From: Andrew Jones <drjones@redhat.com>
To: pbonzini@redhat.com
Cc: kvmarm@lists.cs.columbia.edu, kvm@vger.kernel.org
Subject: [PULL kvm-unit-tests 04/39] arm64: timer: Add ISB before reading the counter value
Date: Sat, 4 Apr 2020 16:36:56 +0200 [thread overview]
Message-ID: <20200404143731.208138-5-drjones@redhat.com> (raw)
In-Reply-To: <20200404143731.208138-1-drjones@redhat.com>
From: Alexandru Elisei <alexandru.elisei@arm.com>
Reads of the physical counter and the virtual counter registers "can occur
speculatively and out of order relative to other instructions executed on
the same PE" [1, 2].
There is no theoretical limit to the number of instructions that the CPU
can reorder and we use the counter value to program the timer to fire in
the future. Add an ISB before reading the counter to make sure the read
instruction is not reordered too long in the past with regard to the
instruction that programs the timer alarm, thus causing the timer to fire
unexpectedly. This matches what Linux does (see
arch/arm64/include/asm/arch_timer.h).
Because we use the counter value to program the timer, we create a register
dependency [3] between the value that we read and the value that we write to
CVAL and thus we don't need a barrier after the read. Linux does things
differently because the read needs to be ordered with regard to a memory
load (more information in commit 75a19a0202db ("arm64: arch_timer: Ensure
counter register reads occur with seqlock held")).
This also matches what we already do in get_cntvct from
lib/arm{,64}/asm/processor.h.
[1] ARM DDI 0487E.a, section D11.2.1
[2] ARM DDI 0487E.a, section D11.2.2
[3] ARM DDI 0486E.a, section B2.3.2
Signed-off-by: Alexandru Elisei <alexandru.elisei@arm.com>
Signed-off-by: Andrew Jones <drjones@redhat.com>
---
arm/timer.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arm/timer.c b/arm/timer.c
index c6ea108cfa4b..e758e84855c3 100644
--- a/arm/timer.c
+++ b/arm/timer.c
@@ -30,6 +30,7 @@ static void ptimer_unsupported_handler(struct pt_regs *regs, unsigned int esr)
static u64 read_vtimer_counter(void)
{
+ isb();
return read_sysreg(cntvct_el0);
}
@@ -68,6 +69,7 @@ static void write_vtimer_ctl(u64 val)
static u64 read_ptimer_counter(void)
{
+ isb();
return read_sysreg(cntpct_el0);
}
--
2.25.1
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
next prev parent reply other threads:[~2020-04-04 14:38 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-04 14:36 [PULL kvm-unit-tests 00/39] arm/arm64: The old and new Andrew Jones
2020-04-04 14:36 ` [PULL kvm-unit-tests 01/39] Makefile: Use no-stack-protector compiler options Andrew Jones
2020-04-04 14:36 ` [PULL kvm-unit-tests 02/39] arm/arm64: psci: Don't run C code without stack or vectors Andrew Jones
2020-04-04 14:36 ` [PULL kvm-unit-tests 03/39] arm64: timer: Add ISB after register writes Andrew Jones
2020-04-04 14:36 ` Andrew Jones [this message]
2020-04-04 14:36 ` [PULL kvm-unit-tests 05/39] arm64: timer: Make irq_received volatile Andrew Jones
2020-04-04 14:36 ` [PULL kvm-unit-tests 06/39] arm64: timer: EOIR the interrupt after masking the timer Andrew Jones
2020-04-04 14:36 ` [PULL kvm-unit-tests 07/39] arm64: timer: Wait for the GIC to sample timer interrupt state Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 08/39] arm64: timer: Check the " Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 09/39] arm64: timer: Test behavior when timer disabled or masked Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 10/39] arm/arm64: Perform dcache clean + invalidate after turning MMU off Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 11/39] arm/arm64: gic: Move gic_state enumeration to asm/gic.h Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 12/39] arm64: timer: Use the proper RDist register name in GICv3 Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 13/39] arm64: timer: Use existing helpers to access counter/timers Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 14/39] arm64: timer: Speed up gic-timer-state check Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 15/39] arm64: Provide read/write_sysreg_s Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 16/39] arm: pmu: Let pmu tests take a sub-test parameter Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 17/39] arm: pmu: Don't check PMCR.IMP anymore Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 18/39] arm: pmu: Add a pmu struct Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 19/39] arm: pmu: Introduce defines for PMU versions Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 20/39] arm: pmu: Check Required Event Support Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 21/39] arm: pmu: Basic event counter Tests Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 22/39] arm: pmu: Test SW_INCR event count Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 23/39] arm: pmu: Test chained counters Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 24/39] arm: pmu: test 32-bit <-> 64-bit transitions Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 25/39] arm: gic: Introduce gic_irq_set_clr_enable() helper Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 26/39] arm: pmu: Test overflow interrupts Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 27/39] libcflat: Add other size defines Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 28/39] page_alloc: Introduce get_order() Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 29/39] arm/arm64: gic: Introduce setup_irq() helper Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 30/39] arm/arm64: gicv3: Add some re-distributor defines Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 31/39] arm/arm64: gicv3: Set the LPI config and pending tables Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 32/39] arm/arm64: ITS: Introspection tests Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 33/39] arm/arm64: ITS: its_enable_defaults Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 34/39] arm/arm64: ITS: Device and collection Initialization Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 35/39] arm/arm64: ITS: Commands Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 36/39] arm/arm64: ITS: INT functional tests Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 37/39] arm/run: Allow Migration tests Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 38/39] arm/arm64: ITS: migration tests Andrew Jones
2020-04-04 14:37 ` [PULL kvm-unit-tests 39/39] arm/arm64: ITS: pending table migration test Andrew Jones
2020-04-07 15:28 ` [PULL kvm-unit-tests 00/39] arm/arm64: The old and new Paolo Bonzini
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=20200404143731.208138-5-drjones@redhat.com \
--to=drjones@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=pbonzini@redhat.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