summaryrefslogtreecommitdiff
path: root/lib/Target/X86/X86JITInfo.cpp
diff options
context:
space:
mode:
authorJeffrey Yasskin <jyasskin@google.com>2009-11-16 22:41:33 +0000
committerJeffrey Yasskin <jyasskin@google.com>2009-11-16 22:41:33 +0000
commitd1ba06bf131a9d217426529d2e28af1f2eeed47a (patch)
tree72ae565430358edb4e81b988c8725938c1f60763 /lib/Target/X86/X86JITInfo.cpp
parentda589a3a963e6cc179d850c5fd395d3e10ce741c (diff)
downloadllvm-d1ba06bf131a9d217426529d2e28af1f2eeed47a.tar.gz
llvm-d1ba06bf131a9d217426529d2e28af1f2eeed47a.tar.bz2
llvm-d1ba06bf131a9d217426529d2e28af1f2eeed47a.tar.xz
Make X86-64 in the Large model always emit 64-bit calls.
The large code model is documented at http://www.x86-64.org/documentation/abi.pdf and says that calls should assume their target doesn't live within the 32-bit pc-relative offset that fits in the call instruction. To do this, we turn off the global-address->target-global-address conversion in X86TargetLowering::LowerCall(). The first attempt at this broke the lazy JIT because it can separate the movabs(imm->reg) from the actual call instruction. The lazy JIT receives the address of the movabs as a relocation and needs to record the return address from the call; and then when that call happens, it needs to patch the movabs with the newly-compiled target. We could thread the call instruction into the relocation and record the movabs<->call mapping explicitly, but that seems to require at least as much new complication in the code generator as this change. To fix this, we make lazy functions _always_ go through a call stub. You'd think we'd only have to force lazy calls through a stub on difficult platforms, but that turns out to break indirect calls through a function pointer. The right fix for that is to distinguish between calls and address-of operations on uncompiled functions, but that's complex enough to leave for someone else to do. Another attempt at this defined a new CALL64i pseudo-instruction, which expanded to a 2-instruction sequence in the assembly output and was special-cased in the X86CodeEmitter's emitInstruction() function. That broke indirect calls in the same way as above. This patch also removes a hack forcing Darwin to the small code model. Without far-call-stubs, the small code model requires things of the JITMemoryManager that the DefaultJITMemoryManager can't provide. Thanks to echristo for lots of testing! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@88984 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/Target/X86/X86JITInfo.cpp')
-rw-r--r--lib/Target/X86/X86JITInfo.cpp5
1 files changed, 3 insertions, 2 deletions
diff --git a/lib/Target/X86/X86JITInfo.cpp b/lib/Target/X86/X86JITInfo.cpp
index 62ca47ff78..0792bdd4dd 100644
--- a/lib/Target/X86/X86JITInfo.cpp
+++ b/lib/Target/X86/X86JITInfo.cpp
@@ -367,8 +367,9 @@ X86CompilationCallback2(intptr_t *StackPtr, intptr_t RetAddr) {
// Rewrite the call target... so that we don't end up here every time we
// execute the call.
#if defined (X86_64_JIT)
- if (!isStub)
- *(intptr_t *)(RetAddr - 0xa) = NewVal;
+ assert(isStub &&
+ "X86-64 doesn't support rewriting non-stub lazy compilation calls:"
+ " the call instruction varies too much.");
#else
*(intptr_t *)RetAddr = (intptr_t)(NewVal-RetAddr-4);
#endif