The compiler is not able to do constant folding on "asm volatile" code.
Evaluate whether or not the function argument is a constant expression
and if this is the case, return an equivalent builtin expression.
On linux 6.7 with an allyesconfig and clang 17.0.6, it saves roughly
4 KB.
$ size --format=GNU vmlinux.before vmlinux.after
text data bss total filename
4827900 1798340 364057 6990297 vmlinux.before
4827072 1795060 364057 6986189 vmlinux.after
Reference: commit fdb6649ab7c1 ("x86/asm/bitops: Use __builtin_ctzl() to evaluate constant expressions")
Link: https://git.kernel.org/torvalds/c/fdb6649ab7c1
Signed-off-by: Vincent Mailhol <mailhol.vincent@xxxxxxxxxx>
---
arch/hexagon/include/asm/bitops.h | 15 +++++++++++++++
1 file changed, 15 insertions(+)
diff --git a/arch/hexagon/include/asm/bitops.h b/arch/hexagon/include/asm/bitops.h
index e856d6dbfe16..c74a639c84f3 100644
--- a/arch/hexagon/include/asm/bitops.h
+++ b/arch/hexagon/include/asm/bitops.h
@@ -204,6 +204,9 @@ static __always_inline long ffz(int x)
{
int r;
+ if (__builtin_constant_p(x))
+ return __builtin_ctzl(~x);
+
asm("%0 = ct1(%1);\n"
: "=&r" (r)
: "r" (x));
@@ -221,6 +224,9 @@ static __always_inline int fls(unsigned int x)
{
int r;
+ if (__builtin_constant_p(x))
+ return x ? BITS_PER_TYPE(x) - __builtin_clz(x) : 0;
+
asm("{ %0 = cl0(%1);}\n"
"%0 = sub(#32,%0);\n"
: "=&r" (r)
@@ -242,6 +248,9 @@ static __always_inline int ffs(int x)
{
int r;
+ if (__builtin_constant_p(x))
+ return __builtin_ffs(x);
+
asm("{ P0 = cmp.eq(%1,#0); %0 = ct0(%1);}\n"
"{ if (P0) %0 = #0; if (!P0) %0 = add(%0,#1);}\n"
: "=&r" (r)
@@ -264,6 +273,9 @@ static __always_inline unsigned long __ffs(unsigned long word)
{
int num;
+ if (__builtin_constant_p(word))
+ return __builtin_ctzl(word);
+
asm("%0 = ct0(%1);\n"
: "=&r" (num)
: "r" (word));
@@ -282,6 +294,9 @@ static __always_inline unsigned long __fls(unsigned long word)
{
int num;
+ if (__builtin_constant_p(word))
+ return BITS_PER_LONG - 1 - __builtin_clzl(word);
+
asm("%0 = cl0(%1);\n"
"%0 = sub(#31,%0);\n"
: "=&r" (num)
--
2.43.0
