From: Daniel Henrique Barboza <danielhb413@gmail.com>
To: qemu-devel@nongnu.org
Cc: gustavo.romero@linaro.org,
Daniel Henrique Barboza <danielhb413@gmail.com>,
richard.henderson@linaro.org, groug@kaod.org,
qemu-ppc@nongnu.org, clg@kaod.org, david@gibson.dropbear.id.au
Subject: [PATCH v2 14/16] target/ppc: PMU: insns counter negative overflow support
Date: Tue, 24 Aug 2021 13:30:30 -0300 [thread overview]
Message-ID: <20210824163032.394099-15-danielhb413@gmail.com> (raw)
In-Reply-To: <20210824163032.394099-1-danielhb413@gmail.com>
Enabling counter negative overflow for the PMCs that are counting
instructions is simpler than when counting cycles. Instruction
counting is done via helper_insns_inc(), which is called every
time a TB ends.
Firing a performance monitor alert due to a counter negative overflow
in this case is a matter of checking if the counter value is over
0x80000000 each time the counters are incremented and, if counter negative
events are enabled for that specific counter, trigger the PM alert.
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
---
target/ppc/power8_pmu.c | 23 +++++++++++++++++++++--
target/ppc/translate.c | 8 ++++++++
2 files changed, 29 insertions(+), 2 deletions(-)
diff --git a/target/ppc/power8_pmu.c b/target/ppc/power8_pmu.c
index 690476051d..28db086225 100644
--- a/target/ppc/power8_pmu.c
+++ b/target/ppc/power8_pmu.c
@@ -252,9 +252,8 @@ static void start_cycle_count_session(CPUPPCState *env)
}
}
-static void cpu_ppc_pmu_timer_cb(void *opaque)
+static void fire_PMC_interrupt(PowerPCCPU *cpu)
{
- PowerPCCPU *cpu = opaque;
CPUPPCState *env = &cpu->env;
if (!(env->spr[SPR_POWER_MMCR0] & MMCR0_EBE)) {
@@ -284,6 +283,13 @@ static void cpu_ppc_pmu_timer_cb(void *opaque)
ppc_set_irq(cpu, PPC_INTERRUPT_PMC, 1);
}
+static void cpu_ppc_pmu_timer_cb(void *opaque)
+{
+ PowerPCCPU *cpu = opaque;
+
+ fire_PMC_interrupt(cpu);
+}
+
void cpu_ppc_pmu_timer_init(CPUPPCState *env)
{
PowerPCCPU *cpu = env_archcpu(env);
@@ -377,6 +383,8 @@ static bool pmc_counting_insns(CPUPPCState *env, int sprn,
/* This helper assumes that the PMC is running. */
void helper_insns_inc(CPUPPCState *env, uint32_t num_insns)
{
+ bool overflow_triggered = false;
+ PowerPCCPU *cpu;
int sprn;
for (sprn = SPR_POWER_PMC1; sprn <= SPR_POWER_PMC5; sprn++) {
@@ -390,8 +398,19 @@ void helper_insns_inc(CPUPPCState *env, uint32_t num_insns)
} else {
env->spr[sprn] += num_insns;
}
+
+ if (env->spr[sprn] >= COUNTER_NEGATIVE_VAL &&
+ pmc_counter_negative_enabled(env, sprn)) {
+ overflow_triggered = true;
+ env->spr[sprn] = COUNTER_NEGATIVE_VAL;
+ }
}
}
+
+ if (overflow_triggered) {
+ cpu = env_archcpu(env);
+ fire_PMC_interrupt(cpu);
+ }
}
#endif /* defined(TARGET_PPC64) && !defined(CONFIG_USER_ONLY) */
diff --git a/target/ppc/translate.c b/target/ppc/translate.c
index d4cfc567cf..8302022852 100644
--- a/target/ppc/translate.c
+++ b/target/ppc/translate.c
@@ -4437,6 +4437,14 @@ static void pmu_count_insns(DisasContext *ctx)
tcg_gen_andi_tl(t0, t0, MMCR0_FC);
tcg_gen_brcond_tl(TCG_COND_EQ, t0, t_mmcr0FC, l_exit);
+ /*
+ * The PMU insns_inc() helper stops the internal PMU timer if a
+ * counter overflows happens. In that case, if the guest is
+ * running with icount and we do not handle it beforehand,
+ * the helper can trigger a 'bad icount read'.
+ */
+ gen_icount_io_start(ctx);
+
gen_helper_insns_inc(cpu_env, tcg_constant_i32(ctx->base.num_insns));
gen_set_label(l_exit);
--
2.31.1
next prev parent reply other threads:[~2021-08-24 16:45 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-24 16:30 [PATCH v2 00/16] PMU-EBB support for PPC64 TCG Daniel Henrique Barboza
2021-08-24 16:30 ` [PATCH v2 01/16] target/ppc: add user write access control for PMU SPRs Daniel Henrique Barboza
2021-08-25 4:26 ` David Gibson
2021-08-24 16:30 ` [PATCH v2 02/16] target/ppc: add user read functions for MMCR0 and MMCR2 Daniel Henrique Barboza
2021-08-25 4:30 ` David Gibson
2021-08-25 12:25 ` Paul A. Clarke
2021-08-25 13:35 ` David Gibson
2021-08-24 16:30 ` [PATCH v2 03/16] target/ppc: add exclusive user write function for MMCR0 Daniel Henrique Barboza
2021-08-25 4:37 ` David Gibson
2021-08-25 14:01 ` Daniel Henrique Barboza
2021-08-24 16:30 ` [PATCH v2 04/16] target/ppc: PMU basic cycle count for pseries TCG Daniel Henrique Barboza
2021-08-25 5:19 ` David Gibson
2021-08-25 14:05 ` Daniel Henrique Barboza
2021-08-24 16:30 ` [PATCH v2 05/16] target/ppc/power8_pmu.c: enable PMC1-PMC4 events Daniel Henrique Barboza
2021-08-25 5:23 ` David Gibson
2021-08-24 16:30 ` [PATCH v2 06/16] target/ppc: PMU: add instruction counting Daniel Henrique Barboza
2021-08-25 5:31 ` David Gibson
2021-08-25 14:10 ` Daniel Henrique Barboza
2021-08-24 16:30 ` [PATCH v2 07/16] target/ppc/power8_pmu.c: add PM_RUN_INST_CMPL (0xFA) event Daniel Henrique Barboza
2021-08-25 5:32 ` David Gibson
2021-08-24 16:30 ` [PATCH v2 08/16] target/ppc/power8_pmu.c: add PMC14/PMC56 counter freeze bits Daniel Henrique Barboza
2021-08-24 16:30 ` [PATCH v2 09/16] PPC64/TCG: Implement 'rfebb' instruction Daniel Henrique Barboza
2021-08-30 12:12 ` Matheus K. Ferst
2021-08-30 18:41 ` Daniel Henrique Barboza
2021-08-24 16:30 ` [PATCH v2 10/16] target/ppc: PMU Event-Based exception support Daniel Henrique Barboza
2021-08-25 5:37 ` David Gibson
2021-08-30 19:09 ` Daniel Henrique Barboza
2021-08-24 16:30 ` [PATCH v2 11/16] target/ppc/excp_helper.c: EBB handling adjustments Daniel Henrique Barboza
2021-08-24 16:30 ` [PATCH v2 12/16] target/ppc/power8_pmu.c: enable PMC1 counter negative overflow Daniel Henrique Barboza
2021-08-24 16:30 ` [PATCH v2 13/16] target/ppc/power8_pmu.c: cycles overflow with all PMCs Daniel Henrique Barboza
2021-08-24 16:30 ` Daniel Henrique Barboza [this message]
2021-08-24 16:30 ` [PATCH v2 15/16] target/ppc/translate: PMU: handle setting of PMCs while running Daniel Henrique Barboza
2021-08-24 16:30 ` [PATCH v2 16/16] target/ppc/power8_pmu.c: handle overflow bits when PMU is running Daniel Henrique Barboza
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=20210824163032.394099-15-danielhb413@gmail.com \
--to=danielhb413@gmail.com \
--cc=clg@kaod.org \
--cc=david@gibson.dropbear.id.au \
--cc=groug@kaod.org \
--cc=gustavo.romero@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
--cc=richard.henderson@linaro.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).