From: Arnd Bergmann <arnd(a)arndb.de>
Export the this_cpu_has_cap() for use by modules. This is
used by TRBE driver. Without this patch, TRBE will fail
to build as a module :
ERROR: modpost: "this_cpu_has_cap" [drivers/hwtracing/coresight/coresight-trbe.ko] undefined!
Fixes: 8a1065127d95 ("coresight: trbe: Add infrastructure for Errata handling")
Cc: Will Deacon <will(a)kernel.org>
Cc: Catalin Marinas <catalin.marinas(a)arm.com>
Cc: Mathieu Poirier <mathieu.poirier(a)linaro.org>
Cc: Anshuman Khandual <anshuman.khandual(a)arm.com>
Signed-off-by: Arnd Bergmann <arnd(a)arndb.de>
[ change to EXPORT_SYMBOL_GPL ]
Acked-by: Catalin Marinas <catalin.marinas(a)arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose(a)arm.com>
---
arch/arm64/kernel/cpufeature.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
index f8a3067d10c6..82e68c69bb99 100644
--- a/arch/arm64/kernel/cpufeature.c
+++ b/arch/arm64/kernel/cpufeature.c
@@ -2839,6 +2839,7 @@ bool this_cpu_has_cap(unsigned int n)
return false;
}
+EXPORT_SYMBOL_GPL(this_cpu_has_cap);
/*
* This helper function is used in a narrow window when,
--
2.25.4
On 21/10/2021 08:38, Tao Zhang wrote:
> Add TPDA and TPDM support to DTS for RB5 board. This change is a
> sample for validating. After applying this patch, the new TPDM and
> TPDA nodes can be observed at the coresight devices path. TPDM and
> TPDA hardware can be operated by commands.
>
> List the commands for validating this series patches as below.
> echo 1 > /sys/bus/coresight/devices/tmc_etf0/enable_sink
> echo 1 > /sys/bus/coresight/devices/tpdm0/enable_source
> echo 1 > /sys/bus/coresight/devices/tpdm0/integration_test
> echo 2 > /sys/bus/coresight/devices/tpdm0/integration_test
> cat /dev/tmc_etf0 > /data/etf-tpdm0.bin
> echo 0 > /sys/bus/coresight/devices/tpdm0/enable_source
> echo 0 > /sys/bus/coresight/devices/tmc_etf0/enable_sink
> echo 1 > /sys/bus/coresight/devices/tmc_etf0/enable_sink
> echo 1 > /sys/bus/coresight/devices/tpdm1/enable_source
> echo 1 > /sys/bus/coresight/devices/tpdm1/integration_test
> echo 2 > /sys/bus/coresight/devices/tpdm1/integration_test
> cat /dev/tmc_etf0 > /data/etf-tpdm1.bin
> echo 0 > /sys/bus/coresight/devices/tpdm1/enable_source
> echo 0 > /sys/bus/coresight/devices/tmc_etf0/enable_sink
> echo 1 > /sys/bus/coresight/devices/tmc_etf0/enable_sink
> echo 1 > /sys/bus/coresight/devices/tpdm2/enable_source
> echo 1 > /sys/bus/coresight/devices/tpdm2/integration_test
> echo 2 > /sys/bus/coresight/devices/tpdm2/integration_test
> cat /dev/tmc_etf0 > /data/etf-tpdm2.bin
> echo 0 > /sys/bus/coresight/devices/tpdm2/enable_source
> echo 0 > /sys/bus/coresight/devices/tmc_etf0/enable_sink
>
> If the data from TPDMs can be obtained from the ETF, it means
> that the TPDMs verification is successful. At the same time,
How can we decode the TPDM trace ? Is there a public decoder
available ?
> since TPDM0, TPDM1 and TPDM2 are all connected to the same
> funnel "funnel@6c2d000" and output via different output ports,
> it also means that the following patches verification is
> successful.
> coresight: add support to enable more coresight paths
> coresight: funnel: add support for multiple output ports
>
> Signed-off-by: Tao Zhang <quic_taozha(a)quicinc.com>
> ---
> arch/arm64/boot/dts/qcom/qrb5165-rb5.dts | 439 +++++++++++++++++++++++
> 1 file changed, 439 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/qcom/qrb5165-rb5.dts b/arch/arm64/boot/dts/qcom/qrb5165-rb5.dts
> index 8ac96f8e79d4..bcec8b181e11 100644
> --- a/arch/arm64/boot/dts/qcom/qrb5165-rb5.dts
> +++ b/arch/arm64/boot/dts/qcom/qrb5165-rb5.dts
> @@ -222,6 +222,445 @@
> +
> + funnel@6b04000 {
> + compatible = "arm,coresight-dynamic-funnel", "arm,primecell";
> + arm,primecell-periphid = <0x000bb908>;
> +
> + reg = <0 0x6b04000 0 0x1000>;
> + reg-names = "funnel-base";
> +
> + clocks = <&aoss_qmp>;
> + clock-names = "apb_pclk";
> +
> + out-ports {
> + port {
> + merge_funnel_out: endpoint {
> + remote-endpoint =
> + <&etf_in>;
> + };
> + };
> + };
> +
> + in-ports {
> + #address-cells = <1>;
> + #size-cells = <0>;
> +
> + port@7 {
> + reg = <7>;
> + swao_funnel_in7: endpoint {
> + slave-mode;
This is obsolete, with the new in-ports/out-ports construct.
Suzuki
Hi Leo,
Would you be ok with the current patch the way it is? In case it's of
any help, I'm sharing the testing steps that James and I went through
when testing this internally, if you want to add to it
- Test that only a portion of the buffer is saved until there is a wraparound
$ ./perf record -vvv -e arm_spe/period=148576/u -S -- taskset --cpu-list 0 stress --cpu 1 & while true; do sleep 0.2; killall -s USR2 perf; done
- Test snapshot mode in CPU mode
$ sudo ./perf record -vvv -C 0 -e arm_spe/period=148576/u -S -- taskset --cpu-list 0 stress --cpu 1 &
- Test that auxtrace buffers correspond to an aux record
- Test snapshot default sizes in sudo and user modes
- Test small snapshot size
$ ./perf record -vvv -e arm_spe/period=148576/u -S1000 -m16,16 -- taskset --cpu-list 0 stress --cpu 1 &
If there are any concerns with the patches, please let me know and I
will try to address them.
Thanks,
German
On 13/10/2021 08:51, Will Deacon wrote:
> On Wed, Oct 13, 2021 at 08:39:16AM +0800, Leo Yan wrote:
>> On Mon, Oct 11, 2021 at 04:55:37PM +0100, German Gomez wrote:
>>> On 06/10/2021 10:51, Leo Yan wrote:
>>>> On Wed, Oct 06, 2021 at 10:35:20AM +0100, German Gomez wrote:
>>>>
>>>> [...]
>>>>
>>>>>> So simply say, I think the head pointer monotonically increasing is
>>>>>> the right thing to do in Arm SPE driver.
>>>>> I will talk to James about how we can proceed on this.
>>>> Thanks!
>>> I took this offline with James and, though it looks possible to patch
>>> the SPE driver to have a monotonically increasing head pointer in order
>>> to simplify the handling in the perf tool, it could be a breaking change
>>> for users of the perf_event_open syscall that currently rely on the way
>>> it works now.
>> Here I cannot create the connection between AUX head pointer and the
>> breakage of calling perf_event_open().
>>
>> Could you elaborate what's the reason the monotonical increasing head
>> pointer will lead to the breakage for perf_event_open()?
> It's a user-visible change in behaviour, isn't it? Therefore we risk
> breaking applications that rely on the current behaviour if we change it
> unconditionally.
>
> Given that the driver has always worked like this and it doesn't sound like
> it's the end of the world to deal with it in userspace (after all, it's
> aligned with intel-pt), then I don't think we should change it.
>
> Will