|
| From: | Daniel Henrique Barboza |
| Subject: | Re: [RISC-V][tech-server-soc] [RFC 2/2] target/riscv: Add server platform reference cpu |
| Date: | Tue, 12 Mar 2024 09:52:33 -0300 |
| User-agent: | Mozilla Thunderbird |
On 3/12/24 09:33, Wu, Fei wrote:
On 3/8/2024 3:15 AM, Daniel Henrique Barboza wrote:On 3/7/24 04:36, Wu, Fei wrote:On 3/6/2024 9:26 PM, Wu, Fei wrote:On 3/5/2024 1:58 PM, Wu, Fei wrote:On 3/5/2024 3:43 AM, Daniel Henrique Barboza wrote:On 3/4/24 07:25, Fei Wu wrote:The harts requirements of RISC-V server platform [1] require RVA23 ISA profile support, plus Sv48, Svadu, H, Sscofmpf etc. This patch provides a virt CPU type (rvsp-ref) as compliant as possible. [1] https://github.com/riscv-non-isa/riscv-server-platform/blob/main/server_platform_requirements.adoc Signed-off-by: Fei Wu <fei2.wu@intel.com> ---> hw/riscv/server_platform_ref.c | 6 +++- target/riscv/cpu-qom.h | 1 + target/riscv/cpu.c | 62 ++++++++++++++++++++++++++++++++++ 3 files changed, 68 insertions(+), 1 deletion(-) diff --git a/hw/riscv/server_platform_ref.c b/hw/riscv/server_platform_ref.c index ae90c4b27a..52ec607cee 100644 --- a/hw/riscv/server_platform_ref.c +++ b/hw/riscv/server_platform_ref.c @@ -1205,11 +1205,15 @@ static void rvsp_ref_machine_class_init(ObjectClass *oc, void *data) { char str[128]; MachineClass *mc = MACHINE_CLASS(oc); + static const char * const valid_cpu_types[] = { + TYPE_RISCV_CPU_RVSP_REF, + }; mc->desc = "RISC-V Server SoC Reference board"; mc->init = rvsp_ref_machine_init; mc->max_cpus = RVSP_CPUS_MAX; - mc->default_cpu_type = TYPE_RISCV_CPU_BASE; + mc->default_cpu_type = TYPE_RISCV_CPU_RVSP_REF; + mc->valid_cpu_types = valid_cpu_types;I suggest introducing this patch first, then the new machine type that will use it as a default CPU. The reason is to facilitate future bisects. If we introduce the board first, a future bisect might hit the previous patch, the board will be run using RV64 instead of the correct CPU, and we'll have different results because of it.Good suggestion.mc->pci_allow_0_address = true; mc->default_nic = "e1000e"; mc->possible_cpu_arch_ids = riscv_numa_possible_cpu_arch_ids; diff --git a/target/riscv/cpu-qom.h b/target/riscv/cpu-qom.h index 3670cfe6d9..adb934d19e 100644 --- a/target/riscv/cpu-qom.h +++ b/target/riscv/cpu-qom.h @@ -49,6 +49,7 @@ #define TYPE_RISCV_CPU_SIFIVE_U54 RISCV_CPU_TYPE_NAME("sifive-u54") #define TYPE_RISCV_CPU_THEAD_C906 RISCV_CPU_TYPE_NAME("thead-c906") #define TYPE_RISCV_CPU_VEYRON_V1 RISCV_CPU_TYPE_NAME("veyron-v1") +#define TYPE_RISCV_CPU_RVSP_REF RISCV_CPU_TYPE_NAME("rvsp-ref") #define TYPE_RISCV_CPU_HOST RISCV_CPU_TYPE_NAME("host") OBJECT_DECLARE_CPU_TYPE(RISCVCPU, RISCVCPUClass, RISCV_CPU) diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index 5ff0192c52..bc91be702b 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -2282,6 +2282,67 @@ static void rva22s64_profile_cpu_init(Object *obj) RVA22S64.enabled = true; } + +static void rv64_rvsp_ref_cpu_init(Object *obj) +{ + CPURISCVState *env = &RISCV_CPU(obj)->env; + RISCVCPU *cpu = RISCV_CPU(obj); + + riscv_cpu_set_misa_ext(env, RVG | RVC | RVS | RVU | RVH | RVV); + + /* FIXME: change to 1.13 */ + env->priv_ver = PRIV_VERSION_1_12_0; + + /* RVA22U64 */ + cpu->cfg.mmu = true; + cpu->cfg.ext_zifencei = true; + cpu->cfg.ext_zicsr = true; + cpu->cfg.ext_zicntr = true; + cpu->cfg.ext_zihpm = true; + cpu->cfg.ext_zihintpause = true; + cpu->cfg.ext_zba = true; + cpu->cfg.ext_zbb = true; + cpu->cfg.ext_zbs = true; + cpu->cfg.zic64b = true; + cpu->cfg.ext_zicbom = true; + cpu->cfg.ext_zicbop = true; + cpu->cfg.ext_zicboz = true; + cpu->cfg.cbom_blocksize = 64; + cpu->cfg.cbop_blocksize = 64; + cpu->cfg.cboz_blocksize = 64; + cpu->cfg.ext_zfhmin = true; + cpu->cfg.ext_zkt = true;You can change this whole block with: RVA22U64.enabled = true; riscv_cpu_add_profiles() will check if we have a profile enabled and, if that's the case, we'll enable all its extensions in the CPU. In the near future, when we implement a proper RVA23 support, we'll be able to just do a single RVA23S64.enabled = true in this cpu_init(). But for now we can at least declare RVA22U64 (perhaps RVA22S64) support for this CPU.Hi Daniel, I'm not sure if it's a regression or the usage has been changed. I'm not able to use '-cpu rva22s64' on latest qemu (db596ae190).I did a quick git bisect and found that commit d06f28db6 "target/riscv: move 'mmu' to riscv_cpu_properties[]" disabled mmu by default, so that an explicit mmu option should be added to qemu command line like '-cpu rva22s64,mmu=true', I think rva22s64 should enable it by default.This is fixed in alistair/riscv-to-apply.next by this commit:Hi Daniel, The following still doesn't work for me with this commit (already upstreamed now) RVA22S64.enabled = true; cpu_set_profile() does nothing for vendor cpu as it checks riscv_cpu_is_vendor() at the beginning, and it's still not working even if it's removed. I think rvsp-ref is supposed to be TYPE_RISCV_VENDOR_CPU, changing to other type such as DYNAMIC doesn't work either, how to make this work?
Hmm yeah, set_profile() is filtering vendor CPUs because, although we are
exposing
profiles as command line options, users would be able to enable profiles in
vendor
CPUs, pottentialy enable extensions that the CPU doesn't have, and we don't want
that.
One way to keep this behavior but allowing the usage I mentioned for rvsp-ref is
to put everything that's after this point in cpu_set_profile():
if (!visit_type_bool(v, name, &value, errp)) {
return;
}
in it's own helper, and then we would be able to enable profile extensions for
vendor CPUs outside of this setter.
This is out of scope for your RFC though. I thought this was a use case we can
use for
rvsp-ref right now, but I was wrong. I believe we can keep what you've already
done
(i.e. setting all exts by hand). We can rework in a generic way to enable
profile
extensions for all CPUs later.
Thanks,
Daniel
Thanks, Fei.commit 5b8d66e6bf7904535ee277e9c70b332c4462f10a Author: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Date: Thu Feb 15 19:39:50 2024 -0300 target/riscv/tcg: set 'mmu' with 'satp' in cpu_set_profile() Recent changes in options handling removed the 'mmu' default the bare CPUs had, meaning that we must enable 'mmu' by hand when using the rva22s64 profile CPU. Given that this profile is setting a satp mode, it already implies that we need a 'mmu'. Enable the 'mmu' in this case. Thanks, DanielThanks, Fei.-- latest qemu, cannot see linux boot message and blocked $Q -machine virt -nographic -m 2G -smp 2 \ -cpu rva22s64 \ -bios /usr/lib/riscv64-linux-gnu/opensbi/generic/fw_jump.elf \ -kernel $Kernel Boot HART ID : 1 Boot HART Domain : root Boot HART Priv Version : v1.12 Boot HART Base ISA : rv64imafdc Boot HART ISA Extensions : time Boot HART PMP Count : 0 Boot HART PMP Granularity : 0 Boot HART PMP Address Bits: 0 Boot HART MHPM Count : 16 Boot HART MIDELEG : 0x0000000000000222 Boot HART MEDELEG : 0x000000000000b109 -- latest qemu, w/o rva22s64, looks good $Q -machine virt -nographic -m 2G -smp 2 \ -bios /usr/lib/riscv64-linux-gnu/opensbi/generic/fw_jump.elf \ -kernel $Kernel Boot HART ID : 0 Boot HART Domain : root Boot HART Priv Version : v1.12 Boot HART Base ISA : rv64imafdch Boot HART ISA Extensions : time,sstc Boot HART PMP Count : 16 Boot HART PMP Granularity : 4 Boot HART PMP Address Bits: 54 Boot HART MHPM Count : 16 Boot HART MIDELEG : 0x0000000000001666 Boot HART MEDELEG : 0x0000000000f0b509 [ 0.000000] Linux version 6.8.0-rc6+ (box@riscv-sw-lvm-1) (riscv64-linux-gnu-gcc (Ubuntu 9.4.0-1ubuntu1~20.04) 9.4.0, GNU ld (GNU Binutils for Ubuntu) 2.34) #17 SMP Wed Feb 28 08:38:42 UTC 2024 -- commit dfa3c4c57, patch to enable rva22s64, looks good $Q -machine virt -nographic -m 2G -smp 2 \ -cpu rva22s64 \ -bios /usr/lib/riscv64-linux-gnu/opensbi/generic/fw_jump.elf \ -kernel $Kernel Boot HART ID : 0 Boot HART Domain : root Boot HART Priv Version : v1.12 Boot HART Base ISA : rv64imafdc Boot HART ISA Extensions : time Boot HART PMP Count : 16 Boot HART PMP Granularity : 4 Boot HART PMP Address Bits: 54 Boot HART MHPM Count : 16 Boot HART MIDELEG : 0x0000000000000222 Boot HART MEDELEG : 0x000000000000b109 [ 0.000000] Linux version 6.8.0-rc6+ (box@riscv-sw-lvm-1) (riscv64-linux-gnu-gcc (Ubuntu 9.4.0-1ubuntu1~20.04) 9.4.0, GNU ld (GNU Binutils for Ubuntu) 2.34) #17 SMP Wed Feb 28 08:38:42 UTC 2024 Thanks, FeiLet me try. Thanks, Fei.Thanks, Daniel+ + /* RVA23U64 */ + cpu->cfg.ext_zvfhmin = true; + cpu->cfg.ext_zvbb = true; + cpu->cfg.ext_zvkt = true; + cpu->cfg.ext_zihintntl = true; + cpu->cfg.ext_zicond = true; + cpu->cfg.ext_zcb = true; + cpu->cfg.ext_zfa = true; + cpu->cfg.ext_zawrs = true; + + /* RVA23S64 */ + cpu->cfg.ext_zifencei = true; + cpu->cfg.svade = true; + cpu->cfg.ext_svpbmt = true; + cpu->cfg.ext_svinval = true; + cpu->cfg.ext_svnapot = true; + cpu->cfg.ext_sstc = true; + cpu->cfg.ext_sscofpmf = true; + cpu->cfg.ext_smstateen = true; + + cpu->cfg.ext_smaia = true; + cpu->cfg.ext_ssaia = true; + + /* Server Platform */ +#ifndef CONFIG_USER_ONLY + set_satp_mode_max_supported(cpu, VM_1_10_SV48); +#endif + cpu->cfg.ext_svadu = true; + cpu->cfg.ext_zkr = true; +} #endif static const gchar *riscv_gdb_arch_name(CPUState *cs) @@ -2547,6 +2608,7 @@ static const TypeInfo riscv_cpu_type_infos[] = { DEFINE_BARE_CPU(TYPE_RISCV_CPU_RV64E, MXL_RV64, rv64e_bare_cpu_init), DEFINE_PROFILE_CPU(TYPE_RISCV_CPU_RVA22U64, MXL_RV64, rva22u64_profile_cpu_init), DEFINE_PROFILE_CPU(TYPE_RISCV_CPU_RVA22S64, MXL_RV64, rva22s64_profile_cpu_init), + DEFINE_VENDOR_CPU(TYPE_RISCV_CPU_RVSP_REF, MXL_RV64, rv64_rvsp_ref_cpu_init), #endif };-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#135): https://lists.riscv.org/g/tech-server-soc/message/135 Mute This Topic: https://lists.riscv.org/mt/104719379/7152971 Group Owner: tech-server-soc+owner@lists.riscv.org Unsubscribe: https://lists.riscv.org/g/tech-server-soc/leave/12737993/7152971/1793629631/xyzzy [fei2.wu@intel.com] -=-=-=-=-=-=-=-=-=-=-=-
| [Prev in Thread] | Current Thread | [Next in Thread] |