s390x: fix flat file load on 32 bit systems
PATCH v2 1/1 s390x: fix flat file load on 32 bit systems
pc-bios/s390-zipl.rom is a flat image so it's expected that loading it as elf will fail.
It should fall back on loading a flat file, but doesn't on 32 bit systems, instead it fails printing:
qemu: hardware error: could not load bootloader 's390-zipl.rom'
The result is boot failure.
The reason is that a 64 bit unsigned interger which is set to -1 on error is compared to -1UL which on a 32 bit system with gcc is a 32 bit unsigned interger.
Since both are unsigned, no sign extension takes place and comparison evaluates to non-equal.
There's no reason to do clever tricks: all functions we call actually return int so just use int. In fact ram_addr_t dos not make any sense - it's meaning is "memory handle for migration".
And then we can use == -1 everywhere, consistently.
---
Changes from v1:
better fix: use int everywhere
fix all places with same bug (e.g. -kernel was broken too)
hw/s390x/ipl.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)
hw/s390x/ipl.c +10 -8
@@ -62,10 +62,9 @@ typedef struct S390IPLState {static int s390_ipl_init (SysBusDevice *dev) {S390IPLState *ipl =S390_IPL (dev); -ram_addr_t kernel_size =0 ;if (!ipl->kernel) { -ram_addr_t bios_size =0 ; +int bios_size;char *bios_filename;/* Load zipl bootloader */
@@ -80,7 +79,7 @@ static int s390_ipl_init(SysBusDevice *dev) bios_size =load_elf (bios_filename,NULL ,NULL , &ipl->start_addr,NULL ,NULL ,1 ,ELF_MACHINE ,0 ); -if (bios_size ==-1UL ) { +if (bios_size ==-1 ) { bios_size =load_image_targphys (bios_filename,ZIPL_IMAGE_START ,4096 ); ipl->start_addr =ZIPL_IMAGE_START ;
@@ -90,17 +89,19 @@ static int s390_ipl_init(SysBusDevice *dev) }g_free (bios_filename); -if ((long )bios_size <0 ) { +if (bios_size ==-1 ) {hw_error ("could not load bootloader '%s'\n" , bios_name); }return 0 ; }else { +int kernel_size; + kernel_size =load_elf (ipl->kernel,NULL ,NULL ,NULL ,NULL ,NULL ,1 ,ELF_MACHINE ,0 ); -if (kernel_size ==-1UL ) { +if (kernel_size ==-1 ) { kernel_size =load_image_targphys (ipl->kernel,0 , ram_size); } -if (kernel_size ==-1UL ) { +if (kernel_size ==-1 ) {fprintf (stderr,"could not load kernel '%s'\n" , ipl->kernel);return -1 ; }
@@ -115,7 +116,8 @@ static int s390_ipl_init(SysBusDevice *dev) ipl->start_addr =KERN_IMAGE_START ; }if (ipl->initrd) { -ram_addr_t initrd_offset, initrd_size; +hwaddr initrd_offset; +int initrd_size; initrd_offset =INITRD_START ;while (kernel_size +0x100000 > initrd_offset) {
@@ -123,7 +125,7 @@ static int s390_ipl_init(SysBusDevice *dev) } initrd_size =load_image_targphys (ipl->initrd, initrd_offset, ram_size - initrd_offset); -if (initrd_size ==-1UL ) { +if (initrd_size ==-1 ) {fprintf (stderr,"qemu: could not load initrd '%s'\n" , ipl->initrd);exit (1 ); }
Signature · 1 line
MST
Quoted 14 lines · Michael S. Tsirkin
pc-bios/s390-zipl.rom is a flat image so it's expected that loading it as elf will fail.
It should fall back on loading a flat file, but doesn't on 32 bit systems, instead it fails printing:
qemu: hardware error: could not load bootloader 's390-zipl.rom'The result is boot failure.
The reason is that a 64 bit unsigned interger which is set to -1 on error is compared to -1UL which on a 32 bit system with gcc is a 32 bit unsigned interger.
Since both are unsigned, no sign extension takes place and comparison evaluates to non-equal.
Quoted 8 lines
There's no reason to do clever tricks: all functions we call actually return int so just use int. In fact
ram_addr_tdos not make any sense - it's meaning is "memory handle for migration".And then we can use == -1 everywhere, consistently.
Signed-off-by Michael S. Tsirkin <[email protected]>
Alex
Quoted 22 lines · On Thu, 21 Nov 2013 14:52:02 +0200 "Michael S. Tsirkin" <[email protected]> wrote:
pc-bios/s390-zipl.rom is a flat image so it's expected that loading it as elf will fail.
It should fall back on loading a flat file, but doesn't on 32 bit systems, instead it fails printing:
qemu: hardware error: could not load bootloader 's390-zipl.rom'The result is boot failure.
The reason is that a 64 bit unsigned interger which is set to -1 on error is compared to -1UL which on a 32 bit system with gcc is a 32 bit unsigned interger.
Since both are unsigned, no sign extension takes place and comparison evaluates to non-equal.There's no reason to do clever tricks: all functions we call actually return int so just use int. In fact
ram_addr_tdos not make any sense - it's meaning is "memory handle for migration".And then we can use == -1 everywhere, consistently.
Signed-off-by Michael S. Tsirkin <[email protected]>
Quoted 9 lines
---
Changes from v1:
better fix: use int everywhere
fix all places with same bug (e.g. -kernel was broken too)hw/s390x/ipl.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)