Re: [PATCH 4/8] gdbserver/linux-aarch64: Factor out function to get aarch64_features

[Thiago Jung Bauermann <thiago.bauermann@linaro.org>]

Luis Machado <luis.machado@arm.com> writes:

> On 9/8/22 07:41, Thiago Jung Bauermann via Gdb-patches wrote:
>> It will be used in a subsequent commit.  There's no functional change.
>> ---
>>   gdbserver/linux-aarch64-low.cc | 45 ++++++++++++++++++++++------------
>>   1 file changed, 29 insertions(+), 16 deletions(-)
>> diff --git a/gdbserver/linux-aarch64-low.cc b/gdbserver/linux-aarch64-low.cc
>> index 576925838f49..9b57be73818e 100644
>> --- a/gdbserver/linux-aarch64-low.cc
>> +++ b/gdbserver/linux-aarch64-low.cc
>> @@ -652,6 +652,28 @@ aarch64_target::low_delete_process (arch_process_info *info)
>>     xfree (info);
>>   }
>>   +/* Matches HWCAP_PACA in kernel header arch/arm64/include/uapi/asm/hwcap.h.  */
>> +#define AARCH64_HWCAP_PACA (1 << 30)
>> +
>> +static gdb::optional<struct aarch64_features>
>> +aarch64_get_arch_features (const thread_info *thread)
>> +{
>> +  struct aarch64_features features;
>> +  int vq = aarch64_sve_get_vq (thread->id.lwp ());
>> +
>> +  if (vq < 0)
>> +    return {};
>
> If we can't identify SVE properly, should we carry on and disable SVE
> support but enable all the other features we can find?

Yes, that's what happens in v2 now. aarch64_sve_get_vq isn't changed by
the patch series anymore so it continues to return vq = 0 if ptrace
fails. Because of that, in v2 this function now returns an
aarch64_features struct directly.

>> +
>> +  features.vq = vq;
>> +  /* A-profile PAC is 64-bit only.  */
>> +  features.pauth = linux_get_hwcap (thread, 8) & AARCH64_HWCAP_PACA;
>> +  /* A-profile MTE is 64-bit only.  */
>> +  features.mte = linux_get_hwcap2 (thread, 8) & HWCAP2_MTE;
>> +  features.tls = true;
>> +
>> +  return features;
>> +}

-- 
Thiago