From: Alistair Francis <alistair23@gmail.com>
To: Jose Martins <josemartins90@gmail.com>
Cc: "open list:RISC-V TCG CPUs" <qemu-riscv@nongnu.org>,
Sagar Karandikar <sagark@eecs.berkeley.edu>,
Bastian Koppelmann <kbastian@mail.uni-paderborn.de>,
"qemu-devel@nongnu.org Developers" <qemu-devel@nongnu.org>,
Alistair Francis <Alistair.Francis@wdc.com>,
Palmer Dabbelt <palmer@dabbelt.com>,
LIU Zhiwei <zhiwei_liu@c-sky.com>
Subject: Re: [PATCH v3] target/riscv: fix VS interrupts forwarding to HS
Date: Mon, 18 Oct 2021 11:13:42 +1000 [thread overview]
Message-ID: <CAKmqyKMb-VNiQZv-ZFeCMwPQ3hkm1ODn_w55pGdBgpxtwg8tnw@mail.gmail.com> (raw)
In-Reply-To: <CAC41xo12s+uHEyLkc6vO07iBAC84o6aHBFFdqxDm3NQjBYjuAw@mail.gmail.com>
On Mon, Oct 18, 2021 at 6:30 AM Jose Martins <josemartins90@gmail.com> wrote:
>
> Hello Zhiwei and Alistair,
>
> I noticed this patch did not make it upstream, contrarily to a couple
> other patches I submitted around the same time. Is there something
> else needed from my side to push this forward?
From your last response I thought you sent a different series that
replaces this patch. If that's not the case do you mind sending this
patch again?
Alistair
>
> Best,
> José
>
> On Wed, 2 Jun 2021 at 20:14, Jose Martins <josemartins90@gmail.com> wrote:
> >
> > Hello Zhiwei and Alistair,
> >
> > I went for a middle-ground solution. I divided the patch into two: one
> > fixes the vs interrupt forwarding to the hypervisor, the other removes
> > the unnecessary force exception stuff. I just submitted the patch
> > series. I hope it's ok with you.
> >
> > José
> >
> > On Fri, 28 May 2021 at 01:36, LIU Zhiwei <zhiwei_liu@c-sky.com> wrote:
> > >
> > >
> > > On 5/28/21 6:34 AM, Alistair Francis wrote:
> > > > On Sun, May 23, 2021 at 1:45 AM Jose Martins <josemartins90@gmail.com> wrote:
> > > >> VS interrupts (2, 6, 10) were not correctly forwarded to hs-mode when
> > > >> not delegated in hideleg (which was not being taken into account). This
> > > >> was mainly because hs level sie was not always considered enabled when
> > > >> it should. The spec states that "Interrupts for higher-privilege modes,
> > > >> y>x, are always globally enabled regardless of the setting of the global
> > > >> yIE bit for the higher-privilege mode." and also "For purposes of
> > > >> interrupt global enables, HS-mode is considered more privileged than
> > > >> VS-mode, and VS-mode is considered more privileged than VU-mode".
> > > >>
> > > >> These interrupts should be treated the same as any other kind of
> > > >> exception. Therefore, there is no need to "force an hs exception" as the
> > > >> current privilege level, the state of the global ie and of the
> > > >> delegation registers should be enough to route the interrupt to the
> > > >> appropriate privilege level in riscv_cpu_do_interrupt. Also, these
> > > >> interrupts never target m-mode, which is guaranteed by the hardwiring
> > > >> of the corresponding bits in mideleg. The same is true for synchronous
> > > >> exceptions, specifically, guest page faults which must be hardwired in
> > > >> to zero hedeleg. As such the hs_force_except mechanism can be removed.
> > > >>
> > > >> Signed-off-by: Jose Martins <josemartins90@gmail.com>
> > > > This looks right to me, but this was one of the most confusing parts
> > > > of the implementation. I also don't think the patch is too long as
> > > > it's mostly just deleting stuff.
> > > >
> > > > I don't fully understand your concerns Zhiwei, do you mind stating them again?
> > >
> > > Hi Alistair,
> > >
> > > My main concern is the commit message is to complex.
> > >
> > > I also have a question why force hs exception in current code.
> > > Then we can give a brief commit message.
> > >
> > > Best Regards,
> > > Zhiwei
> > >
> > > >
> > > > Alistair
> > > >
> > > >> ---
> > > >> This version of the patch also removes the uneeded hs_force_except
> > > >> functions, variables and macro.
> > > >>
> > > >> target/riscv/cpu.h | 2 --
> > > >> target/riscv/cpu_bits.h | 6 -----
> > > >> target/riscv/cpu_helper.c | 54 +++++++--------------------------------
> > > >> 3 files changed, 9 insertions(+), 53 deletions(-)
> > > >>
> > > >> diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> > > >> index 0a33d387ba..a30a64241a 100644
> > > >> --- a/target/riscv/cpu.h
> > > >> +++ b/target/riscv/cpu.h
> > > >> @@ -337,8 +337,6 @@ bool riscv_cpu_exec_interrupt(CPUState *cs, int interrupt_request);
> > > >> bool riscv_cpu_fp_enabled(CPURISCVState *env);
> > > >> bool riscv_cpu_virt_enabled(CPURISCVState *env);
> > > >> void riscv_cpu_set_virt_enabled(CPURISCVState *env, bool enable);
> > > >> -bool riscv_cpu_force_hs_excep_enabled(CPURISCVState *env);
> > > >> -void riscv_cpu_set_force_hs_excep(CPURISCVState *env, bool enable);
> > > >> bool riscv_cpu_two_stage_lookup(int mmu_idx);
> > > >> int riscv_cpu_mmu_index(CPURISCVState *env, bool ifetch);
> > > >> hwaddr riscv_cpu_get_phys_page_debug(CPUState *cpu, vaddr addr);
> > > >> diff --git a/target/riscv/cpu_bits.h b/target/riscv/cpu_bits.h
> > > >> index caf4599207..7322f54157 100644
> > > >> --- a/target/riscv/cpu_bits.h
> > > >> +++ b/target/riscv/cpu_bits.h
> > > >> @@ -462,12 +462,6 @@
> > > >>
> > > >> /* Virtulisation Register Fields */
> > > >> #define VIRT_ONOFF 1
> > > >> -/* This is used to save state for when we take an exception. If this is set
> > > >> - * that means that we want to force a HS level exception (no matter what the
> > > >> - * delegation is set to). This will occur for things such as a second level
> > > >> - * page table fault.
> > > >> - */
> > > >> -#define FORCE_HS_EXCEP 2
> > > >>
> > > >> /* RV32 satp CSR field masks */
> > > >> #define SATP32_MODE 0x80000000
> > > >> diff --git a/target/riscv/cpu_helper.c b/target/riscv/cpu_helper.c
> > > >> index 21c54ef561..babe3d844b 100644
> > > >> --- a/target/riscv/cpu_helper.c
> > > >> +++ b/target/riscv/cpu_helper.c
> > > >> @@ -38,36 +38,24 @@ int riscv_cpu_mmu_index(CPURISCVState *env, bool ifetch)
> > > >> #ifndef CONFIG_USER_ONLY
> > > >> static int riscv_cpu_local_irq_pending(CPURISCVState *env)
> > > >> {
> > > >> - target_ulong irqs;
> > > >> + target_ulong virt_enabled = riscv_cpu_virt_enabled(env);
> > > >>
> > > >> target_ulong mstatus_mie = get_field(env->mstatus, MSTATUS_MIE);
> > > >> target_ulong mstatus_sie = get_field(env->mstatus, MSTATUS_SIE);
> > > >> - target_ulong hs_mstatus_sie = get_field(env->mstatus_hs, MSTATUS_SIE);
> > > >>
> > > >> - target_ulong pending = env->mip & env->mie &
> > > >> - ~(MIP_VSSIP | MIP_VSTIP | MIP_VSEIP);
> > > >> - target_ulong vspending = (env->mip & env->mie &
> > > >> - (MIP_VSSIP | MIP_VSTIP | MIP_VSEIP));
> > > >> + target_ulong pending = env->mip & env->mie;
> > > >>
> > > >> target_ulong mie = env->priv < PRV_M ||
> > > >> (env->priv == PRV_M && mstatus_mie);
> > > >> target_ulong sie = env->priv < PRV_S ||
> > > >> (env->priv == PRV_S && mstatus_sie);
> > > >> - target_ulong hs_sie = env->priv < PRV_S ||
> > > >> - (env->priv == PRV_S && hs_mstatus_sie);
> > > >> + target_ulong hsie = virt_enabled || sie;
> > > >> + target_ulong vsie = virt_enabled && sie;
> > > >>
> > > >> - if (riscv_cpu_virt_enabled(env)) {
> > > >> - target_ulong pending_hs_irq = pending & -hs_sie;
> > > >> -
> > > >> - if (pending_hs_irq) {
> > > >> - riscv_cpu_set_force_hs_excep(env, FORCE_HS_EXCEP);
> > > >> - return ctz64(pending_hs_irq);
> > > >> - }
> > > >> -
> > > >> - pending = vspending;
> > > >> - }
> > > >> -
> > > >> - irqs = (pending & ~env->mideleg & -mie) | (pending & env->mideleg & -sie);
> > > >> + target_ulong irqs =
> > > >> + (pending & ~env->mideleg & -mie) |
> > > >> + (pending & env->mideleg & ~env->hideleg & -hsie) |
> > > >> + (pending & env->mideleg & env->hideleg & -vsie);
> > > >>
> > > >> if (irqs) {
> > > >> return ctz64(irqs); /* since non-zero */
> > > >> @@ -190,24 +178,6 @@ void riscv_cpu_set_virt_enabled(CPURISCVState *env, bool enable)
> > > >> env->virt = set_field(env->virt, VIRT_ONOFF, enable);
> > > >> }
> > > >>
> > > >> -bool riscv_cpu_force_hs_excep_enabled(CPURISCVState *env)
> > > >> -{
> > > >> - if (!riscv_has_ext(env, RVH)) {
> > > >> - return false;
> > > >> - }
> > > >> -
> > > >> - return get_field(env->virt, FORCE_HS_EXCEP);
> > > >> -}
> > > >> -
> > > >> -void riscv_cpu_set_force_hs_excep(CPURISCVState *env, bool enable)
> > > >> -{
> > > >> - if (!riscv_has_ext(env, RVH)) {
> > > >> - return;
> > > >> - }
> > > >> -
> > > >> - env->virt = set_field(env->virt, FORCE_HS_EXCEP, enable);
> > > >> -}
> > > >> -
> > > >> bool riscv_cpu_two_stage_lookup(int mmu_idx)
> > > >> {
> > > >> return mmu_idx & TB_FLAGS_PRIV_HYP_ACCESS_MASK;
> > > >> @@ -896,7 +866,6 @@ void riscv_cpu_do_interrupt(CPUState *cs)
> > > >>
> > > >> RISCVCPU *cpu = RISCV_CPU(cs);
> > > >> CPURISCVState *env = &cpu->env;
> > > >> - bool force_hs_execp = riscv_cpu_force_hs_excep_enabled(env);
> > > >> uint64_t s;
> > > >>
> > > >> /* cs->exception is 32-bits wide unlike mcause which is XLEN-bits wide
> > > >> @@ -925,8 +894,6 @@ void riscv_cpu_do_interrupt(CPUState *cs)
> > > >> case RISCV_EXCP_INST_GUEST_PAGE_FAULT:
> > > >> case RISCV_EXCP_LOAD_GUEST_ACCESS_FAULT:
> > > >> case RISCV_EXCP_STORE_GUEST_AMO_ACCESS_FAULT:
> > > >> - force_hs_execp = true;
> > > >> - /* fallthrough */
> > > >> case RISCV_EXCP_INST_ADDR_MIS:
> > > >> case RISCV_EXCP_INST_ACCESS_FAULT:
> > > >> case RISCV_EXCP_LOAD_ADDR_MIS:
> > > >> @@ -985,8 +952,7 @@ void riscv_cpu_do_interrupt(CPUState *cs)
> > > >> env->hstatus = set_field(env->hstatus, HSTATUS_GVA, 0);
> > > >> }
> > > >>
> > > >> - if (riscv_cpu_virt_enabled(env) && ((hdeleg >> cause) & 1) &&
> > > >> - !force_hs_execp) {
> > > >> + if (riscv_cpu_virt_enabled(env) && ((hdeleg >> cause) & 1)) {
> > > >> /* Trap to VS mode */
> > > >> /*
> > > >> * See if we need to adjust cause. Yes if its VS mode interrupt
> > > >> @@ -1008,7 +974,6 @@ void riscv_cpu_do_interrupt(CPUState *cs)
> > > >> htval = env->guest_phys_fault_addr;
> > > >>
> > > >> riscv_cpu_set_virt_enabled(env, 0);
> > > >> - riscv_cpu_set_force_hs_excep(env, 0);
> > > >> } else {
> > > >> /* Trap into HS mode */
> > > >> env->hstatus = set_field(env->hstatus, HSTATUS_SPV, false);
> > > >> @@ -1044,7 +1009,6 @@ void riscv_cpu_do_interrupt(CPUState *cs)
> > > >>
> > > >> /* Trapping to M mode, virt is disabled */
> > > >> riscv_cpu_set_virt_enabled(env, 0);
> > > >> - riscv_cpu_set_force_hs_excep(env, 0);
> > > >> }
> > > >>
> > > >> s = env->mstatus;
> > > >> --
> > > >> 2.30.2
> > > >>
> > > >>
next prev parent reply other threads:[~2021-10-18 1:16 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-22 15:44 [PATCH v3] target/riscv: fix VS interrupts forwarding to HS Jose Martins
2021-05-26 7:02 ` LIU Zhiwei
2021-05-26 11:50 ` Jose Martins
2021-05-27 8:40 ` LIU Zhiwei
2021-05-27 22:34 ` Alistair Francis
2021-05-28 0:36 ` LIU Zhiwei
2021-06-02 19:14 ` Jose Martins
2021-10-17 20:30 ` Jose Martins
2021-10-18 1:13 ` Alistair Francis [this message]
2021-10-25 9:47 ` Jose Martins
2021-10-26 6:57 ` Alistair Francis
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=CAKmqyKMb-VNiQZv-ZFeCMwPQ3hkm1ODn_w55pGdBgpxtwg8tnw@mail.gmail.com \
--to=alistair23@gmail.com \
--cc=Alistair.Francis@wdc.com \
--cc=josemartins90@gmail.com \
--cc=kbastian@mail.uni-paderborn.de \
--cc=palmer@dabbelt.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-riscv@nongnu.org \
--cc=sagark@eecs.berkeley.edu \
--cc=zhiwei_liu@c-sky.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).