[ovs-discuss] [PATCH net-next v2] fast_hash: clobber registers correctly for inline function use

[Hannes Frederic Sowa]

In case the arch_fast_hash call gets inlined we need to tell gcc which
registers are clobbered with. rhashtable was fine, because it used
arch_fast_hash via function pointer and thus the compiler took care of
that. In case of openvswitch the call got inlined and arch_fast_hash
touched registeres which gcc didn't know about.

Also don't use conditional compilation inside arguments, as this confuses
sparse.

Reported-by: Jay Vosburgh <jay.vosburgh at canonical.com>
Cc: Pravin Shelar <pshelar at nicira.com>
Cc: Jesse Gross <jesse at nicira.com>
Signed-off-by: Hannes Frederic Sowa <hannes at stressinduktion.org>
---

v2)
After studying gcc documentation again, it occured to me that I need to
specificy all input operands in the clobber section, too. Otherwise gcc
can expect that the inline assembler section won't modify the inputs,
which is not true.

Bye,
Hannes


 arch/x86/include/asm/hash.h | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/arch/x86/include/asm/hash.h b/arch/x86/include/asm/hash.h
index a881d78..a25c45a 100644
--- a/arch/x86/include/asm/hash.h
+++ b/arch/x86/include/asm/hash.h
@@ -23,11 +23,15 @@ static inline u32 arch_fast_hash(const void *data, u32 len, u32 seed)
 {
 	u32 hash;
 
-	alternative_call(__jhash, __intel_crc4_2_hash, X86_FEATURE_XMM4_2,
 #ifdef CONFIG_X86_64
-			 "=a" (hash), "D" (data), "S" (len), "d" (seed));
+	alternative_call(__jhash, __intel_crc4_2_hash, X86_FEATURE_XMM4_2,
+			 "=a" (hash), "D" (data), "S" (len), "d" (seed)
+			 : "rdi", "rsi", "rdx", "rcx", "r8", "r9", "r10", "r11",
+			   "cc", "memory");
 #else
-			 "=a" (hash), "a" (data), "d" (len), "c" (seed));
+	alternative_call(__jhash, __intel_crc4_2_hash, X86_FEATURE_XMM4_2,
+			 "=a" (hash), "a" (data), "d" (len), "c" (seed)
+			 : "edx", "ecx", "cc", "memory");
 #endif
 	return hash;
 }
@@ -36,11 +40,15 @@ static inline u32 arch_fast_hash2(const u32 *data, u32 len, u32 seed)
 {
 	u32 hash;
 
-	alternative_call(__jhash2, __intel_crc4_2_hash2, X86_FEATURE_XMM4_2,
 #ifdef CONFIG_X86_64
-			 "=a" (hash), "D" (data), "S" (len), "d" (seed));
+	alternative_call(__jhash2, __intel_crc4_2_hash2, X86_FEATURE_XMM4_2,
+			 "=a" (hash), "D" (data), "S" (len), "d" (seed)
+			 : "rdi", "rsi", "rdx", "rcx", "r8", "r9", "r10", "r11",
+			   "cc", "memory");
 #else
-			 "=a" (hash), "a" (data), "d" (len), "c" (seed));
+	alternative_call(__jhash2, __intel_crc4_2_hash2, X86_FEATURE_XMM4_2,
+			 "=a" (hash), "a" (data), "d" (len), "c" (seed)
+			 : "edx", "ecx", "cc", "memory");
 #endif
 	return hash;
 }
-- 
1.9.3