[U-Boot] cmd_sf warning on size_t
Hi Mike,
I got this warning on cmd_sf.c,
cmd_sf.c: In function `spi_flash_update_block': cmd_sf.c:130: warning: unsigned int format, size_t arg (arg 4) cmd_sf.c:135: warning: unsigned int format, size_t arg (arg 3)
It was related to our arch/avr32|blackfin|nios2/asm/posix_types.h typedef unsigned long __kernel_size_t;
While most other 32 bits archs use unsigned int, we three use unsigned long.
It is said in linux-2.6/include/asm-generic/posix_types.h /* * Most 32 bit architectures use "unsigned int" size_t, * and all 64 bit architectures use "unsigned long" size_t. */
Shall we fix the cmd_sf.c, or shall I change size_t to unsigned int as other 32 bits archs do?
Best regards, Thomas
Dear Thomas Chou,
On 14.12.2011 07:36, Thomas Chou wrote:
Hi Mike,
I got this warning on cmd_sf.c,
cmd_sf.c: In function `spi_flash_update_block': cmd_sf.c:130: warning: unsigned int format, size_t arg (arg 4) cmd_sf.c:135: warning: unsigned int format, size_t arg (arg 3)
I tumble over this these days too.
It was related to our arch/avr32|blackfin|nios2/asm/posix_types.h typedef unsigned long __kernel_size_t;
While most other 32 bits archs use unsigned int, we three use unsigned long.
It is said in linux-2.6/include/asm-generic/posix_types.h /*
- Most 32 bit architectures use "unsigned int" size_t,
- and all 64 bit architectures use "unsigned long" size_t.
*/
Shall we fix the cmd_sf.c, or shall I change size_t to unsigned int as other 32 bits archs do?
I tend to fix the cmd_sf.c and arch/avr32/cpu/at32ap700x/mmu.c -> mmu.c:25: warning: format '%08x' expects type 'unsigned int', but argument 2 has type 'uintptr_t'
I have already a patch, and would send it.
best regards
Andreas Bießmann
From: Andreas Bießmann biessmann@corscience.de
Some architectures define size_t to unsigned long even though they are 32 bit architectures.
This patch fixes several warnings when running MAKEALL for avr32 architecture:
---8<--- mmu.c: In function 'mmu_init_r': mmu.c:25: warning: format '%08x' expects type 'unsigned int', but argument 2 has type 'uintptr_t' fat.c: In function 'do_fat_read': fat.c:879: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int' cmd_sf.c: In function 'spi_flash_update_block': cmd_sf.c:130: warning: format '%#x' expects type 'unsigend int', but argument 4 has type 'size_t' cmd_sf.c:135: warning: format '%x' expects type 'unsigned int', but argument 3 has type 'size_t' --->8---
Signed-off-by: Andreas Bießmann biessmann@corscience.de cc: Mike Frysinger vapier@gentoo.org cc: Thomas Chou thomas@wytron.com.tw --- arch/avr32/cpu/at32ap700x/mmu.c | 2 +- common/cmd_sf.c | 4 ++-- fs/fat/fat.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/avr32/cpu/at32ap700x/mmu.c b/arch/avr32/cpu/at32ap700x/mmu.c index c3a1b93..fda06a9 100644 --- a/arch/avr32/cpu/at32ap700x/mmu.c +++ b/arch/avr32/cpu/at32ap700x/mmu.c @@ -22,7 +22,7 @@ void mmu_init_r(unsigned long dest_addr) */ vmr_table_addr = (uintptr_t)&mmu_vmr_table; sysreg_write(PTBR, vmr_table_addr); - printf("VMR table @ 0x%08x\n", vmr_table_addr); + printf("VMR table @ %#lx\n", vmr_table_addr);
/* Enable paging */ sysreg_write(MMUCR, SYSREG_BF(DRP, 1) | SYSREG_BF(DLA, 1) diff --git a/common/cmd_sf.c b/common/cmd_sf.c index 7225656..612fd18 100644 --- a/common/cmd_sf.c +++ b/common/cmd_sf.c @@ -127,12 +127,12 @@ static int do_spi_flash_probe(int argc, char * const argv[]) static const char *spi_flash_update_block(struct spi_flash *flash, u32 offset, size_t len, const char *buf, char *cmp_buf, size_t *skipped) { - debug("offset=%#x, sector_size=%#x, len=%#x\n", + debug("offset=%#x, sector_size=%#x, len=%#zx\n", offset, flash->sector_size, len); if (spi_flash_read(flash, offset, len, cmp_buf)) return "read"; if (memcmp(cmp_buf, buf, len) == 0) { - debug("Skip region %x size %x: no change\n", + debug("Skip region %x size %zx: no change\n", offset, len); *skipped += len; return NULL; diff --git a/fs/fat/fat.c b/fs/fat/fat.c index 9a29458..dbb8db9 100644 --- a/fs/fat/fat.c +++ b/fs/fat/fat.c @@ -876,7 +876,7 @@ do_fat_read (const char *filename, void *buffer, unsigned long maxsize, while (1) { int i;
- debug("FAT read sect=%d, clust_size=%d, DIRENTSPERBLOCK=%d\n", + debug("FAT read sect=%d, clust_size=%d, DIRENTSPERBLOCK=%zd\n", cursect, mydata->clust_size, DIRENTSPERBLOCK);
if (disk_read(cursect,
On 12/14/2011 06:04 PM, Andreas Bießmann wrote:
From: Andreas Bießmannbiessmann@corscience.de
Some architectures define size_t to unsigned long even though they are 32 bit architectures.
This patch fixes several warnings when running MAKEALL for avr32 architecture:
Hi Andreas,
Thanks. I have tested cmd_sf and fat.
Tested-by: Thomas Chou thomas@wytron.com.tw
I would suggest splitting this patch into three, as they related to different custodians.
Best regards, Thomas
Hi Thomas,
On Dec 15, 2011 1:22 PM, "Thomas Chou" thomas@wytron.com.tw wrote:
On 12/14/2011 06:04 PM, Andreas Bießmann wrote:
From: Andreas Bießmannbiessmann@corscience.de
Some architectures define size_t to unsigned long even though they are
32 bit
architectures.
This patch fixes several warnings when running MAKEALL for avr32
architecture:
Hi Andreas,
Thanks. I have tested cmd_sf and fat.
Tested-by: Thomas Chou thomas@wytron.com.tw
I would suggest splitting this patch into three, as they related to
different custodians.
No, that is not true - the patch is a single logical unit
The three maintainers can each provide an Ack
Regards,
Graeme
On Wednesday 14 December 2011 21:39:48 Graeme Russ wrote:
On Dec 15, 2011 1:22 PM, "Thomas Chou" wrote:
I would suggest splitting this patch into three, as they related to different custodians.
No, that is not true - the patch is a single logical unit
meh. no need to keep them as one. -mike
On Wednesday 14 December 2011 05:04:07 Andreas Bießmann wrote:
--- a/arch/avr32/cpu/at32ap700x/mmu.c +++ b/arch/avr32/cpu/at32ap700x/mmu.c
- printf("VMR table @ 0x%08x\n", vmr_table_addr);
- printf("VMR table @ %#lx\n", vmr_table_addr);
this isn't the same. probably should be %08lx.
--- a/common/cmd_sf.c +++ b/common/cmd_sf.c
- debug("offset=%#x, sector_size=%#x, len=%#x\n",
- debug("offset=%#x, sector_size=%#x, len=%#zx\n",
debug("Skip region %x size %x: no change\n",
debug("Skip region %x size %zx: no change\n",--- a/fs/fat/fat.c +++ b/fs/fat/fat.c
debug("FAT read sect=%d, clust_size=%d, DIRENTSPERBLOCK=%d\n",
debug("FAT read sect=%d, clust_size=%d, DIRENTSPERBLOCK=%zd\n",
these look good. -mike
Dear Mike Frysinger,
On 15.12.2011 06:35, Mike Frysinger wrote:
On Wednesday 14 December 2011 05:04:07 Andreas Bießmann wrote:
--- a/arch/avr32/cpu/at32ap700x/mmu.c +++ b/arch/avr32/cpu/at32ap700x/mmu.c
- printf("VMR table @ 0x%08x\n", vmr_table_addr);
- printf("VMR table @ %#lx\n", vmr_table_addr);
this isn't the same. probably should be %08lx.
You are right, but as you may know the memory map for at32ap7000 has its SDRAM physically at 0x10000000. Therefore the result will be the same for 0x%08lx and %#lx (in all currently available boards). But maybe one will locate the vmr_table sometimes in SRAM some day?
I will provide v2 which change this back, split off in three patches and add the respective custodian in cc.
best regards
Andreas Bießmann
On Thursday 15 December 2011 03:30:37 Andreas Bießmann wrote:
On 15.12.2011 06:35, Mike Frysinger wrote:
On Wednesday 14 December 2011 05:04:07 Andreas Bießmann wrote:
--- a/arch/avr32/cpu/at32ap700x/mmu.c +++ b/arch/avr32/cpu/at32ap700x/mmu.c
- printf("VMR table @ 0x%08x\n", vmr_table_addr);
- printf("VMR table @ %#lx\n", vmr_table_addr);
this isn't the same. probably should be %08lx.
You are right, but as you may know the memory map for at32ap7000 has its SDRAM physically at 0x10000000. Therefore the result will be the same for 0x%08lx and %#lx (in all currently available boards).
i don't know anything about AVR32 ;)
But maybe one will locate the vmr_table sometimes in SRAM some day?
seems safer, although probably unlikely as i think the AVR32 is pretty much dead nowadays ? -mike
From: Andreas Bießmann biessmann@corscience.de
Andreas Bießmann (3): cmd_sf.c: fix printf() length modifier fat.c: fix printf() length modifier avr32:mmu.c: fix printf() length modifier
arch/avr32/cpu/at32ap700x/mmu.c | 2 +- common/cmd_sf.c | 4 ++-- fs/fat/fat.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-)
From: Andreas Bießmann biessmann@corscience.de
size_t is not always 'unsigned int', use corret length modifer.
This patch fixes following warning:
---8<--- cmd_sf.c: In function 'spi_flash_update_block': cmd_sf.c:130: warning: format '%#x' expects type 'unsigend int', but argument 4 has type 'size_t' cmd_sf.c:135: warning: format '%x' expects type 'unsigned int', but argument 3 has type 'size_t' --->8---
Signed-off-by: Andreas Bießmann biessmann@corscience.de cc: Mike Frysinger vapier@gentoo.org cc: Thomas Chou thomas@wytron.com.tw --- changes since v1: split off into single patches
total: 0 errors, 0 warnings, 14 lines checked
NOTE: Ignored message types: COMPLEX_MACRO CONSIDER_KSTRTO MINMAX MULTISTATEMENT_MACRO_USE_DO_WHILE
0001-cmd_sf.c-fix-printf-length-modifier.patch has no obvious style problems and is ready for submission.
common/cmd_sf.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/common/cmd_sf.c b/common/cmd_sf.c index 7225656..612fd18 100644 --- a/common/cmd_sf.c +++ b/common/cmd_sf.c @@ -127,12 +127,12 @@ static int do_spi_flash_probe(int argc, char * const argv[]) static const char *spi_flash_update_block(struct spi_flash *flash, u32 offset, size_t len, const char *buf, char *cmp_buf, size_t *skipped) { - debug("offset=%#x, sector_size=%#x, len=%#x\n", + debug("offset=%#x, sector_size=%#x, len=%#zx\n", offset, flash->sector_size, len); if (spi_flash_read(flash, offset, len, cmp_buf)) return "read"; if (memcmp(cmp_buf, buf, len) == 0) { - debug("Skip region %x size %x: no change\n", + debug("Skip region %x size %zx: no change\n", offset, len); *skipped += len; return NULL;
Acked-by: Mike Frysinger vapier@gentoo.org
if it doesn't get picked up by someone else, i'll push via my sf branch -mike
Dear =?UTF-8?q?Andreas=20Bie=C3=9Fmann?=,
In message 1323939415-21743-2-git-send-email-andreas.devel@googlemail.com you wrote:
From: Andreas Bießmann biessmann@corscience.de
size_t is not always 'unsigned int', use corret length modifer.
This patch fixes following warning:
---8<--- cmd_sf.c: In function 'spi_flash_update_block': cmd_sf.c:130: warning: format '%#x' expects type 'unsigend int', but argument 4 has type 'size_t' cmd_sf.c:135: warning: format '%x' expects type 'unsigned int', but argument 3 has type 'size_t' --->8---
Signed-off-by: Andreas Bießmann biessmann@corscience.de cc: Mike Frysinger vapier@gentoo.org cc: Thomas Chou thomas@wytron.com.tw
changes since v1: split off into single patches
total: 0 errors, 0 warnings, 14 lines checked
NOTE: Ignored message types: COMPLEX_MACRO CONSIDER_KSTRTO MINMAX MULTISTATEMENT_MACRO_USE_DO_WHILE
0001-cmd_sf.c-fix-printf-length-modifier.patch has no obvious style problems and is ready for submission.
common/cmd_sf.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)
Applied to "next" branch, thanks.
Best regards,
Wolfgang Denk
From: Andreas Bießmann biessmann@corscience.de
The DIRENTSPERBLOCK utilizes sizeof() which will return a size_t which has no fixed size. Therefor use correct length modifer for printf() statement to prevent compiler warnings.
This patch fixes following warning:
---8<--- fat.c: In function 'do_fat_read': fat.c:879: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int' --->8---
Signed-off-by: Andreas Bießmann biessmann@corscience.de cc: Mike Frysinger vapier@gentoo.org cc: Thomas Chou thomas@wytron.com.tw cc: rjones@nexus-tech.net cc: kharris@nexus-tech.net --- changes since v1: split off into single patches
total: 0 errors, 0 warnings, 8 lines checked
NOTE: Ignored message types: COMPLEX_MACRO CONSIDER_KSTRTO MINMAX MULTISTATEMENT_MACRO_USE_DO_WHILE
0002-fat.c-fix-printf-length-modifier.patch has no obvious style problems and is ready for submission.
fs/fat/fat.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/fs/fat/fat.c b/fs/fat/fat.c index 9a29458..dbb8db9 100644 --- a/fs/fat/fat.c +++ b/fs/fat/fat.c @@ -876,7 +876,7 @@ do_fat_read (const char *filename, void *buffer, unsigned long maxsize, while (1) { int i;
- debug("FAT read sect=%d, clust_size=%d, DIRENTSPERBLOCK=%d\n", + debug("FAT read sect=%d, clust_size=%d, DIRENTSPERBLOCK=%zd\n", cursect, mydata->clust_size, DIRENTSPERBLOCK);
if (disk_read(cursect,
Acked-by: Mike Frysinger vapier@gentoo.org -mike
Dear =?UTF-8?q?Andreas=20Bie=C3=9Fmann?=,
In message 1323939415-21743-3-git-send-email-andreas.devel@googlemail.com you wrote:
From: Andreas Bießmann biessmann@corscience.de
The DIRENTSPERBLOCK utilizes sizeof() which will return a size_t which has no fixed size. Therefor use correct length modifer for printf() statement to prevent compiler warnings.
This patch fixes following warning:
---8<--- fat.c: In function 'do_fat_read': fat.c:879: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int' --->8---
Signed-off-by: Andreas Bießmann biessmann@corscience.de cc: Mike Frysinger vapier@gentoo.org cc: Thomas Chou thomas@wytron.com.tw cc: rjones@nexus-tech.net cc: kharris@nexus-tech.net
changes since v1: split off into single patches
total: 0 errors, 0 warnings, 8 lines checked
NOTE: Ignored message types: COMPLEX_MACRO CONSIDER_KSTRTO MINMAX MULTISTATEMENT_MACRO_USE_DO_WHILE
0002-fat.c-fix-printf-length-modifier.patch has no obvious style problems and is ready for submission.
fs/fat/fat.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
Applied to "next" branch, thanks.
Best regards,
Wolfgang Denk
From: Andreas Bießmann biessmann@corscience.de
avr32 uses unsigned long addresses, fix the printf() length modifier for that fact.
Before this patch following warning occours:
---8<--- mmu.c: In function 'mmu_init_r': mmu.c:25: warning: format '%08x' expects type 'unsigned int', but argument 2 has type 'uintptr_t' --->8---
Signed-off-by: Andreas Bießmann biessmann@corscience.de cc: Mike Frysinger vapier@gentoo.org cc: Thomas Chou thomas@wytron.com.tw cc: Reinhard Meyer u-boot@emk-elektronik.de --- changes since v1: split off into single patches
total: 0 errors, 0 warnings, 8 lines checked
NOTE: Ignored message types: COMPLEX_MACRO CONSIDER_KSTRTO MINMAX MULTISTATEMENT_MACRO_USE_DO_WHILE
0003-avr32-mmu.c-fix-printf-length-modifier.patch has no obvious style problems and is ready for submission.
arch/avr32/cpu/at32ap700x/mmu.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/arch/avr32/cpu/at32ap700x/mmu.c b/arch/avr32/cpu/at32ap700x/mmu.c index c3a1b93..0e28b21 100644 --- a/arch/avr32/cpu/at32ap700x/mmu.c +++ b/arch/avr32/cpu/at32ap700x/mmu.c @@ -22,7 +22,7 @@ void mmu_init_r(unsigned long dest_addr) */ vmr_table_addr = (uintptr_t)&mmu_vmr_table; sysreg_write(PTBR, vmr_table_addr); - printf("VMR table @ 0x%08x\n", vmr_table_addr); + printf("VMR table @ 0x%08lx\n", vmr_table_addr);
/* Enable paging */ sysreg_write(MMUCR, SYSREG_BF(DRP, 1) | SYSREG_BF(DLA, 1)
Acked-by: Mike Frysinger vapier@gentoo.org -mike
Dear =?UTF-8?q?Andreas=20Bie=C3=9Fmann?=,
In message 1323939415-21743-4-git-send-email-andreas.devel@googlemail.com you wrote:
From: Andreas Bießmann biessmann@corscience.de
avr32 uses unsigned long addresses, fix the printf() length modifier for that fact.
Before this patch following warning occours:
---8<--- mmu.c: In function 'mmu_init_r': mmu.c:25: warning: format '%08x' expects type 'unsigned int', but argument 2 has type 'uintptr_t' --->8---
Signed-off-by: Andreas Bießmann biessmann@corscience.de cc: Mike Frysinger vapier@gentoo.org cc: Thomas Chou thomas@wytron.com.tw cc: Reinhard Meyer u-boot@emk-elektronik.de
changes since v1: split off into single patches
total: 0 errors, 0 warnings, 8 lines checked
NOTE: Ignored message types: COMPLEX_MACRO CONSIDER_KSTRTO MINMAX MULTISTATEMENT_MACRO_USE_DO_WHILE
0003-avr32-mmu.c-fix-printf-length-modifier.patch has no obvious style problems and is ready for submission.
arch/avr32/cpu/at32ap700x/mmu.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
Applied to "next" branch, thanks.
Best regards,
Wolfgang Denk
participants (5)
-
Andreas Bießmann -
Graeme Russ -
Mike Frysinger -
Thomas Chou -
Wolfgang Denk