[U-Boot-Users] HELP, trying to remove complier warnings

[U-Boot-Users] HELP, trying to remove complier warnings

[Kumar Gala]

I'm running with gcc-4.3 from fedora 9.  I get the following warnings:

dlmalloc.c: In function ?malloc_extend_top?:
dlmalloc.c:1971: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:1999: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2029: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2031: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2042: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c: In function ?malloc?:
dlmalloc.c:2245: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2245: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2253: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2258: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2263: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2318: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2327: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2329: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2344: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2356: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2360: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2362: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2363: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c: In function ?free?:
dlmalloc.c:2431: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2444: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2483: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2483: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c: In function ?realloc?:
dlmalloc.c:2594: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2599: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2604: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2605: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2637: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2646: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2647: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c: In function ?calloc?:
dlmalloc.c:2896: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2897: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c: In function ?malloc_trim?:
dlmalloc.c:2987: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:2997: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:3008: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:3012: error: dereferencing type-punned pointer will break  
strict-aliasing rules
dlmalloc.c:3021: error: dereferencing type-punned pointer will break  
strict-aliasing rules

They all result from the bin_at macro in the file.  I can't seem to do  
anything to the code to get rid of them.

I see these both on x86 or ppc host builds:

x86:

gcc -g  -Os -m32    -D__KERNEL__ -DTEXT_BASE=0x387c0000 -I/local/home/ 
galak/u-boot/include -fno-builtin -ffreestanding -nostdinc -isystem / 
usr/lib/gcc/x86_64-redhat-linux/4.3.0/include -pipe  -DCONFIG_I386 - 
D__I386__ -march=i386 -Werror -Wall -Wstrict-prototypes -fno-stack- 
protector -c -o dlmalloc.o dlmalloc.c

ppc:

gcc -g  -Os   -fPIC -ffixed-r14 -meabi -D__KERNEL__ - 
DTEXT_BASE=0xeff80000 -I/local/home/galak/git/u-boot-85xx/include -fno- 
builtin -ffreestanding -nostdinc -isystem /usr/lib/gcc/ppc64-redhat- 
linux/4.3.0/include -pipe  -DCONFIG_PPC -D__powerpc__ -DCONFIG_MPC85xx  
-DCONFIG_E500 -ffixed-r2 -Wa,-me500 -msoft-float -mno-string -mno-spe - 
DCONFIG_E500=1 -DCONFIG_MPC85xx=1 -DCONFIG_PPC_P4080=1 -Wall -Wstrict- 
prototypes -fno-stack-protector  -c -o dlmalloc.o dlmalloc.c

- k	

[U-Boot-Users] HELP, trying to remove complier warnings

[Howard, Marc]

Kumar,

Try adding -fno-strict-aliasing

Marc Howard 

> -----Original Message-----
> From: u-boot-users-bounces at lists.sourceforge.net 
> [mailto:u-boot-users-bounces at lists.sourceforge.net] On Behalf 
> Of Kumar Gala
> Sent: Thursday, July 10, 2008 9:59 AM
> To: u-boot-users at lists.sourceforge.net Users
> Subject: [U-Boot-Users] HELP, trying to remove complier warnings
> 
> I'm running with gcc-4.3 from fedora 9.  I get the following warnings:
> 
> dlmalloc.c: In function 'malloc_extend_top':
> dlmalloc.c:1971: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:1999: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2029: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2031: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2042: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c: In function 'malloc':
> dlmalloc.c:2245: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2245: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2253: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2258: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2263: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2318: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2327: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2329: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2344: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2356: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2360: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2362: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2363: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c: In function 'free':
> dlmalloc.c:2431: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2444: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2483: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2483: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c: In function 'realloc':
> dlmalloc.c:2594: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2599: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2604: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2605: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2637: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2646: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2647: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c: In function 'calloc':
> dlmalloc.c:2896: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2897: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c: In function 'malloc_trim':
> dlmalloc.c:2987: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:2997: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:3008: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:3012: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> dlmalloc.c:3021: error: dereferencing type-punned pointer will break  
> strict-aliasing rules
> 
> They all result from the bin_at macro in the file.  I can't 
> seem to do  
> anything to the code to get rid of them.
> 
> I see these both on x86 or ppc host builds:
> 
> x86:
> 
> gcc -g  -Os -m32    -D__KERNEL__ -DTEXT_BASE=0x387c0000 
> -I/local/home/ 
> galak/u-boot/include -fno-builtin -ffreestanding -nostdinc -isystem / 
> usr/lib/gcc/x86_64-redhat-linux/4.3.0/include -pipe  -DCONFIG_I386 - 
> D__I386__ -march=i386 -Werror -Wall -Wstrict-prototypes -fno-stack- 
> protector -c -o dlmalloc.o dlmalloc.c
> 
> ppc:
> 
> gcc -g  -Os   -fPIC -ffixed-r14 -meabi -D__KERNEL__ - 
> DTEXT_BASE=0xeff80000 
> -I/local/home/galak/git/u-boot-85xx/include -fno- 
> builtin -ffreestanding -nostdinc -isystem /usr/lib/gcc/ppc64-redhat- 
> linux/4.3.0/include -pipe  -DCONFIG_PPC -D__powerpc__ 
> -DCONFIG_MPC85xx  
> -DCONFIG_E500 -ffixed-r2 -Wa,-me500 -msoft-float -mno-string 
> -mno-spe - 
> DCONFIG_E500=1 -DCONFIG_MPC85xx=1 -DCONFIG_PPC_P4080=1 -Wall 
> -Wstrict- 
> prototypes -fno-stack-protector  -c -o dlmalloc.o dlmalloc.c
> 
> - k	
> 
> --------------------------------------------------------------
> -----------
> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
> Studies have shown that voting for your favorite open source project,
> along with a healthy diet, reduces your potential for chronic lameness
> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
> _______________________________________________
> U-Boot-Users mailing list
> U-Boot-Users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/u-boot-users
> 

[U-Boot-Users] HELP, trying to remove complier warnings

[Wolfgang Denk]

In message <921B6E79C3E18642BFFC1C8633C430DB01023B9A@CA1EXCLV07.adcorp.kla-tencor.com> you wrote:
> 
> Try adding -fno-strict-aliasing

No, we don't want to hush up compiler warnings, we want to fix the
problems instead.

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
The first thing we do is kill all the lawyers.
(Shakespeare. II Henry VI, Act IV, scene ii)

[U-Boot-Users] HELP, trying to remove complier warnings

[Scott Wood]

Wolfgang Denk wrote:
> In message <921B6E79C3E18642BFFC1C8633C430DB01023B9A@CA1EXCLV07.adcorp.kla-tencor.com> you wrote:
>> Try adding -fno-strict-aliasing
> 
> No, we don't want to hush up compiler warnings, we want to fix the
> problems instead.

It's not silencing a warning (if it were, it'd be a -W flag); it's 
disabling an incompatible optimization.

-Scott

[U-Boot-Users] HELP, trying to remove complier warnings

[Wolfgang Denk]

In message <48765849.5050306@freescale.com> you wrote:
> Wolfgang Denk wrote:
> > In message <921B6E79C3E18642BFFC1C8633C430DB01023B9A@CA1EXCLV07.adcorp.kla-tencor.com> you wrote:
> >> Try adding -fno-strict-aliasing
> > 
> > No, we don't want to hush up compiler warnings, we want to fix the
> > problems instead.
> 
> It's not silencing a warning (if it were, it'd be a -W flag); it's 
> disabling an incompatible optimization.

OK.

Then let me rephrase: We do not want to disable optimizations, ...

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
The more sins you confess, the more books you will sell.

[U-Boot-Users] HELP, trying to remove complier warnings

[Scott Wood]

Wolfgang Denk wrote:
> In message <48765849.5050306@freescale.com> you wrote:
>> Wolfgang Denk wrote:
>>> In message <921B6E79C3E18642BFFC1C8633C430DB01023B9A@CA1EXCLV07.adcorp.kla-tencor.com> you wrote:
>>>> Try adding -fno-strict-aliasing
>>> No, we don't want to hush up compiler warnings, we want to fix the
>>> problems instead.
>> It's not silencing a warning (if it were, it'd be a -W flag); it's 
>> disabling an incompatible optimization.
> 
> OK.
> 
> Then let me rephrase: We do not want to disable optimizations, ...

If you want to figure out what that code is doing and rewrite it to be 
compliant with strict aliasing, go ahead (I tried, and it made my brain 
hurt).  In the meantime, we shouldn't be enabling the optimization on 
code that was written for an older version of the C language where such 
an optimization was prohibited, and where the compiler is letting us 
know that it thinks the code depends on the older semantics.

-Scott

[U-Boot-Users] HELP, trying to remove complier warnings

[Kumar Gala]

On Jul 10, 2008, at 1:43 PM, Scott Wood wrote:

> Wolfgang Denk wrote:
>> In message <921B6E79C3E18642BFFC1C8633C430DB01023B9A@CA1EXCLV07.adcorp.kla-tencor.com 
>> > you wrote:
>>> Try adding -fno-strict-aliasing
>>
>> No, we don't want to hush up compiler warnings, we want to fix the
>> problems instead.
>
> It's not silencing a warning (if it were, it'd be a -W flag); it's
> disabling an incompatible optimization.

Did we come to any resolution on this?

I'm with Scott in that trying to re-write the code isn't worth the  
pain and its reasonable to disable the optimization until someone is  
willing to update to a newer dlmalloc which might not have the issue.

- k

[U-Boot-Users] HELP, trying to remove complier warnings

[Jerry Van Baren]

Kumar Gala wrote:
> On Jul 10, 2008, at 1:43 PM, Scott Wood wrote:
> 
>> Wolfgang Denk wrote:
>>> In message <921B6E79C3E18642BFFC1C8633C430DB01023B9A@CA1EXCLV07.adcorp.kla-tencor.com 
>>>> you wrote:
>>>> Try adding -fno-strict-aliasing
>>> No, we don't want to hush up compiler warnings, we want to fix the
>>> problems instead.
>> It's not silencing a warning (if it were, it'd be a -W flag); it's
>> disabling an incompatible optimization.
> 
> Did we come to any resolution on this?
> 
> I'm with Scott in that trying to re-write the code isn't worth the  
> pain and its reasonable to disable the optimization until someone is  
> willing to update to a newer dlmalloc which might not have the issue.
> 
> - k

FWIIW, +1 for me too.

gvb