>                  INFO: lockdep is turned off.