diff --git a/README.md b/README.md
index 51bef5e2..46983a28 100644
--- a/README.md
+++ b/README.md
@@ -103,9 +103,7 @@ rely on CI.
 - [x] aarch64/chkstk.S
 - [x] adddf3.c
 - [x] addsf3.c
-- [x] arm/adddf3vfp.S
 - [x] arm/addsf3.S
-- [x] arm/addsf3vfp.S
 - [x] arm/aeabi_dcmp.S
 - [x] arm/aeabi_fcmp.S
 - [x] arm/aeabi_idivmod.S
@@ -116,45 +114,13 @@ rely on CI.
 - [x] arm/aeabi_uidivmod.S
 - [x] arm/aeabi_uldivmod.S
 - [ ] arm/chkstk.S
-- [x] arm/divdf3vfp.S
 - [ ] arm/divmodsi4.S (generic version is done)
-- [x] arm/divsf3vfp.S
 - [ ] arm/divsi3.S (generic version is done)
-- [x] arm/eqdf2vfp.S
-- [x] arm/eqsf2vfp.S
-- [x] arm/extendsfdf2vfp.S
-- [ ] arm/fixdfsivfp.S
-- [ ] arm/fixsfsivfp.S
-- [ ] arm/fixunsdfsivfp.S
-- [ ] arm/fixunssfsivfp.S
-- [ ] arm/floatsidfvfp.S
-- [ ] arm/floatsisfvfp.S
-- [ ] arm/floatunssidfvfp.S
-- [ ] arm/floatunssisfvfp.S
-- [x] arm/gedf2vfp.S
-- [x] arm/gesf2vfp.S
-- [x] arm/gtdf2vfp.S
-- [x] arm/gtsf2vfp.S
-- [x] arm/ledf2vfp.S
-- [x] arm/lesf2vfp.S
-- [x] arm/ltdf2vfp.S
-- [x] arm/ltsf2vfp.S
 - [ ] arm/modsi3.S (generic version is done)
-- [x] arm/muldf3vfp.S
-- [x] arm/mulsf3vfp.S
-- [x] arm/nedf2vfp.S
-- [ ] arm/negdf2vfp.S
-- [ ] arm/negsf2vfp.S
-- [x] arm/nesf2vfp.S
 - [x] arm/softfloat-alias.list
-- [x] arm/subdf3vfp.S
-- [x] arm/subsf3vfp.S
-- [x] arm/truncdfsf2vfp.S
 - [ ] arm/udivmodsi4.S (generic version is done)
 - [ ] arm/udivsi3.S (generic version is done)
 - [ ] arm/umodsi3.S (generic version is done)
-- [ ] arm/unorddf2vfp.S
-- [ ] arm/unordsf2vfp.S
 - [x] ashldi3.c
 - [x] ashrdi3.c
 - [ ] avr/divmodhi4.S
@@ -501,6 +467,43 @@ Floating-point implementations of builtins that are only called from soft-float
 - ~~x86_64/floatdidf.c~~
 - ~~x86_64/floatdisf.c~~
 
+Unsupported in any current target: used on old versions of 32-bit iOS with ARMv5.
+
+- ~~arm/adddf3vfp.S~~
+- ~~arm/addsf3vfp.S~~
+- ~~arm/divdf3vfp.S~~
+- ~~arm/divsf3vfp.S~~
+- ~~arm/eqdf2vfp.S~~
+- ~~arm/eqsf2vfp.S~~
+- ~~arm/extendsfdf2vfp.S~~
+- ~~arm/fixdfsivfp.S~~
+- ~~arm/fixsfsivfp.S~~
+- ~~arm/fixunsdfsivfp.S~~
+- ~~arm/fixunssfsivfp.S~~
+- ~~arm/floatsidfvfp.S~~
+- ~~arm/floatsisfvfp.S~~
+- ~~arm/floatunssidfvfp.S~~
+- ~~arm/floatunssisfvfp.S~~
+- ~~arm/gedf2vfp.S~~
+- ~~arm/gesf2vfp.S~~
+- ~~arm/gtdf2vfp.S~~
+- ~~arm/gtsf2vfp.S~~
+- ~~arm/ledf2vfp.S~~
+- ~~arm/lesf2vfp.S~~
+- ~~arm/ltdf2vfp.S~~
+- ~~arm/ltsf2vfp.S~~
+- ~~arm/muldf3vfp.S~~
+- ~~arm/mulsf3vfp.S~~
+- ~~arm/nedf2vfp.S~~
+- ~~arm/negdf2vfp.S~~
+- ~~arm/negsf2vfp.S~~
+- ~~arm/nesf2vfp.S~~
+- ~~arm/subdf3vfp.S~~
+- ~~arm/subsf3vfp.S~~
+- ~~arm/truncdfsf2vfp.S~~
+- ~~arm/unorddf2vfp.S~~
+- ~~arm/unordsf2vfp.S~~
+
 ## License
 
 The compiler-builtins crate is dual licensed under both the University of
diff --git a/src/float/add.rs b/src/float/add.rs
index 03ed131a..bceef7b0 100644
--- a/src/float/add.rs
+++ b/src/float/add.rs
@@ -208,14 +208,4 @@ intrinsics! {
     pub extern "C" fn __addtf3(a: f128, b: f128) -> f128 {
         add(a, b)
     }
-
-    #[cfg(target_arch = "arm")]
-    pub extern "C" fn __addsf3vfp(a: f32, b: f32) -> f32 {
-        a + b
-    }
-
-    #[cfg(target_arch = "arm")]
-    pub extern "C" fn __adddf3vfp(a: f64, b: f64) -> f64 {
-        a + b
-    }
 }
diff --git a/src/float/cmp.rs b/src/float/cmp.rs
index 1901ca4b..bb7d4b49 100644
--- a/src/float/cmp.rs
+++ b/src/float/cmp.rs
@@ -258,55 +258,4 @@ intrinsics! {
     pub extern "aapcs" fn __aeabi_dcmpgt(a: f64, b: f64) -> i32 {
         (__gtdf2(a, b) > 0) as i32
     }
-
-    // On hard-float targets LLVM will use native instructions
-    // for all VFP intrinsics below
-
-    pub extern "C" fn __gesf2vfp(a: f32, b: f32) -> i32 {
-        (a >= b) as i32
-    }
-
-    pub extern "C" fn __gedf2vfp(a: f64, b: f64) -> i32 {
-        (a >= b) as i32
-    }
-
-    pub extern "C" fn __gtsf2vfp(a: f32, b: f32) -> i32 {
-        (a > b) as i32
-    }
-
-    pub extern "C" fn __gtdf2vfp(a: f64, b: f64) -> i32 {
-        (a > b) as i32
-    }
-
-    pub extern "C" fn __ltsf2vfp(a: f32, b: f32) -> i32 {
-        (a < b) as i32
-    }
-
-    pub extern "C" fn __ltdf2vfp(a: f64, b: f64) -> i32 {
-        (a < b) as i32
-    }
-
-    pub extern "C" fn __lesf2vfp(a: f32, b: f32) -> i32 {
-        (a <= b) as i32
-    }
-
-    pub extern "C" fn __ledf2vfp(a: f64, b: f64) -> i32 {
-        (a <= b) as i32
-    }
-
-    pub extern "C" fn __nesf2vfp(a: f32, b: f32) -> i32 {
-        (a != b) as i32
-    }
-
-    pub extern "C" fn __nedf2vfp(a: f64, b: f64) -> i32 {
-        (a != b) as i32
-    }
-
-    pub extern "C" fn __eqsf2vfp(a: f32, b: f32) -> i32 {
-        (a == b) as i32
-    }
-
-    pub extern "C" fn __eqdf2vfp(a: f64, b: f64) -> i32 {
-        (a == b) as i32
-    }
 }
diff --git a/src/float/div.rs b/src/float/div.rs
index c0d780b6..2a57ee1a 100644
--- a/src/float/div.rs
+++ b/src/float/div.rs
@@ -926,14 +926,4 @@ intrinsics! {
     pub extern "C" fn __divdf3(a: f64, b: f64) -> f64 {
         div64(a, b)
     }
-
-    #[cfg(target_arch = "arm")]
-    pub extern "C" fn __divsf3vfp(a: f32, b: f32) -> f32 {
-        a / b
-    }
-
-    #[cfg(target_arch = "arm")]
-    pub extern "C" fn __divdf3vfp(a: f64, b: f64) -> f64 {
-        a / b
-    }
 }
diff --git a/src/float/extend.rs b/src/float/extend.rs
index 9fabcde2..997475c8 100644
--- a/src/float/extend.rs
+++ b/src/float/extend.rs
@@ -76,11 +76,6 @@ intrinsics! {
     pub extern "C" fn  __extendsfdf2(a: f32) -> f64 {
         extend(a)
     }
-
-    #[cfg(target_arch = "arm")]
-    pub extern "C" fn  __extendsfdf2vfp(a: f32) -> f64 {
-        a as f64 // LLVM generate 'fcvtds'
-    }
 }
 
 intrinsics! {
diff --git a/src/float/mul.rs b/src/float/mul.rs
index cb0fcdfa..a4c69ea8 100644
--- a/src/float/mul.rs
+++ b/src/float/mul.rs
@@ -199,14 +199,4 @@ intrinsics! {
     pub extern "C" fn __multf3(a: f128, b: f128) -> f128 {
         mul(a, b)
     }
-
-    #[cfg(target_arch = "arm")]
-    pub extern "C" fn __mulsf3vfp(a: f32, b: f32) -> f32 {
-        a * b
-    }
-
-    #[cfg(target_arch = "arm")]
-    pub extern "C" fn __muldf3vfp(a: f64, b: f64) -> f64 {
-        a * b
-    }
 }
diff --git a/src/float/sub.rs b/src/float/sub.rs
index d33016ea..7e8a8945 100644
--- a/src/float/sub.rs
+++ b/src/float/sub.rs
@@ -23,14 +23,4 @@ intrinsics! {
 
         __addtf3(a, f128::from_repr(b.repr() ^ f128::SIGN_MASK))
     }
-
-    #[cfg(target_arch = "arm")]
-    pub extern "C" fn __subsf3vfp(a: f32, b: f32) -> f32 {
-        a - b
-    }
-
-    #[cfg(target_arch = "arm")]
-    pub extern "C" fn __subdf3vfp(a: f64, b: f64) -> f64 {
-        a - b
-    }
 }
diff --git a/src/float/trunc.rs b/src/float/trunc.rs
index 5c17cd96..a25b6eab 100644
--- a/src/float/trunc.rs
+++ b/src/float/trunc.rs
@@ -124,11 +124,6 @@ intrinsics! {
     pub extern "C" fn __truncdfsf2(a: f64) -> f32 {
         trunc(a)
     }
-
-    #[cfg(target_arch = "arm")]
-    pub extern "C" fn __truncdfsf2vfp(a: f64) -> f32 {
-        a as f32
-    }
 }
 
 intrinsics! {
diff --git a/testcrate/tests/addsub.rs b/testcrate/tests/addsub.rs
index 1ba2df74..284a2bf5 100644
--- a/testcrate/tests/addsub.rs
+++ b/testcrate/tests/addsub.rs
@@ -139,13 +139,3 @@ mod float_addsub_f128_ppc {
         f128, __addkf3, __subkf3, Quad, not(feature = "no-sys-f128");
     }
 }
-
-#[cfg(target_arch = "arm")]
-mod float_addsub_arm {
-    use super::*;
-
-    float_sum! {
-        f32, __addsf3vfp, __subsf3vfp, Single, all();
-        f64, __adddf3vfp, __subdf3vfp, Double, all();
-    }
-}
diff --git a/testcrate/tests/cmp.rs b/testcrate/tests/cmp.rs
index 7e973e7e..e3161f37 100644
--- a/testcrate/tests/cmp.rs
+++ b/testcrate/tests/cmp.rs
@@ -156,7 +156,6 @@ mod float_comparisons_arm {
     fn cmp_f32() {
         use compiler_builtins::float::cmp::{
             __aeabi_fcmpeq, __aeabi_fcmpge, __aeabi_fcmpgt, __aeabi_fcmple, __aeabi_fcmplt,
-            __eqsf2vfp, __gesf2vfp, __gtsf2vfp, __lesf2vfp, __ltsf2vfp, __nesf2vfp,
         };
 
         fuzz_float_2(N, |x: f32, y: f32| {
@@ -166,12 +165,6 @@ mod float_comparisons_arm {
                 0, x == y, __aeabi_fcmpeq;
                 0, x >= y, __aeabi_fcmpge;
                 0, x > y, __aeabi_fcmpgt;
-                0, x < y, __ltsf2vfp;
-                0, x <= y, __lesf2vfp;
-                0, x == y, __eqsf2vfp;
-                0, x >= y, __gesf2vfp;
-                0, x > y, __gtsf2vfp;
-                1, x != y, __nesf2vfp;
             );
         });
     }
@@ -180,7 +173,6 @@ mod float_comparisons_arm {
     fn cmp_f64() {
         use compiler_builtins::float::cmp::{
             __aeabi_dcmpeq, __aeabi_dcmpge, __aeabi_dcmpgt, __aeabi_dcmple, __aeabi_dcmplt,
-            __eqdf2vfp, __gedf2vfp, __gtdf2vfp, __ledf2vfp, __ltdf2vfp, __nedf2vfp,
         };
 
         fuzz_float_2(N, |x: f64, y: f64| {
@@ -190,12 +182,6 @@ mod float_comparisons_arm {
                 0, x == y, __aeabi_dcmpeq;
                 0, x >= y, __aeabi_dcmpge;
                 0, x > y, __aeabi_dcmpgt;
-                0, x < y, __ltdf2vfp;
-                0, x <= y, __ledf2vfp;
-                0, x == y, __eqdf2vfp;
-                0, x >= y, __gedf2vfp;
-                0, x > y, __gtdf2vfp;
-                1, x != y, __nedf2vfp;
             );
         });
     }
diff --git a/testcrate/tests/conv.rs b/testcrate/tests/conv.rs
index ce1f64e6..24f3a04a 100644
--- a/testcrate/tests/conv.rs
+++ b/testcrate/tests/conv.rs
@@ -258,12 +258,6 @@ mod extend {
         f32 => f64, Single => Double, __extendsfdf2, all();
     }
 
-    #[cfg(target_arch = "arm")]
-    f_to_f! {
-        extend,
-        f32 => f64, Single => Double, __extendsfdf2vfp, all();
-    }
-
     #[cfg(all(f16_enabled, f128_enabled))]
     #[cfg(not(any(target_arch = "powerpc", target_arch = "powerpc64")))]
     f_to_f! {
@@ -293,12 +287,6 @@ mod trunc {
         f64 => f32, Double => Single, __truncdfsf2, all();
     }
 
-    #[cfg(target_arch = "arm")]
-    f_to_f! {
-        trunc,
-        f64 => f32, Double => Single, __truncdfsf2vfp, all();
-    }
-
     #[cfg(all(f16_enabled, f128_enabled))]
     #[cfg(not(any(target_arch = "powerpc", target_arch = "powerpc64")))]
     f_to_f! {
diff --git a/testcrate/tests/div_rem.rs b/testcrate/tests/div_rem.rs
index ff78b4f5..418e9c18 100644
--- a/testcrate/tests/div_rem.rs
+++ b/testcrate/tests/div_rem.rs
@@ -156,13 +156,3 @@ mod float_div {
         f64, __divdf3, Double, all();
     }
 }
-
-#[cfg(target_arch = "arm")]
-mod float_div_arm {
-    use super::*;
-
-    float! {
-        f32, __divsf3vfp, Single, all();
-        f64, __divdf3vfp, Double, all();
-    }
-}
diff --git a/testcrate/tests/mul.rs b/testcrate/tests/mul.rs
index 867622fd..449d1948 100644
--- a/testcrate/tests/mul.rs
+++ b/testcrate/tests/mul.rs
@@ -153,13 +153,3 @@ mod float_mul_f128_ppc {
         f128, __mulkf3, Quad, not(feature = "no-sys-f128");
     }
 }
-
-#[cfg(target_arch = "arm")]
-mod float_mul_arm {
-    use super::*;
-
-    float_mul! {
-        f32, __mulsf3vfp, Single, all();
-        f64, __muldf3vfp, Double, all();
-    }
-}