Re: [dpdk-dev] [PATCH v1] build: add more implementers' IDs and PNs for Arm platforms

[Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>]

Hi Herbert,

On Wed, Jan 31, 2018 at 03:39:19PM +0800, Herbert Guan wrote:
> 1) Add native PN option '-march=native' to allow automatic detection.
>    Set 'arm_force_native_march' to 'true' in config/arm/meson.build
>    to use native PN option.
> 2) Add implementer_pn option for part num selection in cross compile
> 3) Add known Arm cortex PN support
> 4) Add known implementers' IDs (use generic flags/archs by default)
> 5) Sync build options with config/common_armv8a_linuxapp
>
> Signed-off-by: Herbert Guan <herbert.guan@arm.com>
> ---
<snip>
> +
>  machine_args_generic = [
> -	['default', ['-march=armv8-a+crc+crypto']]]
> +	['default', ['-march=armv8-a']],

Any specific reason for this change?
Traditional make uses
MACHINE_CFLAGS += -march=armv8-a+crc+crypto
found at mk/machine/armv8a/rte.vars.mk

> +	['native', ['-march=native']],
> +	['0xd03', ['-mcpu=cortex-a53']],
> +	['0xd04', ['-mcpu=cortex-a35']],
> +	['0xd07', ['-mcpu=cortex-a57']],
> +	['0xd08', ['-mcpu=cortex-a72']],
> +	['0xd09', ['-mcpu=cortex-a73']],
> +	['0xd0a', ['-mcpu=cortex-a75']],
> +]
>  machine_args_cavium = [
>  	['default', ['-march=armv8-a+crc+crypto','-mcpu=thunderx']],
> +	['native', ['-march=native']],
>  	['0xa1', ['-mcpu=thunderxt88']],
>  	['0xa2', ['-mcpu=thunderxt81']],
>  	['0xa3', ['-mcpu=thunderxt83']]]
>
> -flags_generic = [[]]
> +flags_common_default = [
> +	# Accelarate rte_memcpy. Be sure to run unit test (memcpy_perf_autotest)
> +	# to determine the best threshold in code. Refer to notes in source file
> +	# (lib/librte_eal/common/include/arch/arm/rte_memcpy_64.h) for more info.
> +	['RTE_ARCH_ARM64_MEMCPY', false],
> +	#	['RTE_ARM64_MEMCPY_ALIGNED_THRESHOLD', 2048],
> +	#	['RTE_ARM64_MEMCPY_UNALIGNED_THRESHOLD', 512],
> +	# Leave below RTE_ARM64_MEMCPY_xxx options commented out, unless there're
> +	# strong reasons.
> +	#	['RTE_ARM64_MEMCPY_SKIP_GCC_VER_CHECK', false],
> +	#	['RTE_ARM64_MEMCPY_ALIGN_MASK', 0xF],
> +	#	['RTE_ARM64_MEMCPY_STRICT_ALIGN', false],
> +
> +	['RTE_LIBRTE_FM10K_PMD', false],
> +	['RTE_LIBRTE_SFC_EFX_PMD', false],
> +	['RTE_LIBRTE_AVP_PMD', false],
> +
> +	['RTE_SCHED_VECTOR', false],
> +]
> +
> +flags_generic = [
> +	['RTE_MACHINE', '"armv8a"'],
> +	['RTE_CACHE_LINE_SIZE', 128]]
>  flags_cavium = [
>  	['RTE_MACHINE', '"thunderx"'],
>  	['RTE_CACHE_LINE_SIZE', 128],
> @@ -22,8 +55,21 @@ flags_cavium = [
>  	['RTE_MAX_VFIO_GROUPS', 128],
>  	['RTE_RING_USE_C11_MEM_MODEL', false]]
>
> +## Arm implementer ID (ARM DDI 0487C.a, Section G7.2.106, Page G7-5321)
>  impl_generic = ['Generic armv8', flags_generic, machine_args_generic]
> +impl_0x41 = ['Arm', flags_generic, machine_args_generic]
> +impl_0x42 = ['Broadcom', flags_generic, machine_args_generic]
>  impl_0x43 = ['Cavium', flags_cavium, machine_args_cavium]
> +impl_0x44 = ['DEC', flags_generic, machine_args_generic]
> +impl_0x49 = ['Infineon', flags_generic, machine_args_generic]
> +impl_0x4d = ['Motorola', flags_generic, machine_args_generic]
> +impl_0x4e = ['NVIDIA', flags_generic, machine_args_generic]
> +impl_0x50 = ['AppliedMicro', flags_generic, machine_args_generic]
> +impl_0x51 = ['Qualcomm', flags_generic, machine_args_generic]
> +impl_0x53 = ['Samsung', flags_generic, machine_args_generic]
> +impl_0x56 = ['Marvell', flags_generic, machine_args_generic]
> +impl_0x69 = ['Intel', flags_generic, machine_args_generic]
> +
>

One minor concern here is DPAA/DPAA2 use cacheline size og 64B unlike
traditional 128B armv8. found at config/defconfig_arm64-dpaa/2-linuxapp-gcc
maybe Hemanth could comment on this.

>  if cc.get_define('__clang__') != ''
>  	dpdk_conf.set_quoted('RTE_TOOLCHAIN', 'clang')
> @@ -55,19 +101,31 @@ else
>  				meson.current_source_dir(), 'armv8_machine.py'))
>  		cmd = run_command(detect_vendor.path())
>  		if cmd.returncode() == 0
> -			cmd_output = cmd.stdout().strip().split(' ')
> +			cmd_output = cmd.stdout().to_lower().strip().split(' ')
>  		endif

<snip>

Verified on thunderx with gcc 5.3.0/7.2.1 and clang 5.0.1

Regards,
Pavan

> @@ -79,22 +137,19 @@ else
>  	# for gcc versions > 7
>  	if cc.version().version_compare(
>  			'<7.0') or cmd_output.length() == 0
> -		foreach marg: machine[2]
> -			if marg[0] == 'default'
> -				foreach f: marg[1]
> -					machine_args += f
> -				endforeach
> -			endif
> -		endforeach
> -	else
> -		foreach marg: machine[2]
> -			if marg[0] == cmd_output[3]
> -				foreach f: marg[1]
> -					machine_args += f
> -				endforeach
> -			endif
> -		endforeach
> +		if not meson.is_cross_build() and arm_force_native_march == true
> +			impl_pn = 'native'
> +		else
> +			impl_pn = 'default'
> +		endif
>  	endif
> +	foreach marg: machine[2]
> +		if marg[0] == impl_pn
> +			foreach f: marg[1]
> +				machine_args += f
> +			endforeach
> +		endif
> +	endforeach
>  endif
>  message(machine_args)
>
> --
> 1.8.3.1
>