Merge tag 'powerpc-5.14-6' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux

Pull powerpc fixes from Michael Ellerman:

 - Fix random crashes on some 32-bit CPUs by adding isync() after
   locking/unlocking KUEP

 - Fix intermittent crashes when loading modules with strict module RWX

 - Fix a section mismatch introduce by a previous fix.

Thanks to Christophe Leroy, Fabiano Rosas, Laurent Vivier, Murilo
Opsfelder Araújo, Nathan Chancellor, and Stan Johnson.

h# -----BEGIN PGP SIGNATURE-----

* tag 'powerpc-5.14-6' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc/mm: Fix set_memory_*() against concurrent accesses
  powerpc/32s: Fix random crashes by adding isync() after locking/unlocking KUEP
  powerpc/xive: Do not mark xive_request_ipi() as __init

3 files changed, 31 insertions, 14 deletions

diff --git a/arch/powerpc/include/asm/book3s/32/kup.h b/arch/powerpc/include/asm/book3s/32/kup.h
index 64201125a287..d4b145b279f6 100644
--- a/arch/powerpc/include/asm/book3s/32/kup.h
+++ b/arch/powerpc/include/asm/book3s/32/kup.h
@@ -4,6 +4,8 @@
 
 #include <asm/bug.h>
 #include <asm/book3s/32/mmu-hash.h>
+#include <asm/mmu.h>
+#include <asm/synch.h>
 
 #ifndef __ASSEMBLY__
 
@@ -28,6 +30,15 @@ static inline void kuep_lock(void)
 		return;
 
 	update_user_segments(mfsr(0) | SR_NX);
+	/*
+	 * This isync() shouldn't be necessary as the kernel is not excepted to
+	 * run any instruction in userspace soon after the update of segments,
+	 * but hash based cores (at least G3) seem to exhibit a random
+	 * behaviour when the 'isync' is not there. 603 cores don't have this
+	 * behaviour so don't do the 'isync' as it saves several CPU cycles.
+	 */
+	if (mmu_has_feature(MMU_FTR_HPTE_TABLE))
+		isync();	/* Context sync required after mtsr() */
 }
 
 static inline void kuep_unlock(void)
@@ -36,6 +47,15 @@ static inline void kuep_unlock(void)
 		return;
 
 	update_user_segments(mfsr(0) & ~SR_NX);
+	/*
+	 * This isync() shouldn't be necessary as a 'rfi' will soon be executed
+	 * to return to userspace, but hash based cores (at least G3) seem to
+	 * exhibit a random behaviour when the 'isync' is not there. 603 cores
+	 * don't have this behaviour so don't do the 'isync' as it saves several
+	 * CPU cycles.
+	 */
+	if (mmu_has_feature(MMU_FTR_HPTE_TABLE))
+		isync();	/* Context sync required after mtsr() */
 }
 
 #ifdef CONFIG_PPC_KUAP
diff --git a/arch/powerpc/mm/pageattr.c b/arch/powerpc/mm/pageattr.c
index 0876216ceee6..edea388e9d3f 100644
--- a/arch/powerpc/mm/pageattr.c
+++ b/arch/powerpc/mm/pageattr.c
@@ -18,16 +18,12 @@
 /*
  * Updates the attributes of a page in three steps:
  *
- * 1. invalidate the page table entry
- * 2. flush the TLB
- * 3. install the new entry with the updated attributes
- *
- * Invalidating the pte means there are situations where this will not work
- * when in theory it should.
- * For example:
- * - removing write from page whilst it is being executed
- * - setting a page read-only whilst it is being read by another CPU
+ * 1. take the page_table_lock
+ * 2. install the new entry with the updated attributes
+ * 3. flush the TLB
  *
+ * This sequence is safe against concurrent updates, and also allows updating the
+ * attributes of a page currently being executed or accessed.
  */
 static int change_page_attr(pte_t *ptep, unsigned long addr, void *data)
 {
@@ -36,9 +32,7 @@ static int change_page_attr(pte_t *ptep, unsigned long addr, void *data)
 
 	spin_lock(&init_mm.page_table_lock);
 
-	/* invalidate the PTE so it's safe to modify */
-	pte = ptep_get_and_clear(&init_mm, addr, ptep);
-	flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
+	pte = ptep_get(ptep);
 
 	/* modify the PTE bits as desired, then apply */
 	switch (action) {
@@ -59,11 +53,14 @@ static int change_page_attr(pte_t *ptep, unsigned long addr, void *data)
 		break;
 	}
 
-	set_pte_at(&init_mm, addr, ptep, pte);
+	pte_update(&init_mm, addr, ptep, ~0UL, pte_val(pte), 0);
 
 	/* See ptesync comment in radix__set_pte_at() */
 	if (radix_enabled())
 		asm volatile("ptesync": : :"memory");
+
+	flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
+
 	spin_unlock(&init_mm.page_table_lock);
 
 	return 0;
diff --git a/arch/powerpc/sysdev/xive/common.c b/arch/powerpc/sysdev/xive/common.c
index 943fd30095af..8183ca343675 100644
--- a/arch/powerpc/sysdev/xive/common.c
+++ b/arch/powerpc/sysdev/xive/common.c
@@ -1170,7 +1170,7 @@ out:
 	return ret;
 }
 
-static int __init xive_request_ipi(unsigned int cpu)
+static int xive_request_ipi(unsigned int cpu)
 {
 	struct xive_ipi_desc *xid = &xive_ipis[early_cpu_to_node(cpu)];
 	int ret;