xen/x86: Work around Clang code generation bug with asm parameters
PATCH 1/1 xen/x86: Work around Clang code generation bug with asm parameters
Clang 9 and later don't handle the clobber of %r10 correctly in _hypercall64_4(). See https://bugs.llvm.org/show_bug.cgi?id=48122
Rewrite the logic to use the "+r" form, rather than using separate input and output specifiers, which works around the issue. For consistency, adjust all operand handling.
---
CC: Jan Beulich <[email protected]>
CC: Roger Pau Monné <[email protected]>
CC: Wei Liu <[email protected]>
--- xen/include/asm-x86/guest/xen-hcall.h | 35 ++++++++++++++++------------------- 1 file changed, 16 insertions(+), 19 deletions(-)
xen/include/asm-x86/guest/xen-hcall.h +16 -19
@@ -39,53 +39,50 @@#define _hypercall64_1 (type, hcall, a1) \ ({ \ - long res, tmp__; \ + long res, _a1 = (long)(a1); \ asm volatile ( \ "call hypercall_page + %c[offset]" \ - : "=a" (res), "=D" (tmp__) ASM_CALL_CONSTRAINT \ - : [offset] "i" (hcall * 32), \ - "1" ((long)(a1)) \ + : "=a" (res), "+D" (_a1) ASM_CALL_CONSTRAINT \ + : [offset] "i" (hcall * 32) \ : "memory" ); \ (type)res; \ })#define _hypercall64_2 (type, hcall, a1, a2) \ ({ \ - long res, tmp__; \ + long res, _a1 = (long)(a1), _a2 = (long)(a2); \ asm volatile ( \ "call hypercall_page + %c[offset]" \ - : "=a" (res), "=D" (tmp__), "=S" (tmp__) \ + : "=a" (res), "+D" (_a1), "+S" (_a2) \ ASM_CALL_CONSTRAINT \ - : [offset] "i" (hcall * 32), \ - "1" ((long)(a1)), "2" ((long)(a2)) \ + : [offset] "i" (hcall * 32) \ : "memory" ); \ (type)res; \ })#define _hypercall64_3 (type, hcall, a1, a2, a3) \ ({ \ - long res, tmp__; \ + long res, _a1 = (long)(a1), _a2 = (long)(a2), \ + _a3 = (long)(a3); \ asm volatile ( \ "call hypercall_page + %c[offset]" \ - : "=a" (res), "=D" (tmp__), "=S" (tmp__), "=d" (tmp__) \ + : "=a" (res), "+D" (_a1), "+S" (_a2), "+d" (_a3) \ ASM_CALL_CONSTRAINT \ - : [offset] "i" (hcall * 32), \ - "1" ((long)(a1)), "2" ((long)(a2)), "3" ((long)(a3)) \ + : [offset] "i" (hcall * 32) \ : "memory" ); \ (type)res; \ })#define _hypercall64_4 (type, hcall, a1, a2, a3, a4) \ ({ \ - long res, tmp__; \ - register long _a4 asm ("r10") = ((long)(a4)); \ + long res, _a1 = (long)(a1), _a2 = (long)(a2), \ + _a3 = (long)(a3); \ + register long _a4 asm ("r10") = (long)(a4); \ asm volatile ( \ "call hypercall_page + %c[offset]" \ - : "=a" (res), "=D" (tmp__), "=S" (tmp__), "=d" (tmp__), \ - "=&r" (tmp__) ASM_CALL_CONSTRAINT \ - : [offset] "i" (hcall * 32), \ - "1" ((long)(a1)), "2" ((long)(a2)), "3" ((long)(a3)), \ - "4" (_a4) \ + : "=a" (res), "+D" (_a1), "+S" (_a2), "+d" (_a3) \ + "+r" (_a4) ASM_CALL_CONSTRAINT \ + : [offset] "i" (hcall * 32) \ : "memory" ); \ (type)res; \ })
Signature · 1 line
2.11.0
Quoted 2 lines · Andrew Cooper
Clang 9 and later don't handle the clobber of %r10 correctly in
_hypercall64_4(). See https://bugs.llvm.org/show_bug.cgi?id=48122
Are you sure this is a bug? With ...
xen/include/asm-x86/guest/xen-hcall.h line 85
Quoted 11 lines
#define _hypercall64_4(type, hcall, a1, a2, a3, a4) \ ({ \xen/include/asm-x86/guest/xen-hcall.h +3
-4-long res, tmp__; \ - register long _a4 asm ("r10") = ((long)(a4)); \ +long res, _a1 = (long )(a1), _a2 = (long )(a2), \ + _a3 = (long )(a3); \ + register long _a4 asm ("r10") = (long)(a4); \ asmvolatile ( \ "call hypercall_page + %c[offset]" \ - : "=a" (res), "=D" (tmp__), "=S" (tmp__), "=d" (tmp__), \ - "=&r" (tmp__) ASM_CALL_CONSTRAINT \
... this we've requested "any register", while with ...
xen/include/asm-x86/guest/xen-hcall.h line 88
Quoted 3 lines
xen/include/asm-x86/guest/xen-hcall.h +0
-3- : [offset] "i" (hcall * 32), \ - "1" ((long)(a1)), "2" ((long)(a2)), "3" ((long)(a3)), \ -"4" (_a4) \
... this we've asked for that specific register to be initialized from r10 (and without telling the compiler that r10 is going to change).
Also, by what I would have hoped is convention meanwhile, the new macro local variables' names shouldn't start with an underscore, but instead perhaps end in one. But to be honest, despite knowing of the latent (albeit highly hypothetical) issue, each time I find myself making such a comment I'm one tiny step closer to giving up.
Anyway, with at least title and description changed (or your view clarified verbally)
Jan
Quoted 4 lines · Jan Beulich
Quoted 2 lines · Andrew Cooper
Clang 9 and later don't handle the clobber of %r10 correctly in
_hypercall64_4(). See https://bugs.llvm.org/show_bug.cgi?id=48122Are you sure this is a bug?
Yes.
Quoted 13 lines
With ...
xen/include/asm-x86/guest/xen-hcall.h line 85
Quoted 11 lines
#define _hypercall64_4(type, hcall, a1, a2, a3, a4) \ ({ \xen/include/asm-x86/guest/xen-hcall.h +3
-4-long res, tmp__; \ - register long _a4 asm ("r10") = ((long)(a4)); \ +long res, _a1 = (long )(a1), _a2 = (long )(a2), \ + _a3 = (long )(a3); \ + register long _a4 asm ("r10") = (long)(a4); \ asmvolatile ( \ "call hypercall_page + %c[offset]" \ - : "=a" (res), "=D" (tmp__), "=S" (tmp__), "=d" (tmp__), \ - "=&r" (tmp__) ASM_CALL_CONSTRAINT \
Quoted 8 lines
... this we've requested "any register", while with ...
xen/include/asm-x86/guest/xen-hcall.h line 88
Quoted 3 lines
xen/include/asm-x86/guest/xen-hcall.h +0
-3- : [offset] "i" (hcall * 32), \ - "1" ((long)(a1)), "2" ((long)(a2)), "3" ((long)(a3)), \ -"4" (_a4) \... this we've asked for that specific register to be initialized from r10 (and without telling the compiler that r10 is going to change).
Consider applying that same reasoning to "1" instead of "4". In that case, a1 would no longer be bound to %rdi.
The use of "4" explicitly binds the input and the output, which includes requiring them to be the same register.
Furthermore, LLVM tends to consider "not behaving in the same was as GCC" a bug.
Quoted 6 lines
Also, by what I would have hoped is convention meanwhile, the new macro local variables' names shouldn't start with an underscore, but instead perhaps end in one. But to be honest, despite knowing of the latent (albeit highly hypothetical) issue, each time I find myself making such a comment I'm one tiny step closer to giving up.
Convention is not created by you getting irritated at others getting irritated at you for requesting bizarre and unusual changes in submitted code, or by continuing to point things out, knowing that others specifically disagree with your reasoning in this case.
Convention is created when there is general agreement over the technical merits of writing code in one particular way vs the alternatives, *and* its written down somewhere, so this argument does not continue any further.
There is no restrictions or guidance in the coding style to prefer one form over another, therefore anything is acceptable.
~Andrew
Quoted 26 lines · Andrew Cooper
Quoted 4 lines · Jan Beulich
Quoted 2 lines · Andrew Cooper
Clang 9 and later don't handle the clobber of %r10 correctly in
_hypercall64_4(). See https://bugs.llvm.org/show_bug.cgi?id=48122Are you sure this is a bug?
Yes.
Quoted 18 lines
With ...
xen/include/asm-x86/guest/xen-hcall.h line 85
Quoted 11 lines
#define _hypercall64_4(type, hcall, a1, a2, a3, a4) \ ({ \xen/include/asm-x86/guest/xen-hcall.h +3
-4-long res, tmp__; \ - register long _a4 asm ("r10") = ((long)(a4)); \ +long res, _a1 = (long )(a1), _a2 = (long )(a2), \ + _a3 = (long )(a3); \ + register long _a4 asm ("r10") = (long)(a4); \ asmvolatile ( \ "call hypercall_page + %c[offset]" \ - : "=a" (res), "=D" (tmp__), "=S" (tmp__), "=d" (tmp__), \ - "=&r" (tmp__) ASM_CALL_CONSTRAINT \... this we've requested "any register", while with ...
Quoted 3 lines
xen/include/asm-x86/guest/xen-hcall.h +0
-3- : [offset] "i" (hcall * 32), \ - "1" ((long)(a1)), "2" ((long)(a2)), "3" ((long)(a3)), \ -"4" (_a4) \
Quoted 6 lines
Quoted 3 lines
... this we've asked for that specific register to be initialized from r10 (and without telling the compiler that r10 is going to change).
Consider applying that same reasoning to "1" instead of "4". In that case, a1 would no longer be bound to %rdi.
That's different: "=D" specifies the register, and "1" says "use the same register as input". Whereas, as said, "=&r" says "use any register" with "1" saying "use the same register" and (_a4) specifying where the value is to come from.
Quoted 5 lines
The use of "4" explicitly binds the input and the output, which includes requiring them to be the same register.
Furthermore, LLVM tends to consider "not behaving in the same was as GCC" a bug.
That's a fair statement, but then still the description wants re-wording. Plus of course future gcc is free to change their behavior to that currently observed with clang.
Consider the following example (on an arch where "f" is a floating point register and there are ways to copy directly between GPR and floating point registers:
int i;
register float f asm("f7") = <input>;
asm("..." : "=r" (i) : "0" (f));In this case obviously f7 can't be used for i (as it doesn't match "r"). It's merely that the initial value of i is to come from f7. In fact for Arm64 this
extern float flt;
int test(void) {
int i;
register float f asm("s7") = flt;
asm("add %0,%0,5" : "=r" (i) : "0" (f));
return i;
}behaves exactly as described:
test:
adrp x0, flt
ldr s7, [x0, @lo12(flt)]
fmov w0, s7
add x0, x0, #5ret
(Whether fmov is a sensible choice here is a different question; I'd have expected some fcvt*.)
Similarly a constant initial value would demonstrate this (or one necessarily coming from memory), albeit in a less obvious
way: It doesn't say "this constant lives in the register", but "this constant is to be loaded into the register".
Quoted 12 lines
Quoted 6 lines
Also, by what I would have hoped is convention meanwhile, the new macro local variables' names shouldn't start with an underscore, but instead perhaps end in one. But to be honest, despite knowing of the latent (albeit highly hypothetical) issue, each time I find myself making such a comment I'm one tiny step closer to giving up.
Convention is not created by you getting irritated at others getting irritated at you for requesting bizarre and unusual changes in submitted code, or by continuing to point things out, knowing that others specifically disagree with your reasoning in this case.
Quoted 6 lines
Convention is created when there is general agreement over the technical merits of writing code in one particular way vs the alternatives, *and* its written down somewhere, so this argument does not continue any further.
There is no restrictions or guidance in the coding style to prefer one form over another, therefore anything is acceptable.
Our coding style document imo ought to describe things not already specified by the standard. What scopes certain identifiers are to be considered reserved in is, however, written down there already.
Jan
Quoted 73 lines · Jan Beulich
Quoted 26 lines · Andrew Cooper
Quoted 4 lines · Jan Beulich
Quoted 2 lines · Andrew Cooper
Clang 9 and later don't handle the clobber of %r10 correctly in
_hypercall64_4(). See https://bugs.llvm.org/show_bug.cgi?id=48122Are you sure this is a bug?
Yes.
Quoted 18 lines
With ...
xen/include/asm-x86/guest/xen-hcall.h line 85
Quoted 11 lines
#define _hypercall64_4(type, hcall, a1, a2, a3, a4) \ ({ \xen/include/asm-x86/guest/xen-hcall.h +3
-4-long res, tmp__; \ - register long _a4 asm ("r10") = ((long)(a4)); \ +long res, _a1 = (long )(a1), _a2 = (long )(a2), \ + _a3 = (long )(a3); \ + register long _a4 asm ("r10") = (long)(a4); \ asmvolatile ( \ "call hypercall_page + %c[offset]" \ - : "=a" (res), "=D" (tmp__), "=S" (tmp__), "=d" (tmp__), \ - "=&r" (tmp__) ASM_CALL_CONSTRAINT \... this we've requested "any register", while with ...
Quoted 3 lines
xen/include/asm-x86/guest/xen-hcall.h +0
-3- : [offset] "i" (hcall * 32), \ - "1" ((long)(a1)), "2" ((long)(a2)), "3" ((long)(a3)), \ -"4" (_a4) \Quoted 6 lines
Quoted 3 lines
... this we've asked for that specific register to be initialized from r10 (and without telling the compiler that r10 is going to change).
Consider applying that same reasoning to "1" instead of "4". In that case, a1 would no longer be bound to %rdi.
That's different: "=D" specifies the register, and "1" says "use the same register as input". Whereas, as said, "=&r" says "use any register" with "1" saying "use the same register" and (
_a4) specifying where the value is to come from.Quoted 5 lines
The use of "4" explicitly binds the input and the output, which includes requiring them to be the same register.
Furthermore, LLVM tends to consider "not behaving in the same was as GCC" a bug.
That's a fair statement, but then still the description wants re-wording. Plus of course future gcc is free to change their behavior to that currently observed with clang.
Consider the following example (on an arch where "f" is a floating point register and there are ways to copy directly between GPR and floating point registers:
int i; register float f asm("f7") = <input>; asm("..." : "=r" (i) : "0" (f));In this case obviously f7 can't be used for i (as it doesn't match "r"). It's merely that the initial value of i is to come from f7. In fact for Arm64 this
extern float flt;
int test(void) { int i; register float f asm("s7") = flt; asm("add %0,%0,5" : "=r" (i) : "0" (f)); return i; }behaves exactly as described:
test:
Quoted 8 lines
adrp x0, flt ldr s7, [x0, @lo12(flt)] fmov w0, s7 add x0, x0, #5ret
(Whether fmov is a sensible choice here is a different question; I'd have expected some fcvt*.)
Meanwhile I've realized that I neither need to resort to Arm here, nor to floating point, e.g.
int test2(int in) {
int i;
register int ri asm("ecx") = in;
asm("nop %0" : "=r" (i) : "0" (ri));
return i;
}You'll find that the resulting code (at -O2; gcc 10.2.0) doesn't use %ecx at all - %edi gets moved directly to %eax.
Jan