xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Xen-devel <xen-devel@lists.xen.org>
Cc: Keir Fraser <keir@xen.org>,
	Ian Campbell <Ian.Campbell@citrix.com>,
	Andrew Cooper <andrew.cooper3@citrix.com>,
	Ian Jackson <Ian.Jackson@eu.citrix.com>, Tim Deegan <tim@xen.org>,
	Jan Beulich <JBeulich@suse.com>
Subject: [PATCH v2 1/3] hvmloader/tests: Improvements to shadow_gs_test()
Date: Tue, 29 Jul 2014 18:49:02 +0100	[thread overview]
Message-ID: <1406656144-24882-2-git-send-email-andrew.cooper3@citrix.com> (raw)
In-Reply-To: <1406656144-24882-1-git-send-email-andrew.cooper3@citrix.com>

Use .code64 and .code32, and real mnemonics instead of bare opcodes.  No
functional change to the test.

Correct register constraints.  PD_START doesn't necessarily need to be in eax,
but eax does need to identified as clobbered.

As GAS is unable to express a register as an input and also clobbered, opt for
an explicit eax clobber and allow some choice for the register used to pass
PD_START (and indeed, objdump indicates that esi is chosen).  ebx gains an
early clobber to remove it from the candidate list for PD_START.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
CC: Keir Fraser <keir@xen.org>
CC: Jan Beulich <JBeulich@suse.com>
CC: Tim Deegan <tim@xen.org>
CC: Ian Campbell <Ian.Campbell@citrix.com>
CC: Ian Jackson <Ian.Jackson@eu.citrix.com>

---
v2: Include register constraint fixes as suggested by Jan
---
 tools/firmware/hvmloader/tests.c |   18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/tools/firmware/hvmloader/tests.c b/tools/firmware/hvmloader/tests.c
index 52772aa..931ed2f 100644
--- a/tools/firmware/hvmloader/tests.c
+++ b/tools/firmware/hvmloader/tests.c
@@ -135,7 +135,7 @@ static int rep_io_test(void)
 static int shadow_gs_test(void)
 {
     uint64_t *pd = (uint64_t *)PD_START;
-    uint32_t i, eax, ebx, ecx, edx;
+    uint32_t i, eax, ebx, ecx, edx, dummy;
 
     /* Skip this test if the CPU does not support long mode. */
     cpuid(0x80000000, &eax, &ebx, &ecx, &edx);
@@ -158,7 +158,7 @@ static int shadow_gs_test(void)
         "mov $0x20,%%ebx; "
         "mov %%ebx,%%cr4; "
         /* CR3 */
-        "mov %%eax,%%cr3; "
+        "mov %0,%%cr3; "
         /* EFER.LME=1 */
         "mov $0xc0000080,%%ecx; rdmsr; btsl $8,%%eax; wrmsr; "
         /* CR0.PG=1 */
@@ -170,12 +170,14 @@ static int shadow_gs_test(void)
         /* Push LRETQ stack frame. */
         "pushl $0; pushl $"STR(SEL_CODE32)"; pushl $0; pushl $2f; "
         /* Jump to 64-bit mode. */
-        "ljmp $"STR(SEL_CODE64)",$1f; 1: "
+        "ljmp $"STR(SEL_CODE64)",$1f; "
+        ".code64; 1: "
         /* Swap GS_BASE and SHADOW_GS_BASE */
-        ".byte 0x0f,0x01,0xf8; " /* SWAPGS */
+        "swapgs; "
         /* Jump to 32-bit mode. */
-        ".byte 0x89, 0xe4; "     /* MOV ESP,ESP */
-        ".byte 0x48, 0xcb; 2: "  /* LRETQ */
+        "movl %%esp,%%esp; "
+        "lretq; "
+        ".code32; 2:"
         /* Read SHADOW_GS_BASE: should now contain 2 */
         "mov $0xc0000102,%%ecx; rdmsr; mov %%eax,%%ebx; "
         /* CR0.PG=0 */
@@ -185,7 +187,9 @@ static int shadow_gs_test(void)
         "mov $0xc0000080,%%ecx; rdmsr; btcl $8,%%eax; wrmsr; "
         /* CR4.PAE=0 */
         "xor %%eax,%%eax; mov %%eax,%%cr4; "
-        : "=b" (ebx) : "a" (PD_START) : "ecx", "edx", "memory" );
+        : "=r" (dummy), "=&b" (ebx)
+        : "0" (PD_START)
+        : "eax", "ecx", "edx", "memory" );
 
     return (ebx == 2) ? TEST_PASS : TEST_FAIL;
 }
-- 
1.7.10.4

  reply	other threads:[~2014-07-29 17:49 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-29 17:49 [PATCH v2 0/3] HVMLoader test improvements Andrew Cooper
2014-07-29 17:49 ` Andrew Cooper [this message]
2014-07-30  7:37   ` [PATCH v2 1/3] hvmloader/tests: Improvements to shadow_gs_test() Jan Beulich
2014-07-29 17:49 ` [PATCH v2 2/3] hvmloader: Introduce cpuid_count() helper function Andrew Cooper
2014-07-29 17:49 ` [PATCH v2 3/3] hvmloader/tests: Introduce WRFSBASE test Andrew Cooper

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=1406656144-24882-2-git-send-email-andrew.cooper3@citrix.com \
    --to=andrew.cooper3@citrix.com \
    --cc=Ian.Campbell@citrix.com \
    --cc=Ian.Jackson@eu.citrix.com \
    --cc=JBeulich@suse.com \
    --cc=keir@xen.org \
    --cc=tim@xen.org \
    --cc=xen-devel@lists.xen.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).