[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <02ad7973-f1d9-573b-8986-9a13e51aa661@loongson.cn>
Date: Tue, 19 Aug 2025 09:01:35 +0800
From: Bibo Mao <maobibo@...ngson.cn>
To: Andrew Jones <ajones@...tanamicro.com>
Cc: Dong Yang <dayss1224@...il.com>, pbonzini@...hat.com, shuah@...nel.org,
kvm@...r.kernel.org, linux-kselftest@...r.kernel.org,
linux-kernel@...r.kernel.org, chenhaucai@...nel.org,
Quan Zhou <zhouquan@...as.ac.cn>
Subject: Re: [PATCH] KVM: loongarch: selftests: Remove common tests built by
TEST_GEN_PROGS_COMMON
On 2025/8/19 上午8:30, Andrew Jones wrote:
> On Mon, Aug 11, 2025 at 06:49:07PM +0800, Bibo Mao wrote:
>> Hi Dong,
>>
>> Thanks for you patch.
>>
>> On 2025/8/11 下午4:24, Dong Yang wrote:
>>> Remove the common KVM test cases already added to TEST_GEN_PROGS_COMMON
>>> as following:
>>>
>>> demand_paging_test
>>> dirty_log_test
>>> guest_print_test
>>> kvm_binary_stats_test
>>> kvm_create_max_vcpus
>>> kvm_page_table_test
>>> set_memory_region_test
>>>
>>> Fixes: a867688c8cbb ("KVM: selftests: Add supported test cases for LoongArch")
>>> Signed-off-by: Quan Zhou <zhouquan@...as.ac.cn>
>>> Signed-off-by: Dong Yang <dayss1224@...il.com>
>>> ---
>>> tools/testing/selftests/kvm/Makefile.kvm | 7 -------
>>> 1 file changed, 7 deletions(-)
>>>
>>> diff --git a/tools/testing/selftests/kvm/Makefile.kvm b/tools/testing/selftests/kvm/Makefile.kvm
>>> index 38b95998e1e6..d2ad85a8839f 100644
>>> --- a/tools/testing/selftests/kvm/Makefile.kvm
>>> +++ b/tools/testing/selftests/kvm/Makefile.kvm
>>> @@ -199,17 +199,10 @@ TEST_GEN_PROGS_riscv += get-reg-list
>>> TEST_GEN_PROGS_riscv += steal_time
>> TEST_GEN_PROGS_loongarch = $(TEST_GEN_PROGS_COMMON) is missing.
>>
>> BTW irqfd_test in TEST_GEN_PROGS_COMMON fails to run on LoongArch, does this
>> test case pass to run on Riscv?
>
> It appears to. It outputs the vm mode created and then exits with a zero
> exit code.
Here is output of irqfd test on LoongArch
[root@...-131 kvm]# ./irqfd_test
Random seed: 0x6b8b4567
==== Test Assertion Failure ====
include/kvm_util.h:527: !ret
pid=4016 tid=4016 errno=11 - Resource temporarily unavailable
1 0x00000001200027ab: kvm_irqfd at kvm_util.h:527
2 0x00000001200020d7: main at irqfd_test.c:100
3 0x00007ffff38a8707: ?? ??:0
4 0x00007ffff38a87ef: ?? ??:0
5 0x00000001200023b7: _start at ??:?
KVM_IRQFD failed, rc: -1 errno: 11 (Resource temporarily unavailable)
The problem is that kernel irqchip is not created with irqfd_test, and
function kvm_arch_intc_initialized() returns false on LongArch.
/*
* returns true if the virtual interrupt controller is initialized and
* ready to accept virtual IRQ. On some architectures virtual interrupt
* controller is dynamically instantiated and this is not always true.
*/
bool kvm_arch_intc_initialized(struct kvm *kvm);
On LoongArch virtual irqchip is dynamically created by VMM.
Regards
Bibo Mao
>
> Thanks,
> drew
>
>>
>> Regards
>> Bibo Mao
>>> TEST_GEN_PROGS_loongarch += coalesced_io_test
>>> -TEST_GEN_PROGS_loongarch += demand_paging_test
>>> TEST_GEN_PROGS_loongarch += dirty_log_perf_test
>>> -TEST_GEN_PROGS_loongarch += dirty_log_test
>>> -TEST_GEN_PROGS_loongarch += guest_print_test
>>> TEST_GEN_PROGS_loongarch += hardware_disable_test
>>> -TEST_GEN_PROGS_loongarch += kvm_binary_stats_test
>>> -TEST_GEN_PROGS_loongarch += kvm_create_max_vcpus
>>> -TEST_GEN_PROGS_loongarch += kvm_page_table_test
>>> TEST_GEN_PROGS_loongarch += memslot_modification_stress_test
>>> TEST_GEN_PROGS_loongarch += memslot_perf_test
>>> -TEST_GEN_PROGS_loongarch += set_memory_region_test
>>> SPLIT_TESTS += arch_timer
>>> SPLIT_TESTS += get-reg-list
>>>
>>
Powered by blists - more mailing lists