| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CAEn-LTqaK3pvjkTeNztDF8keJmr=GtMnCOFwZ=v4mHZd7fdMpw@mail.gmail.com>
Date: Tue, 21 Jun 2022 11:28:05 +0300
From: David Abdurachmanov <david.abdurachmanov@...il.com>
To: Nikita Shubin <nikita.shubin@...uefel.me>
Cc: Anup Patel <apatel@...tanamicro.com>,
Palmer Dabbelt <palmer@...belt.com>,
Paul Walmsley <paul.walmsley@...ive.com>,
Arnd Bergmann <arnd@...db.de>,
Atish Patra <atishp@...shpatra.org>,
Heinrich Schuchardt <heinrich.schuchardt@...onical.com>,
Anup Patel <anup@...infault.org>,
linux-riscv <linux-riscv@...ts.infradead.org>,
"linux-kernel@...r.kernel.org List" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] RISC-V: Add mvendorid, marchid, and mimpid to
/proc/cpuinfo output
On Tue, Jun 21, 2022 at 10:53 AM Nikita Shubin
<nikita.shubin@...uefel.me> wrote:
>
> Hello Anup!
>
> On Mon, 20 Jun 2022 17:25:49 +0530
> Anup Patel <apatel@...tanamicro.com> wrote:
>
> > Identifying the underlying RISC-V implementation can be important
> > for some of the user space applications. For example, the perf tool
> > uses arch specific CPU implementation id (i.e. CPUID) to select a
> > JSON file describing custom perf events on a CPU.
> >
> > Currently, there is no way to identify RISC-V implementation so we
> > add mvendorid, marchid, and mimpid to /proc/cpuinfo output.
>
> Tested on Sifive Unmatched:
>
> localhost / # cat /proc/cpuinfo
> processor : 0
> hart : 4
> isa : rv64imafdc
> mmu : sv39
> uarch : sifive,bullet0
> mvendorid : 0x489
> marchid : 0x8000000000000007
> mimpid : 0x20181004
>
> processor : 1
> hart : 1
> isa : rv64imafdc
> mmu : sv39
> uarch : sifive,bullet0
> mvendorid : 0x489
> marchid : 0x8000000000000007
> mimpid : 0x20181004
>
> processor : 2
> hart : 2
> isa : rv64imafdc
> mmu : sv39
> uarch : sifive,bullet0
> mvendorid : 0x489
> marchid : 0x8000000000000007
> mimpid : 0x20181004
>
> processor : 3
> hart : 3
> isa : rv64imafdc
> mmu : sv39
> uarch : sifive,bullet0
> mvendorid : 0x489
> marchid : 0x8000000000000007
> mimpid : 0x20181004
>
> mvendorid, marchid values match the register description in u74 manual.
>
> mimpid seems to be ok, through i can't find exact in U74/Unmatched docs.
SiFive main website -> Products -> Documentation -> U74-MC Core
Complex Manual -> Page 274: B.2 Machine Implementation ID Register
(mimpid).
This mimpid value is not in the table thus it's most likely not part
of a particular U74 generator release. Looking at this value we could
assume that U74 cores within FU740 predates the general availability.
Cheers,
david
>
> Tested-by: Nikita Shubin <n.shubin@...ro.com>
>
> >
> > Signed-off-by: Anup Patel <apatel@...tanamicro.com>
> > ---
> > arch/riscv/kernel/cpu.c | 51
> > +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 51
> > insertions(+)
> >
> > diff --git a/arch/riscv/kernel/cpu.c b/arch/riscv/kernel/cpu.c
> > index fba9e9f46a8c..c037b8691bbb 100644
> > --- a/arch/riscv/kernel/cpu.c
> > +++ b/arch/riscv/kernel/cpu.c
> > @@ -3,10 +3,13 @@
> > * Copyright (C) 2012 Regents of the University of California
> > */
> >
> > +#include <linux/cpu.h>
> > #include <linux/init.h>
> > #include <linux/seq_file.h>
> > #include <linux/of.h>
> > +#include <asm/csr.h>
> > #include <asm/hwcap.h>
> > +#include <asm/sbi.h>
> > #include <asm/smp.h>
> > #include <asm/pgtable.h>
> >
> > @@ -64,6 +67,50 @@ int riscv_of_parent_hartid(struct device_node
> > *node) }
> >
> > #ifdef CONFIG_PROC_FS
> > +
> > +struct riscv_cpuinfo {
> > + unsigned long mvendorid;
> > + unsigned long marchid;
> > + unsigned long mimpid;
> > +};
> > +static DEFINE_PER_CPU(struct riscv_cpuinfo, riscv_cpuinfo);
> > +
> > +static int riscv_cpuinfo_starting(unsigned int cpu)
> > +{
> > + struct riscv_cpuinfo *ci = this_cpu_ptr(&riscv_cpuinfo);
> > +
> > +#if defined(CONFIG_RISCV_SBI)
> > + ci->mvendorid = sbi_spec_is_0_1() ? 0 : sbi_get_mvendorid();
> > + ci->marchid = sbi_spec_is_0_1() ? 0 : sbi_get_marchid();
> > + ci->mimpid = sbi_spec_is_0_1() ? 0 : sbi_get_mimpid();
> > +#elif defined(CONFIG_RISCV_M_MODE)
> > + ci->mvendorid = csr_read(CSR_MVENDORID);
> > + ci->marchid = csr_read(CSR_MARCHID);
> > + ci->mimpid = csr_read(CSR_MIMPID);
> > +#else
> > + ci->mvendorid = 0;
> > + ci->marchid = 0;
> > + ci->mimpid = 0;
> > +#endif
> > +
> > + return 0;
> > +}
> > +
> > +static int __init riscv_cpuinfo_init(void)
> > +{
> > + int ret;
> > +
> > + ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN,
> > "riscv/cpuinfo:starting",
> > + riscv_cpuinfo_starting, NULL);
> > + if (ret < 0) {
> > + pr_err("cpuinfo: failed to register hotplug
> > callbacks.\n");
> > + return ret;
> > + }
> > +
> > + return 0;
> > +}
> > +device_initcall(riscv_cpuinfo_init);
> > +
> > #define __RISCV_ISA_EXT_DATA(UPROP, EXTID) \
> > { \
> > .uprop = #UPROP, \
> > @@ -178,6 +225,7 @@ static int c_show(struct seq_file *m, void *v)
> > {
> > unsigned long cpu_id = (unsigned long)v - 1;
> > struct device_node *node = of_get_cpu_node(cpu_id, NULL);
> > + struct riscv_cpuinfo *ci = per_cpu_ptr(&riscv_cpuinfo,
> > cpu_id); const char *compat, *isa;
> >
> > seq_printf(m, "processor\t: %lu\n", cpu_id);
> > @@ -188,6 +236,9 @@ static int c_show(struct seq_file *m, void *v)
> > if (!of_property_read_string(node, "compatible", &compat)
> > && strcmp(compat, "riscv"))
> > seq_printf(m, "uarch\t\t: %s\n", compat);
> > + seq_printf(m, "mvendorid\t: 0x%lx\n", ci->mvendorid);
> > + seq_printf(m, "marchid\t\t: 0x%lx\n", ci->marchid);
> > + seq_printf(m, "mimpid\t\t: 0x%lx\n", ci->mimpid);
> > seq_puts(m, "\n");
> > of_node_put(node);
> >
>
>
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@...ts.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv
Powered by blists - more mailing lists