All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] x86emul: correct stub invocation constraints
@ 2017-04-25  9:04 Jan Beulich
  2017-04-25 15:00 ` Andrew Cooper
  2017-04-26 14:01 ` Boris Ostrovsky
  0 siblings, 2 replies; 8+ messages in thread
From: Jan Beulich @ 2017-04-25  9:04 UTC (permalink / raw)
  To: xen-devel; +Cc: Andrew Cooper, Julien Grall

[-- Attachment #1: Type: text/plain, Size: 1513 bytes --]

Stub invocations need to have the space the stub occupies as an input,
to prevent the compiler from re-ordering (or omitting) writes to it.

Signed-off-by: Jan Beulich <jbeulich@suse.com>

--- a/xen/arch/x86/x86_emulate/x86_emulate.c
+++ b/xen/arch/x86/x86_emulate/x86_emulate.c
@@ -837,7 +837,8 @@ do{ asm volatile (
                    ".popsection\n\t"                                    \
                    _ASM_EXTABLE(.Lret%=, .Lfix%=)                       \
                    : [exn] "+g" (res_), constraints,                    \
-                     [stub] "rm" (stub.func) );                         \
+                     [stub] "rm" (stub.func),                           \
+                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.ptr) );   \
     if ( unlikely(~res_.raw) )                                          \
     {                                                                   \
         gprintk(XENLOG_WARNING,                                         \
@@ -853,7 +854,8 @@ do{ asm volatile (
 #else
 # define invoke_stub(pre, post, constraints...)                         \
     asm volatile ( pre "\n\tcall *%[stub]\n\t" post                     \
-                   : constraints, [stub] "rm" (stub.func) )
+                   : constraints, [stub] "rm" (stub.func),              \
+                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.buf) )
 #endif
 
 #define emulate_stub(dst, src...) do {                                  \




[-- Attachment #2: x86emul-stub-constraints.patch --]
[-- Type: text/plain, Size: 1555 bytes --]

x86emul: correct stub invocation constraints

Stub invocations need to have the space the stub occupies as an input,
to prevent the compiler from re-ordering (or omitting) writes to it.

Signed-off-by: Jan Beulich <jbeulich@suse.com>

--- a/xen/arch/x86/x86_emulate/x86_emulate.c
+++ b/xen/arch/x86/x86_emulate/x86_emulate.c
@@ -837,7 +837,8 @@ do{ asm volatile (
                    ".popsection\n\t"                                    \
                    _ASM_EXTABLE(.Lret%=, .Lfix%=)                       \
                    : [exn] "+g" (res_), constraints,                    \
-                     [stub] "rm" (stub.func) );                         \
+                     [stub] "rm" (stub.func),                           \
+                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.ptr) );   \
     if ( unlikely(~res_.raw) )                                          \
     {                                                                   \
         gprintk(XENLOG_WARNING,                                         \
@@ -853,7 +854,8 @@ do{ asm volatile (
 #else
 # define invoke_stub(pre, post, constraints...)                         \
     asm volatile ( pre "\n\tcall *%[stub]\n\t" post                     \
-                   : constraints, [stub] "rm" (stub.func) )
+                   : constraints, [stub] "rm" (stub.func),              \
+                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.buf) )
 #endif
 
 #define emulate_stub(dst, src...) do {                                  \

[-- Attachment #3: Type: text/plain, Size: 127 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86emul: correct stub invocation constraints
  2017-04-25  9:04 [PATCH] x86emul: correct stub invocation constraints Jan Beulich
@ 2017-04-25 15:00 ` Andrew Cooper
  2017-04-25 16:48   ` Julien Grall
  2017-04-26 14:01 ` Boris Ostrovsky
  1 sibling, 1 reply; 8+ messages in thread
From: Andrew Cooper @ 2017-04-25 15:00 UTC (permalink / raw)
  To: Jan Beulich, xen-devel; +Cc: Julien Grall

On 25/04/17 10:04, Jan Beulich wrote:
> Stub invocations need to have the space the stub occupies as an input,
> to prevent the compiler from re-ordering (or omitting) writes to it.
>
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86emul: correct stub invocation constraints
  2017-04-25 15:00 ` Andrew Cooper
@ 2017-04-25 16:48   ` Julien Grall
  0 siblings, 0 replies; 8+ messages in thread
From: Julien Grall @ 2017-04-25 16:48 UTC (permalink / raw)
  To: Andrew Cooper, Jan Beulich, xen-devel



On 25/04/17 16:00, Andrew Cooper wrote:
> On 25/04/17 10:04, Jan Beulich wrote:
>> Stub invocations need to have the space the stub occupies as an input,
>> to prevent the compiler from re-ordering (or omitting) writes to it.
>>
>> Signed-off-by: Jan Beulich <jbeulich@suse.com>
>
> Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>

Release-acked-by: Julien Grall <julien.grall@arm.com>

>

-- 
Julien Grall

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86emul: correct stub invocation constraints
  2017-04-25  9:04 [PATCH] x86emul: correct stub invocation constraints Jan Beulich
  2017-04-25 15:00 ` Andrew Cooper
@ 2017-04-26 14:01 ` Boris Ostrovsky
  2017-04-26 14:19   ` Jan Beulich
  2017-04-27  8:07   ` Jan Beulich
  1 sibling, 2 replies; 8+ messages in thread
From: Boris Ostrovsky @ 2017-04-26 14:01 UTC (permalink / raw)
  To: Jan Beulich, xen-devel; +Cc: Andrew Cooper, Julien Grall

On 04/25/2017 05:04 AM, Jan Beulich wrote:
> Stub invocations need to have the space the stub occupies as an input,
> to prevent the compiler from re-ordering (or omitting) writes to it.
>
> Signed-off-by: Jan Beulich <jbeulich@suse.com>
>
> --- a/xen/arch/x86/x86_emulate/x86_emulate.c
> +++ b/xen/arch/x86/x86_emulate/x86_emulate.c
> @@ -837,7 +837,8 @@ do{ asm volatile (
>                     ".popsection\n\t"                                    \
>                     _ASM_EXTABLE(.Lret%=, .Lfix%=)                       \
>                     : [exn] "+g" (res_), constraints,                    \
> -                     [stub] "rm" (stub.func) );                         \
> +                     [stub] "rm" (stub.func),                           \
> +                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.ptr) );   \
>      if ( unlikely(~res_.raw) )                                          \
>      {                                                                   \
>          gprintk(XENLOG_WARNING,                                         \
> @@ -853,7 +854,8 @@ do{ asm volatile (
>  #else
>  # define invoke_stub(pre, post, constraints...)                         \
>      asm volatile ( pre "\n\tcall *%[stub]\n\t" post                     \
> -                   : constraints, [stub] "rm" (stub.func) )
> +                   : constraints, [stub] "rm" (stub.func),              \
> +                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.buf) )
>  #endif
>  
>  #define emulate_stub(dst, src...) do {                                  \
>


This breaks on old compilers:

FC-64
<build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_emulator>
gcc --version
gcc (GCC) 4.4.4 20100503 (Red Hat 4.4.4-2)
Copyright (C) 2010 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

FC-64
<build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_emulator>
pwd
/home/build/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_emulator
FC-64
<build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_emulator>
gcc  -m64 -DBUILD_ID -fno-strict-aliasing -std=gnu99 -Wall
-Wstrict-prototypes -Wdeclaration-after-statement
-Wno-unused-but-set-variable   -g3 -O0 -fno-omit-frame-pointer
-D__XEN_INTERFACE_VERSION__=__XEN_LATEST_INTERFACE_VERSION__ -MMD -MF
.x86_emulate.o.d -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE  
-I/home/build/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_emulator/../../../tools/include
-D__XEN_TOOLS__ -I.  -c -o x86_emulate.o x86_emulate.c
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c: In function ‘x86_emulate’:
x86_emulate/x86_emulate.c:4085: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:4161: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:4226: error: memory input 5 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:4229: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:4279: error: memory input 5 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:4288: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:4353: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:4402: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:4465: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:4516: error: memory input 5 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:4522: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:5632: error: memory input 5 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:5679: error: memory input 8 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:5863: error: memory input 3 is not directly
addressable
x86_emulate/x86_emulate.c:6069: error: memory input 4 is not directly
addressable
x86_emulate/x86_emulate.c:6213: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:7029: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:7178: error: memory input 6 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:7328: error: memory input 7 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:7362: error: memory input 6 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:7482: error: memory input 3 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:7603: error: memory input 9 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:7607: error: memory input 9 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:7676: error: memory input 6 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:7715: error: memory input 6 is not directly
addressable
In file included from x86_emulate.c:157:
x86_emulate/x86_emulate.c:7795: error: memory input 4 is not directly
addressable
x86_emulate/x86_emulate.c:7798: error: memory input 3 is not directly
addressable
FC-64
<build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_emulator>



-boris

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86emul: correct stub invocation constraints
  2017-04-26 14:01 ` Boris Ostrovsky
@ 2017-04-26 14:19   ` Jan Beulich
  2017-04-26 14:29     ` Boris Ostrovsky
  2017-04-27  8:07   ` Jan Beulich
  1 sibling, 1 reply; 8+ messages in thread
From: Jan Beulich @ 2017-04-26 14:19 UTC (permalink / raw)
  To: Boris Ostrovsky; +Cc: Andrew Cooper, Julien Grall, xen-devel

>>> On 26.04.17 at 16:01, <boris.ostrovsky@oracle.com> wrote:
> On 04/25/2017 05:04 AM, Jan Beulich wrote:
>> Stub invocations need to have the space the stub occupies as an input,
>> to prevent the compiler from re-ordering (or omitting) writes to it.
>>
>> Signed-off-by: Jan Beulich <jbeulich@suse.com>
>>
>> --- a/xen/arch/x86/x86_emulate/x86_emulate.c
>> +++ b/xen/arch/x86/x86_emulate/x86_emulate.c
>> @@ -837,7 +837,8 @@ do{ asm volatile (
>>                     ".popsection\n\t"                                    \
>>                     _ASM_EXTABLE(.Lret%=, .Lfix%=)                       \
>>                     : [exn] "+g" (res_), constraints,                    \
>> -                     [stub] "rm" (stub.func) );                         \
>> +                     [stub] "rm" (stub.func),                           \
>> +                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.ptr) );   \
>>      if ( unlikely(~res_.raw) )                                          \
>>      {                                                                   \
>>          gprintk(XENLOG_WARNING,                                         \
>> @@ -853,7 +854,8 @@ do{ asm volatile (
>>  #else
>>  # define invoke_stub(pre, post, constraints...)                         \
>>      asm volatile ( pre "\n\tcall *%[stub]\n\t" post                     \
>> -                   : constraints, [stub] "rm" (stub.func) )
>> +                   : constraints, [stub] "rm" (stub.func),              \
>> +                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.buf) )
>>  #endif
>>  
>>  #define emulate_stub(dst, src...) do {                                  \
>>
> 
> 
> This breaks on old compilers:
> 
> FC-64
> <build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_em
> ulator>
> gcc --version
> gcc (GCC) 4.4.4 20100503 (Red Hat 4.4.4-2)

I did try with 4.3.x, fwiw (but I'm afraid I've lost that machine just
now, and will hardly set it up again using an old distro). Also I can't
immediately see what the compiler dislikes and hence how a fix may
look like (short of adding memory clobbers instead).

Jan

> Copyright (C) 2010 Free Software Foundation, Inc.
> This is free software; see the source for copying conditions.  There is NO
> warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
> 
> FC-64
> <build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_em
> ulator>
> pwd
> /home/build/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_emulator
> FC-64
> <build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_em
> ulator>
> gcc  -m64 -DBUILD_ID -fno-strict-aliasing -std=gnu99 -Wall
> -Wstrict-prototypes -Wdeclaration-after-statement
> -Wno-unused-but-set-variable   -g3 -O0 -fno-omit-frame-pointer
> -D__XEN_INTERFACE_VERSION__=__XEN_LATEST_INTERFACE_VERSION__ -MMD -MF
> .x86_emulate.o.d -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE  
> -I/home/build/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_emulat
> or/../../../tools/include
> -D__XEN_TOOLS__ -I.  -c -o x86_emulate.o x86_emulate.c
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c: In function ‘x86_emulate’:
> x86_emulate/x86_emulate.c:4085: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:4161: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:4226: error: memory input 5 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:4229: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:4279: error: memory input 5 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:4288: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:4353: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:4402: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:4465: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:4516: error: memory input 5 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:4522: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:5632: error: memory input 5 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:5679: error: memory input 8 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:5863: error: memory input 3 is not directly
> addressable
> x86_emulate/x86_emulate.c:6069: error: memory input 4 is not directly
> addressable
> x86_emulate/x86_emulate.c:6213: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:7029: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:7178: error: memory input 6 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:7328: error: memory input 7 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:7362: error: memory input 6 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:7482: error: memory input 3 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:7603: error: memory input 9 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:7607: error: memory input 9 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:7676: error: memory input 6 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:7715: error: memory input 6 is not directly
> addressable
> In file included from x86_emulate.c:157:
> x86_emulate/x86_emulate.c:7795: error: memory input 4 is not directly
> addressable
> x86_emulate/x86_emulate.c:7798: error: memory input 3 is not directly
> addressable
> FC-64
> <build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_em
> ulator>
> 
> 
> 
> -boris



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86emul: correct stub invocation constraints
  2017-04-26 14:19   ` Jan Beulich
@ 2017-04-26 14:29     ` Boris Ostrovsky
  0 siblings, 0 replies; 8+ messages in thread
From: Boris Ostrovsky @ 2017-04-26 14:29 UTC (permalink / raw)
  To: Jan Beulich; +Cc: Andrew Cooper, Julien Grall, xen-devel


>> This breaks on old compilers:
>>
>> FC-64
>> <build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_em
>> ulator>
>> gcc --version
>> gcc (GCC) 4.4.4 20100503 (Red Hat 4.4.4-2)
> I did try with 4.3.x, fwiw (but I'm afraid I've lost that machine just
> now, and will hardly set it up again using an old distro). Also I can't
> immediately see what the compiler dislikes and hence how a fix may
> look like (short of adding memory clobbers instead).

This is Fedora 13, if that helps.

I can probably arrange your access to that environment (copying Konrad
who owns it).

-boris



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86emul: correct stub invocation constraints
  2017-04-26 14:01 ` Boris Ostrovsky
  2017-04-26 14:19   ` Jan Beulich
@ 2017-04-27  8:07   ` Jan Beulich
  2017-04-27  8:30     ` Jan Beulich
  1 sibling, 1 reply; 8+ messages in thread
From: Jan Beulich @ 2017-04-27  8:07 UTC (permalink / raw)
  To: Boris Ostrovsky; +Cc: Andrew Cooper, Julien Grall, xen-devel

>>> On 26.04.17 at 16:01, <boris.ostrovsky@oracle.com> wrote:
> On 04/25/2017 05:04 AM, Jan Beulich wrote:
>> Stub invocations need to have the space the stub occupies as an input,
>> to prevent the compiler from re-ordering (or omitting) writes to it.
>>
>> Signed-off-by: Jan Beulich <jbeulich@suse.com>
>>
>> --- a/xen/arch/x86/x86_emulate/x86_emulate.c
>> +++ b/xen/arch/x86/x86_emulate/x86_emulate.c
>> @@ -837,7 +837,8 @@ do{ asm volatile (
>>                     ".popsection\n\t"                                    \
>>                     _ASM_EXTABLE(.Lret%=, .Lfix%=)                       \
>>                     : [exn] "+g" (res_), constraints,                    \
>> -                     [stub] "rm" (stub.func) );                         \
>> +                     [stub] "rm" (stub.func),                           \
>> +                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.ptr) );   \
>>      if ( unlikely(~res_.raw) )                                          \
>>      {                                                                   \
>>          gprintk(XENLOG_WARNING,                                         \
>> @@ -853,7 +854,8 @@ do{ asm volatile (
>>  #else
>>  # define invoke_stub(pre, post, constraints...)                         \
>>      asm volatile ( pre "\n\tcall *%[stub]\n\t" post                     \
>> -                   : constraints, [stub] "rm" (stub.func) )
>> +                   : constraints, [stub] "rm" (stub.func),              \
>> +                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.buf) )
>>  #endif
>>  
>>  #define emulate_stub(dst, src...) do {                                  \
>>
> 
> 
> This breaks on old compilers:
> 
> FC-64
> <build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_em
> ulator>
> gcc --version
> gcc (GCC) 4.4.4 20100503 (Red Hat 4.4.4-2)

Btw., I've just realized that I did use an old gcc only on the
hypervisor build. Do you see the same issue there, or is this tools
side specific?

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86emul: correct stub invocation constraints
  2017-04-27  8:07   ` Jan Beulich
@ 2017-04-27  8:30     ` Jan Beulich
  0 siblings, 0 replies; 8+ messages in thread
From: Jan Beulich @ 2017-04-27  8:30 UTC (permalink / raw)
  To: Boris Ostrovsky; +Cc: Andrew Cooper, Julien Grall, xen-devel

>>> On 27.04.17 at 10:07, <JBeulich@suse.com> wrote:
>>>> On 26.04.17 at 16:01, <boris.ostrovsky@oracle.com> wrote:
>> On 04/25/2017 05:04 AM, Jan Beulich wrote:
>>> Stub invocations need to have the space the stub occupies as an input,
>>> to prevent the compiler from re-ordering (or omitting) writes to it.
>>>
>>> Signed-off-by: Jan Beulich <jbeulich@suse.com>
>>>
>>> --- a/xen/arch/x86/x86_emulate/x86_emulate.c
>>> +++ b/xen/arch/x86/x86_emulate/x86_emulate.c
>>> @@ -837,7 +837,8 @@ do{ asm volatile (
>>>                     ".popsection\n\t"                                    \
>>>                     _ASM_EXTABLE(.Lret%=, .Lfix%=)                       \
>>>                     : [exn] "+g" (res_), constraints,                    \
>>> -                     [stub] "rm" (stub.func) );                         \
>>> +                     [stub] "rm" (stub.func),                           \
>>> +                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.ptr) );   \
>>>      if ( unlikely(~res_.raw) )                                          \
>>>      {                                                                   \
>>>          gprintk(XENLOG_WARNING,                                         \
>>> @@ -853,7 +854,8 @@ do{ asm volatile (
>>>  #else
>>>  # define invoke_stub(pre, post, constraints...)                         \
>>>      asm volatile ( pre "\n\tcall *%[stub]\n\t" post                     \
>>> -                   : constraints, [stub] "rm" (stub.func) )
>>> +                   : constraints, [stub] "rm" (stub.func),              \
>>> +                     "m" (*(uint8_t(*)[MAX_INST_LEN + 1])stub.buf) )
>>>  #endif
>>>  
>>>  #define emulate_stub(dst, src...) do {                                  \
>>>
>> 
>> 
>> This breaks on old compilers:
>> 
>> FC-64
>> <build@build-mk2:~/xtt-x86_64/bootstrap/xen.git/tools/fuzz/x86_instruction_em
>> ulator>
>> gcc --version
>> gcc (GCC) 4.4.4 20100503 (Red Hat 4.4.4-2)
> 
> Btw., I've just realized that I did use an old gcc only on the
> hypervisor build. Do you see the same issue there, or is this tools
> side specific?

And now that I've extracted it into a smaller example and thus was
able to try, I can see the issue with 4.3.x. The problem is that with

struct stub {
    unsigned long addr;
    void *ptr;
    unsigned char buf[16];
};

void test(const struct stub*ptr) {
	asm("" :: "m" (*(unsigned char(*)[16])ptr->addr));
	asm("" :: "m" (*(unsigned char(*)[16])ptr->ptr));
	asm("" :: "m" (*(unsigned char(*)[16])ptr->buf));
	asm("" :: "m" (*(unsigned char(*)[16])&ptr->buf));
	asm("" :: "m" (*(unsigned char(*)[16])&ptr->buf[0]));
}

none of the last three work, so we'll have to resort to using the
first. I'll have to verify that this is good enough for the case
where I did actually observe things to break without the extra
constraint (with a not yet submitted patch).

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2017-04-27  8:30 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-04-25  9:04 [PATCH] x86emul: correct stub invocation constraints Jan Beulich
2017-04-25 15:00 ` Andrew Cooper
2017-04-25 16:48   ` Julien Grall
2017-04-26 14:01 ` Boris Ostrovsky
2017-04-26 14:19   ` Jan Beulich
2017-04-26 14:29     ` Boris Ostrovsky
2017-04-27  8:07   ` Jan Beulich
2017-04-27  8:30     ` Jan Beulich

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.