| Message ID | 20220120200735.2739543-2-atishp@rivosinc.com |
|---|---|
| State | New |
| Headers | show |
| Series | Privilege version update | expand |
On 1/21/22 7:07 AM, Atish Patra wrote: > Add the definition for ratified privileged specification version v1.12 > > Signed-off-by: Atish Patra <atishp@rivosinc.com> > --- > target/riscv/cpu.h | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h > index 4d630867650a..671f65100b1a 100644 > --- a/target/riscv/cpu.h > +++ b/target/riscv/cpu.h > @@ -82,6 +82,7 @@ enum { > > #define PRIV_VERSION_1_10_0 0x00011000 > #define PRIV_VERSION_1_11_0 0x00011100 > +#define PRIV_VERSION_1_12_0 0x00011200 Is there any good reason for defining things this way, as opposed to a simple enumeration? A simple enum would eliminate the need for > + /* The default privilege specification version supported is 1.10 */ > + if (!csr_min_priv) { > + csr_min_priv = PRIV_VERSION_1_10_0; > + } in patch 5. r~
On Sun, Jan 23, 2022 at 11:59 PM Richard Henderson < richard.henderson@linaro.org> wrote: > On 1/21/22 7:07 AM, Atish Patra wrote: > > Add the definition for ratified privileged specification version v1.12 > > > > Signed-off-by: Atish Patra <atishp@rivosinc.com> > > --- > > target/riscv/cpu.h | 1 + > > 1 file changed, 1 insertion(+) > > > > diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h > > index 4d630867650a..671f65100b1a 100644 > > --- a/target/riscv/cpu.h > > +++ b/target/riscv/cpu.h > > @@ -82,6 +82,7 @@ enum { > > > > #define PRIV_VERSION_1_10_0 0x00011000 > > #define PRIV_VERSION_1_11_0 0x00011100 > > +#define PRIV_VERSION_1_12_0 0x00011200 > > Is there any good reason for defining things this way, as opposed to a > simple enumeration? > A simple enum would eliminate the need for > > Agreed. A simple enum would be much nicer. I was just following the previous definition of PRIV_VERSION_1_10_0 & PRIV_VERSION_1_11_0. I am not sure about the reason behind this scheme. @Alistair Francis <alistair.francis@wdc.com> Is there any history behind this scheme ? or Are you okay if I change it ? > > + /* The default privilege specification version supported is 1.10 */ > > + if (!csr_min_priv) { > > + csr_min_priv = PRIV_VERSION_1_10_0; > > + } > > in patch 5. > > > r~ >
On Sat, Jan 29, 2022 at 10:57 AM Atish Kumar Patra <atishp@rivosinc.com> wrote: > > > > On Sun, Jan 23, 2022 at 11:59 PM Richard Henderson <richard.henderson@linaro.org> wrote: >> >> On 1/21/22 7:07 AM, Atish Patra wrote: >> > Add the definition for ratified privileged specification version v1.12 >> > >> > Signed-off-by: Atish Patra <atishp@rivosinc.com> >> > --- >> > target/riscv/cpu.h | 1 + >> > 1 file changed, 1 insertion(+) >> > >> > diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h >> > index 4d630867650a..671f65100b1a 100644 >> > --- a/target/riscv/cpu.h >> > +++ b/target/riscv/cpu.h >> > @@ -82,6 +82,7 @@ enum { >> > >> > #define PRIV_VERSION_1_10_0 0x00011000 >> > #define PRIV_VERSION_1_11_0 0x00011100 >> > +#define PRIV_VERSION_1_12_0 0x00011200 >> >> Is there any good reason for defining things this way, as opposed to a simple enumeration? >> A simple enum would eliminate the need for >> > > Agreed. A simple enum would be much nicer. I was just following the previous definition of > PRIV_VERSION_1_10_0 & PRIV_VERSION_1_11_0. > > I am not sure about the reason behind this scheme. > > @Alistair Francis Is there any history behind this scheme ? I don't think so > or Are you okay if I change it ? Yep :) Alistair
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index 4d630867650a..671f65100b1a 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -82,6 +82,7 @@ enum { #define PRIV_VERSION_1_10_0 0x00011000 #define PRIV_VERSION_1_11_0 0x00011100 +#define PRIV_VERSION_1_12_0 0x00011200 #define VEXT_VERSION_1_00_0 0x00010000
Add the definition for ratified privileged specification version v1.12 Signed-off-by: Atish Patra <atishp@rivosinc.com> --- target/riscv/cpu.h | 1 + 1 file changed, 1 insertion(+)