Le 13/02/2020 à 06:23, Daniel Axtens a écrit :
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index e05e6dd67ae6..ef455da7efa3 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -285,18 +285,25 @@ void __init early_setup(unsigned long dt_ptr)
/* -------- printk is _NOT_ safe to use here ! ------- */
- /* Try new device tree based feature discovery ... */
- if (!dt_cpu_ftrs_init(__va(dt_ptr)))
- /* Otherwise use the old style CPU table */
- identify_cpu(0, mfspr(SPRN_PVR));
+ /* Assume we're on cpu 0 for now. Don't write to the paca yet!
+ *
+ * We need to load a PACA very early if we are using kcov. kcov will
+ * call in_task() in its instrumentation, which relies on the current
+ * task from the PACA. dt_cpu_ftrs_init is coveraged-enabled and also
+ * calls into the coverage-enabled generic dt library.
+ */
Checkpatch won't be happy I guess. For multiline comments, there should
be no text on the first line starting with /*
Christophe
