* [PATCH 0/3] Initial support for Power
@ 2023-06-07 15:06 Shawn Anastasio
2023-06-07 15:06 ` [PATCH 1/3] xen: Add files needed for minimal Power build Shawn Anastasio
` (3 more replies)
0 siblings, 4 replies; 27+ messages in thread
From: Shawn Anastasio @ 2023-06-07 15:06 UTC (permalink / raw)
To: xen-devel
Cc: tpearson, Shawn Anastasio, Andrew Cooper, George Dunlap,
Jan Beulich, Julien Grall, Stefano Stabellini, Wei Liu
Hello all,
This patch series adds support for building a minimal image
(head.o-only) for Power ISA 2.07B+ (POWER8+) systems. The first patch
boots to an infinite loop and the second adds early serial console
support on pseries VMs, with bare metal support planned next.
Since Xen previously had support for a much older version of the ISA in
version 3.2.3, we were able to carry over some headers and support
routines from that version. Unlike that initial port though, this effort
focuses solely on POWER8+ CPUs that are capable of running in Little
Endian mode.
With an appropriate powerpc64le-linux-gnu cross-toolchain, the minimal
image can be built with:
$ make XEN_TARGET_ARCH=ppc64 -C xen openpower_defconfig
$ make XEN_TARGET_ARCH=ppc64 SUBSYSTEMS=xen -C xen TARGET=ppc64/head.o
The resulting head.o can then be booted in a standard QEMU/pseries VM:
$ qemu-system-ppc64 -M pseries-5.2 -m 256M -kernel xen/ppc64/head.o \
-vga none -serial mon:stdio -nographic
Thanks,
Shawn
Shawn Anastasio (3):
xen: Add files needed for minimal Power build
xen/ppc: Implement early serial printk on PaPR/pseries
maintainers: Add PPC64 maintainer
MAINTAINERS | 3 +
config/ppc64.mk | 5 +
xen/Makefile | 5 +-
xen/arch/ppc/Kconfig | 42 +++++
xen/arch/ppc/Kconfig.debug | 5 +
xen/arch/ppc/Makefile | 18 ++
xen/arch/ppc/Rules.mk | 0
xen/arch/ppc/arch.mk | 11 ++
xen/arch/ppc/boot_of.c | 122 +++++++++++++
xen/arch/ppc/configs/openpower_defconfig | 14 ++
xen/arch/ppc/early_printk.c | 36 ++++
xen/arch/ppc/include/asm/boot.h | 31 ++++
xen/arch/ppc/include/asm/bug.h | 6 +
xen/arch/ppc/include/asm/byteorder.h | 74 ++++++++
xen/arch/ppc/include/asm/cache.h | 6 +
xen/arch/ppc/include/asm/config.h | 66 +++++++
xen/arch/ppc/include/asm/early_printk.h | 14 ++
xen/arch/ppc/include/asm/msr.h | 67 +++++++
xen/arch/ppc/include/asm/page-bits.h | 7 +
xen/arch/ppc/include/asm/processor.h | 223 +++++++++++++++++++++++
xen/arch/ppc/include/asm/reg_defs.h | 218 ++++++++++++++++++++++
xen/arch/ppc/include/asm/string.h | 6 +
xen/arch/ppc/include/asm/types.h | 64 +++++++
xen/arch/ppc/ppc64/Makefile | 1 +
xen/arch/ppc/ppc64/asm-offsets.c | 55 ++++++
xen/arch/ppc/ppc64/head.S | 126 +++++++++++++
xen/arch/ppc/setup.c | 32 ++++
xen/arch/ppc/xen.lds.S | 173 ++++++++++++++++++
28 files changed, 1428 insertions(+), 2 deletions(-)
create mode 100644 config/ppc64.mk
create mode 100644 xen/arch/ppc/Kconfig
create mode 100644 xen/arch/ppc/Kconfig.debug
create mode 100644 xen/arch/ppc/Makefile
create mode 100644 xen/arch/ppc/Rules.mk
create mode 100644 xen/arch/ppc/arch.mk
create mode 100644 xen/arch/ppc/boot_of.c
create mode 100644 xen/arch/ppc/configs/openpower_defconfig
create mode 100644 xen/arch/ppc/early_printk.c
create mode 100644 xen/arch/ppc/include/asm/boot.h
create mode 100644 xen/arch/ppc/include/asm/bug.h
create mode 100644 xen/arch/ppc/include/asm/byteorder.h
create mode 100644 xen/arch/ppc/include/asm/cache.h
create mode 100644 xen/arch/ppc/include/asm/config.h
create mode 100644 xen/arch/ppc/include/asm/early_printk.h
create mode 100644 xen/arch/ppc/include/asm/msr.h
create mode 100644 xen/arch/ppc/include/asm/page-bits.h
create mode 100644 xen/arch/ppc/include/asm/processor.h
create mode 100644 xen/arch/ppc/include/asm/reg_defs.h
create mode 100644 xen/arch/ppc/include/asm/string.h
create mode 100644 xen/arch/ppc/include/asm/types.h
create mode 100644 xen/arch/ppc/ppc64/Makefile
create mode 100644 xen/arch/ppc/ppc64/asm-offsets.c
create mode 100644 xen/arch/ppc/ppc64/head.S
create mode 100644 xen/arch/ppc/setup.c
create mode 100644 xen/arch/ppc/xen.lds.S
--
2.30.2
^ permalink raw reply [flat|nested] 27+ messages in thread* [PATCH 1/3] xen: Add files needed for minimal Power build 2023-06-07 15:06 [PATCH 0/3] Initial support for Power Shawn Anastasio @ 2023-06-07 15:06 ` Shawn Anastasio 2023-06-09 9:15 ` Jan Beulich 2023-06-07 15:06 ` [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries Shawn Anastasio ` (2 subsequent siblings) 3 siblings, 1 reply; 27+ messages in thread From: Shawn Anastasio @ 2023-06-07 15:06 UTC (permalink / raw) To: xen-devel Cc: tpearson, Shawn Anastasio, Andrew Cooper, George Dunlap, Jan Beulich, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio Add the build system changes required to build for ppc64le (POWER8+). Following in the footsteps of the initial riscv port, only building the head.o target, which boots to an infinite loop, is supported: $ make XEN_TARGET_ARCH=ppc64 -C xen openpower_defconfig $ make XEN_TARGET_ARCH=ppc64 SUBSYSTEMS=xen -C xen TARGET=ppc64/head.o This port targets POWER8+ CPUs running in Little Endian mode specifically, and does not boot on older machines. Additionally, this initial skeleton only implements the PaPR/pseries boot protocol which allows it to be booted in a standard QEMU virtual machine: $ qemu-system-ppc64 -M pseries-5.2 -m 256M -kernel ppc64/head.o Where possible, this patch uses header definitions and support routines from the original Xen PowerPC port present in Xen 3.2.3. Though we are targeting a much newer ISA than that original port did, some of the definitions have remained similar enough for reuse. Signed-off-by: Shawn Anastasio <shawnanastasio@raptorengineering.com> --- config/ppc64.mk | 5 + xen/Makefile | 5 +- xen/arch/ppc/Kconfig | 42 +++++ xen/arch/ppc/Kconfig.debug | 0 xen/arch/ppc/Makefile | 17 ++ xen/arch/ppc/Rules.mk | 0 xen/arch/ppc/arch.mk | 11 ++ xen/arch/ppc/configs/openpower_defconfig | 13 ++ xen/arch/ppc/include/asm/config.h | 63 +++++++ xen/arch/ppc/include/asm/msr.h | 67 +++++++ xen/arch/ppc/include/asm/page-bits.h | 7 + xen/arch/ppc/include/asm/processor.h | 173 ++++++++++++++++++ xen/arch/ppc/include/asm/reg_defs.h | 218 +++++++++++++++++++++++ xen/arch/ppc/ppc64/Makefile | 1 + xen/arch/ppc/ppc64/asm-offsets.c | 0 xen/arch/ppc/ppc64/head.S | 67 +++++++ xen/arch/ppc/setup.c | 14 ++ xen/arch/ppc/xen.lds.S | 173 ++++++++++++++++++ 18 files changed, 874 insertions(+), 2 deletions(-) create mode 100644 config/ppc64.mk create mode 100644 xen/arch/ppc/Kconfig create mode 100644 xen/arch/ppc/Kconfig.debug create mode 100644 xen/arch/ppc/Makefile create mode 100644 xen/arch/ppc/Rules.mk create mode 100644 xen/arch/ppc/arch.mk create mode 100644 xen/arch/ppc/configs/openpower_defconfig create mode 100644 xen/arch/ppc/include/asm/config.h create mode 100644 xen/arch/ppc/include/asm/msr.h create mode 100644 xen/arch/ppc/include/asm/page-bits.h create mode 100644 xen/arch/ppc/include/asm/processor.h create mode 100644 xen/arch/ppc/include/asm/reg_defs.h create mode 100644 xen/arch/ppc/ppc64/Makefile create mode 100644 xen/arch/ppc/ppc64/asm-offsets.c create mode 100644 xen/arch/ppc/ppc64/head.S create mode 100644 xen/arch/ppc/setup.c create mode 100644 xen/arch/ppc/xen.lds.S diff --git a/config/ppc64.mk b/config/ppc64.mk new file mode 100644 index 0000000000..a46de35cf1 --- /dev/null +++ b/config/ppc64.mk @@ -0,0 +1,5 @@ +CONFIG_PPC64 := y +CONFIG_PPC64_64 := y +CONFIG_PPC64_$(XEN_OS) := y + +CONFIG_XEN_INSTALL_SUFFIX := diff --git a/xen/Makefile b/xen/Makefile index e89fc461fc..db5454fb58 100644 --- a/xen/Makefile +++ b/xen/Makefile @@ -38,7 +38,7 @@ EFI_MOUNTPOINT ?= $(BOOT_DIR)/efi ARCH=$(XEN_TARGET_ARCH) SRCARCH=$(shell echo $(ARCH) | \ sed -e 's/x86.*/x86/' -e s'/arm\(32\|64\)/arm/g' \ - -e s'/riscv.*/riscv/g') + -e s'/riscv.*/riscv/g' -e s'/ppc.*/ppc/g') export ARCH SRCARCH # Allow someone to change their config file @@ -244,7 +244,7 @@ include $(XEN_ROOT)/Config.mk export TARGET_SUBARCH := $(XEN_TARGET_ARCH) export TARGET_ARCH := $(shell echo $(XEN_TARGET_ARCH) | \ sed -e 's/x86.*/x86/' -e s'/arm\(32\|64\)/arm/g' \ - -e s'/riscv.*/riscv/g') + -e s'/riscv.*/riscv/g' -e s'/ppc.*/ppc/g') export CONFIG_SHELL := $(SHELL) export CC CXX LD NM OBJCOPY OBJDUMP ADDR2LINE @@ -563,6 +563,7 @@ _clean: $(Q)$(MAKE) $(clean)=xsm $(Q)$(MAKE) $(clean)=crypto $(Q)$(MAKE) $(clean)=arch/arm + $(Q)$(MAKE) $(clean)=arch/ppc $(Q)$(MAKE) $(clean)=arch/riscv $(Q)$(MAKE) $(clean)=arch/x86 $(Q)$(MAKE) $(clean)=test diff --git a/xen/arch/ppc/Kconfig b/xen/arch/ppc/Kconfig new file mode 100644 index 0000000000..e80a5fb036 --- /dev/null +++ b/xen/arch/ppc/Kconfig @@ -0,0 +1,42 @@ +config PPC + def_bool y + +config PPC64 + def_bool y + select 64BIT + +config ARCH_DEFCONFIG + string + default "arch/ppc/configs/openpower_defconfig" + +menu "Architecture Features" + +source "arch/Kconfig" + +endmenu + +menu "ISA Selection" + +choice + prompt "Base ISA" + default POWER_ISA_2_07B if PPC64 + help + This selects the base ISA version that Xen will target. + +config POWER_ISA_2_07B + bool "POWER ISA 2.07B+" + help + Target version 2.07B+ of the POWER ISA (POWER8+) + +config POWER_ISA_3_00 + bool "POWER ISA 3.00+" + help + Target version 3.00+ of the POWER ISA (POWER9+) + +endchoice + +endmenu + +source "common/Kconfig" + +source "drivers/Kconfig" diff --git a/xen/arch/ppc/Kconfig.debug b/xen/arch/ppc/Kconfig.debug new file mode 100644 index 0000000000..e69de29bb2 diff --git a/xen/arch/ppc/Makefile b/xen/arch/ppc/Makefile new file mode 100644 index 0000000000..b3ad837d4b --- /dev/null +++ b/xen/arch/ppc/Makefile @@ -0,0 +1,17 @@ +obj-$(CONFIG_PPC64) += ppc64/ +obj-y += setup.o + +$(TARGET): $(TARGET)-syms + cp -f $< $@ + +$(TARGET)-syms: $(objtree)/prelink.o $(obj)/xen.lds + $(LD) $(XEN_LDFLAGS) -T $(obj)/xen.lds -N $< $(build_id_linker) -o $@ + $(NM) -pa --format=sysv $(@D)/$(@F) \ + | $(objtree)/tools/symbols --all-symbols --xensyms --sysv --sort \ + >$(@D)/$(@F).map + +$(obj)/xen.lds: $(src)/xen.lds.S FORCE + $(call if_changed_dep,cpp_lds_S) + +.PHONY: include +include: diff --git a/xen/arch/ppc/Rules.mk b/xen/arch/ppc/Rules.mk new file mode 100644 index 0000000000..e69de29bb2 diff --git a/xen/arch/ppc/arch.mk b/xen/arch/ppc/arch.mk new file mode 100644 index 0000000000..c185a5028a --- /dev/null +++ b/xen/arch/ppc/arch.mk @@ -0,0 +1,11 @@ +######################################## +# Power-specific definitions + +ppc-march-$(CONFIG_POWER_ISA_2_07B) := power8 +ppc-march-$(CONFIG_POWER_ISA_3_00) := power9 + +CFLAGS += -mcpu=$(ppc-march-y) -mstrict-align -mcmodel=large -mabi=elfv2 -mno-altivec -mno-vsx + +# TODO: Drop override when more of the build is working +override ALL_OBJS-y = arch/$(TARGET_ARCH)/built_in.o +override ALL_LIBS-y = diff --git a/xen/arch/ppc/configs/openpower_defconfig b/xen/arch/ppc/configs/openpower_defconfig new file mode 100644 index 0000000000..8783eb3488 --- /dev/null +++ b/xen/arch/ppc/configs/openpower_defconfig @@ -0,0 +1,13 @@ +# CONFIG_SCHED_CREDIT is not set +# CONFIG_SCHED_RTDS is not set +# CONFIG_SCHED_NULL is not set +# CONFIG_SCHED_ARINC653 is not set +# CONFIG_TRACEBUFFER is not set +# CONFIG_HYPFS is not set +# CONFIG_GRANT_TABLE is not set +# CONFIG_SPECULATIVE_HARDEN_ARRAY is not set + +CONFIG_PPC64=y +CONFIG_DEBUG=y +CONFIG_DEBUG_INFO=y +CONFIG_EXPERT=y diff --git a/xen/arch/ppc/include/asm/config.h b/xen/arch/ppc/include/asm/config.h new file mode 100644 index 0000000000..7a2862ef7a --- /dev/null +++ b/xen/arch/ppc/include/asm/config.h @@ -0,0 +1,63 @@ +#ifndef __PPC_CONFIG_H__ +#define __PPC_CONFIG_H__ + +#include <xen/const.h> +#include <xen/page-size.h> + +#if defined(CONFIG_PPC64) +#define LONG_BYTEORDER 3 +#define ELFSIZE 64 +#define MAX_VIRT_CPUS 1024u +#else +#error "Unsupported PowerPC variant" +#endif + +#define BYTES_PER_LONG (1 << LONG_BYTEORDER) +#define BITS_PER_LONG (BYTES_PER_LONG << 3) +#define POINTER_ALIGN BYTES_PER_LONG + +#define BITS_PER_LLONG 64 + +/* xen_ulong_t is always 64 bits */ +#define BITS_PER_XEN_ULONG 64 + +#define CONFIG_PPC_L1_CACHE_SHIFT 7 +#define CONFIG_PAGEALLOC_MAX_ORDER 18 +#define CONFIG_DOMU_MAX_ORDER 9 +#define CONFIG_HWDOM_MAX_ORDER 10 + +#define OPT_CONSOLE_STR "dtuart" +#define INVALID_VCPU_ID MAX_VIRT_CPUS + +/* Linkage for PPC */ +#ifdef __ASSEMBLY__ +#define ALIGN .align 2 + +#define ENTRY(name) \ + .globl name; \ + ALIGN; \ + name: +#endif + +#define XEN_VIRT_START _AT(UL, 0x400000) + +#define SMP_CACHE_BYTES (1 << 6) + +#define STACK_ORDER 2 +#define STACK_SIZE (PAGE_SIZE << STACK_ORDER) + +/* 288 bytes below the stack pointer must be preserved by interrupt handlers */ +#define STACK_VOLATILE_AREA 288 + +/* size of minimum stack frame; C code can write into the caller's stack */ +#define STACK_FRAME_OVERHEAD 32 + +#endif /* __PPC_CONFIG_H__ */ +/* + * Local variables: + * mode: C + * c-file-style: "BSD" + * c-basic-offset: 4 + * indent-tabs-mode: nil + * End: + */ diff --git a/xen/arch/ppc/include/asm/msr.h b/xen/arch/ppc/include/asm/msr.h new file mode 100644 index 0000000000..7ef0b5e3ad --- /dev/null +++ b/xen/arch/ppc/include/asm/msr.h @@ -0,0 +1,67 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright (C) IBM Corp. 2005 + * + * Authors: Jimi Xenidis <jimix@watson.ibm.com> + */ + +#ifndef _POWERPC_MSR_H +#define _POWERPC_MSR_H + +#ifdef __ASSEMBLY__ +#define ULL(x) x +#else +#define ULL(x) x ## ULL +#endif + +/* Flags in MSR: */ +#define MSR_SF ULL(0x8000000000000000) +#define MSR_TA ULL(0x4000000000000000) +#define MSR_ISF ULL(0x2000000000000000) +#define MSR_HV ULL(0x1000000000000000) +#define MSR_VMX ULL(0x0000000002000000) +#define MSR_MER ULL(0x0000000000200000) +#define MSR_POW ULL(0x0000000000040000) +#define MSR_ILE ULL(0x0000000000010000) +#define MSR_EE ULL(0x0000000000008000) +#define MSR_PR ULL(0x0000000000004000) +#define MSR_FP ULL(0x0000000000002000) +#define MSR_ME ULL(0x0000000000001000) +#define MSR_FE0 ULL(0x0000000000000800) +#define MSR_SE ULL(0x0000000000000400) +#define MSR_BE ULL(0x0000000000000200) +#define MSR_FE1 ULL(0x0000000000000100) +#define MSR_IP ULL(0x0000000000000040) +#define MSR_IR ULL(0x0000000000000020) +#define MSR_DR ULL(0x0000000000000010) +#define MSR_PMM ULL(0x0000000000000004) +#define MSR_RI ULL(0x0000000000000002) +#define MSR_LE ULL(0x0000000000000001) + +/* MSR bits set on the systemsim simulator */ +#define MSR_SIM ULL(0x0000000020000000) +#define MSR_SYSTEMSIM ULL(0x0000000010000000) + +/* On a trap, srr1's copy of msr defines some bits as follows: */ +#define MSR_TRAP_FE ULL(0x0000000000100000) /* Floating Point Exception */ +#define MSR_TRAP_IOP ULL(0x0000000000080000) /* Illegal Instruction */ +#define MSR_TRAP_PRIV ULL(0x0000000000040000) /* Privileged Instruction */ +#define MSR_TRAP ULL(0x0000000000020000) /* Trap Instruction */ +#define MSR_TRAP_NEXT ULL(0x0000000000010000) /* PC is next instruction */ +#define MSR_TRAP_BITS (MSR_TRAP_FE|MSR_TRAP_IOP|MSR_TRAP_PRIV|MSR_TRAP) + +#endif /* _POWERPC_MSR_H */ diff --git a/xen/arch/ppc/include/asm/page-bits.h b/xen/arch/ppc/include/asm/page-bits.h new file mode 100644 index 0000000000..4c01bf9716 --- /dev/null +++ b/xen/arch/ppc/include/asm/page-bits.h @@ -0,0 +1,7 @@ +#ifndef __PPC_PAGE_BITS_H__ +#define __PPC_PAGE_BITS_H__ + +#define PAGE_SHIFT 16 /* 64 KiB Pages */ +#define PADDR_BITS 48 + +#endif /* __PPC_PAGE_BITS_H__ */ diff --git a/xen/arch/ppc/include/asm/processor.h b/xen/arch/ppc/include/asm/processor.h new file mode 100644 index 0000000000..0ab7bfc9df --- /dev/null +++ b/xen/arch/ppc/include/asm/processor.h @@ -0,0 +1,173 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright IBM Corp. 2005, 2006, 2007 + * Copyright Raptor Engineering, LLC + * + * Authors: Hollis Blanchard <hollisb@us.ibm.com> + * Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com> + * Timothy Pearson <tpearson@raptorengineering.com> + * Shawn Anastasio <sanastasio@raptorengineering.com> + */ + +#ifndef _ASM_PROCESSOR_H_ +#define _ASM_PROCESSOR_H_ + +#include <xen/config.h> +#include <asm/reg_defs.h> +#include <asm/msr.h> + +#define IOBMP_BYTES 8192 +#define IOBMP_INVALID_OFFSET 0x8000 + +/* most assembler do not know this instruction */ +#define HRFID .long 0x4c000224 + +/* Processor Version Register (PVR) field extraction */ + +#define PVR_VER(pvr) (((pvr) >> 16) & 0xFFFF) /* Version field */ +#define PVR_REV(pvr) (((pvr) >> 0) & 0xFFFF) /* Revison field */ + +#define __is_processor(pv) (PVR_VER(mfspr(SPRN_PVR)) == (pv)) + +/* + * IBM has further subdivided the standard PowerPC 16-bit version and + * revision subfields of the PVR for the PowerPC 403s into the following: + */ + +#define PVR_FAM(pvr) (((pvr) >> 20) & 0xFFF) /* Family field */ +#define PVR_MEM(pvr) (((pvr) >> 16) & 0xF) /* Member field */ +#define PVR_CORE(pvr) (((pvr) >> 12) & 0xF) /* Core field */ +#define PVR_CFG(pvr) (((pvr) >> 8) & 0xF) /* Configuration field */ +#define PVR_MAJ(pvr) (((pvr) >> 4) & 0xF) /* Major revision field */ +#define PVR_MIN(pvr) (((pvr) >> 0) & 0xF) /* Minor revision field */ + +/* Processor Version Numbers */ + +#define PVR_403GA 0x00200000 +#define PVR_403GB 0x00200100 +#define PVR_403GC 0x00200200 +#define PVR_403GCX 0x00201400 +#define PVR_405GP 0x40110000 +#define PVR_STB03XXX 0x40310000 +#define PVR_NP405H 0x41410000 +#define PVR_NP405L 0x41610000 +#define PVR_601 0x00010000 +#define PVR_602 0x00050000 +#define PVR_603 0x00030000 +#define PVR_603e 0x00060000 +#define PVR_603ev 0x00070000 +#define PVR_603r 0x00071000 +#define PVR_604 0x00040000 +#define PVR_604e 0x00090000 +#define PVR_604r 0x000A0000 +#define PVR_620 0x00140000 +#define PVR_740 0x00080000 +#define PVR_750 PVR_740 +#define PVR_740P 0x10080000 +#define PVR_750P PVR_740P +#define PVR_7400 0x000C0000 +#define PVR_7410 0x800C0000 +#define PVR_7450 0x80000000 +#define PVR_8540 0x80200000 +#define PVR_8560 0x80200000 +/* + * For the 8xx processors, all of them report the same PVR family for + * the PowerPC core. The various versions of these processors must be + * differentiated by the version number in the Communication Processor + * Module (CPM). + */ +#define PVR_821 0x00500000 +#define PVR_823 PVR_821 +#define PVR_850 PVR_821 +#define PVR_860 PVR_821 +#define PVR_8240 0x00810100 +#define PVR_8245 0x80811014 +#define PVR_8260 PVR_8240 + +/* 64-bit processors */ +#define PVR_NORTHSTAR 0x0033 +#define PVR_PULSAR 0x0034 +#define PVR_POWER4 0x0035 +#define PVR_ICESTAR 0x0036 +#define PVR_SSTAR 0x0037 +#define PVR_POWER4p 0x0038 +#define PVR_970 0x0039 +#define PVR_POWER5 0x003A +#define PVR_POWER5p 0x003B +#define PVR_970FX 0x003C +#define PVR_630 0x0040 +#define PVR_630p 0x0041 +#define PVR_970MP 0x0044 +#define PVR_BE 0x0070 + +#ifdef __ASSEMBLY__ + +#define LOADADDR(rn, name) \ + lis rn, name##@highest; \ + ori rn, rn, name##@higher; \ + rldicr rn, rn, 32, 31; \ + oris rn, rn, name##@h; \ + ori rn, rn, name##@l + +#define SET_REG_TO_CONST(reg, value) \ + lis reg, (((value) >> 48) & 0xFFFF); \ + ori reg, reg, (((value) >> 32) & 0xFFFF); \ + rldicr reg, reg, 32, 31; \ + oris reg, reg, (((value) >> 16) & 0xFFFF); \ + ori reg, reg, ((value) &0xFFFF); + +#define SET_REG_TO_LABEL(reg, label) \ + lis reg, (label) @highest; \ + ori reg, reg, (label) @higher; \ + rldicr reg, reg, 32, 31; \ + oris reg, reg, (label) @h; \ + ori reg, reg, (label) @l; + +/* Taken from Linux kernel source (arch/powerpc/boot/ppc_asm.h) */ +#define FIXUP_ENDIAN \ + tdi 0, 0, 0x48; /* Reverse endian of b . + 8 */ \ + b $ + 44; /* Skip trampoline if endian is good */ \ + .long 0xa600607d; /* mfmsr r11 */ \ + .long 0x01006b69; /* xori r11,r11,1 */ \ + .long 0x00004039; /* li r10,0 */ \ + .long 0x6401417d; /* mtmsrd r10,1 */ \ + .long 0x05009f42; /* bcl 20,31,$+4 */ \ + .long 0xa602487d; /* mflr r10 */ \ + .long 0x14004a39; /* addi r10,r10,20 */ \ + .long 0xa6035a7d; /* mtsrr0 r10 */ \ + .long 0xa6037b7d; /* mtsrr1 r11 */ \ + .long 0x2400004c /* rfid */ + +#define _GLOBAL(name) \ + .section ".text"; \ + .align 2; \ + .globl name; \ + .type name, @function; \ + name: + +#define _STATIC(name) \ + .section ".text"; \ + .align 2; \ + .type name, @function; \ + name: + +#define _ENTRY(name) name + +#else /* __ASSEMBLY__ */ +#endif /* __ASSEMBLY__ */ + +#endif diff --git a/xen/arch/ppc/include/asm/reg_defs.h b/xen/arch/ppc/include/asm/reg_defs.h new file mode 100644 index 0000000000..ac435517cc --- /dev/null +++ b/xen/arch/ppc/include/asm/reg_defs.h @@ -0,0 +1,218 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright IBM Corp. 2005, 2007 + * + * Authors: Jimi Xenidis <jimix@watson.ibm.com> + * Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com> + */ + +#ifndef _ASM_REG_DEFS_H_ +#define _ASM_REG_DEFS_H_ + +#ifdef __ASSEMBLY__ + +#define UL(x) x + +/* Condition Register Bit Fields */ + +#define cr0 0 +#define cr1 1 +#define cr2 2 +#define cr3 3 +#define cr4 4 +#define cr5 5 +#define cr6 6 +#define cr7 7 + + +/* General Purpose Registers (GPRs) */ + +#define r0 0 +#define r1 1 +#define r2 2 +#define r3 3 +#define r4 4 +#define r5 5 +#define r6 6 +#define r7 7 +#define r8 8 +#define r9 9 +#define r10 10 +#define r11 11 +#define r12 12 +#define r13 13 +#define r14 14 +#define r15 15 +#define r16 16 +#define r17 17 +#define r18 18 +#define r19 19 +#define r20 20 +#define r21 21 +#define r22 22 +#define r23 23 +#define r24 24 +#define r25 25 +#define r26 26 +#define r27 27 +#define r28 28 +#define r29 29 +#define r30 30 +#define r31 31 + +/* Floating Point Registers (FPRs) */ +#define fr0 0 +#define fr1 1 +#define fr2 2 +#define fr3 3 +#define fr4 4 +#define fr5 5 +#define fr6 6 +#define fr7 7 +#define fr8 8 +#define fr9 9 +#define fr10 10 +#define fr11 11 +#define fr12 12 +#define fr13 13 +#define fr14 14 +#define fr15 15 +#define fr16 16 +#define fr17 17 +#define fr18 18 +#define fr19 19 +#define fr20 20 +#define fr21 21 +#define fr22 22 +#define fr23 23 +#define fr24 24 +#define fr25 25 +#define fr26 26 +#define fr27 27 +#define fr28 28 +#define fr29 29 +#define fr30 30 +#define fr31 31 + +/* Vector Registers (FPRs) */ +#define vr0 0 +#define vr1 1 +#define vr2 2 +#define vr3 3 +#define vr4 4 +#define vr5 5 +#define vr6 6 +#define vr7 7 +#define vr8 8 +#define vr9 9 +#define vr10 10 +#define vr11 11 +#define vr12 12 +#define vr13 13 +#define vr14 14 +#define vr15 15 +#define vr16 16 +#define vr17 17 +#define vr18 18 +#define vr19 19 +#define vr20 20 +#define vr21 21 +#define vr22 22 +#define vr23 23 +#define vr24 24 +#define vr25 25 +#define vr26 26 +#define vr27 27 +#define vr28 28 +#define vr29 29 +#define vr30 30 +#define vr31 31 + +#else /* defined(__ASSEMBLY__) */ + +#define UL(x) x ## UL + +#endif + +/* Special Purpose Registers */ +#define SPRN_VRSAVE 256 +#define SPRN_DSISR 18 +#define SPRN_DAR 19 +#define SPRN_DEC 22 +#define SPRN_SRR0 26 +#define SPRN_SRR1 27 +#define SPRN_TBRL 268 +#define SPRN_TBRU 269 +#define SPRN_SPRG0 272 +#define SPRN_SPRG1 273 +#define SPRN_SPRG2 274 +#define SPRN_SPRG3 275 +#define SPRN_TBWL 284 +#define SPRN_TBWU 285 + +#define SPRN_HSPRG0 304 +#define SPRN_HSPRG1 305 +#define SPRN_HDEC 310 +#define SPRN_HIOR 311 +#define SPRN_RMOR 312 +#define SPRN_HRMOR 313 +#define SPRN_HSRR0 314 +#define SPRN_HSRR1 315 +#define SPRN_LPCR 318 +#define SPRN_LPIDR 319 + +/* Performance monitor spr encodings */ +#define SPRN_MMCRA 786 +#define MMCRA_SAMPHV UL(0x10000000) /* state of MSR HV when SIAR set */ +#define MMCRA_SAMPPR UL(0x08000000) /* state of MSR PR when SIAR set */ +#define MMCRA_SAMPLE_ENABLE UL(0x00000001) /* enable sampling */ +#define NUM_PMCS 8 +#define SPRN_PMC1 787 +#define SPRN_PMC2 788 +#define SPRN_PMC3 789 +#define SPRN_PMC4 790 +#define SPRN_PMC5 791 +#define SPRN_PMC6 792 +#define SPRN_PMC7 793 +#define SPRN_PMC8 794 +#define SPRN_MMCR0 795 +#define MMCR0_FC UL(0x80000000) /* freeze counters */ +#define MMCR0_FCS UL(0x40000000) /* freeze in supervisor state */ +#define MMCR0_FCP UL(0x20000000) /* freeze in problem state */ +#define MMCR0_FCM1 UL(0x10000000) /* freeze counters while MSR mark = 1 */ +#define MMCR0_FCM0 UL(0x08000000) /* freeze counters while MSR mark = 0 */ +#define MMCR0_PMAE UL(0x04000000) /* performance monitor alert enabled */ +#define MMCR0_PMAO UL(0x00000080) /* performance monitor alert occurred */ +#define MMCR0_FCH UL(0x00000001) /* freeze conditions in hypervisor */ +#define SPRN_SIAR 796 +#define SPRN_SDAR 797 +#define SPRN_MMCR1 798 + +/* As defined for PU G4 */ +#define SPRN_HID0 1008 +#define SPRN_HID1 1009 +#define SPRN_HID4 1012 + +#define SPRN_DABR 1013 +#define SPRN_HID5 1014 +#define SPRN_DABRX 1015 +#define SPRN_HID6 1017 +#define SPRN_HID7 1018 +#define SPRN_HID8 1019 +#define SPRN_PIR 1023 + +#endif /* _ASM_REG_DEFS_H_ */ diff --git a/xen/arch/ppc/ppc64/Makefile b/xen/arch/ppc/ppc64/Makefile new file mode 100644 index 0000000000..24aaeb84df --- /dev/null +++ b/xen/arch/ppc/ppc64/Makefile @@ -0,0 +1 @@ +obj-y += head.o \ No newline at end of file diff --git a/xen/arch/ppc/ppc64/asm-offsets.c b/xen/arch/ppc/ppc64/asm-offsets.c new file mode 100644 index 0000000000..e69de29bb2 diff --git a/xen/arch/ppc/ppc64/head.S b/xen/arch/ppc/ppc64/head.S new file mode 100644 index 0000000000..d91b0972ae --- /dev/null +++ b/xen/arch/ppc/ppc64/head.S @@ -0,0 +1,67 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Copyright (C) 2005, 2006 IBM Corp. + * Copyright (C) 2023 Raptor Engineering, LLC + * + * Authors: Jimi Xenidis <jimix@watson.ibm.com> + * Hollis Blanchard <hollisb@us.ibm.com> + * Timothy Pearson <tpearson@raptorengineering.com> + * Shawn Anastasio <sanastasio@raptorengineering.com> + */ + +#include <asm/config.h> +#include <asm/msr.h> +#include <asm/processor.h> +#include <asm/asm-offsets.h> + +.section .text.header, "ax", %progbits + +ENTRY(start) + /* NOTE: (OF only) r3-r7 contain data passed to kernel from fw */ + FIXUP_ENDIAN + + /* Set MSR_HV bit */ + mfmsr r8 + li r9, MSR_HV >> 48 + sldi r9, r9, 48 + or r8, r8, r9 + mtmsrd r8 + + /* set up the initial stack */ + SET_REG_TO_LABEL(r1, cpu0_boot_stack) + li r8, 0 + std r8, 0(r1) + + /* call the init function */ + LOADADDR(r12,start_xen) + mtctr r12 + bctrl + + /* should never return */ + trap + + /* Note! GDB 6.3 makes the very stupid assumption that PC > SP means we are + * in a Linux signal trampoline, and it begins groping for a struct + * rt_sigframe on the stack. Naturally, this fails miserably for our + * backtrace. To work around this behavior, we must make certain that our + * stack is always above our text, e.g. in the data section. */ + .data /* DO NOT REMOVE; see GDB note above */ + .align 4 +cpu0_boot_stack_bottom: + .space STACK_SIZE +cpu0_boot_stack: + .space STACK_FRAME_OVERHEAD diff --git a/xen/arch/ppc/setup.c b/xen/arch/ppc/setup.c new file mode 100644 index 0000000000..c9509fb6a6 --- /dev/null +++ b/xen/arch/ppc/setup.c @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +#include <xen/compile.h> +#include <xen/init.h> + +void __init noreturn start_xen(void) +{ + for ( ;; ) + { + // Set current hardware thread to very low priority + asm volatile("or 31,31,31"); + } + + unreachable(); +} diff --git a/xen/arch/ppc/xen.lds.S b/xen/arch/ppc/xen.lds.S new file mode 100644 index 0000000000..89fcdc9c00 --- /dev/null +++ b/xen/arch/ppc/xen.lds.S @@ -0,0 +1,173 @@ +#include <xen/xen.lds.h> + +#undef ENTRY +#undef ALIGN + +OUTPUT_ARCH(powerpc:common64) +ENTRY(start) + +PHDRS +{ + text PT_LOAD ; +#if defined(BUILD_ID) + note PT_NOTE ; +#endif +} + +/** + * OF's base load address is 0x400000 (XEN_VIRT_START). + * By defining sections this way, we can keep our virtual address base at 0x400000 + * while keeping the physical base at 0x0. + * + * Without this, OF incorrectly loads .text at 0x400000 + 0x400000 = 0x800000. + * Taken from x86/xen.lds.S + */ +#ifdef CONFIG_LD_IS_GNU +# define DECL_SECTION(x) x : AT(ADDR(#x) - XEN_VIRT_START) +#else +# define DECL_SECTION(x) x : AT(ADDR(x) - XEN_VIRT_START) +#endif + +SECTIONS +{ + . = XEN_VIRT_START; + + DECL_SECTION(.text) { + _stext = .; /* Text section */ + *(.text.header) + + *(.text.cold) + *(.text.unlikely .text.*_unlikely .text.unlikely.*) + + *(.text) +#ifdef CONFIG_CC_SPLIT_SECTIONS + *(.text.*) +#endif + + *(.fixup) + *(.gnu.warning) + . = ALIGN(POINTER_ALIGN); + _etext = .; /* End of text section */ + } :text + + . = ALIGN(PAGE_SIZE); + DECL_SECTION(.rodata) { + _srodata = .; /* Read-only data */ + *(.rodata) + *(.rodata.*) + *(.data.rel.ro) + *(.data.rel.ro.*) + + VPCI_ARRAY + + . = ALIGN(POINTER_ALIGN); + _erodata = .; /* End of read-only data */ + } :text + + #if defined(BUILD_ID) + . = ALIGN(4); + DECL_SECTION(.note.gnu.build-id) { + __note_gnu_build_id_start = .; + *(.note.gnu.build-id) + __note_gnu_build_id_end = .; + } :note :text + #endif + _erodata = .; /* End of read-only data */ + + . = ALIGN(PAGE_SIZE); + DECL_SECTION(.data.ro_after_init) { + __ro_after_init_start = .; + *(.data.ro_after_init) + . = ALIGN(PAGE_SIZE); + __ro_after_init_end = .; + } : text + + DECL_SECTION(.data.read_mostly) { + *(.data.read_mostly) + } :text + + . = ALIGN(PAGE_SIZE); + DECL_SECTION(.data) { /* Data */ + *(.data.page_aligned) + . = ALIGN(8); + __start_schedulers_array = .; + *(.data.schedulers) + __end_schedulers_array = .; + + HYPFS_PARAM + + *(.data .data.*) + CONSTRUCTORS + } :text + + . = ALIGN(PAGE_SIZE); /* Init code and data */ + __init_begin = .; + DECL_SECTION(.init.text) { + _sinittext = .; + *(.init.text) + _einittext = .; + . = ALIGN(PAGE_SIZE); /* Avoid mapping alt insns executable */ + } :text + + . = ALIGN(PAGE_SIZE); + DECL_SECTION(.init.data) { + *(.init.rodata) + *(.init.rodata.*) + + . = ALIGN(POINTER_ALIGN); + __setup_start = .; + *(.init.setup) + __setup_end = .; + + __initcall_start = .; + *(.initcallpresmp.init) + __presmp_initcall_end = .; + *(.initcall1.init) + __initcall_end = .; + + LOCK_PROFILE_DATA + + *(.init.data) + *(.init.data.rel) + *(.init.data.rel.*) + + . = ALIGN(8); + __ctors_start = .; + *(.ctors) + *(.init_array) + *(SORT(.init_array.*)) + __ctors_end = .; + } :text + . = ALIGN(POINTER_ALIGN); + __init_end = .; + + DECL_SECTION(.bss) { /* BSS */ + __bss_start = .; + *(.bss.stack_aligned) + . = ALIGN(PAGE_SIZE); + *(.bss.page_aligned) + . = ALIGN(PAGE_SIZE); + __per_cpu_start = .; + *(.bss.percpu.page_aligned) + *(.bss.percpu) + . = ALIGN(SMP_CACHE_BYTES); + *(.bss.percpu.read_mostly) + . = ALIGN(SMP_CACHE_BYTES); + __per_cpu_data_end = .; + *(.bss .bss.*) + . = ALIGN(POINTER_ALIGN); + __bss_end = .; + } :text + _end = . ; + + /* Section for the device tree blob (if any). */ + DECL_SECTION(.dtb) { *(.dtb) } :text + + DWARF2_DEBUG_SECTIONS + + DISCARD_SECTIONS + + STABS_DEBUG_SECTIONS + + ELF_DETAILS_SECTIONS +} -- 2.30.2 ^ permalink raw reply related [flat|nested] 27+ messages in thread
* Re: [PATCH 1/3] xen: Add files needed for minimal Power build 2023-06-07 15:06 ` [PATCH 1/3] xen: Add files needed for minimal Power build Shawn Anastasio @ 2023-06-09 9:15 ` Jan Beulich 2023-06-09 15:34 ` Shawn Anastasio 0 siblings, 1 reply; 27+ messages in thread From: Jan Beulich @ 2023-06-09 9:15 UTC (permalink / raw) To: Shawn Anastasio Cc: tpearson, Andrew Cooper, George Dunlap, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel On 07.06.2023 17:06, Shawn Anastasio wrote: > Add the build system changes required to build for ppc64le (POWER8+). > Following in the footsteps of the initial riscv port, only building > the head.o target, which boots to an infinite loop, is supported: > > $ make XEN_TARGET_ARCH=ppc64 -C xen openpower_defconfig > $ make XEN_TARGET_ARCH=ppc64 SUBSYSTEMS=xen -C xen TARGET=ppc64/head.o > > This port targets POWER8+ CPUs running in Little Endian mode specifically, > and does not boot on older machines. Additionally, this initial skeleton > only implements the PaPR/pseries boot protocol which allows it to be > booted in a standard QEMU virtual machine: > > $ qemu-system-ppc64 -M pseries-5.2 -m 256M -kernel ppc64/head.o > > Where possible, this patch uses header definitions and support routines > from the original Xen PowerPC port present in Xen 3.2.3. Though we are > targeting a much newer ISA than that original port did, some of the > definitions have remained similar enough for reuse. > > Signed-off-by: Shawn Anastasio <shawnanastasio@raptorengineering.com> Just a few small remarks, as following Andrew's comments I expect the patch will shrink quite a bit: > --- /dev/null > +++ b/config/ppc64.mk > @@ -0,0 +1,5 @@ > +CONFIG_PPC64 := y > +CONFIG_PPC64_64 := y > +CONFIG_PPC64_$(XEN_OS) := y The first of the 64-s here are a little odd; looking at RISC-V's counterpart, wouldn't this want to be CONFIG_PPC := y CONFIG_PPC_64 := y CONFIG_PPC_$(XEN_OS) := y > --- /dev/null > +++ b/xen/arch/ppc/Kconfig > @@ -0,0 +1,42 @@ > +config PPC > + def_bool y Is this necessary? Iirc PPC is frequently used as a name for 32-bit PPC (but then also elsewhere as covering both 32- and 64-bit), so I'm not sure we want this without having a need for it. > +config PPC64 > + def_bool y > + select 64BIT > + > +config ARCH_DEFCONFIG > + string > + default "arch/ppc/configs/openpower_defconfig" > + > +menu "Architecture Features" > + > +source "arch/Kconfig" > + > +endmenu > + > +menu "ISA Selection" > + > +choice > + prompt "Base ISA" > + default POWER_ISA_2_07B if PPC64 > + help > + This selects the base ISA version that Xen will target. > + > +config POWER_ISA_2_07B > + bool "POWER ISA 2.07B+" > + help > + Target version 2.07B+ of the POWER ISA (POWER8+) > + > +config POWER_ISA_3_00 > + bool "POWER ISA 3.00+" > + help > + Target version 3.00+ of the POWER ISA (POWER9+) What are the + in here meant to indicate? Since this is about a baseline ISA, I find such a use (presumably standing for "or newer") ambiguous. Jan ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 1/3] xen: Add files needed for minimal Power build 2023-06-09 9:15 ` Jan Beulich @ 2023-06-09 15:34 ` Shawn Anastasio 0 siblings, 0 replies; 27+ messages in thread From: Shawn Anastasio @ 2023-06-09 15:34 UTC (permalink / raw) To: Jan Beulich Cc: tpearson, Andrew Cooper, George Dunlap, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel On Fri Jun 9, 2023 at 4:15 AM CDT, Jan Beulich wrote: > > --- /dev/null > > +++ b/config/ppc64.mk > > @@ -0,0 +1,5 @@ > > +CONFIG_PPC64 := y > > +CONFIG_PPC64_64 := y > > +CONFIG_PPC64_$(XEN_OS) := y > > The first of the 64-s here are a little odd; looking at RISC-V's > counterpart, wouldn't this want to be > > CONFIG_PPC := y > CONFIG_PPC_64 := y > CONFIG_PPC_$(XEN_OS) := y> Good point, this was definitely an oversight on my part. I'll clean this up in v2. > > --- /dev/null > > +++ b/xen/arch/ppc/Kconfig > > @@ -0,0 +1,42 @@ > > +config PPC > > + def_bool y > > Is this necessary? Iirc PPC is frequently used as a name for 32-bit PPC > (but then also elsewhere as covering both 32- and 64-bit), so I'm not > sure we want this without having a need for it. This was mostly modeled after riscv/Kconfig. The idea was that this option guards PPC support in general (both 32- and 64-bit) and the subsequent option specifically opts in to 64-bit. Given that this effort is only targeting 64-bit mode, though, the argument could definitely be made that having the two options is redundant, but I believe that is also the case for the risc-v port. > > +config PPC64 > > + def_bool y > > + select 64BIT > > + > > +config ARCH_DEFCONFIG > > + string > > + default "arch/ppc/configs/openpower_defconfig" > > + > > +menu "Architecture Features" > > + > > +source "arch/Kconfig" > > + > > +endmenu > > + > > +menu "ISA Selection" > > + > > +choice > > + prompt "Base ISA" > > + default POWER_ISA_2_07B if PPC64 > > + help > > + This selects the base ISA version that Xen will target. > > + > > +config POWER_ISA_2_07B > > + bool "POWER ISA 2.07B+" > > + help > > + Target version 2.07B+ of the POWER ISA (POWER8+) > > + > > +config POWER_ISA_3_00 > > + bool "POWER ISA 3.00+" > > + help > > + Target version 3.00+ of the POWER ISA (POWER9+) > > What are the + in here meant to indicate? Since this is about a baseline > ISA, I find such a use (presumably standing for "or newer") ambiguous. Fair point. The + was intended as an "or newer" but I agree that it's ambiguous (for instance, it might be interpreted as a part of the ISA's name). I'll remove it in v2, along with a renaming of "POWER ISA" to "Power ISA" which I originally overlooked. > Jan Thanks, Shawn ^ permalink raw reply [flat|nested] 27+ messages in thread
* [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-07 15:06 [PATCH 0/3] Initial support for Power Shawn Anastasio 2023-06-07 15:06 ` [PATCH 1/3] xen: Add files needed for minimal Power build Shawn Anastasio @ 2023-06-07 15:06 ` Shawn Anastasio 2023-06-09 9:22 ` Jan Beulich 2023-06-07 15:06 ` [PATCH 3/3] maintainers: Add PPC64 maintainer Shawn Anastasio 2023-06-07 18:07 ` [PATCH 0/3] Initial support for Power Andrew Cooper 3 siblings, 1 reply; 27+ messages in thread From: Shawn Anastasio @ 2023-06-07 15:06 UTC (permalink / raw) To: xen-devel Cc: tpearson, Shawn Anastasio, Andrew Cooper, George Dunlap, Jan Beulich, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio On typical Power VMs (e.g. QEMU's -M pseries), a variety of services are provided by OpenFirmware, including an early serial console. Implement the required interfaces to call into OpenFirmware and write to the serial console. Since OpenFirmware runs in 32-bit Big Endian mode and Xen runs in 64-bit Little Endian mode, a thunk is required to save/restore any potentially-clobbered registers as well as to perform the required endianness switch. Thankfully, linux already has such a routine, which was imported into head.S. Support for bare metal (PowerNV) will be implemented in a future patch. Signed-off-by: Shawn Anastasio <shawnanastasio@raptorengineering.com> --- xen/arch/ppc/Kconfig.debug | 5 + xen/arch/ppc/Makefile | 3 +- xen/arch/ppc/boot_of.c | 122 +++++++++++++++++++++++ xen/arch/ppc/configs/openpower_defconfig | 1 + xen/arch/ppc/early_printk.c | 36 +++++++ xen/arch/ppc/include/asm/boot.h | 31 ++++++ xen/arch/ppc/include/asm/bug.h | 6 ++ xen/arch/ppc/include/asm/byteorder.h | 74 ++++++++++++++ xen/arch/ppc/include/asm/cache.h | 6 ++ xen/arch/ppc/include/asm/config.h | 3 + xen/arch/ppc/include/asm/early_printk.h | 14 +++ xen/arch/ppc/include/asm/processor.h | 54 +++++++++- xen/arch/ppc/include/asm/string.h | 6 ++ xen/arch/ppc/include/asm/types.h | 64 ++++++++++++ xen/arch/ppc/ppc64/asm-offsets.c | 55 ++++++++++ xen/arch/ppc/ppc64/head.S | 59 +++++++++++ xen/arch/ppc/setup.c | 20 +++- 17 files changed, 555 insertions(+), 4 deletions(-) create mode 100644 xen/arch/ppc/boot_of.c create mode 100644 xen/arch/ppc/early_printk.c create mode 100644 xen/arch/ppc/include/asm/boot.h create mode 100644 xen/arch/ppc/include/asm/bug.h create mode 100644 xen/arch/ppc/include/asm/byteorder.h create mode 100644 xen/arch/ppc/include/asm/cache.h create mode 100644 xen/arch/ppc/include/asm/early_printk.h create mode 100644 xen/arch/ppc/include/asm/string.h create mode 100644 xen/arch/ppc/include/asm/types.h diff --git a/xen/arch/ppc/Kconfig.debug b/xen/arch/ppc/Kconfig.debug index e69de29bb2..3c25a446e8 100644 --- a/xen/arch/ppc/Kconfig.debug +++ b/xen/arch/ppc/Kconfig.debug @@ -0,0 +1,5 @@ +config EARLY_PRINTK + bool "Enable early printk" + default DEBUG + help + Enables early printk debug messages \ No newline at end of file diff --git a/xen/arch/ppc/Makefile b/xen/arch/ppc/Makefile index b3ad837d4b..00b3b7baf3 100644 --- a/xen/arch/ppc/Makefile +++ b/xen/arch/ppc/Makefile @@ -1,5 +1,6 @@ obj-$(CONFIG_PPC64) += ppc64/ -obj-y += setup.o +obj-y += setup.o boot_of.o +obj-$(CONFIG_EARLY_PRINTK) += early_printk.o $(TARGET): $(TARGET)-syms cp -f $< $@ diff --git a/xen/arch/ppc/boot_of.c b/xen/arch/ppc/boot_of.c new file mode 100644 index 0000000000..9447d92661 --- /dev/null +++ b/xen/arch/ppc/boot_of.c @@ -0,0 +1,122 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright IBM Corp. 2005, 2006, 2007 + * Copyright Raptor Engineering, LLC + * + * Authors: Jimi Xenidis <jimix@watson.ibm.com> + * Hollis Blanchard <hollisb@us.ibm.com> + * Shawn Anastasio <sanastasio@raptorengineering.com> + */ + +#include <xen/kernel.h> +#include <xen/init.h> +#include <xen/lib.h> +#include <asm/byteorder.h> +#include <asm/boot.h> + +#define ADDR(x) (uint32_t)(unsigned long)(x) + +/* OF entrypoint*/ +static unsigned long of_vec; + +/* OF device handles*/ +static int bof_chosen; +static int of_out; + +static int of_call(const char *service, uint32_t nargs, uint32_t nrets, + int32_t rets[], ...) +{ + int rc; + va_list args; + int i; + struct of_service s = { 0 }; + + s.ofs_service = cpu_to_be32(ADDR(service)); + s.ofs_nargs = cpu_to_be32(nargs); + s.ofs_nrets = cpu_to_be32(nrets); + + /* copy all the params into the args array */ + va_start(args, rets); + + for ( i = 0; i < nargs; i++ ) + { + s.ofs_args[i] = cpu_to_be32(va_arg(args, uint32_t)); + } + + va_end(args); + + rc = enter_prom(&s, of_vec); + + /* yes always to the copy, just in case */ + for ( i = 0; i < nrets; i++ ) + { + rets[i] = be32_to_cpu(s.ofs_args[i + nargs]); + } + + return rc; +} + +static int of_finddevice(const char *devspec) +{ + int rets[1] = { OF_FAILURE }; + + of_call("finddevice", 1, 1, rets, devspec); + if ( rets[0] == OF_FAILURE ) + return OF_FAILURE; + + return rets[0]; +} + +static int of_getprop(int ph, const char *name, void *buf, uint32_t buflen) +{ + int rets[1] = { OF_FAILURE }; + + of_call("getprop", 4, 1, rets, ph, ADDR(name), buf, buflen); + + if ( rets[0] == OF_FAILURE ) + return OF_FAILURE; + + return rets[0]; +} + +int of_write(int ih, const char *addr, uint32_t len) +{ + int rets[1] = { OF_FAILURE }; + + if ( of_call("write", 3, 1, rets, ih, addr, len) == OF_FAILURE ) + return OF_FAILURE; + + return rets[0]; +} + +void of_putchar(char c) +{ + if ( c == '\n' ) + { + char buf = '\r'; + (void) of_write(of_out, &buf, 1); + } + (void) of_write(of_out, &c, 1); +} + +void boot_of_init(unsigned long vec) +{ + of_vec = vec; + bof_chosen = of_finddevice("/chosen"); + of_getprop(bof_chosen, "stdout", &of_out, sizeof(of_out)); + of_out = be32_to_cpu(of_out); +} \ No newline at end of file diff --git a/xen/arch/ppc/configs/openpower_defconfig b/xen/arch/ppc/configs/openpower_defconfig index 8783eb3488..f7cc075e45 100644 --- a/xen/arch/ppc/configs/openpower_defconfig +++ b/xen/arch/ppc/configs/openpower_defconfig @@ -10,4 +10,5 @@ CONFIG_PPC64=y CONFIG_DEBUG=y CONFIG_DEBUG_INFO=y +CONFIG_EARLY_PRINTK=y CONFIG_EXPERT=y diff --git a/xen/arch/ppc/early_printk.c b/xen/arch/ppc/early_printk.c new file mode 100644 index 0000000000..ab9213d801 --- /dev/null +++ b/xen/arch/ppc/early_printk.c @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Power early printk using OpenFirmware or OPAL + * + * Copyright Raptor Engineering, LLC + * + * Authors: Shawn Anastasio <sanastasio@raptorengineering.com> + */ + +#include <xen/lib.h> +#include <asm/boot.h> + +static void (*putchar_func)(char); + +void early_printk_init(bool is_of) +{ + putchar_func = is_of ? of_putchar : NULL; +} + +void early_puts(const char *s, size_t nr) +{ + if ( !putchar_func ) + return; + + while ( nr-- > 0 ) + putchar_func(*s++); +} + +void early_printk(const char *s) +{ + if ( !putchar_func ) + return; + + while ( *s ) + putchar_func(*s++); +} \ No newline at end of file diff --git a/xen/arch/ppc/include/asm/boot.h b/xen/arch/ppc/include/asm/boot.h new file mode 100644 index 0000000000..96d56f9650 --- /dev/null +++ b/xen/arch/ppc/include/asm/boot.h @@ -0,0 +1,31 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright Raptor Engineering, LLC + * + * Authors: Shawn Anastasio <@raptorengineering.com> + */ + +#ifndef _ASM_PPC_BOOT_H +#define _ASM_PPC_BOOT_H + +#include <xen/types.h> + +/* a collection of interfaces used during boot. */ +enum { + OF_FAILURE = -1, + OF_SUCCESS = 0, +}; + +struct of_service { + __be32 ofs_service; + __be32 ofs_nargs; + __be32 ofs_nrets; + __be32 ofs_args[10]; +}; + +extern int enter_prom(struct of_service *args, unsigned long entry); + +void boot_of_init(unsigned long); +void of_putchar(char c); + +#endif /* _ASM_PPC_BOOT_H */ \ No newline at end of file diff --git a/xen/arch/ppc/include/asm/bug.h b/xen/arch/ppc/include/asm/bug.h new file mode 100644 index 0000000000..40022a757b --- /dev/null +++ b/xen/arch/ppc/include/asm/bug.h @@ -0,0 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef _ASM_PPC_BUG_H +#define _ASM_PPC_BUG_H + +#endif /* _ASM_PPC_BUG_H */ \ No newline at end of file diff --git a/xen/arch/ppc/include/asm/byteorder.h b/xen/arch/ppc/include/asm/byteorder.h new file mode 100644 index 0000000000..ac840f9b1c --- /dev/null +++ b/xen/arch/ppc/include/asm/byteorder.h @@ -0,0 +1,74 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ASM_POWERPC_BYTEORDER_H +#define _ASM_POWERPC_BYTEORDER_H + +#include <asm/types.h> +#include <xen/compiler.h> + +static inline __u16 ld_le16(const volatile __u16 *addr) +{ + __u16 val; + + asm volatile ("lhbrx %0,0,%1" : "=r" (val) : "r" (addr), "m" (*addr)); + return val; +} + +static inline void st_le16(volatile __u16 *addr, const __u16 val) +{ + asm volatile ("sthbrx %1,0,%2" : "=m" (*addr) : "r" (val), "r" (addr)); +} + +static inline __u32 ld_le32(const volatile __u32 *addr) +{ + __u32 val; + + asm volatile ("lwbrx %0,0,%1" : "=r" (val) : "r" (addr), "m" (*addr)); + return val; +} + +static inline void st_le32(volatile __u32 *addr, const __u32 val) +{ + asm volatile ("stwbrx %1,0,%2" : "=m" (*addr) : "r" (val), "r" (addr)); +} + +static inline __attribute_const__ __u16 ___arch__swab16(__u16 value) +{ + __u16 result; + + asm("rlwimi %0,%1,8,16,23" + : "=r" (result) + : "r" (value), "0" (value >> 8)); + return result; +} + +static inline __attribute_const__ __u32 ___arch__swab32(__u32 value) +{ + __u32 result; + + asm("rlwimi %0,%1,24,16,23\n\t" + "rlwimi %0,%1,8,8,15\n\t" + "rlwimi %0,%1,24,0,7" + : "=r" (result) + : "r" (value), "0" (value >> 24)); + return result; +} + +#define __arch__swab16(x) ___arch__swab16(x) +#define __arch__swab32(x) ___arch__swab32(x) + +/* The same, but returns converted value from the location pointer by addr. */ +#define __arch__swab16p(addr) ld_le16(addr) +#define __arch__swab32p(addr) ld_le32(addr) + +/* The same, but do the conversion in situ, ie. put the value back to addr. */ +#define __arch__swab16s(addr) st_le16(addr,*addr) +#define __arch__swab32s(addr) st_le32(addr,*addr) + +#define __BYTEORDER_HAS_U64__ +#ifndef __powerpc64__ +#define __SWAB_64_THRU_32__ +#endif /* __powerpc64__ */ + +#include <xen/byteorder/little_endian.h> + +#endif /* _ASM_POWERPC_BYTEORDER_H */ diff --git a/xen/arch/ppc/include/asm/cache.h b/xen/arch/ppc/include/asm/cache.h new file mode 100644 index 0000000000..677a2b3915 --- /dev/null +++ b/xen/arch/ppc/include/asm/cache.h @@ -0,0 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef _ASM_PPC_CACHE_H +#define _ASM_PPC_CACHE_H + +#endif /* _ASM_PPC_CACHE_H */ \ No newline at end of file diff --git a/xen/arch/ppc/include/asm/config.h b/xen/arch/ppc/include/asm/config.h index 7a2862ef7a..a3438de92b 100644 --- a/xen/arch/ppc/include/asm/config.h +++ b/xen/arch/ppc/include/asm/config.h @@ -52,6 +52,9 @@ /* size of minimum stack frame; C code can write into the caller's stack */ #define STACK_FRAME_OVERHEAD 32 +/* size of minimum stack frame that can hold an entire cpu_user_regs struct */ +#define STACK_SWITCH_FRAME_SIZE (UREGS_sizeof + STACK_FRAME_OVERHEAD) + #endif /* __PPC_CONFIG_H__ */ /* * Local variables: diff --git a/xen/arch/ppc/include/asm/early_printk.h b/xen/arch/ppc/include/asm/early_printk.h new file mode 100644 index 0000000000..a9566a1240 --- /dev/null +++ b/xen/arch/ppc/include/asm/early_printk.h @@ -0,0 +1,14 @@ +#ifndef __EARLY_PRINTK_H__ +#define __EARLY_PRINTK_H__ + +#include <xen/early_printk.h> + +#ifdef CONFIG_EARLY_PRINTK +void early_printk_init(bool is_of); +void early_printk(const char *str); +#else +static inline void early_printk_init(bool is_of) {} +static inline void early_printk(const char *s) {} +#endif + +#endif /* __EARLY_PRINTK_H__ */ \ No newline at end of file diff --git a/xen/arch/ppc/include/asm/processor.h b/xen/arch/ppc/include/asm/processor.h index 0ab7bfc9df..63409f003a 100644 --- a/xen/arch/ppc/include/asm/processor.h +++ b/xen/arch/ppc/include/asm/processor.h @@ -114,7 +114,42 @@ #define PVR_970MP 0x0044 #define PVR_BE 0x0070 -#ifdef __ASSEMBLY__ +#ifndef __ASSEMBLY__ + +#include <xen/types.h> + +/* User-accessible registers: nost of these need to be saved/restored + * for every nested Xen invocation. */ +struct cpu_user_regs +{ + uint64_t gprs[32]; + uint64_t lr; + uint64_t ctr; + uint64_t srr0; + uint64_t srr1; + uint64_t pc; + uint64_t msr; + uint64_t fpscr; /* XXX Is this necessary */ + uint64_t xer; + uint64_t hid4; /* debug only */ + uint64_t dar; /* debug only */ + uint32_t dsisr; /* debug only */ + uint32_t cr; + uint32_t __pad; /* good spot for another 32bit reg */ + uint32_t entry_vector; +}; + +static __inline__ void sync(void) +{ + __asm__ __volatile__("sync"); +} + +static __inline__ void isync(void) +{ + __asm__ __volatile__("isync"); +} + +#else #define LOADADDR(rn, name) \ lis rn, name##@highest; \ @@ -152,6 +187,22 @@ .long 0xa6037b7d; /* mtsrr1 r11 */ \ .long 0x2400004c /* rfid */ +/* Taken from Linux kernel source (arch/powerpc/boot/crt0.S) */ +.macro OP_REGS op, width, start, end, base, offset + .Lreg=\start + .rept (\end - \start + 1) + \op .Lreg,\offset+\width*.Lreg(\base) + .Lreg=.Lreg+1 + .endr +.endm + +#define SAVE_GPRS(start, end, base) OP_REGS std, 8, start, end, base, 0 +#define REST_GPRS(start, end, base) OP_REGS ld, 8, start, end, base, 0 +#define SAVE_GPR(n, base) SAVE_GPRS(n, n, base) +#define REST_GPR(n, base) REST_GPRS(n, n, base) +#define SAVE_NVGPRS(base) SAVE_GPRS(14, 31, base) +#define REST_NVGPRS(base) REST_GPRS(14, 31, base) + #define _GLOBAL(name) \ .section ".text"; \ .align 2; \ @@ -167,7 +218,6 @@ #define _ENTRY(name) name -#else /* __ASSEMBLY__ */ #endif /* __ASSEMBLY__ */ #endif diff --git a/xen/arch/ppc/include/asm/string.h b/xen/arch/ppc/include/asm/string.h new file mode 100644 index 0000000000..aed83b0d47 --- /dev/null +++ b/xen/arch/ppc/include/asm/string.h @@ -0,0 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef _ASM_PPC_STRING_H +#define _ASM_PPC_STRING_H + +#endif /* _ASM_PPC_STRING_H */ \ No newline at end of file diff --git a/xen/arch/ppc/include/asm/types.h b/xen/arch/ppc/include/asm/types.h new file mode 100644 index 0000000000..44b5df49f2 --- /dev/null +++ b/xen/arch/ppc/include/asm/types.h @@ -0,0 +1,64 @@ +/* from xen/include/asm-x86/types.h */ + +#ifndef _PPC_TYPES_H +#define _PPC_TYPES_H + +#ifndef __ASSEMBLY__ +typedef unsigned short umode_t; + +/* + * __xx is ok: it doesn't pollute the POSIX namespace. Use these in the + * header files exported to user space + */ + +typedef __signed__ char __s8; +typedef unsigned char __u8; + +typedef __signed__ short __s16; +typedef unsigned short __u16; + +typedef __signed__ int __s32; +typedef unsigned int __u32; + +#if defined(__GNUC__) && !defined(__STRICT_ANSI__) +#if defined(__ppc__) +typedef __signed__ long long __s64; +typedef unsigned long long __u64; + +#elif defined(__PPC64__) +typedef __signed__ long __s64; +typedef unsigned long __u64; +#endif +#endif + +typedef signed char s8; +typedef unsigned char u8; + +typedef signed short s16; +typedef unsigned short u16; + +typedef signed int s32; +typedef unsigned int u32; + +#if defined(__ppc__) +typedef signed long long s64; +typedef unsigned long long u64; +typedef unsigned int size_t; +#elif defined(__PPC64__) +typedef signed long s64; +typedef unsigned long u64; +typedef unsigned long size_t; +#endif + +typedef unsigned long paddr_t; +#define PRIpaddr "08lx" + +/* DMA addresses come in generic and 64-bit flavours. */ + +typedef unsigned long dma_addr_t; +typedef u64 dma64_addr_t; + +typedef unsigned short xmem_bufctl_t; + +#endif /* __ASSEMBLY__ */ +#endif diff --git a/xen/arch/ppc/ppc64/asm-offsets.c b/xen/arch/ppc/ppc64/asm-offsets.c index e69de29bb2..a6de2b2768 100644 --- a/xen/arch/ppc/ppc64/asm-offsets.c +++ b/xen/arch/ppc/ppc64/asm-offsets.c @@ -0,0 +1,55 @@ +/* + * Generate definitions needed by assembly language modules. + * This code generates raw asm output which is post-processed + * to extract and format the required data. + */ + +#include <asm/processor.h> + +#define DEFINE(_sym, _val) \ + asm volatile ("\n.ascii\"==>#define " #_sym " %0 /* " #_val " */<==\"" \ + : : "i" (_val) ) +#define BLANK() \ + asm volatile ( "\n.ascii\"==><==\"" : : ) +#define OFFSET(_sym, _str, _mem) \ + DEFINE(_sym, offsetof(_str, _mem)); + +/* base-2 logarithm */ +#define __L2(_x) (((_x) & 0x00000002) ? 1 : 0) +#define __L4(_x) (((_x) & 0x0000000c) ? ( 2 + __L2( (_x)>> 2)) : __L2( _x)) +#define __L8(_x) (((_x) & 0x000000f0) ? ( 4 + __L4( (_x)>> 4)) : __L4( _x)) +#define __L16(_x) (((_x) & 0x0000ff00) ? ( 8 + __L8( (_x)>> 8)) : __L8( _x)) +#define LOG_2(_x) (((_x) & 0xffff0000) ? (16 + __L16((_x)>>16)) : __L16(_x)) + +void __dummy__(void) +{ + DEFINE(GPR_WIDTH, sizeof(unsigned long)); + DEFINE(FPR_WIDTH, sizeof(double)); + + OFFSET(UREGS_gprs, struct cpu_user_regs, gprs); + OFFSET(UREGS_r0, struct cpu_user_regs, gprs[0]); + OFFSET(UREGS_r1, struct cpu_user_regs, gprs[1]); + OFFSET(UREGS_r13, struct cpu_user_regs, gprs[13]); + OFFSET(UREGS_srr0, struct cpu_user_regs, srr0); + OFFSET(UREGS_srr1, struct cpu_user_regs, srr1); + OFFSET(UREGS_pc, struct cpu_user_regs, pc); + OFFSET(UREGS_msr, struct cpu_user_regs, msr); + OFFSET(UREGS_lr, struct cpu_user_regs, lr); + OFFSET(UREGS_ctr, struct cpu_user_regs, ctr); + OFFSET(UREGS_xer, struct cpu_user_regs, xer); + OFFSET(UREGS_hid4, struct cpu_user_regs, hid4); + OFFSET(UREGS_dar, struct cpu_user_regs, dar); + OFFSET(UREGS_dsisr, struct cpu_user_regs, dsisr); + OFFSET(UREGS_cr, struct cpu_user_regs, cr); + OFFSET(UREGS_fpscr, struct cpu_user_regs, fpscr); + DEFINE(UREGS_sizeof, sizeof(struct cpu_user_regs)); +} + +/* + * Local variables: + * mode: C + * c-file-style: "BSD" + * c-basic-offset: 4 + * indent-tabs-mode: nil + * End: + */ diff --git a/xen/arch/ppc/ppc64/head.S b/xen/arch/ppc/ppc64/head.S index d91b0972ae..d5ca1c08ad 100644 --- a/xen/arch/ppc/ppc64/head.S +++ b/xen/arch/ppc/ppc64/head.S @@ -65,3 +65,62 @@ cpu0_boot_stack_bottom: .space STACK_SIZE cpu0_boot_stack: .space STACK_FRAME_OVERHEAD + +/* Adapted from Linux kernel source (arch/powerpc/boot/crt0.S) */ +_GLOBAL(enter_prom) + mflr r0 + std r0,16(r1) + stdu r1,-STACK_SWITCH_FRAME_SIZE(r1) /* Save SP and create stack space */ + + /* Because PROM is running in 32b mode, it clobbers the high order half + * of all registers that it saves. We therefore save those registers + * PROM might touch to the stack. (r0, r3-r13 are caller saved) + */ + SAVE_GPR(2, r1) + SAVE_GPR(13, r1) + SAVE_NVGPRS(r1) + mfcr r10 + mfmsr r11 + std r10,UREGS_cr(r1) + std r11,UREGS_msr(r1) + + /* Put PROM address in SRR0 */ + mtsrr0 r4 + + /* Setup our trampoline return addr in LR */ + bcl 20,31,$+4 +0: mflr r4 + addi r4,r4,(1f - 0b) + mtlr r4 + + /* Prepare a 32-bit mode big endian MSR + */ + SET_REG_TO_CONST(r12, MSR_SF | MSR_LE) + andc r11,r11,r12 + mtsrr1 r11 + rfid + +1: /* Return from OF */ + FIXUP_ENDIAN + + /* Just make sure that r1 top 32 bits didn't get + * corrupt by OF + */ + rldicl r1,r1,0,32 + + /* Restore the MSR (back to 64 bits) */ + ld r0,UREGS_msr(r1) + mtmsrd r0 + isync + + /* Restore other registers */ + REST_GPR(2, r1) + REST_GPR(13, r1) + REST_NVGPRS(r1) + ld r4,UREGS_cr(r1) + mtcr r4 + + addi r1,r1,STACK_SWITCH_FRAME_SIZE + ld r0,16(r1) + mtlr r0 + blr diff --git a/xen/arch/ppc/setup.c b/xen/arch/ppc/setup.c index c9509fb6a6..fb12d4df34 100644 --- a/xen/arch/ppc/setup.c +++ b/xen/arch/ppc/setup.c @@ -1,9 +1,27 @@ /* SPDX-License-Identifier: GPL-2.0-or-later */ #include <xen/compile.h> #include <xen/init.h> +#include <asm/boot.h> +#include <asm/early_printk.h> -void __init noreturn start_xen(void) +void __init noreturn start_xen(unsigned long r3, unsigned long r4, + unsigned long r5, unsigned long r6, + unsigned long r7) { + if ( r5 ) + { + /* OF boot protocol */ + boot_of_init(r5); + early_printk_init(true); + } + else + { + /* kexec boot: Unimplemented */ + __builtin_trap(); + } + + early_printk("Hello, ppc64le!\n"); + for ( ;; ) { // Set current hardware thread to very low priority -- 2.30.2 ^ permalink raw reply related [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-07 15:06 ` [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries Shawn Anastasio @ 2023-06-09 9:22 ` Jan Beulich 2023-06-09 9:29 ` Andrew Cooper 2023-06-09 15:36 ` Shawn Anastasio 0 siblings, 2 replies; 27+ messages in thread From: Jan Beulich @ 2023-06-09 9:22 UTC (permalink / raw) To: Shawn Anastasio Cc: tpearson, Andrew Cooper, George Dunlap, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel On 07.06.2023 17:06, Shawn Anastasio wrote: > On typical Power VMs (e.g. QEMU's -M pseries), a variety of services > are provided by OpenFirmware, including an early serial console. > Implement the required interfaces to call into OpenFirmware and write > to the serial console. > > Since OpenFirmware runs in 32-bit Big Endian mode and Xen runs in > 64-bit Little Endian mode, a thunk is required to save/restore > any potentially-clobbered registers as well as to perform the > required endianness switch. Thankfully, linux already has such > a routine, which was imported into head.S. > > Support for bare metal (PowerNV) will be implemented in a future > patch. > > Signed-off-by: Shawn Anastasio <shawnanastasio@raptorengineering.com> Just a couple of nits: > xen/arch/ppc/Kconfig.debug | 5 + > xen/arch/ppc/Makefile | 3 +- > xen/arch/ppc/boot_of.c | 122 +++++++++++++++++++++++ > xen/arch/ppc/configs/openpower_defconfig | 1 + > xen/arch/ppc/early_printk.c | 36 +++++++ > xen/arch/ppc/include/asm/boot.h | 31 ++++++ > xen/arch/ppc/include/asm/bug.h | 6 ++ > xen/arch/ppc/include/asm/byteorder.h | 74 ++++++++++++++ > xen/arch/ppc/include/asm/cache.h | 6 ++ > xen/arch/ppc/include/asm/config.h | 3 + > xen/arch/ppc/include/asm/early_printk.h | 14 +++ > xen/arch/ppc/include/asm/processor.h | 54 +++++++++- > xen/arch/ppc/include/asm/string.h | 6 ++ > xen/arch/ppc/include/asm/types.h | 64 ++++++++++++ > xen/arch/ppc/ppc64/asm-offsets.c | 55 ++++++++++ > xen/arch/ppc/ppc64/head.S | 59 +++++++++++ > xen/arch/ppc/setup.c | 20 +++- > 17 files changed, 555 insertions(+), 4 deletions(-) > create mode 100644 xen/arch/ppc/boot_of.c Unless required, in new additions we tend to prefer dashes over underscores. In filenames it is pretty rare that dashes really need avoiding. > --- a/xen/arch/ppc/Kconfig.debug > +++ b/xen/arch/ppc/Kconfig.debug > @@ -0,0 +1,5 @@ > +config EARLY_PRINTK > + bool "Enable early printk" > + default DEBUG > + help > + Enables early printk debug messages > \ No newline at end of file There are many examples of this throughout the patch, which you want to take care of. > --- /dev/null > +++ b/xen/arch/ppc/boot_of.c > @@ -0,0 +1,122 @@ > +/* SPDX-License-Identifier: GPL-2.0-or-later */ By default we mean to use ... > --- /dev/null > +++ b/xen/arch/ppc/early_printk.c > @@ -0,0 +1,36 @@ > +/* SPDX-License-Identifier: GPL-2.0 */ ... the more modern form of this (GPL-2.0-only). Anything deviating from that may want justifying in the description. Jan ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 9:22 ` Jan Beulich @ 2023-06-09 9:29 ` Andrew Cooper 2023-06-09 9:38 ` Jan Beulich 2023-06-09 15:36 ` Shawn Anastasio 1 sibling, 1 reply; 27+ messages in thread From: Andrew Cooper @ 2023-06-09 9:29 UTC (permalink / raw) To: Jan Beulich, Shawn Anastasio Cc: tpearson, George Dunlap, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel On 09/06/2023 10:22 am, Jan Beulich wrote: >> --- /dev/null >> +++ b/xen/arch/ppc/boot_of.c >> @@ -0,0 +1,122 @@ >> +/* SPDX-License-Identifier: GPL-2.0-or-later */ > By default we mean to use ... > >> --- /dev/null >> +++ b/xen/arch/ppc/early_printk.c >> @@ -0,0 +1,36 @@ >> +/* SPDX-License-Identifier: GPL-2.0 */ > ... the more modern form of this (GPL-2.0-only). Anything deviating from > that may want justifying in the description. GPL-2.0-or-later is fine. It's only the un-qualified GPL-2.0 form which is deprecated, and should be replaced with an -or-later or -only suffix, as chosen by the copyright holder. ~Andrew ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 9:29 ` Andrew Cooper @ 2023-06-09 9:38 ` Jan Beulich 2023-06-09 9:43 ` Julien Grall 2023-06-09 9:43 ` Andrew Cooper 0 siblings, 2 replies; 27+ messages in thread From: Jan Beulich @ 2023-06-09 9:38 UTC (permalink / raw) To: Andrew Cooper Cc: tpearson, George Dunlap, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel, Shawn Anastasio On 09.06.2023 11:29, Andrew Cooper wrote: > On 09/06/2023 10:22 am, Jan Beulich wrote: >>> --- /dev/null >>> +++ b/xen/arch/ppc/boot_of.c >>> @@ -0,0 +1,122 @@ >>> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >> By default we mean to use ... >> >>> --- /dev/null >>> +++ b/xen/arch/ppc/early_printk.c >>> @@ -0,0 +1,36 @@ >>> +/* SPDX-License-Identifier: GPL-2.0 */ >> ... the more modern form of this (GPL-2.0-only). Anything deviating from >> that may want justifying in the description. > > GPL-2.0-or-later is fine. Hmm, I was merely following https://lists.xen.org/archives/html/xen-devel/2023-06/msg00415.html. The text at the top of ./COPYING looks to suggest -only, and I'm unaware of any other place where our default is actually written down. Jan > It's only the un-qualified GPL-2.0 form which is deprecated, and should > be replaced with an -or-later or -only suffix, as chosen by the > copyright holder. > > ~Andrew ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 9:38 ` Jan Beulich @ 2023-06-09 9:43 ` Julien Grall 2023-06-09 9:43 ` Andrew Cooper 1 sibling, 0 replies; 27+ messages in thread From: Julien Grall @ 2023-06-09 9:43 UTC (permalink / raw) To: Jan Beulich, Andrew Cooper Cc: tpearson, George Dunlap, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel, Shawn Anastasio Hi, On 09/06/2023 10:38, Jan Beulich wrote: > On 09.06.2023 11:29, Andrew Cooper wrote: >> On 09/06/2023 10:22 am, Jan Beulich wrote: >>>> --- /dev/null >>>> +++ b/xen/arch/ppc/boot_of.c >>>> @@ -0,0 +1,122 @@ >>>> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >>> By default we mean to use ... >>> >>>> --- /dev/null >>>> +++ b/xen/arch/ppc/early_printk.c >>>> @@ -0,0 +1,36 @@ >>>> +/* SPDX-License-Identifier: GPL-2.0 */ >>> ... the more modern form of this (GPL-2.0-only). Anything deviating from >>> that may want justifying in the description. >> >> GPL-2.0-or-later is fine. > > Hmm, I was merely following > https://lists.xen.org/archives/html/xen-devel/2023-06/msg00415.html. > The text at the top of ./COPYING looks to suggest -only, and I'm > unaware of any other place where our default is actually written down. We had several discussion in the past about using GPLv2+ license in Xen (see [1], [2]). It has always been unclear what is the impact on contribution with such license. So I think we should stick with GPL-2.0 for new code unless there is a reason to diverge. Cheers, [1] https://patchwork.kernel.org/project/xen-devel/patch/1474985810-12289-1-git-send-email-lars.kurth@citrix.com/#19650817 [2] https://lore.kernel.org/all/alpine.DEB.2.22.394.2208231140140.15247@ubuntu-linux-20-04-desktop/ -- Julien Grall ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 9:38 ` Jan Beulich 2023-06-09 9:43 ` Julien Grall @ 2023-06-09 9:43 ` Andrew Cooper 2023-06-09 9:46 ` Julien Grall 1 sibling, 1 reply; 27+ messages in thread From: Andrew Cooper @ 2023-06-09 9:43 UTC (permalink / raw) To: Jan Beulich Cc: tpearson, George Dunlap, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel, Shawn Anastasio On 09/06/2023 10:38 am, Jan Beulich wrote: > On 09.06.2023 11:29, Andrew Cooper wrote: >> On 09/06/2023 10:22 am, Jan Beulich wrote: >>>> --- /dev/null >>>> +++ b/xen/arch/ppc/boot_of.c >>>> @@ -0,0 +1,122 @@ >>>> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >>> By default we mean to use ... >>> >>>> --- /dev/null >>>> +++ b/xen/arch/ppc/early_printk.c >>>> @@ -0,0 +1,36 @@ >>>> +/* SPDX-License-Identifier: GPL-2.0 */ >>> ... the more modern form of this (GPL-2.0-only). Anything deviating from >>> that may want justifying in the description. >> GPL-2.0-or-later is fine. > Hmm, I was merely following > https://lists.xen.org/archives/html/xen-devel/2023-06/msg00415.html. > The text at the top of ./COPYING looks to suggest -only, and I'm > unaware of any other place where our default is actually written down. The license is chosen by the submitter/copyright holder, based on their preferences/wishes. It's fine for Xen to say "if you've got no vested interest, we recommend GPL-2.0-only", but that is strictly a recommendation and no more. If the submitter chooses GPL-2.0-or-later, that is their prerogative. We have plenty of GPL-2.0-or-later code in Xen. ~Andrew ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 9:43 ` Andrew Cooper @ 2023-06-09 9:46 ` Julien Grall 2023-06-09 9:54 ` Andrew Cooper 0 siblings, 1 reply; 27+ messages in thread From: Julien Grall @ 2023-06-09 9:46 UTC (permalink / raw) To: Andrew Cooper, Jan Beulich Cc: tpearson, George Dunlap, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel, Shawn Anastasio Hi Andrew, On 09/06/2023 10:43, Andrew Cooper wrote: > On 09/06/2023 10:38 am, Jan Beulich wrote: >> On 09.06.2023 11:29, Andrew Cooper wrote: >>> On 09/06/2023 10:22 am, Jan Beulich wrote: >>>>> --- /dev/null >>>>> +++ b/xen/arch/ppc/boot_of.c >>>>> @@ -0,0 +1,122 @@ >>>>> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >>>> By default we mean to use ... >>>> >>>>> --- /dev/null >>>>> +++ b/xen/arch/ppc/early_printk.c >>>>> @@ -0,0 +1,36 @@ >>>>> +/* SPDX-License-Identifier: GPL-2.0 */ >>>> ... the more modern form of this (GPL-2.0-only). Anything deviating from >>>> that may want justifying in the description. >>> GPL-2.0-or-later is fine. >> Hmm, I was merely following >> https://lists.xen.org/archives/html/xen-devel/2023-06/msg00415.html. >> The text at the top of ./COPYING looks to suggest -only, and I'm >> unaware of any other place where our default is actually written down. > > The license is chosen by the submitter/copyright holder, based on their > preferences/wishes. > > It's fine for Xen to say "if you've got no vested interest, we recommend > GPL-2.0-only", but that is strictly a recommendation and no more. > > If the submitter chooses GPL-2.0-or-later, that is their prerogative. > We have plenty of GPL-2.0-or-later code in Xen. From my past experience, the submitters tend to just copy the license from an existing file in Xen rather than explicitly choosing it. So I think it is fair to ask the question because our original and default license is GPLv2 nor GPLv2+. Cheers, -- Julien Grall ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 9:46 ` Julien Grall @ 2023-06-09 9:54 ` Andrew Cooper 2023-06-09 10:12 ` Julien Grall 0 siblings, 1 reply; 27+ messages in thread From: Andrew Cooper @ 2023-06-09 9:54 UTC (permalink / raw) To: Julien Grall, Jan Beulich Cc: tpearson, George Dunlap, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel, Shawn Anastasio On 09/06/2023 10:46 am, Julien Grall wrote: > On 09/06/2023 10:43, Andrew Cooper wrote: >> On 09/06/2023 10:38 am, Jan Beulich wrote: >>> On 09.06.2023 11:29, Andrew Cooper wrote: >>>> On 09/06/2023 10:22 am, Jan Beulich wrote: >>>>>> --- /dev/null >>>>>> +++ b/xen/arch/ppc/boot_of.c >>>>>> @@ -0,0 +1,122 @@ >>>>>> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >>>>> By default we mean to use ... >>>>> >>>>>> --- /dev/null >>>>>> +++ b/xen/arch/ppc/early_printk.c >>>>>> @@ -0,0 +1,36 @@ >>>>>> +/* SPDX-License-Identifier: GPL-2.0 */ >>>>> ... the more modern form of this (GPL-2.0-only). Anything >>>>> deviating from >>>>> that may want justifying in the description. >>>> GPL-2.0-or-later is fine. >>> Hmm, I was merely following >>> https://lists.xen.org/archives/html/xen-devel/2023-06/msg00415.html. >>> The text at the top of ./COPYING looks to suggest -only, and I'm >>> unaware of any other place where our default is actually written down. >> >> The license is chosen by the submitter/copyright holder, based on their >> preferences/wishes. >> >> It's fine for Xen to say "if you've got no vested interest, we recommend >> GPL-2.0-only", but that is strictly a recommendation and no more. >> >> If the submitter chooses GPL-2.0-or-later, that is their prerogative. >> We have plenty of GPL-2.0-or-later code in Xen. > > From my past experience, the submitters tend to just copy the license > from an existing file in Xen rather than explicitly choosing it. So I > think it is fair to ask the question because our original and default > license is GPLv2 nor GPLv2+. Did you read the bit in the cover letter about part of this code being derived from the out-of-tree port years ago? You're blindly assuming that there is even a choice of license available to be used. The submitter chooses the license to use. You can request that they justify it, but you cannot demand that they change it. ~Andrew ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 9:54 ` Andrew Cooper @ 2023-06-09 10:12 ` Julien Grall 2023-06-09 15:01 ` Shawn Anastasio 0 siblings, 1 reply; 27+ messages in thread From: Julien Grall @ 2023-06-09 10:12 UTC (permalink / raw) To: Andrew Cooper, Jan Beulich Cc: tpearson, George Dunlap, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel, Shawn Anastasio Hi Andrew, On 09/06/2023 10:54, Andrew Cooper wrote: > On 09/06/2023 10:46 am, Julien Grall wrote: >> On 09/06/2023 10:43, Andrew Cooper wrote: >>> On 09/06/2023 10:38 am, Jan Beulich wrote: >>>> On 09.06.2023 11:29, Andrew Cooper wrote: >>>>> On 09/06/2023 10:22 am, Jan Beulich wrote: >>>>>>> --- /dev/null >>>>>>> +++ b/xen/arch/ppc/boot_of.c >>>>>>> @@ -0,0 +1,122 @@ >>>>>>> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >>>>>> By default we mean to use ... >>>>>> >>>>>>> --- /dev/null >>>>>>> +++ b/xen/arch/ppc/early_printk.c >>>>>>> @@ -0,0 +1,36 @@ >>>>>>> +/* SPDX-License-Identifier: GPL-2.0 */ >>>>>> ... the more modern form of this (GPL-2.0-only). Anything >>>>>> deviating from >>>>>> that may want justifying in the description. >>>>> GPL-2.0-or-later is fine. >>>> Hmm, I was merely following >>>> https://lists.xen.org/archives/html/xen-devel/2023-06/msg00415.html. >>>> The text at the top of ./COPYING looks to suggest -only, and I'm >>>> unaware of any other place where our default is actually written down. >>> >>> The license is chosen by the submitter/copyright holder, based on their >>> preferences/wishes. >>> >>> It's fine for Xen to say "if you've got no vested interest, we recommend >>> GPL-2.0-only", but that is strictly a recommendation and no more. >>> >>> If the submitter chooses GPL-2.0-or-later, that is their prerogative. >>> We have plenty of GPL-2.0-or-later code in Xen. >> >> From my past experience, the submitters tend to just copy the license >> from an existing file in Xen rather than explicitly choosing it. So I >> think it is fair to ask the question because our original and default >> license is GPLv2 nor GPLv2+. > > Did you read the bit in the cover letter about part of this code being > derived from the out-of-tree port years ago? Yes I read it... But I didn't check the original license and ... > > You're blindly assuming that there is even a choice of license available > to be used. ... I didn't assume anything here. I made a generic statement because your e-mail lead to think that all the submitter made a conscious decision. Note that, from past discussion, we agreed that it would be fine to re-license from gplv2+ to gplv2-only without requesting the original author. So technically there is a choice. As a side note, "blindly" is not very inclusive. We may have different view, but it doesn't mean yours is better than mine (and vice-versa). You could have express your opinion without saying "blindly" and it would have come across less rude. > The submitter chooses the license to use. You can request that they > justify it, but you cannot demand that they change it. Strictly speaking we can refuse any code. That count for license as well. Anyway, I didn't request a change here. I merely pointed out that any use of GPLv2+ should be justified because on Arm most of the people don't pay attention on the license and pick the one from an existing file. Cheers, -- Julien Grall ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 10:12 ` Julien Grall @ 2023-06-09 15:01 ` Shawn Anastasio 2023-06-09 16:07 ` Julien Grall 0 siblings, 1 reply; 27+ messages in thread From: Shawn Anastasio @ 2023-06-09 15:01 UTC (permalink / raw) To: Julien Grall, Andrew Cooper, Jan Beulich Cc: tpearson, George Dunlap, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel On Fri Jun 9, 2023 at 5:12 AM CDT, Julien Grall wrote: > Strictly speaking we can refuse any code. That count for license as > well. Anyway, I didn't request a change here. I merely pointed out that > any use of GPLv2+ should be justified because on Arm most of the people > don't pay attention on the license and pick the one from an existing file. Hi Julien, The choice of GPLv2+ for many of the files in this patchset was indeed inherited from old IBM-written Xen code that the files in question were derived from. I did not realize it was permissible or even desirable to relicense those to GPLv2-only. As for the new files, GPLv2+ was chosen to remain consistent and to open the door for future derivations from GPLv2+ licensed code, either from the older Xen tree or from the Linux ppc tree, much of which is also licensed as GPLv2+. If it would reduce friction, these files could be relicensed to GPLv2-only. Thanks, Shawn ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 15:01 ` Shawn Anastasio @ 2023-06-09 16:07 ` Julien Grall 2023-06-09 16:20 ` Shawn Anastasio 2023-06-12 15:19 ` George Dunlap 0 siblings, 2 replies; 27+ messages in thread From: Julien Grall @ 2023-06-09 16:07 UTC (permalink / raw) To: Shawn Anastasio, Andrew Cooper, Jan Beulich Cc: tpearson, George Dunlap, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel Hi Shawn, On 09/06/2023 16:01, Shawn Anastasio wrote: > On Fri Jun 9, 2023 at 5:12 AM CDT, Julien Grall wrote: >> Strictly speaking we can refuse any code. That count for license as >> well. Anyway, I didn't request a change here. I merely pointed out that >> any use of GPLv2+ should be justified because on Arm most of the people >> don't pay attention on the license and pick the one from an existing file. > > Hi Julien, > > The choice of GPLv2+ for many of the files in this patchset was indeed > inherited from old IBM-written Xen code that the files in question were > derived from. I did not realize it was permissible or even desirable to > relicense those to GPLv2-only. > > As for the new files, GPLv2+ was chosen to remain consistent and to open > the door for future derivations from GPLv2+ licensed code, either from > the older Xen tree or from the Linux ppc tree, much of which is also > licensed as GPLv2+. If it would reduce friction, these files could be > relicensed to GPLv2-only. (Before someone points out, I know this is already a problem on other part of Xen. But it would be ideal if we avoid spreading this mess on new architectures :). Thanks for the explanations. To clarify, are you saying that all the files will be GPLv2+ or just some? If the latter, then my concern would be that if you need to import GPLv2-only code, then you may need to write your code in a different file. This may become messy to handle and some developer may end up to be confused. I am not a lawyer though, so you may want to check the implications here. Cheers, -- Julien Grall ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 16:07 ` Julien Grall @ 2023-06-09 16:20 ` Shawn Anastasio 2023-06-12 15:19 ` George Dunlap 1 sibling, 0 replies; 27+ messages in thread From: Shawn Anastasio @ 2023-06-09 16:20 UTC (permalink / raw) To: Julien Grall, Andrew Cooper, Jan Beulich Cc: tpearson, George Dunlap, Stefano Stabellini, Wei Liu, xen-devel On Fri Jun 9, 2023 at 11:07 AM CDT, Julien Grall wrote: > Thanks for the explanations. To clarify, are you saying that all the > files will be GPLv2+ or just some? My idea was to license any file where I expect there to derivations from existing code as GPLv2+, and fall back to GPLv2-only for newly-written files for which I expect there to be no reuse of existing GPLv2+ code. > If the latter, then my concern would be that if you need to import > GPLv2-only code, then you may need to write your code in a different > file. This may become messy to handle and some developer may end up to > be confused. Agreed, and I definitely want to minimize confusion. Ultimately I think situations like this will be unavoidable, though, since derivations will likely need to be made from both GPLv2-only and GPLv2+ code. > Cheers, Thanks, Shawn ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 16:07 ` Julien Grall 2023-06-09 16:20 ` Shawn Anastasio @ 2023-06-12 15:19 ` George Dunlap 2023-06-12 15:31 ` Julien Grall 2023-06-13 14:59 ` Shawn Anastasio 1 sibling, 2 replies; 27+ messages in thread From: George Dunlap @ 2023-06-12 15:19 UTC (permalink / raw) To: Julien Grall Cc: Shawn Anastasio, Andrew Cooper, Jan Beulich, tpearson, George Dunlap, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel [-- Attachment #1: Type: text/plain, Size: 4195 bytes --] On Fri, Jun 9, 2023 at 5:07 PM Julien Grall <julien@xen.org> wrote: > Hi Shawn, > > On 09/06/2023 16:01, Shawn Anastasio wrote: > > On Fri Jun 9, 2023 at 5:12 AM CDT, Julien Grall wrote: > >> Strictly speaking we can refuse any code. That count for license as > >> well. Anyway, I didn't request a change here. I merely pointed out that > >> any use of GPLv2+ should be justified because on Arm most of the people > >> don't pay attention on the license and pick the one from an existing > file. > > > > Hi Julien, > > > > The choice of GPLv2+ for many of the files in this patchset was indeed > > inherited from old IBM-written Xen code that the files in question were > > derived from. I did not realize it was permissible or even desirable to > > relicense those to GPLv2-only. > > > > As for the new files, GPLv2+ was chosen to remain consistent and to open > > the door for future derivations from GPLv2+ licensed code, either from > > the older Xen tree or from the Linux ppc tree, much of which is also > > licensed as GPLv2+. If it would reduce friction, these files could be > > relicensed to GPLv2-only. > > (Before someone points out, I know this is already a problem on other > part of Xen. But it would be ideal if we avoid spreading this mess on > new architectures :). > > Thanks for the explanations. To clarify, are you saying that all the > files will be GPLv2+ or just some? > > If the latter, then my concern would be that if you need to import > GPLv2-only code, then you may need to write your code in a different > file. This may become messy to handle and some developer may end up to > be confused. > > I am not a lawyer though, so you may want to check the implications here. > Shawn, Again sorry that you've sort of bumped a hornet's nest here. Just to clarify, the situation as I understand it is: 1. Large parts of Xen, being inherited from the Linux Kernel, are GPLv2-only; and the documentation clearly states that code is GPLv2-only unless explicitly stated otherwise. 2. Some individual files in Xen are labelled as GPLv2-or-later; but as they rely on the "only" files, Xen as a whole can only be compiled under a GPLv2 license. 3. New contributions to a file assumed to have the same license as the header of the file; i.e., the code contained in patches to GPLv2-or-later files is assumed to be granted according to a GPLv2-or-later license. 4. In the past, the legal teams of some contributors -- namely ARM -- were wary of the GPLv3; specifically the patent grant. Since ARM doesn't make anything themselves, their patents are literally their product; they need to be very careful of not accidentally granting them to the world. I think one thing ARM may have been afraid of at some point is one of their engineers accidentally submitting a patch to a GPLv2-or-later file which would, when taken with a GPLv3 (or GPLv4 license, once it comes out) cause them to lose too much control over their IP. My understanding is that Julien is afraid that if the "GPLv2-or-later" files start to proliferate, that companies like ARM will start to become more wary of contributing; and so has been generally trying to encourage new files to be labelled "GPLv2-only" unless there's a good reason to do otherwise. (Other issues like copying code from GPLv2-only are potential pitfalls as well, but probably less important.) HOWEVER, as Andrew says, there is no official policy at this point; all the document say is that GPLv2-only is the default unless explicitly stated otherwise. Furthermore, the concerns raised by ARM's legal team were nearly a decade ago; it's not clear to me whether they still care that much. All that to say: If you don't mind and feel that you can do so legally, then consider switching to GPLv2-only; but if you don't want to and/or feel that you can't do so legally, feel free to leave it as-is. Additionally, I think it would be good if the community *did* have a discussion about whether we want an official policy; so that either we can point people to the relevant doc (with explanation), or stop bothering about it. :-) -George [-- Attachment #2: Type: text/html, Size: 5004 bytes --] ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-12 15:19 ` George Dunlap @ 2023-06-12 15:31 ` Julien Grall 2023-06-13 14:59 ` Shawn Anastasio 1 sibling, 0 replies; 27+ messages in thread From: Julien Grall @ 2023-06-12 15:31 UTC (permalink / raw) To: George Dunlap Cc: Shawn Anastasio, Andrew Cooper, Jan Beulich, tpearson, George Dunlap, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel Hi George, Thanks for the summary! A couple of comments below. On 12/06/2023 16:19, George Dunlap wrote: > On Fri, Jun 9, 2023 at 5:07 PM Julien Grall <julien@xen.org> wrote: > >> Hi Shawn, >> >> On 09/06/2023 16:01, Shawn Anastasio wrote: >>> On Fri Jun 9, 2023 at 5:12 AM CDT, Julien Grall wrote: >>>> Strictly speaking we can refuse any code. That count for license as >>>> well. Anyway, I didn't request a change here. I merely pointed out that >>>> any use of GPLv2+ should be justified because on Arm most of the people >>>> don't pay attention on the license and pick the one from an existing >> file. >>> >>> Hi Julien, >>> >>> The choice of GPLv2+ for many of the files in this patchset was indeed >>> inherited from old IBM-written Xen code that the files in question were >>> derived from. I did not realize it was permissible or even desirable to >>> relicense those to GPLv2-only. >>> >>> As for the new files, GPLv2+ was chosen to remain consistent and to open >>> the door for future derivations from GPLv2+ licensed code, either from >>> the older Xen tree or from the Linux ppc tree, much of which is also >>> licensed as GPLv2+. If it would reduce friction, these files could be >>> relicensed to GPLv2-only. >> >> (Before someone points out, I know this is already a problem on other >> part of Xen. But it would be ideal if we avoid spreading this mess on >> new architectures :). >> >> Thanks for the explanations. To clarify, are you saying that all the >> files will be GPLv2+ or just some? >> >> If the latter, then my concern would be that if you need to import >> GPLv2-only code, then you may need to write your code in a different >> file. This may become messy to handle and some developer may end up to >> be confused. >> >> I am not a lawyer though, so you may want to check the implications here. >> > > Shawn, > > Again sorry that you've sort of bumped a hornet's nest here. > > Just to clarify, the situation as I understand it is: > > 1. Large parts of Xen, being inherited from the Linux Kernel, are > GPLv2-only; and the documentation clearly states that code is GPLv2-only > unless explicitly stated otherwise. > > 2. Some individual files in Xen are labelled as GPLv2-or-later; but as they > rely on the "only" files, Xen as a whole can only be compiled under a GPLv2 > license. > > 3. New contributions to a file assumed to have the same license as the > header of the file; i.e., the code contained in patches to GPLv2-or-later > files is assumed to be granted according to a GPLv2-or-later license. The new contribution here could be code imported from Linux that would be GPLv2-only in GPLv2-or-later file. It is not clear to me what would be the legal implication. > > 4. In the past, the legal teams of some contributors -- namely ARM -- were > wary of the GPLv3; specifically the patent grant. Since ARM doesn't make > anything themselves, their patents are literally their product; they need > to be very careful of not accidentally granting them to the world. I think > one thing ARM may have been afraid of at some point is one of their > engineers accidentally submitting a patch to a GPLv2-or-later file which > would, when taken with a GPLv3 (or GPLv4 license, once it comes out) cause > them to lose too much control over their IP. > > My understanding is that Julien is afraid that if the "GPLv2-or-later" > files start to proliferate, that companies like ARM will start to become > more wary of contributing; and so has been generally trying to encourage > new files to be labelled "GPLv2-only" unless there's a good reason to do > otherwise. (Other issues like copying code from GPLv2-only are potential > pitfalls as well, but probably less important.) There is that and also the fact that we now need to be more careful when importing code from Linux. In Shawn's case this is mitigated by the fact that the license in Xen files should match the one in Linux. > Additionally, I think it would be good if the community *did* have a > discussion about whether we want an official policy; so that either we can > point people to the relevant doc (with explanation), or stop bothering > about it. :-) +1. Do you think that would be a good topic for Xen Summit? Cheers, -- Julien Grall ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-12 15:19 ` George Dunlap 2023-06-12 15:31 ` Julien Grall @ 2023-06-13 14:59 ` Shawn Anastasio 1 sibling, 0 replies; 27+ messages in thread From: Shawn Anastasio @ 2023-06-13 14:59 UTC (permalink / raw) To: George Dunlap, Julien Grall Cc: Andrew Cooper, Jan Beulich, tpearson, George Dunlap, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel On Mon Jun 12, 2023 at 10:19 AM CDT, George Dunlap wrote: > Shawn, > > Again sorry that you've sort of bumped a hornet's nest here. > > Just to clarify, the situation as I understand it is: Hi George, No problem, and thank you for the detailed explanation. > HOWEVER, as Andrew says, there is no official policy at this point; all the > document say is that GPLv2-only is the default unless explicitly stated > otherwise. > > Furthermore, the concerns raised by ARM's legal team were nearly a decade > ago; it's not clear to me whether they still care that much. > > All that to say: If you don't mind and feel that you can do so legally, > then consider switching to GPLv2-only; but if you don't want to and/or feel > that you can't do so legally, feel free to leave it as-is. I'll definitely keep this in mind when creating new files that don't need to include GPLv2+ code. > -George Thanks, Shawn ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries 2023-06-09 9:22 ` Jan Beulich 2023-06-09 9:29 ` Andrew Cooper @ 2023-06-09 15:36 ` Shawn Anastasio 1 sibling, 0 replies; 27+ messages in thread From: Shawn Anastasio @ 2023-06-09 15:36 UTC (permalink / raw) To: Jan Beulich Cc: tpearson, Andrew Cooper, George Dunlap, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel On Fri Jun 9, 2023 at 4:22 AM CDT, Jan Beulich wrote: > > xen/arch/ppc/Kconfig.debug | 5 + > > xen/arch/ppc/Makefile | 3 +- > > xen/arch/ppc/boot_of.c | 122 +++++++++++++++++++++++ > > xen/arch/ppc/configs/openpower_defconfig | 1 + > > xen/arch/ppc/early_printk.c | 36 +++++++ > > xen/arch/ppc/include/asm/boot.h | 31 ++++++ > > xen/arch/ppc/include/asm/bug.h | 6 ++ > > xen/arch/ppc/include/asm/byteorder.h | 74 ++++++++++++++ > > xen/arch/ppc/include/asm/cache.h | 6 ++ > > xen/arch/ppc/include/asm/config.h | 3 + > > xen/arch/ppc/include/asm/early_printk.h | 14 +++ > > xen/arch/ppc/include/asm/processor.h | 54 +++++++++- > > xen/arch/ppc/include/asm/string.h | 6 ++ > > xen/arch/ppc/include/asm/types.h | 64 ++++++++++++ > > xen/arch/ppc/ppc64/asm-offsets.c | 55 ++++++++++ > > xen/arch/ppc/ppc64/head.S | 59 +++++++++++ > > xen/arch/ppc/setup.c | 20 +++- > > 17 files changed, 555 insertions(+), 4 deletions(-) > > create mode 100644 xen/arch/ppc/boot_of.c > > Unless required, in new additions we tend to prefer dashes over > underscores. In filenames it is pretty rare that dashes really need > avoiding. Thanks for pointing that out -- I'll fix this in v2. > > --- a/xen/arch/ppc/Kconfig.debug > > +++ b/xen/arch/ppc/Kconfig.debug > > @@ -0,0 +1,5 @@ > > +config EARLY_PRINTK > > + bool "Enable early printk" > > + default DEBUG > > + help > > + Enables early printk debug messages > > \ No newline at end of file > > There are many examples of this throughout the patch, which you want to > take care of. Ditto. > > --- /dev/null > > +++ b/xen/arch/ppc/boot_of.c > > @@ -0,0 +1,122 @@ > > +/* SPDX-License-Identifier: GPL-2.0-or-later */ > > By default we mean to use ... > > > --- /dev/null > > +++ b/xen/arch/ppc/early_printk.c > > @@ -0,0 +1,36 @@ > > +/* SPDX-License-Identifier: GPL-2.0 */ > > ... the more modern form of this (GPL-2.0-only). Anything deviating from > that may want justifying in the description. Got it -- I'll clean this up as well. > Jan Thanks, Shawn ^ permalink raw reply [flat|nested] 27+ messages in thread
* [PATCH 3/3] maintainers: Add PPC64 maintainer 2023-06-07 15:06 [PATCH 0/3] Initial support for Power Shawn Anastasio 2023-06-07 15:06 ` [PATCH 1/3] xen: Add files needed for minimal Power build Shawn Anastasio 2023-06-07 15:06 ` [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries Shawn Anastasio @ 2023-06-07 15:06 ` Shawn Anastasio 2023-06-09 9:04 ` Jan Beulich 2023-06-07 18:07 ` [PATCH 0/3] Initial support for Power Andrew Cooper 3 siblings, 1 reply; 27+ messages in thread From: Shawn Anastasio @ 2023-06-07 15:06 UTC (permalink / raw) To: xen-devel Cc: tpearson, Shawn Anastasio, Andrew Cooper, George Dunlap, Jan Beulich, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio Signed-off-by: Shawn Anastasio <shawnanastasio@raptorengineering.com> --- MAINTAINERS | 3 +++ 1 file changed, 3 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 1bb7a6a839..bcd36f9019 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -451,6 +451,9 @@ M: Wei Liu <wl@xen.org> S: Supported T: git https://xenbits.xenproject.org/git-http/ovmf.git +PPC64 +M: Shawn Anastasio <sanastasio@raptorengineering.com> + POWER MANAGEMENT M: Jan Beulich <jbeulich@suse.com> S: Supported -- 2.30.2 ^ permalink raw reply related [flat|nested] 27+ messages in thread
* Re: [PATCH 3/3] maintainers: Add PPC64 maintainer 2023-06-07 15:06 ` [PATCH 3/3] maintainers: Add PPC64 maintainer Shawn Anastasio @ 2023-06-09 9:04 ` Jan Beulich 2023-06-09 15:40 ` Shawn Anastasio 0 siblings, 1 reply; 27+ messages in thread From: Jan Beulich @ 2023-06-09 9:04 UTC (permalink / raw) To: Shawn Anastasio Cc: tpearson, Andrew Cooper, George Dunlap, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel On 07.06.2023 17:06, Shawn Anastasio wrote: > --- a/MAINTAINERS > +++ b/MAINTAINERS > @@ -451,6 +451,9 @@ M: Wei Liu <wl@xen.org> > S: Supported > T: git https://xenbits.xenproject.org/git-http/ovmf.git > > +PPC64 > +M: Shawn Anastasio <sanastasio@raptorengineering.com> > + > POWER MANAGEMENT > M: Jan Beulich <jbeulich@suse.com> > S: Supported Nit: You want to move your insertion further down, to maintain alphabetic sorting. You further want at least one F: line; see e.g. RISC-V's entry. Jan ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 3/3] maintainers: Add PPC64 maintainer 2023-06-09 9:04 ` Jan Beulich @ 2023-06-09 15:40 ` Shawn Anastasio 0 siblings, 0 replies; 27+ messages in thread From: Shawn Anastasio @ 2023-06-09 15:40 UTC (permalink / raw) To: Jan Beulich Cc: tpearson, Andrew Cooper, George Dunlap, Julien Grall, Stefano Stabellini, Wei Liu, Shawn Anastasio, xen-devel On Fri Jun 9, 2023 at 4:04 AM CDT, Jan Beulich wrote: > On 07.06.2023 17:06, Shawn Anastasio wrote: > > --- a/MAINTAINERS > > +++ b/MAINTAINERS > > @@ -451,6 +451,9 @@ M: Wei Liu <wl@xen.org> > > S: Supported > > T: git https://xenbits.xenproject.org/git-http/ovmf.git > > > > +PPC64 > > +M: Shawn Anastasio <sanastasio@raptorengineering.com> > > + > > POWER MANAGEMENT > > M: Jan Beulich <jbeulich@suse.com> > > S: Supported > > Nit: You want to move your insertion further down, to maintain alphabetic > sorting. You further want at least one F: line; see e.g. RISC-V's entry. Will fix in v2. > Jan Thanks, Shawn ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 0/3] Initial support for Power 2023-06-07 15:06 [PATCH 0/3] Initial support for Power Shawn Anastasio ` (2 preceding siblings ...) 2023-06-07 15:06 ` [PATCH 3/3] maintainers: Add PPC64 maintainer Shawn Anastasio @ 2023-06-07 18:07 ` Andrew Cooper 2023-06-07 19:01 ` Shawn Anastasio 3 siblings, 1 reply; 27+ messages in thread From: Andrew Cooper @ 2023-06-07 18:07 UTC (permalink / raw) To: Shawn Anastasio, xen-devel Cc: tpearson, George Dunlap, Jan Beulich, Julien Grall, Stefano Stabellini, Wei Liu On 07/06/2023 4:06 pm, Shawn Anastasio wrote: > Hello all, > > This patch series adds support for building a minimal image > (head.o-only) for Power ISA 2.07B+ (POWER8+) systems. The first patch > boots to an infinite loop and the second adds early serial console > support on pseries VMs, with bare metal support planned next. > > Since Xen previously had support for a much older version of the ISA in > version 3.2.3, we were able to carry over some headers and support > routines from that version. Unlike that initial port though, this effort > focuses solely on POWER8+ CPUs that are capable of running in Little > Endian mode. > > With an appropriate powerpc64le-linux-gnu cross-toolchain, the minimal > image can be built with: > > $ make XEN_TARGET_ARCH=ppc64 -C xen openpower_defconfig > $ make XEN_TARGET_ARCH=ppc64 SUBSYSTEMS=xen -C xen TARGET=ppc64/head.o > > The resulting head.o can then be booted in a standard QEMU/pseries VM: > > $ qemu-system-ppc64 -M pseries-5.2 -m 256M -kernel xen/ppc64/head.o \ > -vga none -serial mon:stdio -nographic > > Thanks, > Shawn > > Shawn Anastasio (3): > xen: Add files needed for minimal Power build > xen/ppc: Implement early serial printk on PaPR/pseries > maintainers: Add PPC64 maintainer Oh wow - this is a surprise, but certainly a good one. We've recently done just a similar exercise with RISCV64, starting with getting cross-compilation and a basic smoke test into our CI. We use gitlab CI, and one example is https://gitlab.com/xen-project/xen/-/pipelines/889871648 (search for riscv64 in amongst all the x86 and ARM). The configuration is all in tree, in the automation/ directory. Relevant files to copy/tweak are: automation/build/archlinux/current-riscv64.dockerfile (x86 cross compile container) automation/scripts/qemu-smoke-riscv64.sh (smoke test script) automation/gitlab-ci/{build,test}.yaml (wire the jobs up) The smoke test looks on stdout for "All set up" which can be any string put out via earlyprintk. If you look around in the Xen tree at bb62c25e3e5c and take the makefile override's in particular, you should be able to get `make -C xen build` work without any magic TARGET= overrides, and without most of the headers you've added in patch 1. The trick is to have a head.S which doesn't include any files (except the config.h it gets implicitly). We're still trying to do some re-arranging to the common / arch split to remove unnecessary boilerplate. Having a set of PPC headers too will make it easier to spot the rough edges in the current boundary. Looking to the future, where could XenProject get some real hardware to put into our CI systems? We'd want to do that in due course. I see you've nominated yourself as maintainer, which is of course fine. How much time do you have to put towards this? It is some part of a full time job, or just your own free time? Do you have any suggested reading for those of us who are invariably going to need to learn some of the CPU/platform/architecture details, but aren't really PPC-literate right now? Thanks, and welcome. ~Andrew ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 0/3] Initial support for Power 2023-06-07 18:07 ` [PATCH 0/3] Initial support for Power Andrew Cooper @ 2023-06-07 19:01 ` Shawn Anastasio 2023-06-07 19:30 ` Andrew Cooper 0 siblings, 1 reply; 27+ messages in thread From: Shawn Anastasio @ 2023-06-07 19:01 UTC (permalink / raw) To: Andrew Cooper, xen-devel Cc: tpearson, George Dunlap, Jan Beulich, Julien Grall, Stefano Stabellini, Wei Liu On Wed Jun 7, 2023 at 1:07 PM CDT, Andrew Cooper wrote: > Oh wow - this is a surprise, but certainly a good one. I'm glad to hear that! > We've recently done just a similar exercise with RISCV64, starting with > getting cross-compilation and a basic smoke test into our CI. I see. I used the initial RISC-V patch as a model for patch 1, though I didn't realize that the TARGET= override were unnecessary for getting a minimal image building. > We use gitlab CI, and one example is > https://gitlab.com/xen-project/xen/-/pipelines/889871648 (search for > riscv64 in amongst all the x86 and ARM). > > The configuration is all in tree, in the automation/ directory. > Relevant files to copy/tweak are: > > automation/build/archlinux/current-riscv64.dockerfile (x86 cross compile > container) > automation/scripts/qemu-smoke-riscv64.sh (smoke test script) > automation/gitlab-ci/{build,test}.yaml (wire the jobs up) > > The smoke test looks on stdout for "All set up" which can be any string > put out via earlyprintk. Thanks for the pointer -- I'll use this as a model to implement a similar smoke test for PPC64. > If you look around in the Xen tree at bb62c25e3e5c and take the makefile > override's in particular, you should be able to get `make -C xen build` > work without any magic TARGET= overrides, and without most of the > headers you've added in patch 1. The trick is to have a head.S which > doesn't include any files (except the config.h it gets implicitly). Perfect. I'll start work on a v2 without the TARGET= overrides and without the headers in patch #1. > We're still trying to do some re-arranging to the common / arch split to > remove unnecessary boilerplate. Having a set of PPC headers too will > make it easier to spot the rough edges in the current boundary. Great. I've been using arch/arm and arch/riscv as models for which headers to define (as well as the build errors that inevitably pop up when including xen/lib.h or similar). Is this a reasonable approach for now? > Looking to the future, where could XenProject get some real hardware to > put into our CI systems? We'd want to do that in due course. As an employee of Raptor Engineering, my recommendation for an OpenPOWER system would definitely be the Talos II or Blackbird families from Raptor Computing Systems: https://www.raptorcs.com/ :) > I see you've nominated yourself as maintainer, which is of course fine. > How much time do you have to put towards this? It is some part of a > full time job, or just your own free time? I'm currently working on the port full-time for Raptor Engineering. > Do you have any suggested reading for those of us who are invariably > going to need to learn some of the CPU/platform/architecture details, > but aren't really PPC-literate right now? We have a pretty extensive library of documentation covering the ISA as well as CPU and platform details available here: https://wiki.raptorcs.com/wiki/Category:Documentation > Thanks, and welcome. Thanks! Shawn ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 0/3] Initial support for Power 2023-06-07 19:01 ` Shawn Anastasio @ 2023-06-07 19:30 ` Andrew Cooper 2023-06-08 15:19 ` Shawn Anastasio 0 siblings, 1 reply; 27+ messages in thread From: Andrew Cooper @ 2023-06-07 19:30 UTC (permalink / raw) To: Shawn Anastasio, xen-devel Cc: tpearson, George Dunlap, Jan Beulich, Julien Grall, Stefano Stabellini, Wei Liu On 07/06/2023 8:01 pm, Shawn Anastasio wrote: >> If you look around in the Xen tree at bb62c25e3e5c and take the makefile >> override's in particular, you should be able to get `make -C xen build` >> work without any magic TARGET= overrides, and without most of the >> headers you've added in patch 1. The trick is to have a head.S which >> doesn't include any files (except the config.h it gets implicitly). > Perfect. I'll start work on a v2 without the TARGET= overrides and > without the headers in patch #1. Ok, in which case you probably want a first patch series doing: 1) New docker container (only - this wants to be taken early and set up in CI) 2) one-or-more initial build system + head.S infinite loop patch. Probably the maintainer patch too. 3) Wiring the build(s) into build.yaml You'll want to join the https://gitlab.com/xen-project/ and I can set you up with a place to run the full CI (we have a reasonably custom setup which the communal runners don't like). You may also want to hang out on oftc.org #xendevel and/or the matrix bridge thereof. Then a second patch series doing early printk, and adding the smoke test script and wiring it into test.yaml. ~Andrew ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH 0/3] Initial support for Power 2023-06-07 19:30 ` Andrew Cooper @ 2023-06-08 15:19 ` Shawn Anastasio 0 siblings, 0 replies; 27+ messages in thread From: Shawn Anastasio @ 2023-06-08 15:19 UTC (permalink / raw) To: Andrew Cooper, xen-devel Cc: tpearson, George Dunlap, Jan Beulich, Julien Grall, Stefano Stabellini, Wei Liu On Wed Jun 7, 2023 at 2:30 PM CDT, Andrew Cooper wrote: > Ok, in which case you probably want a first patch series doing: > > 1) New docker container (only - this wants to be taken early and set up > in CI) > 2) one-or-more initial build system + head.S infinite loop patch. > Probably the maintainer patch too. > 3) Wiring the build(s) into build.yaml > > You'll want to join the https://gitlab.com/xen-project/ and I can set > you up with a place to run the full CI (we have a reasonably custom > setup which the communal runners don't like). You may also want to hang > out on oftc.org #xendevel and/or the matrix bridge thereof. > > Then a second patch series doing early printk, and adding the smoke test > script and wiring it into test.yaml. Sounds good. I've gone ahead and requested access to that GitLab organization and will get to work on a series doing the things you've requested. P.S. It seems mail from my server is getting rejected by your server, so I'm unable to respond to you directly. Once I have access to my corporate email this should be resolved, but until then I'll reach out to you from a personal email address. Thanks. ^ permalink raw reply [flat|nested] 27+ messages in thread
end of thread, other threads:[~2023-06-13 14:59 UTC | newest] Thread overview: 27+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2023-06-07 15:06 [PATCH 0/3] Initial support for Power Shawn Anastasio 2023-06-07 15:06 ` [PATCH 1/3] xen: Add files needed for minimal Power build Shawn Anastasio 2023-06-09 9:15 ` Jan Beulich 2023-06-09 15:34 ` Shawn Anastasio 2023-06-07 15:06 ` [PATCH 2/3] xen/ppc: Implement early serial printk on PaPR/pseries Shawn Anastasio 2023-06-09 9:22 ` Jan Beulich 2023-06-09 9:29 ` Andrew Cooper 2023-06-09 9:38 ` Jan Beulich 2023-06-09 9:43 ` Julien Grall 2023-06-09 9:43 ` Andrew Cooper 2023-06-09 9:46 ` Julien Grall 2023-06-09 9:54 ` Andrew Cooper 2023-06-09 10:12 ` Julien Grall 2023-06-09 15:01 ` Shawn Anastasio 2023-06-09 16:07 ` Julien Grall 2023-06-09 16:20 ` Shawn Anastasio 2023-06-12 15:19 ` George Dunlap 2023-06-12 15:31 ` Julien Grall 2023-06-13 14:59 ` Shawn Anastasio 2023-06-09 15:36 ` Shawn Anastasio 2023-06-07 15:06 ` [PATCH 3/3] maintainers: Add PPC64 maintainer Shawn Anastasio 2023-06-09 9:04 ` Jan Beulich 2023-06-09 15:40 ` Shawn Anastasio 2023-06-07 18:07 ` [PATCH 0/3] Initial support for Power Andrew Cooper 2023-06-07 19:01 ` Shawn Anastasio 2023-06-07 19:30 ` Andrew Cooper 2023-06-08 15:19 ` Shawn Anastasio
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).