qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH-for-10.0] system/qtest: Access memory using cpu_physical_memo


From: Philippe Mathieu-Daudé
Subject: Re: [PATCH-for-10.0] system/qtest: Access memory using cpu_physical_memory_rw() API
Date: Mon, 9 Dec 2024 21:42:50 +0100
User-agent: Mozilla Thunderbird

On 9/12/24 21:34, Fabiano Rosas wrote:
Fabiano Rosas <farosas@suse.de> writes:

Philippe Mathieu-Daudé <philmd@linaro.org> writes:

There is no vCPU within the QTest accelerator (well, they
are stubs doing nothing, see dummy_cpu_thread_fn).
Directly use the cpu_physical_memory_rw() API -- which
amusingly prefixed 'cpu_' does not use vCPU -- to access
memory. This reduces accesses to the global 'first_cpu'.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

Queued, thanks!

Scratch that, I see some tests failing:

QTEST_QEMU_BINARY=../build/qemu-system-arm ./tests/qtest/sse-timer-test
# random seed: R02S7b4a25dac8a32a86ee016342d60e78ec
# starting QEMU: exec ../build/qemu-system-arm -qtest unix:/tmp/qtest-7386.sock 
-qtest-log /dev/fd/2 -chardev socket,path=/tmp/qtest-7386.qmp,id=char0 -mon 
chardev=char0,mode=control -display
  none -audio none -machine mps3-an547 -accel qtest
[I 0.000000] OPENED
[R +0.033524] endianness
[S +0.033533] OK little
{"QMP": {"version": {"qemu": {"micro": 93, "minor": 1, "major": 9}, "package": 
"v9.2.0-rc3-23-g3fe88295e8"}, "capabilities": ["oob"]}}
{"execute": "qmp_capabilities"}

{"return": {}}
1..3
# Start of arm tests
# Start of sse-timer tests
[R +0.035240] writel 0x58100000 0x0
[S +0.035250] OK
[R +0.035263] writel 0x4800002c 0x0
[S +0.035265] OK
[R +0.035275] writel 0x4800004c 0x0
[S +0.035278] OK
[R +0.035287] writel 0x58100008 0x0
[S +0.035289] OK
[R +0.035296] writel 0x5810000c 0x0
[S +0.035299] OK
[R +0.035308] clock_step 3125
[S +0.035312] OK 3125
[R +0.035323] readl 0x58100008
[S +0.035326] OK 0x0000000000000000
[R +0.035334] readl 0x5810000c
[S +0.035337] OK 0x0000000000000000
[R +0.035344] writel 0x58100000 0x1
[S +0.035346] OK
[R +0.035354] clock_step 3125
[S +0.035357] OK 6250
[R +0.035364] readl 0x58100008
[S +0.035367] OK 0x0000000000000000
**
ERROR:../tests/qtest/sse-timer-test.c:91:test_counter: assertion failed 
(readl(COUNTER_BASE + CNTCV_LO) == 100): (0 == 100)
Bail out! ERROR:../tests/qtest/sse-timer-test.c:91:test_counter:
assertion failed (readl(COUNTER_BASE + CNTCV_LO) == 100): (0 == 100)

Odd, this should be a no-op. I'll have a look. Thanks for testing!




reply via email to

[Prev in Thread] Current Thread [Next in Thread]