From: David Hildenbrand <david@redhat.com>
To: Thomas Huth <thuth@redhat.com>, kvm@vger.kernel.org
Cc: Janosch Frank <frankja@linux.ibm.com>
Subject: Re: [kvm-unit-tests v2 PATCH] s390x: Add a test for the diagnose 0x308 hypercall
Date: Tue, 16 Apr 2019 13:57:15 +0200 [thread overview]
Message-ID: <ed976365-341d-46fc-7555-e3ce54571530@redhat.com> (raw)
In-Reply-To: <20190415121242.28848-1-thuth@redhat.com>
On 15.04.19 14:12, Thomas Huth wrote:
> The original idea for this test is to have an easy way to check for a
> problem that has been fixed in QEMU recently: QEMU simply aborted if this
> diagnose call was called with an unsupported subcode (e.g. 2). The problem
> has been fixed in QEMU commit 37dbd1f4d4805edcd18d94eb202bb3461b3cd52d
> ("Return specification exception for unimplemented diag 308 subcodes"),
> and this test now should make sure that we do not regress here again.
> While we're at it, also check whether LOAD NORMAL via subcode 1 works
> correctly, whether the diagnose call is blocked correctly in problem state
> and whether subcodes 5 and 6 are generating specification exceptions for
> illegal parameters as expected.
>
> Signed-off-by: Thomas Huth <thuth@redhat.com>
> ---
> v2: Use "epsw" in test_subcode1() to get the right upper bits of the PSW
>
> s390x/Makefile | 1 +
> s390x/diag308.c | 130 ++++++++++++++++++++++++++++++++++++++++++++
> s390x/unittests.cfg | 3 +
> 3 files changed, 134 insertions(+)
> create mode 100644 s390x/diag308.c
>
> diff --git a/s390x/Makefile b/s390x/Makefile
> index af40fd4..1f21ddb 100644
> --- a/s390x/Makefile
> +++ b/s390x/Makefile
> @@ -5,6 +5,7 @@ tests += $(TEST_DIR)/sieve.elf
> tests += $(TEST_DIR)/sthyi.elf
> tests += $(TEST_DIR)/skey.elf
> tests += $(TEST_DIR)/diag10.elf
> +tests += $(TEST_DIR)/diag308.elf
> tests += $(TEST_DIR)/pfmf.elf
> tests += $(TEST_DIR)/cmm.elf
> tests += $(TEST_DIR)/vector.elf
> diff --git a/s390x/diag308.c b/s390x/diag308.c
> new file mode 100644
> index 0000000..70a3b1e
> --- /dev/null
> +++ b/s390x/diag308.c
> @@ -0,0 +1,130 @@
> +/*
> + * Diagnose 0x308 hypercall tests
> + *
> + * Copyright (c) 2019 Thomas Huth, Red Hat Inc.
> + *
> + * This code is free software; you can redistribute it and/or modify it
> + * under the terms of the GNU General Public License version 2, or (at
> + * your option) any later version.
> + */
> +
> +#include <libcflat.h>
> +#include <asm/asm-offsets.h>
> +#include <asm/interrupt.h>
> +
> +/* The diagnose calls should be blocked in problem state */
> +static void test_priv(void)
> +{
> + expect_pgm_int();
> + enter_pstate();
> + asm volatile ("diag %0,%1,0x308" :: "a"(0), "a"(3));
> + check_pgm_int_code(PGM_INT_CODE_PRIVILEGED_OPERATION);
> +}
> +
> +/*
> + * Check that diag308 with subcode 1 loads the PSW at address 0, i.e.
> + * that we can put a pointer into address 4 which then gets executed.
> + */
> +static void test_subcode1(void)
> +{
> + uint64_t saved_psw = *(uint64_t *)0;
> + long subcode = 1;
> + long ret, tmp;
> +
> + asm volatile (
> + " epsw %0,%1\n"
> + " st %0,0\n"
> + " larl %0,0f\n"
> + " oilh %0,0x8000\n"
> + " st %0,4\n"
> + " diag 0,%2,0x308\n"
> + " lghi %0,0\n"
> + " j 1f\n"
> + "0: lghi %0,1\n"
> + "1:"
> + : "=&a"(ret), "=&a"(tmp) : "a"(subcode) : "memory");
> +
> + *(uint64_t *)0 = saved_psw;
> +
> + report("load normal reset done", ret == 1);
> +}
> +
> +/* Expect a specification exception when using an uneven register */
> +static void test_uneven_reg(unsigned int subcode)
> +{
> + register unsigned long sc asm("6") = subcode;
> + register unsigned long r3 asm("9") = 0x2000;
> +
> + report_prefix_push("uneven register");
> + expect_pgm_int();
> + asm volatile ("diag %0,%1,0x308" :: "a"(r3), "a"(sc));
> + check_pgm_int_code(PGM_INT_CODE_SPECIFICATION);
> + report_prefix_pop();
> +}
> +
> +/* Expect a specification exception when using an unaligned address */
> +static void test_unaligned_address(unsigned int subcode)
> +{
> + register unsigned long sc asm("6") = subcode;
> + register unsigned long addr asm("8") = 54321;
> +
> + report_prefix_push("unaligned address");
> + expect_pgm_int();
> + asm volatile ("diag %0,%1,0x308" :: "a"(addr), "a"(sc));
> + check_pgm_int_code(PGM_INT_CODE_SPECIFICATION);
> + report_prefix_pop();
> +}
> +
> +static void test_subcode5(void)
> +{
> + test_uneven_reg(5);
> + test_unaligned_address(5);
> +}
> +
> +static void test_subcode6(void)
> +{
> + test_uneven_reg(6);
> + test_unaligned_address(6);
> +}
> +
> +/* Unsupported subcodes should generate a specification exception */
> +static void test_unsupported_subcode(void)
> +{
> + int subcodes[] = { 2, 0x101, 0xffff, 0x10001, -1 };
> + int idx;
> +
> + for (idx = 0; idx < ARRAY_SIZE(subcodes); idx++) {
> + report_prefix_pushf("0x%04x", subcodes[idx]);
> + expect_pgm_int();
> + asm volatile ("diag %0,%1,0x308" :: "a"(0), "a"(subcodes[idx]));
> + check_pgm_int_code(PGM_INT_CODE_SPECIFICATION);
> + report_prefix_pop();
> + }
> +}
> +
> +static struct {
> + const char *name;
> + void (*func)(void);
> +} tests[] = {
> + { "privileged", test_priv },
> + { "subcode 1", test_subcode1 },
> + { "subcode 5", test_subcode5 },
> + { "subcode 6", test_subcode6 },
> + { "unsupported", test_unsupported_subcode },
> + { NULL, NULL }
> +};
> +
> +int main(int argc, char**argv)
> +{
> + int i;
> +
> + report_prefix_push("diag308");
> + for (i = 0; tests[i].name; i++) {
> + report_prefix_push(tests[i].name);
> + tests[i].func();
> + report_prefix_pop();
> + }
> + report_prefix_pop();
> +
> + return report_summary();
> +}
> diff --git a/s390x/unittests.cfg b/s390x/unittests.cfg
> index e53c365..546b1f2 100644
> --- a/s390x/unittests.cfg
> +++ b/s390x/unittests.cfg
> @@ -44,6 +44,9 @@ file = skey.elf
> [diag10]
> file = diag10.elf
>
> +[diag308]
> +file = diag308.elf
> +
> [pfmf]
> file = pfmf.elf
>
>
You might want to use "g" instead of "a", as we are not dealing with
address registers (base/index). Apart from that, nothing major jumped at me.
--
Thanks,
David / dhildenb
next prev parent reply other threads:[~2019-04-16 11:57 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-15 12:12 [kvm-unit-tests v2 PATCH] s390x: Add a test for the diagnose 0x308 hypercall Thomas Huth
2019-04-16 11:16 ` Christian Borntraeger
2019-04-16 11:57 ` David Hildenbrand [this message]
2019-04-16 12:32 ` Thomas Huth
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=ed976365-341d-46fc-7555-e3ce54571530@redhat.com \
--to=david@redhat.com \
--cc=frankja@linux.ibm.com \
--cc=kvm@vger.kernel.org \
--cc=thuth@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