microblaze: Fix copy_to_user_page macro
copy_to_user_page macro is used in mm/memory.c:access_process_vm
function. This function is called from ptrace code (POKETEXT, POKEDATA)
which write data to memory. Microblaze handle physical address for
caches that's why there is virt_to_phys conversion.
There is potential one location which can caused the problem on WB system.
The important is take a look at write PTRACEs requests
(POKE/TEXT, DATA, USR).
Note:
Majority of Microblaze PTRACE code is moved to generic location
in newer kernel version that's why this solution should work on
the newest kernel version too.
linux/io.h is in cacheflush because of mm/nommu.c
Tested on a WB system - hello world debugging.
Signed-off-by: Michal Simek <monstr@monstr.eu>
diff --git a/arch/microblaze/kernel/ptrace.c b/arch/microblaze/kernel/ptrace.c
index a4a7770..dc03ffc 100644
--- a/arch/microblaze/kernel/ptrace.c
+++ b/arch/microblaze/kernel/ptrace.c
@@ -38,6 +38,8 @@
#include <asm/processor.h>
#include <linux/uaccess.h>
#include <asm/asm-offsets.h>
+#include <asm/cacheflush.h>
+#include <asm/io.h>
/* Returns the address where the register at REG_OFFS in P is stashed away. */
static microblaze_reg_t *reg_save_addr(unsigned reg_offs,
@@ -101,8 +103,21 @@
microblaze_reg_t *reg_addr = reg_save_addr(addr, child);
if (request == PTRACE_PEEKUSR)
val = *reg_addr;
- else
+ else {
+#if 1
*reg_addr = data;
+#else
+ /* MS potential problem on WB system
+ * Be aware that reg_addr is virtual address
+ * virt_to_phys conversion is necessary.
+ * This could be sensible solution.
+ */
+ u32 paddr = virt_to_phys((u32)reg_addr);
+ invalidate_icache_range(paddr, paddr + 4);
+ *reg_addr = data;
+ flush_dcache_range(paddr, paddr + 4);
+#endif
+ }
} else
rval = -EIO;