[Qemu-devel] [PULL 2/3] s390-ccw.img: Fix sporadic errors with ccw boot image - initialize css

[Christian Borntraeger]

We have to set the cssid to 0, otherwise the stsch code will
return an operand exception without the m bit. In the same way
we should set m=0.

This case was triggered in some cases during reboot, if for some
reason the location of blk_schid.cssid contains 1 and m was 0.
Turns out that the qemu elf loader does not zero out the bss section.
On first boot this is ok, but on reboots this might fail.

The symptom was an dump of the old kernel with several areas
overwritten. The bootloader does not register a program check
handler, so bios exception jumped back into the old kernel.

Lets just use a local struct with a designed initializer. That
will guarantee that all other subelements are initialized to 0.

Signed-off-by: Christian Borntraeger <address@hidden>
---
 pc-bios/s390-ccw/main.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/pc-bios/s390-ccw/main.c b/pc-bios/s390-ccw/main.c
index c5d5332..d7fc727 100644
--- a/pc-bios/s390-ccw/main.c
+++ b/pc-bios/s390-ccw/main.c
@@ -10,7 +10,6 @@
 
 #include "s390-ccw.h"
 
-struct subchannel_id blk_schid;
 char stack[PAGE_SIZE * 8] __attribute__((__aligned__(PAGE_SIZE)));
 uint64_t boot_value;
 
@@ -23,13 +22,13 @@ void virtio_panic(const char *string)
 
 static void virtio_setup(uint64_t dev_info)
 {
+    struct subchannel_id blk_schid = { .one = 1};
     struct schib schib;
     int i;
     int r;
     bool found = false;
     bool check_devno = false;
     uint16_t dev_no = -1;
-    blk_schid.one = 1;
 
     if (dev_info != -1) {
         check_devno = true;
-- 
1.8.4.2