From: Alexey Baturo <baturo.alexey@gmail.com>
To: Richard Henderson <richard.henderson@linaro.org>
Cc: zhiwei_liu@linux.alibaba.com, palmer@dabbelt.com,
Alistair.Francis@wdc.com, sagark@eecs.berkeley.edu,
kbastian@mail.uni-paderborn.de, qemu-devel@nongnu.org,
qemu-riscv@nongnu.org
Subject: Re: [PATCH 3/6] target/riscv: Add pointer masking tb flags
Date: Fri, 22 Dec 2023 20:59:32 +0300 [thread overview]
Message-ID: <CAFukJ-DNx0EMpG18vtA2-2DSdROp8nOOOOK_q-EOpu_=Gjp+OA@mail.gmail.com> (raw)
In-Reply-To: <6da02b36-d82f-49ae-ae85-4642486dabe7@linaro.org>
[-- Attachment #1: Type: text/plain, Size: 5403 bytes --]
Hi Richard,
Thanks for the suggestion.
If it's ok to consume another bit(3 bits total) for Pointer Masking flags,
I'll do it.
>so that the translator can see the true width of the address
I guess I'll need a helper to calculate the exact number of bits to
shift(0, 7 or 16) based on those 2 extracted bits. Is it ok with you?
Thanks
пт, 22 дек. 2023 г. в 01:49, Richard Henderson <richard.henderson@linaro.org
>:
> On 12/21/23 21:40, Alexey Baturo wrote:
> > From: Alexey Baturo <baturo.alexey@gmail.com>
> >
> > Signed-off-by: Alexey Baturo <baturo.alexey@gmail.com>
> > ---
> > target/riscv/cpu.h | 19 +++++++++++++------
> > target/riscv/cpu_helper.c | 4 ++++
> > target/riscv/translate.c | 10 ++++++++++
> > 3 files changed, 27 insertions(+), 6 deletions(-)
> >
> > diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> > index f49d4aa52c..2099168950 100644
> > --- a/target/riscv/cpu.h
> > +++ b/target/riscv/cpu.h
> > @@ -390,6 +390,10 @@ struct CPUArchState {
> > target_ulong senvcfg;
> > uint64_t henvcfg;
> > #endif
> > + /* current number of masked top bits by pointer masking */
> > + target_ulong pm_pmlen;
> > + /* if pointer masking should do sign extension */
> > + bool pm_signext;
> >
> > /* Fields from here on are preserved across CPU reset. */
> > QEMUTimer *stimer; /* Internal timer for S-mode interrupt */
> > @@ -538,14 +542,17 @@ FIELD(TB_FLAGS, VILL, 14, 1)
> > FIELD(TB_FLAGS, VSTART_EQ_ZERO, 15, 1)
> > /* The combination of MXL/SXL/UXL that applies to the current cpu
> mode. */
> > FIELD(TB_FLAGS, XL, 16, 2)
> > -FIELD(TB_FLAGS, VTA, 18, 1)
> > -FIELD(TB_FLAGS, VMA, 19, 1)
> > +/* If pointer masking should be applied and address sign extended */
> > +FIELD(TB_FLAGS, PM_ENABLED, 18, 1)
>
> I think it would be better add the entire two bit field here, so that the
> translator can
> see the true width of the address. You can then use tcg_gen_{s}extract_tl
> to perform the
> truncation. At which point the 'target_ulong pm_pmlen' is not required.
>
>
> r~
>
> > +FIELD(TB_FLAGS, PM_SIGNEXTEND, 19, 1)
> > +FIELD(TB_FLAGS, VTA, 20, 1)
> > +FIELD(TB_FLAGS, VMA, 21, 1)
> > /* Native debug itrigger */
> > -FIELD(TB_FLAGS, ITRIGGER, 20, 1)
> > +FIELD(TB_FLAGS, ITRIGGER, 22, 1)
> > /* Virtual mode enabled */
> > -FIELD(TB_FLAGS, VIRT_ENABLED, 21, 1)
> > -FIELD(TB_FLAGS, PRIV, 22, 2)
> > -FIELD(TB_FLAGS, AXL, 24, 2)
> > +FIELD(TB_FLAGS, VIRT_ENABLED, 23, 1)
> > +FIELD(TB_FLAGS, PRIV, 24, 2)
> > +FIELD(TB_FLAGS, AXL, 25, 2)
> >
> > #ifdef TARGET_RISCV32
> > #define riscv_cpu_mxl(env) ((void)(env), MXL_RV32)
> > diff --git a/target/riscv/cpu_helper.c b/target/riscv/cpu_helper.c
> > index a3d477d226..79cddbd930 100644
> > --- a/target/riscv/cpu_helper.c
> > +++ b/target/riscv/cpu_helper.c
> > @@ -135,6 +135,10 @@ void cpu_get_tb_cpu_state(CPURISCVState *env, vaddr
> *pc,
> > flags = FIELD_DP32(flags, TB_FLAGS, VS, vs);
> > flags = FIELD_DP32(flags, TB_FLAGS, XL, env->xl);
> > flags = FIELD_DP32(flags, TB_FLAGS, AXL, cpu_address_xl(env));
> > + if (env->pm_pmlen != 0) {
> > + flags = FIELD_DP32(flags, TB_FLAGS, PM_ENABLED, 1);
> > + }
> > + flags = FIELD_DP32(flags, TB_FLAGS, PM_SIGNEXTEND, env->pm_signext);
> >
> > *pflags = flags;
> > }
> > diff --git a/target/riscv/translate.c b/target/riscv/translate.c
> > index 6b4b9a671c..4c0d526b58 100644
> > --- a/target/riscv/translate.c
> > +++ b/target/riscv/translate.c
> > @@ -42,6 +42,8 @@ static TCGv cpu_gpr[32], cpu_gprh[32], cpu_pc, cpu_vl,
> cpu_vstart;
> > static TCGv_i64 cpu_fpr[32]; /* assume F and D extensions */
> > static TCGv load_res;
> > static TCGv load_val;
> > +/* number of top masked address bits by pointer masking extension */
> > +static TCGv pm_pmlen;
> >
> > /*
> > * If an operation is being performed on less than TARGET_LONG_BITS,
> > @@ -103,6 +105,9 @@ typedef struct DisasContext {
> > bool vl_eq_vlmax;
> > CPUState *cs;
> > TCGv zero;
> > + /* pointer masking extension */
> > + bool pm_enabled;
> > + bool pm_signext;
> > /* Use icount trigger for native debug */
> > bool itrigger;
> > /* FRM is known to contain a valid value. */
> > @@ -1176,6 +1181,8 @@ static void
> riscv_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
> > ctx->xl = FIELD_EX32(tb_flags, TB_FLAGS, XL);
> > ctx->address_xl = FIELD_EX32(tb_flags, TB_FLAGS, AXL);
> > ctx->cs = cs;
> > + ctx->pm_enabled = FIELD_EX32(tb_flags, TB_FLAGS, PM_ENABLED);
> > + ctx->pm_signext = FIELD_EX32(tb_flags, TB_FLAGS, PM_SIGNEXTEND);
> > ctx->itrigger = FIELD_EX32(tb_flags, TB_FLAGS, ITRIGGER);
> > ctx->zero = tcg_constant_tl(0);
> > ctx->virt_inst_excp = false;
> > @@ -1307,4 +1314,7 @@ void riscv_translate_init(void)
> > "load_res");
> > load_val = tcg_global_mem_new(tcg_env, offsetof(CPURISCVState,
> load_val),
> > "load_val");
> > + /* Assign var with number of pointer masking masked bits to tcg
> global */
> > + pm_pmlen = tcg_global_mem_new(tcg_env, offsetof(CPURISCVState,
> pm_pmlen),
> > + "pmlen");
> > }
>
>
[-- Attachment #2: Type: text/html, Size: 6721 bytes --]
next prev parent reply other threads:[~2023-12-22 18:00 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-21 10:40 [PATCH 0/6] Pointer Masking update for Zjpm v0.8 Alexey Baturo
2023-12-21 10:40 ` [PATCH 1/6] target/riscv: Remove obsolete pointer masking extension code Alexey Baturo
2023-12-21 10:40 ` [PATCH 2/6] target/riscv: Add new CSR fields for S{sn, mn, m}pm extensions as part of Zjpm v0.8 Alexey Baturo
2023-12-21 10:40 ` [PATCH 3/6] target/riscv: Add pointer masking tb flags Alexey Baturo
2023-12-21 22:49 ` Richard Henderson
2023-12-22 17:59 ` Alexey Baturo [this message]
2023-12-21 10:40 ` [PATCH 4/6] target/riscv: Add functions to calculate current number of masked bits for pointer masking Alexey Baturo
2023-12-21 10:40 ` [PATCH 5/6] target/riscv: Update address modify functions to take into account " Alexey Baturo
2023-12-21 10:40 ` [PATCH 6/6] target/riscv: Enable updates for pointer masking variables and thus enable pointer masking extension Alexey Baturo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAFukJ-DNx0EMpG18vtA2-2DSdROp8nOOOOK_q-EOpu_=Gjp+OA@mail.gmail.com' \
--to=baturo.alexey@gmail.com \
--cc=Alistair.Francis@wdc.com \
--cc=kbastian@mail.uni-paderborn.de \
--cc=palmer@dabbelt.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-riscv@nongnu.org \
--cc=richard.henderson@linaro.org \
--cc=sagark@eecs.berkeley.edu \
--cc=zhiwei_liu@linux.alibaba.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).