From: Palmer Dabbelt <palmer@sifive.com>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: qemu-riscv@nongnu.org, qemu-devel@nongnu.org,
Michael Clark <mjc@sifive.com>,
Alistair Francis <alistair.francis@wdc.com>,
Palmer Dabbelt <palmer@sifive.com>
Subject: [Qemu-devel] [PULL 2/4] RISC-V: Implement atomic mip/sip CSR updates
Date: Fri, 11 Jan 2019 10:06:28 -0800 [thread overview]
Message-ID: <20190111180630.6433-3-palmer@sifive.com> (raw)
In-Reply-To: <20190111180630.6433-1-palmer@sifive.com>
From: Michael Clark <mjc@sifive.com>
Use the new CSR read/modify/write interface to implement
atomic updates to mip/sip.
Signed-off-by: Michael Clark <mjc@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
---
target/riscv/csr.c | 56 +++++++++++++++++++++++-----------------------
1 file changed, 28 insertions(+), 28 deletions(-)
diff --git a/target/riscv/csr.c b/target/riscv/csr.c
index b61b0ef37971..44ea8b7cb6e8 100644
--- a/target/riscv/csr.c
+++ b/target/riscv/csr.c
@@ -487,25 +487,31 @@ static int write_mbadaddr(CPURISCVState *env, int csrno, target_ulong val)
return 0;
}
-static int read_mip(CPURISCVState *env, int csrno, target_ulong *val)
-{
- *val = atomic_read(&env->mip);
- return 0;
-}
-
-static int write_mip(CPURISCVState *env, int csrno, target_ulong val)
+static int rmw_mip(CPURISCVState *env, int csrno, target_ulong *ret_value,
+ target_ulong new_value, target_ulong write_mask)
{
RISCVCPU *cpu = riscv_env_get_cpu(env);
+ target_ulong mask = write_mask & delegable_ints;
+ uint32_t old_mip;
+
+ /* We can't allow the supervisor to control SEIP as this would allow the
+ * supervisor to clear a pending external interrupt which will result in
+ * lost a interrupt in the case a PLIC is attached. The SEIP bit must be
+ * hardware controlled when a PLIC is attached. This should be an option
+ * for CPUs with software-delegated Supervisor External Interrupts. */
+ mask &= ~MIP_SEIP;
+
+ if (mask) {
+ qemu_mutex_lock_iothread();
+ old_mip = riscv_cpu_update_mip(cpu, mask, (new_value & mask));
+ qemu_mutex_unlock_iothread();
+ } else {
+ old_mip = atomic_read(&env->mip);
+ }
- /*
- * csrs, csrc on mip.SEIP is not decomposable into separate read and
- * write steps, so a different implementation is needed
- */
-
- qemu_mutex_lock_iothread();
- riscv_cpu_update_mip(cpu, MIP_SSIP | MIP_STIP,
- (val & (MIP_SSIP | MIP_STIP)));
- qemu_mutex_unlock_iothread();
+ if (ret_value) {
+ *ret_value = old_mip;
+ }
return 0;
}
@@ -623,17 +629,11 @@ static int write_sbadaddr(CPURISCVState *env, int csrno, target_ulong val)
return 0;
}
-static int read_sip(CPURISCVState *env, int csrno, target_ulong *val)
-{
- *val = atomic_read(&env->mip) & env->mideleg;
- return 0;
-}
-
-static int write_sip(CPURISCVState *env, int csrno, target_ulong val)
+static int rmw_sip(CPURISCVState *env, int csrno, target_ulong *ret_value,
+ target_ulong new_value, target_ulong write_mask)
{
- target_ulong newval = (atomic_read(&env->mip) & ~env->mideleg)
- | (val & env->mideleg);
- return write_mip(env, CSR_MIP, newval);
+ return rmw_mip(env, CSR_MSTATUS, ret_value, new_value,
+ write_mask & env->mideleg);
}
/* Supervisor Protection and Translation */
@@ -812,7 +812,7 @@ static riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = {
[CSR_MEPC] = { read_mepc, write_mepc },
[CSR_MCAUSE] = { read_mcause, write_mcause },
[CSR_MBADADDR] = { read_mbadaddr, write_mbadaddr },
- [CSR_MIP] = { read_mip, write_mip },
+ [CSR_MIP] = { NULL, NULL, rmw_mip },
/* Supervisor Trap Setup */
[CSR_SSTATUS] = { read_sstatus, write_sstatus },
@@ -825,7 +825,7 @@ static riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = {
[CSR_SEPC] = { read_sepc, write_sepc },
[CSR_SCAUSE] = { read_scause, write_scause },
[CSR_SBADADDR] = { read_sbadaddr, write_sbadaddr },
- [CSR_SIP] = { read_sip, write_sip },
+ [CSR_SIP] = { NULL, NULL, rmw_sip },
/* Supervisor Protection and Translation */
[CSR_SATP] = { read_satp, write_satp },
--
2.18.1
next prev parent reply other threads:[~2019-01-11 18:06 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-11 18:06 [Qemu-devel] [PULL] RISC-V Updates for 3.2, Part 2 Palmer Dabbelt
2019-01-11 18:06 ` [Qemu-devel] [PULL 1/4] RISC-V: Implement modular CSR helper interface Palmer Dabbelt
2019-01-11 18:06 ` Palmer Dabbelt [this message]
2019-01-11 18:06 ` [Qemu-devel] [PULL 3/4] RISC-V: Implement existential predicates for CSRs Palmer Dabbelt
2019-01-11 18:06 ` [Qemu-devel] [PULL 4/4] default-configs: Enable USB support for RISC-V machines Palmer Dabbelt
2019-01-14 12:40 ` [Qemu-devel] [PULL] RISC-V Updates for 3.2, Part 2 Peter Maydell
2019-01-24 2:52 ` Palmer Dabbelt
2019-01-24 10:06 ` Peter Maydell
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=20190111180630.6433-3-palmer@sifive.com \
--to=palmer@sifive.com \
--cc=alistair.francis@wdc.com \
--cc=mjc@sifive.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=qemu-riscv@nongnu.org \
/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).