Re: [PULL 20/34] tests/libqos: add riscv/virt machine nodes

From: Thomas Huth <thuth@redhat.com>
To: Christian Schoenebeck <qemu_oss@crudebyte.com>,
	Alistair Francis <alistair23@gmail.com>,
	qemu-devel@nongnu.org
Cc: Alistair Francis <alistair.francis@wdc.com>,
	Greg Kurz <groug@kaod.org>,
	Peter Maydell <peter.maydell@linaro.org>,
	Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Subject: Re: [PULL 20/34] tests/libqos: add riscv/virt machine nodes
Date: Mon, 25 Mar 2024 14:46:21 +0100	[thread overview]
Message-ID: <cfab3a55-4755-45ed-b0a6-9a7b5ab0c005@redhat.com> (raw)
In-Reply-To: <2671306.ea0pUnVCe6@silver>

On 25/03/2024 14.25, Christian Schoenebeck wrote:
> On Monday, March 25, 2024 1:35:52 PM CET Daniel Henrique Barboza wrote:
>> On 3/25/24 06:20, Thomas Huth wrote:
>>> On 08/03/2024 12.11, Alistair Francis wrote:
>>>> From: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
>>>>
>>>> Add a RISC-V 'virt' machine to the graph. This implementation is a
>>>> modified copy of the existing arm machine in arm-virt-machine.c
>>>>
>>>> It contains a virtio-mmio and a generic-pcihost controller. The
>>>> generic-pcihost controller hardcodes assumptions from the ARM 'virt'
>>>> machine, like ecam and pio_base addresses, so we'll add an extra step to
>>>> set its parameters after creating it.
>>>>
>>>> Our command line is incremented with 'aclint' parameters to allow the
>>>> machine to run MSI tests.
>>>>
>>>> Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
>>>> Acked-by: Alistair Francis <alistair.francis@wdc.com>
>>>> Acked-by: Thomas Huth <thuth@redhat.com>
>>>> Message-ID: <20240217192607.32565-7-dbarboza@ventanamicro.com>
>>>> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
>>>> ---
>>>
>>>    Hi!
>>>
>>> I noticed that "make check SPEED=slow" is now failing on the qos-test with both, qemu-system-riscv32 and qemu-system-riscv64. Seems like it fails with the virtio-9p test, when I run the qos-test manually, I get:
>>>
>>> $ MALLOC_PERTURB_=21 V=2 QTEST_QEMU_BINARY=./qemu-system-riscv64 \
>>>      tests/qtest/qos-test -m slow
>>> ...
>>> # Start of local tests
>>> # starting QEMU: exec ./qemu-system-riscv64 -qtest unix:/tmp/qtest-211303.sock -qtest-log /dev/null -chardev socket,path=/tmp/qtest-211303.qmp,id=char0 -mon chardev=char0,mode=control -display none -audio none -M virt,aclint=on,aia=aplic-imsic -fsdev local,id=fsdev0,path='/home/thuth/tmp/qemu-build/qtest-9p-local-MBCML2',security_model=mapped-xattr -device virtio-9p-pci,fsdev=fsdev0,addr=04.0,mount_tag=qtest -accel qtest
>>> ok 168 /riscv64/virt/generic-pcihost/pci-bus-generic/pci-bus/virtio-9p-pci/virtio-9p/virtio-9p-tests/local/config
>>> Received response 7 (RLERROR) instead of 73 (RMKDIR)
>>> Rlerror has errno 17 (File exists)
>>> **
>>> ERROR:../../devel/qemu/tests/qtest/libqos/virtio-9p-client.c:275:v9fs_req_recv: assertion failed (hdr.id == id): (7 == 73)
>>> not ok /riscv64/virt/generic-pcihost/pci-bus-generic/pci-bus/virtio-9p-pci/virtio-9p/virtio-9p-tests/local/create_dir - ERROR:../../devel/qemu/tests/qtest/libqos/virtio-9p-client.c:275:v9fs_req_recv: assertion failed (hdr.id == id): (7 == 73)
>>> Bail out!
>>> Aborted (core dumped)
>>>
>>> Could you please have a look? ... or if it is too cumbersome to fix, could we please always skip the virtio-9p local tests on riscv ?
>>
>> I'll take a look.
>>
>> Do we run these slow tests in the Gitlab pipeline? I don't recall this
>> particular test failing when I first introduced the riscv machine nodes.
> 
> No, the 'local' 9p tests were taken out by moving them to 'slow', because
> these particular tests did not pass in the cloud and gitlab doesn't run
> 'slow':
> 
> commit 558f5c42efded3e0d0b20a90bce2a9a14580d824
> Author: Greg Kurz <groug@kaod.org>
> Date:   Tue Nov 24 08:43:43 2020 +0100
> 
>      tests/9pfs: Mark "local" tests as "slow"
>      
>      The "local" tests can fail on some automated build systems as
>      reported here:
>      
>      https://lists.nongnu.org/archive/html/qemu-devel/2020-11/msg05510.html
>      
>      This will need to be investigated and addressed later. Let's go for a
>      workaround in the meantime : mark the "local" tests as "slow" so that
>      they aren't executed with a simple "make check" like in the case above.
>      
>      Reported-by: Cole Robinson <crobinso@redhat.com>
>      Signed-off-by: Greg Kurz <groug@kaod.org>
>      Reviewed-by: Thomas Huth <thuth@redhat.com>
>      Acked-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
>      Message-Id: <160620382310.1423262.7364287092069513483.stgit@bahia.lan>
>      Signed-off-by: Greg Kurz <groug@kaod.org>
> 
> Could be because the 'local' 9p backend needs xattr support which might not be
> available with gitlab container's filesystem. But I haven't investigated.

I just ran "make check SPEED=slow" locally on my laptop. Only the riscv 
qos-tests were failing, the other targets worked fine. So this must be 
something specific to riscv, I think.

  Thomas