linux-riscv.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Jones <ajones@ventanamicro.com>
To: Haibo Xu <haibo1.xu@intel.com>
Cc: xiaobo55x@gmail.com, Paul Walmsley <paul.walmsley@sifive.com>,
	 Palmer Dabbelt <palmer@dabbelt.com>,
	Albert Ou <aou@eecs.berkeley.edu>,
	 Paolo Bonzini <pbonzini@redhat.com>,
	Shuah Khan <shuah@kernel.org>, Marc Zyngier <maz@kernel.org>,
	 Oliver Upton <oliver.upton@linux.dev>,
	James Morse <james.morse@arm.com>,
	 Suzuki K Poulose <suzuki.poulose@arm.com>,
	Zenghui Yu <yuzenghui@huawei.com>,
	 Anup Patel <anup@brainfault.org>,
	Atish Patra <atishp@atishpatra.org>, Guo Ren <guoren@kernel.org>,
	 Mayuresh Chitale <mchitale@ventanamicro.com>,
	Greentime Hu <greentime.hu@sifive.com>,
	 wchen <waylingii@gmail.com>,
	Conor Dooley <conor.dooley@microchip.com>,
	 Heiko Stuebner <heiko@sntech.de>,
	Minda Chen <minda.chen@starfivetech.com>,
	 Samuel Holland <samuel@sholland.org>,
	Jisheng Zhang <jszhang@kernel.org>,
	 Sean Christopherson <seanjc@google.com>,
	Peter Xu <peterx@redhat.com>, Like Xu <likexu@tencent.com>,
	 Vipin Sharma <vipinsh@google.com>,
	Maciej Wieczor-Retman <maciej.wieczor-retman@intel.com>,
	 Aaron Lewis <aaronlewis@google.com>,
	Thomas Huth <thuth@redhat.com>,
	linux-kernel@vger.kernel.org,  linux-riscv@lists.infradead.org,
	kvm@vger.kernel.org, linux-kselftest@vger.kernel.org,
	 linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev,
	kvm-riscv@lists.infradead.org
Subject: Re: [PATCH v4 11/11] KVM: selftests: Enable tunning of err_margin_us in arch timer test
Date: Wed, 13 Dec 2023 15:27:53 +0100	[thread overview]
Message-ID: <20231213-e11bcfdf39a34099bbc48cea@orel> (raw)
In-Reply-To: <0343a9e4bfa8011fbb6bca0286cee7eab1f17d5d.1702371136.git.haibo1.xu@intel.com>

On Tue, Dec 12, 2023 at 05:31:20PM +0800, Haibo Xu wrote:
> There are intermittent failures occured when stressing the
> arch-timer test in a Qemu VM:
> 
>  Guest assert failed,  vcpu 0; stage; 4; iter: 3
>  ==== Test Assertion Failure ====
>    aarch64/arch_timer.c:196: config_iter + 1 == irq_iter
>    pid=4048 tid=4049 errno=4 - Interrupted system call
>       1  0x000000000040253b: test_vcpu_run at arch_timer.c:248
>       2  0x0000ffffb60dd5c7: ?? ??:0
>       3  0x0000ffffb6145d1b: ?? ??:0
>    0x3 != 0x2 (config_iter + 1 != irq_iter)e
> 
> Further test and debug show that the timeout for an interrupt
> to arrive do have random high fluctuation, espectially when
> testing in an virtual environment.
> 
> To alleviate this issue, just expose the timeout value as user
> configurable and print some hint message to increase the value
> when hitting the failure..
> 
> Signed-off-by: Haibo Xu <haibo1.xu@intel.com>
> ---
>  .../selftests/kvm/aarch64/arch_timer.c        |  8 +++++--
>  tools/testing/selftests/kvm/arch_timer.c      | 22 +++++++++++++------
>  .../selftests/kvm/include/timer_test.h        |  1 +
>  .../testing/selftests/kvm/riscv/arch_timer.c  |  8 +++++--
>  4 files changed, 28 insertions(+), 11 deletions(-)
> 
> diff --git a/tools/testing/selftests/kvm/aarch64/arch_timer.c b/tools/testing/selftests/kvm/aarch64/arch_timer.c
> index 4b421d421c3f..139eecbf77e7 100644
> --- a/tools/testing/selftests/kvm/aarch64/arch_timer.c
> +++ b/tools/testing/selftests/kvm/aarch64/arch_timer.c
> @@ -131,10 +131,14 @@ static void guest_run_stage(struct test_vcpu_shared_data *shared_data,
>  
>  		/* Setup a timeout for the interrupt to arrive */
>  		udelay(msecs_to_usecs(test_args.timer_period_ms) +
> -			TIMER_TEST_ERR_MARGIN_US);
> +			test_args.timer_err_margin_us);
>  
>  		irq_iter = READ_ONCE(shared_data->nr_iter);
> -		GUEST_ASSERT_EQ(config_iter + 1, irq_iter);
> +		__GUEST_ASSERT(config_iter + 1 == irq_iter,
> +			"config_iter + 1 = 0x%lx, irq_iter = 0x%lx.\n"
> +			"  Guest timer interrupt was not trigged within the specified\n"
> +			"  interval, try to increase the error margin by [-e] option.\n",
> +			config_iter + 1, irq_iter);
>  	}
>  }
>  
> diff --git a/tools/testing/selftests/kvm/arch_timer.c b/tools/testing/selftests/kvm/arch_timer.c
> index 60963fce16f2..5050022fd345 100644
> --- a/tools/testing/selftests/kvm/arch_timer.c
> +++ b/tools/testing/selftests/kvm/arch_timer.c
> @@ -5,16 +5,17 @@
>   * The guest's main thread configures the timer interrupt and waits
>   * for it to fire, with a timeout equal to the timer period.
>   * It asserts that the timeout doesn't exceed the timer period plus
> - * an error margin of 100us.
> + * an user configurable error margin(default to 100us).
>   *
>   * On the other hand, upon receipt of an interrupt, the guest's interrupt
>   * handler validates the interrupt by checking if the architectural state
>   * is in compliance with the specifications.
>   *
>   * The test provides command-line options to configure the timer's
> - * period (-p), number of vCPUs (-n), and iterations per stage (-i).
> - * To stress-test the timer stack even more, an option to migrate the
> - * vCPUs across pCPUs (-m), at a particular rate, is also provided.
> + * period (-p), number of vCPUs (-n), iterations per stage (-i), and timer
> + * interrupt arrival error margin (-e). To stress-test the timer stack even
> + * more, an option to migrate the vCPUs across pCPUs (-m), at a particular
> + * rate, is also provided.
>   *
>   * Copyright (c) 2021, Google LLC.
>   */
> @@ -34,6 +35,7 @@ struct test_args test_args = {
>  	.nr_iter = NR_TEST_ITERS_DEF,
>  	.timer_period_ms = TIMER_TEST_PERIOD_MS_DEF,
>  	.migration_freq_ms = TIMER_TEST_MIGRATION_FREQ_MS,
> +	.timer_err_margin_us = TIMER_TEST_ERR_MARGIN_US,
>  	.reserved = 1,
>  };
>  
> @@ -179,8 +181,9 @@ static void test_run(struct kvm_vm *vm)
>  
>  static void test_print_help(char *name)
>  {
> -	pr_info("Usage: %s [-h] [-n nr_vcpus] [-i iterations] [-p timer_period_ms]\n",
> -		name);
> +	pr_info("Usage: %s [-h] [-n nr_vcpus] [-i iterations] [-p timer_period_ms]\n"
> +	        "\t\t    [-m migration_freq_ms] [-o counter_offset]\n"
> +	        "\t\t    [-e timer_err_margin_us]\n", name);
>  	pr_info("\t-n: Number of vCPUs to configure (default: %u; max: %u)\n",
>  		NR_VCPUS_DEF, KVM_MAX_VCPUS);
>  	pr_info("\t-i: Number of iterations per stage (default: %u)\n",
> @@ -190,6 +193,8 @@ static void test_print_help(char *name)
>  	pr_info("\t-m: Frequency (in ms) of vCPUs to migrate to different pCPU. 0 to turn off (default: %u)\n",
>  		TIMER_TEST_MIGRATION_FREQ_MS);
>  	pr_info("\t-o: Counter offset (in counter cycles, default: 0) [aarch64-only]\n");
> +	pr_info("\t-e: Interrupt arrival error margin(in us) of the guest timer (default: %u)\n",
                                                     ^ need space here

> +		TIMER_TEST_ERR_MARGIN_US);
>  	pr_info("\t-h: print this help screen\n");
>  }
>  
> @@ -197,7 +202,7 @@ static bool parse_args(int argc, char *argv[])
>  {
>  	int opt;
>  
> -	while ((opt = getopt(argc, argv, "hn:i:p:m:o:")) != -1) {
> +	while ((opt = getopt(argc, argv, "hn:i:p:m:o:e:")) != -1) {
>  		switch (opt) {
>  		case 'n':
>  			test_args.nr_vcpus = atoi_positive("Number of vCPUs", optarg);
> @@ -216,6 +221,9 @@ static bool parse_args(int argc, char *argv[])
>  		case 'm':
>  			test_args.migration_freq_ms = atoi_non_negative("Frequency", optarg);
>  			break;
> +		case 'e':
> +			test_args.timer_err_margin_us = atoi_non_negative("Error Margin", optarg);
> +			break;
>  		case 'o':
>  			test_args.counter_offset = strtol(optarg, NULL, 0);
>  			test_args.reserved = 0;
> diff --git a/tools/testing/selftests/kvm/include/timer_test.h b/tools/testing/selftests/kvm/include/timer_test.h
> index 968257b893a7..b1d405e7157d 100644
> --- a/tools/testing/selftests/kvm/include/timer_test.h
> +++ b/tools/testing/selftests/kvm/include/timer_test.h
> @@ -22,6 +22,7 @@ struct test_args {
>  	int nr_iter;
>  	int timer_period_ms;
>  	int migration_freq_ms;
> +	int timer_err_margin_us;
>  	/* Members of struct kvm_arm_counter_offset */
>  	uint64_t counter_offset;
>  	uint64_t reserved;
> diff --git a/tools/testing/selftests/kvm/riscv/arch_timer.c b/tools/testing/selftests/kvm/riscv/arch_timer.c
> index 13bf184d1ff5..45a139dc7ce3 100644
> --- a/tools/testing/selftests/kvm/riscv/arch_timer.c
> +++ b/tools/testing/selftests/kvm/riscv/arch_timer.c
> @@ -55,10 +55,14 @@ static void guest_run(struct test_vcpu_shared_data *shared_data)
>  
>  		/* Setup a timeout for the interrupt to arrive */
>  		udelay(msecs_to_usecs(test_args.timer_period_ms) +
> -			TIMER_TEST_ERR_MARGIN_US);
> +			test_args.timer_err_margin_us);
>  
>  		irq_iter = READ_ONCE(shared_data->nr_iter);
> -		GUEST_ASSERT_EQ(config_iter + 1, irq_iter);
> +		__GUEST_ASSERT(config_iter + 1 == irq_iter,
> +			"config_iter + 1 = 0x%lx, irq_iter = 0x%lx.\n"
> +			"  Guest timer interrupt was not trigged within the specified\n"
> +			"  interval, try to increase the error margin by [-e] option.\n",
> +			config_iter + 1, irq_iter);
>  	}
>  }
>  
> -- 
> 2.34.1
>

I probably would have started the series with this patch since you said
you could reproduce the problem on aarch64. Starting the series with
this patch would allow just this patch to get picked up to fix aarch64,
if that's necessary, and also avoids touching a couple places twice
with the code split for riscv.

But anyway, other than the missing space in the help text,

Reviewed-by: Andrew Jones <ajones@ventanamicro.com>

Thanks,
drew

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

  reply	other threads:[~2023-12-13 14:28 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-12  9:31 [PATCH v4 00/11] RISCV: Add kvm Sstc timer selftests Haibo Xu
2023-12-12  9:31 ` [PATCH v4 01/11] selftests/kvm: Fix issues with $(SPLIT_TESTS) Haibo Xu
2023-12-12  9:31 ` [PATCH v4 02/11] KVM: arm64: selftests: Split arch_timer test code Haibo Xu
2023-12-14  6:41   ` Anup Patel
2023-12-12  9:31 ` [PATCH v4 03/11] KVM: selftests: Add CONFIG_64BIT definition for the build Haibo Xu
2023-12-13 13:52   ` Andrew Jones
2023-12-12  9:31 ` [PATCH v4 04/11] tools: riscv: Add header file csr.h Haibo Xu
2023-12-12  9:31 ` [PATCH v4 05/11] tools: riscv: Add header file vdso/processor.h Haibo Xu
2023-12-13 14:02   ` Andrew Jones
2023-12-12  9:31 ` [PATCH v4 06/11] KVM: riscv: selftests: Switch to use macro from csr.h Haibo Xu
2023-12-12  9:31 ` [PATCH v4 07/11] KVM: riscv: selftests: Add exception handling support Haibo Xu
2023-12-12  9:31 ` [PATCH v4 08/11] KVM: riscv: selftests: Add guest helper to get vcpu id Haibo Xu
2023-12-13 14:08   ` Andrew Jones
2023-12-19  2:28     ` Haibo Xu
2023-12-12  9:31 ` [PATCH v4 09/11] KVM: riscv: selftests: Change vcpu_has_ext to a common function Haibo Xu
2023-12-12  9:31 ` [PATCH v4 10/11] KVM: riscv: selftests: Add sstc timer test Haibo Xu
2023-12-13 14:18   ` Andrew Jones
2023-12-12  9:31 ` [PATCH v4 11/11] KVM: selftests: Enable tunning of err_margin_us in arch " Haibo Xu
2023-12-13 14:27   ` Andrew Jones [this message]
2023-12-14  6:44   ` Anup Patel
2023-12-19 18:22   ` Marc Zyngier
2023-12-20  6:50     ` Haibo Xu
2023-12-20  9:00       ` Marc Zyngier
2023-12-20 13:51         ` Haibo Xu
2023-12-20 13:58           ` Marc Zyngier
2023-12-21  2:58             ` Haibo Xu
2023-12-21  9:25               ` Marc Zyngier
2024-01-04 11:09               ` Andrew Jones

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=20231213-e11bcfdf39a34099bbc48cea@orel \
    --to=ajones@ventanamicro.com \
    --cc=aaronlewis@google.com \
    --cc=anup@brainfault.org \
    --cc=aou@eecs.berkeley.edu \
    --cc=atishp@atishpatra.org \
    --cc=conor.dooley@microchip.com \
    --cc=greentime.hu@sifive.com \
    --cc=guoren@kernel.org \
    --cc=haibo1.xu@intel.com \
    --cc=heiko@sntech.de \
    --cc=james.morse@arm.com \
    --cc=jszhang@kernel.org \
    --cc=kvm-riscv@lists.infradead.org \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.linux.dev \
    --cc=likexu@tencent.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=linux-riscv@lists.infradead.org \
    --cc=maciej.wieczor-retman@intel.com \
    --cc=maz@kernel.org \
    --cc=mchitale@ventanamicro.com \
    --cc=minda.chen@starfivetech.com \
    --cc=oliver.upton@linux.dev \
    --cc=palmer@dabbelt.com \
    --cc=paul.walmsley@sifive.com \
    --cc=pbonzini@redhat.com \
    --cc=peterx@redhat.com \
    --cc=samuel@sholland.org \
    --cc=seanjc@google.com \
    --cc=shuah@kernel.org \
    --cc=suzuki.poulose@arm.com \
    --cc=thuth@redhat.com \
    --cc=vipinsh@google.com \
    --cc=waylingii@gmail.com \
    --cc=xiaobo55x@gmail.com \
    --cc=yuzenghui@huawei.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).