When linking the pr110812 units with lto after compiling them with a
riscv(32?) compiler that defaults to the medlow code model, the
compiler uses HI20 relocations that overflow with our linker scripts.
I suppose the test expects PC-relative relocations for symbol
resolution, as in the medany code model, so select that explicitly.
Regstrapped on x86_64-linux-gnu along with other patches in this batch;
also tested with targets riscv64-elf and riscv32-elf on the same host,
both trunk and gcc-15. Ok to install?
for gcc/testsuite/ChangeLog
* gcc.target/riscv/lto/pr110812_0.c: Set cmodel explicitly.
* gcc.target/riscv/lto/pr110812_1.c: Likewise.
---
gcc/testsuite/gcc.target/riscv/lto/pr110812_0.c | 2 +-
gcc/testsuite/gcc.target/riscv/lto/pr110812_1.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/gcc/testsuite/gcc.target/riscv/lto/pr110812_0.c
b/gcc/testsuite/gcc.target/riscv/lto/pr110812_0.c
index a2f70928b1621..ab745e5ffec38 100644
--- a/gcc/testsuite/gcc.target/riscv/lto/pr110812_0.c
+++ b/gcc/testsuite/gcc.target/riscv/lto/pr110812_0.c
@@ -1,5 +1,5 @@
/* { dg-lto-do link } */
-/* { dg-lto-options { { -flto -march=rv64gc -mabi=lp64d } } } */
+/* { dg-lto-options { { -flto -march=rv64gc -mabi=lp64d -mcmodel=medany } } }
*/
/* { dg-extra-ld-options "-nostdlib" } */
void foo();
diff --git a/gcc/testsuite/gcc.target/riscv/lto/pr110812_1.c
b/gcc/testsuite/gcc.target/riscv/lto/pr110812_1.c
index 34d7cbbbcbbb1..a84f92c6627d6 100644
--- a/gcc/testsuite/gcc.target/riscv/lto/pr110812_1.c
+++ b/gcc/testsuite/gcc.target/riscv/lto/pr110812_1.c
@@ -1,4 +1,4 @@
-/* { dg-options { -flto -march=rv64gcv -mabi=lp64d } } */
+/* { dg-options { -flto -march=rv64gcv -mabi=lp64d -mcmodel=medany } } */
#include "riscv_vector.h"
uint8_t *x, *y;
--
Alexandre Oliva, happy hacker https://blog.lx.oliva.nom.br/
Free Software Activist FSFLA co-founder GNU Toolchain Engineer
More tolerance and less prejudice are key for inclusion and diversity.
Excluding neuro-others for not behaving ""normal"" is *not* inclusive!
