From: Paolo Bonzini <pbonzini@redhat.com>
To: Eduardo Habkost <ehabkost@redhat.com>, qemu-devel@nongnu.org
Cc: Xiao Guangrong <guangrong.xiao@linux.intel.com>,
kvm@vger.kernel.org, Richard Henderson <rth@twiddle.net>
Subject: Re: [Qemu-devel] [kvm-unit-test RFC] x86: Memory instructions test case
Date: Wed, 18 Nov 2015 12:23:20 +0100 [thread overview]
Message-ID: <564C5FA8.1040804@redhat.com> (raw)
In-Reply-To: <1446672079-8549-1-git-send-email-ehabkost@redhat.com>
On 04/11/2015 22:21, Eduardo Habkost wrote:
> Quickly hacked test case for memory instructions (clflush, mfence,
> sfence, lfence, clflushopt, clwb, pcommit), that simply checks for #UD
> exceptions.
>
> This was useful to test TCG handling of those instructions.
>
> The "fake clwb" part will probably break once a new instruction use
> those opcodes.
>
> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
> ---
> config/config-x86-common.mak | 2 +
> config/config-x86_64.mak | 2 +-
> x86/memory.c | 88 ++++++++++++++++++++++++++++++++++++++++++++
> 3 files changed, 91 insertions(+), 1 deletion(-)
> create mode 100644 x86/memory.c
>
> diff --git a/config/config-x86-common.mak b/config/config-x86-common.mak
> index c2f9908..b89684d 100644
> --- a/config/config-x86-common.mak
> +++ b/config/config-x86-common.mak
> @@ -108,6 +108,8 @@ $(TEST_DIR)/vmx.elf: $(cstart.o) $(TEST_DIR)/vmx.o $(TEST_DIR)/vmx_tests.o
>
> $(TEST_DIR)/debug.elf: $(cstart.o) $(TEST_DIR)/debug.o
>
> +$(TEST_DIR)/memory.elf: $(cstart.o) $(TEST_DIR)/memory.o
> +
> arch_clean:
> $(RM) $(TEST_DIR)/*.o $(TEST_DIR)/*.flat $(TEST_DIR)/*.elf \
> $(TEST_DIR)/.*.d lib/x86/.*.d
> diff --git a/config/config-x86_64.mak b/config/config-x86_64.mak
> index 7d4eb34..ec4bded 100644
> --- a/config/config-x86_64.mak
> +++ b/config/config-x86_64.mak
> @@ -7,7 +7,7 @@ tests = $(TEST_DIR)/access.flat $(TEST_DIR)/apic.flat \
> $(TEST_DIR)/emulator.flat $(TEST_DIR)/idt_test.flat \
> $(TEST_DIR)/xsave.flat $(TEST_DIR)/rmap_chain.flat \
> $(TEST_DIR)/pcid.flat $(TEST_DIR)/debug.flat \
> - $(TEST_DIR)/ioapic.flat
> + $(TEST_DIR)/ioapic.flat $(TEST_DIR)/memory.flat
> tests += $(TEST_DIR)/svm.flat
> tests += $(TEST_DIR)/vmx.flat
> tests += $(TEST_DIR)/tscdeadline_latency.flat
> diff --git a/x86/memory.c b/x86/memory.c
> new file mode 100644
> index 0000000..cd1eb46
> --- /dev/null
> +++ b/x86/memory.c
> @@ -0,0 +1,88 @@
> +/*
> + * Test for x86 cache and memory instructions
> + *
> + * Copyright (c) 2015 Red Hat Inc
> + *
> + * Authors:
> + * Eduardo Habkost <ehabkost@redhat.com>
> + *
> + * This work is licensed under the terms of the GNU GPL, version 2.
> + */
> +
> +#include "libcflat.h"
> +#include "desc.h"
> +#include "processor.h"
> +
> +static long target;
> +static volatile int ud;
> +static volatile int isize;
> +
> +static void handle_ud(struct ex_regs *regs)
> +{
> + ud = 1;
> + regs->rip += isize;
> +}
> +
> +int main(int ac, char **av)
> +{
> + struct cpuid cpuid7, cpuid1;
> + int xfail;
> +
> + setup_idt();
> + handle_exception(UD_VECTOR, handle_ud);
> +
> + cpuid1 = cpuid(1);
> + cpuid7 = cpuid_indexed(7, 0);
> +
> + /* 3-byte instructions: */
> + isize = 3;
> +
> + xfail = !(cpuid1.d & (1U << 19)); /* CLFLUSH */
> + ud = 0;
> + asm volatile("clflush (%0)" : : "b" (&target));
> + report_xfail("clflush", xfail, ud == 0);
> +
> + xfail = !(cpuid1.d & (1U << 25)); /* SSE */
> + ud = 0;
> + asm volatile("sfence");
> + report_xfail("sfence", xfail, ud == 0);
> +
> + xfail = !(cpuid1.d & (1U << 26)); /* SSE2 */
> + ud = 0;
> + asm volatile("lfence");
> + report_xfail("lfence", xfail, ud == 0);
> +
> + ud = 0;
> + asm volatile("mfence");
> + report_xfail("mfence", xfail, ud == 0);
> +
> + /* 4-byte instructions: */
> + isize = 4;
> +
> + xfail = !(cpuid7.b & (1U << 23)); /* CLFLUSHOPT */
> + ud = 0;
> + /* clflushopt (%rbx): */
> + asm volatile(".byte 0x66, 0x0f, 0xae, 0x3b" : : "b" (&target));
> + report_xfail("clflushopt", xfail, ud == 0);
> +
> + xfail = !(cpuid7.b & (1U << 24)); /* CLWB */
> + ud = 0;
> + /* clwb (%rbx): */
> + asm volatile(".byte 0x66, 0x0f, 0xae, 0x33" : : "b" (&target));
> + report_xfail("clwb", xfail, ud == 0);
> +
> + ud = 0;
> + /* clwb requires a memory operand, the following is NOT a valid
> + * CLWB instruction (modrm == 0xF0).
> + */
> + asm volatile(".byte 0x66, 0x0f, 0xae, 0xf0");
> + report("fake clwb", ud);
> +
> + xfail = !(cpuid7.b & (1U << 22)); /* PCOMMIT */
> + ud = 0;
> + /* pcommit: */
> + asm volatile(".byte 0x66, 0x0f, 0xae, 0xf8");
> + report_xfail("pcommit", xfail, ud == 0);
> +
> + return report_summary();
> +}
>
Applied, thanks.
Paolo
prev parent reply other threads:[~2015-11-18 11:23 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-11-04 21:21 [Qemu-devel] [kvm-unit-test RFC] x86: Memory instructions test case Eduardo Habkost
2015-11-05 3:32 ` Xiao Guangrong
2015-11-18 11:23 ` Paolo Bonzini [this message]
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=564C5FA8.1040804@redhat.com \
--to=pbonzini@redhat.com \
--cc=ehabkost@redhat.com \
--cc=guangrong.xiao@linux.intel.com \
--cc=kvm@vger.kernel.org \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
/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).